U.S. DEPARTMENT OF COMMERCE
                           National Technical Information Service
                           PB-275 950
Diesel  Crankcase  Emissions
Characterization
Southwest Research Inst, San Antonio, Tex
Prepared for

Environmental Protection Agency,  Ann Arbor, Mich  Emission  Control
Technology Div

Sep 77

-------
DIESEL CRANKCASE EMISSIONS
       CHARACTERIZATION
                  by
             Charles T. Hare
            Daniel A. Montalvo
             FINAL REPORT
                  of
               Task No. 4
           Contract 68-03-2196
                  for
       Environmental Protection Agency
     Emission Control Technology Division
           2565 Plymouth Road
         Ann Arbor, Michigan 48105
               May 1977
        SOUTHWEST  RESEARCH  INSTITUTE
        SAN ANTONIO     CORPUS CH RISTI    HOUSTON

-------
                NOTICE






THIS DOCUMENT HAS BEEN REPRODUCED



FROM THE BEST COPY  FURNISHED US BY



THE  SPONSORING AGENCY.  ALTHOUGH IT



IS RECOGNIZED  THAT CERTAIN  PORTIONS



ARE  ILLEGIBLE,  IT IS  BEING RELEASED



IN THE  INTEREST  OF MAKING AVAILABLE



AS MUCH  INFORMATION AS POSSIBLE.

-------
TECHNICAL REPORT DATA - >
Flcasc read Instructions on the reverse before completing _ _ ^ __ _ _^ \
i.
j.
7.
9.
REPORT NO. 2.
EPA-460/3-77-016
TITLE AND SUB TITLE
Diesel Crankcase Emissions Characterization
AUTHOHIS) ,
Charles T. Hare
Daniel A. Montalvo
PERFORMING ORGANIZATION NAME ANO ADDRESS
Southwest Research Institute
P. 0. Drawer 28510
6220 Culebra Road
San Antonio, Texas 78284
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
2565 Plymouth Rd.
Ann Arbor, MI 48105
3. RECIPIENT'S ACCE.SSIONi'JstO/ 1 J
s - -- j -J 7 J U !
S. REPORT DATE f
September 1977 (
6. PERFORMING ORGANIZATION CODE 1
8. PERFORMING ORGANIZATION REPORT NO.
AR-1189 ii
10. PROGRAM ELEMENT NO.
I.
11. CONTRACT/GRANT NO. ,
/.
68-03-2196
1!
13. TYi- £ OF REPORT AND PERIOD COVERED jl
Final Report Task IV I
14. SPONSORING AGENCY CODE ;|
200/05 ;
13. SUPPLEMENTARY NOTES ..
ji
 16. ABSTRACT
          The objective of this project was to characterize crankcase or "blowby"
          emissions from two in-service diesel engines, developing and demonstrating
          methodology for analyses where none existed.  Methodology development and
          demonstration was performed using a dynamometer mounted Detroit Diesel 6V-71,
          and characterization was done on a Cummins NTC-350 and another 6V~71o
Gaseous emissions measured included:  emission mass rate; HC, CO,
                                                                                 and NOx
          (both exhaust and crankcase emissions); individual hydrocarbons; aldehydes;
          and N-dimethylnitrosamine.  Crankcase particulate measurements included:
          mass rate; concentration; size distribution; elemental composition; sulfate;
          total solubles; BaP in solubles; boiling range of solubles; and N-dimethyl-
          nitrosamine.  These measurements were made using variations of the standard
          13-mode diesel procedure.
17. KEY WORDS ANO DOCUMENT ANALYSIS
a. ' DESCRIPTORS
heavy duty vehicles
diesel engines
crankcase emissions
18. DISTRIBUTION STATEMENT
. Unlimited
b.lDENTIFIERS/OPEN ENDED TERMS

19. SECURITY CLASS (This Report}
20. SECURITY CLASS ( This page}
c. COSATI I'i'jld/Croup |
!
21 I
22. PRICE . ,„, f
££LdJLLjLd£LJ
EPA Form 2220-1 (9-73)

-------
      SOUTHWEST  RESEARCH  INSTITUTE
     Post Office Drawer 28510, 6220 Colebra Road
             San Antonio, Texas 78284
DIESEL CRANKCASE EMISSIONS
        CHARACTERIZATION

                    by
              Charles T. Hare
             Daniel A. Montalvo
              FINAL REPORT
                    of
                Task No. 4
            Contract 68-03-2196
                   for
       Environmental Protection Agency
     Emission Control Technology Division
            2565 Plymouth Road
         Ann Arbor, Michigan 48105


                 May 1977
                     Approved:
                                     ->s^_
                     Karl f. Springer, Director
                     Department of Emissions Research
                     Automotive Research Division

-------
                                  FOREWORD
      This report covers Task Order No. 4 of EPA Contract No. 68-03-2196
conducted for the Emissions Control Technology Division, U.S. Environmental
Protection Agency, 2565 Plymouth Road, Ann Arbor, Michigan 48105.  The EPA
Task Officer on this Task Order was Mr. Thomas M. Baines, and the Project
Officer was Mr. Merrill W. Korth.  Principal investigator for Southwest Re-
search Institute was Charles T. Hare, and he was assisted by Daniel A. Montalvo.
Overall supervision was provided by Karl J. Springer.  The project was per-
formed during the period February 1976 tlurough April 1977, and it was iden-
tified within the Institute as Project No. 11-4291-004.
                                      iii

-------
                                  ABSTRACT
      The project described herein was intended to characterize crankcase
or "blowby" emissions from two in-service diesel engines, developing and
demonstrating methodology for analyses where none existed.  These goals were
achieved, and a substantial amount of data is now available as a result.
Methodology development and demonstration was performed using a dynamometer-
mounted Detroit Diesel 6V-71 research engine, and characterization was sub-
sequently done on a truck with a Cummins NTC-350 engine and another 6V-71
which was removed from a city bus and operated on an engine dynamometer.

      Crankcase gaseous emissions measured, with prior methodology develop-
ment where necessary, included:  emission mass rate; HC, CO, CC>2, and NOX
(in both exhaust and crankcase emissions); individual hydrocarbons; aldehydes;
and N-dimethylnitrosamine.  Crankcase particulate measurements included:  mass
rate; concentration; size distribution; elemental composition including car-
bon, hydrogen, nitrogen, sulfur, and metals; sulfate; total solubles; BaP in
solubles; boiling range of solubles; and N-dimethylnitrosamine.  These emis-
sions were measured for either individual modes of the 13-mode diesel proce-
dure or as a composite for the procedure as a whole.
                                     iv

-------
                           TABLE OF CONTENTS


                                                                 Page

FOREWORD                                                          iii

ABSTRACT                                                          iv

LIST OF FIGURES                                                   vii

LIST OF TABLES                                                    ix

I.   SUMMARY AND CONCLUSIONS                                       1

II.  TEST PLAN AND METHODOLOGY DEVELOPMENT                         7

     A.  Development of Test Plan                                  7
     B.  Development of Sampling and Analysis Methodology         10

III. DEMONSTRATION OF METHODOLOGY                                 23

     A.  Measurement of Crankcase Gas  (Blowby) Flowrates          23
     B.  Measurement of HC, CO, C02, NOX, and 02 in Crank-
         case Gases and Comparison to  Gaseous Exhaust Emissions   23
     C.  Measurement of Individual Hydrocarbons in Crankcase
         Gases                                                    24
     D.  Measurement of Aldehydes in Crankcase Gases              26
     E.  Particulate Mass Emissions by Three Techniques           27
     F.  Size Distribution of Particulate                         29
     G.  Amount and Boiling Point Distribution of Organic
         Solubles in Particulate                                  32
     H.  Benzo  (a)Pyrene  (or BaP) in Particulate Organic
         Solubles                                                 32
     I.  Elemental Composition of Particulate                     32
     J.  Sulfate  (SO4=) in Particulate                           34

IV.  CHARACTERIZATION OF  IN-SERVICE CUMMINS NTC-350 ENGINE        35

     A.  Measurement of Crankcase Gas  (Blow-by) Flowrates         35
     B.  Measurement of HC, CO, C02, NOX and 02 in Crank-
         case Gases and Comparison to  Gaseous Exhaust Emissions   36
     C.  Measurement of Individual Hydrocarbons in Crankcase
         Gases                                                    36
     D.  Measurement of Aldehydes in Crankcase Gases              38
     E.  Crankcase Particulate Mass Emissions by Three Tech-
         niques.                                                  38
     F.  Size Distribution of Particulate                         40
     G.  Amount and Boiling Point Distribution of Organic
         Solubles in Particulate                                  40
     H.  Benzo  (a)Pyrene  (or BaP) in Particulate Organic
         Solubles                                                 40
     I.  Elemental Composition of Particulate                     40
     J.  Sulfate  (SO4=) in Particulate                           45

-------
                      TABLE OF CONTENTS  (Cont'd)
V.   CHARACTERIZATION OF IN-SERVICE DETROIT DIESEL 6V-71
     BUS ENGINE                                                   47

     A.  Measurement of Crankcase Gas  (Blowby) Flowrates          47
     B.  Measurement of HC, CO, C02, NO*, and 02 in Crank-
         case Gases and Comparison to Gaseous Exhaust Emissions   47
     C.  Measurement of Individual Hydrocarbons in Crankcase
         Gases                                                    47
     D.  Measurement of Aldehydes in Crankcase Gases
     E.  Crankcase Particulate Mass Emissions by Three Tech-
         niques                                                   51
     F.  Size Distribution of Particulate                         53
     G.  Amount and Boiling Point Distribution of Organic
         Solubles in Particulate                                  53
     H.  Benzo (a)Pyrene (BaP) in Particulate Organic Solubles    56
     I.  Elemental Composition of Particulate                     57
     J.  Sulfate (S04=) in Particulate                            58

VI.  ANALYSIS FOR NITROSAMINES IN CRANKCASE GASES AND PAR-
     TICULATE                                                     59

     A.  Summary of Project Efforts on Nitrosamines               59
     B.  Initial Sampling Systems                                 59
     C.  Analysis of Filter and Impinger Samples taken under
         Initial Scope of Work                                    61
     D.  First Attempt at GC-MS Confirmation of DMNA              62
     E.  Collection and Analysis of Crankcase Gas Samples
         from Mack ETAY(B)  673-A Engine                           62

VII. DISCUSSION OF CRANKCASE EMISSION CONTROL POSSIBILITIES       68

VIII.ANALYSIS OF TEST ENGINE LUBRICATING OILS                     70

LIST OF REFERENCES                                                73

APPENDICES

     A.  Task Order 4 Scope of Work and Modifications

     B.  Gaseous Exhaust Emissions Data on Three Diesel Engines

     C.  Supplemental Nitrosamine Data and Analysis Reports

     D.  Caterpillar Tractor Company Report  to EPA on Diesel
         Crankcase Emissions

     E.  Infrared Spectra  of Lubricating Oils
                                  vi

-------
                              LIST OF FIGURES
Figure                                                              Page

   A    Log-Probability Plot of 13-mode Composite Crankcase
        Particulate Size Distribution  for Three Test Engines           3

   B    Semilog Plot of 13-mode Composite Crankcase Particulate
        Size Distribution  for Three Testi^ngines                       4

   1    Crankcase Emission* Flow Measurement  System                    11

   2    Sampling/Dilution  System  for Impactor  Studies                 11

   3    System for Collection of  Crankcase Gas Samples  for
        Aldehyde Analysis                                             13

   4    System for Collection of  Crankcase Gas Samples  for
        Individual HC Analysis                                        13

   5    Crankcase Emission Total  Particulate Sampler                  14

   6    Sampling System for 47 mm Filters                             16

   7    Hi-Vol Sampler for Total  Crankcase Particulate  Collection    17

   8    Crankcase Particulate Dilution Tunnel  for Use with  Im-
        pactor and 47 mm Sampler                                      17

   9    47 mm Sampler for  Particulate  Collection                      17

   10    Discs and Filter used in  Particle Sizing  Impactor            18

   11    Mock-up SEM Sample Chamber Showing Cold Stage and Liquid
        N2  System Mounted on Door                                   19

   12    General View of SEM                                           19

   13    Diesel Crankcase Particulate Deposit at 800X, 5kV,  and
        -39"C;  Individual  Particles not  Resolved  Due  to Coalesc-
        ing of Liquid Aerosol                                         20

   14    Film of Diesel Crankcase  Particulate at 20X,  5kV, and
        -39°C Showing Effects of  High-Magnification Examination
        of Several Small Areas                                        20

   15    Log-Probability Plot of 6V-71  Development Engine Crank-
        case Particulate Size Distribution,  13-mode Composite        30

   16    Semilog Plot of 6V-71 Development Engine  Crankcase  Par-
        ticulate  Size Distribution, 13-mode  Composite                 31
                                      vii

-------
                          LIST OF FIGURES  (Cont'd)
Figure                                                              Page

  17    Log-Probability Plot of NTC-350 Engine Crankcase Par-
        ticulate Size Distribution, 13-mode Composite                42

  18    Semilog Plot of NTC-350 Engine Crankcase Particulate
        Size Distribution, 13-mode Composite                         43

  19    Log-Probability Plot of 6V-71 Bus Engine Crankcase
        Particulate Size Distribution, 13-mode Composite (Standard
        Speeds)                                                      54

  20    Semilog Plot of 6V-71 Bus Engine Crankcase Particulate Size
        Distribution, 13-mode Composite (Standard Speeds)            55

  21    Impinger System for Collection of Crankcase Gas Samples
        for Nitrosamine Analysis                                     60
                                     viii

-------
                              LIST OF TABLES
Table                                                              Page
 1     Test Plan for Diesel Crankcase Emissions Characterization     8

 2     Identification and Service Data on Characterization Engines   9

 3     Crankcase Flowrates, 6V-71 Development Engine                23

 4     Gaseous Constituents of Crankcase Emissions, 6V-71 Develop-
       ment Engine                                                  24

 5     Crankcase Gaseous Emissions from 6V-71 Development Engine    25

 6     Comparison of Individual HC and Aldehyde Sampling Rates to
       Crankcase Emission Flowrates from 6V-71 Development Engine
       During 13-mode  (3 min/mode) Steady-State Runs                25

 7     Distribution of Crankcase Gas Individual Hydrocarbons,
       13-mode Composite, 6V-71 Development Engine

 8     Crankcase Gas Aldehyde Concentrations for the 6V-71 De-
       velopment Engine, 13-mode Composite                          26

 9     Crankcase Particulate Mass Emissions from 6V-71 Develop-
       ment Engine, Hi-Vol Sampler                                  27

10     Crankcase Particulate Mass Emissions from 6V-71 Develop-
       ment Engine, 47 mm Sampler, 13-mode Composite                28

11     Crankcase Particulate Mass Emissions from 6V-71 Develop-
       ment Engine, Impactor System, 13-mode Composite              28

12     Crankcase Particulate Sizing Data for 6V-71 Development
       Engine, 13-mode Composite                                    29

13     Amount and Boiling Range of Solubles in Crankcase Par-
       ticulate, 6V-71 Development Engine, 13-mode Composite        32

14     BaP Content of Crankcase Particulate, 6V-71 Development
       Engine, 13-mode Composite                                    33

15     Elemental Composition of Crankcase Particulate, 6V-71        33
       Development Engine, 13-mode Composite

16     Sulfate in Crankcase Particulate, 6V-71 Development Engine,
       13-mode Composite                                            34

17     Crankcase Emission Flowrates, Truck with NTC-350 Engine      35

18     Gaseous Constituents of Crankcase Emissions, Truck with
       NTC-350 Engine  (Average of Three Runs)                       36

                                    ix

-------
                          LIST OF  TABLES  (Cont'd)


Table                                                               Page

 19     Crankcase  Mass  Emissions  from  Truck with NTC-350  Engine       37

 20     Crankcase  Gas Individual  Hydrocarbons,  13-mode  Composite,
        NTC-350 Engine                                                37

 21     Crankcase  Gas Aldehyde Concentrations for the NTC-350
        Test Engine, 13-mode Composite                               38

 22     Crankcase  Particulate Mass  Emissions from an NTC-350
        Engine,  Hi-Vol  Sampler, 13-mode Composite                     39

 23     Crankcase  Particulate Mass  Emissions from NTC-350 Engine,
        47 mm Sampler,  13-mode Composite                              39

 24     Crankcase  Particulate Mass  Emissions from NTC-350 Engine,
        Impactor System,  13-mode  Composite                            40

 25     Crankcase  Particulate Sizing Data for NTC-350 Engine,
        13-mode  Composite                                            41

 26     Amount  and Boiling Range  of Solubles in Crankcase Par-
        ticulate,  NTC-350 Engine, 13-mode Composite                   44

 27     BaP  Content of  Crankcase  Particulate, NTC-350 Engine,,
        13-mode  Composite                                            44

 28     Elemental  Composition of  Crankcase Particulate, NTC-350
        Engine,  13-mode Composition                                  45

 29     Sulfate  in Crankcase Particulate, NTC-350 Engine, 13-mode
        Composite                                                     46

 30     Crankcase  Emission Flowrates,  6V-71 City Bus Engine At
        Standard Engine Speeds                                        48

 31     Crankcase  Emission Flowrates,  6V-71 City Bus Engine
        at Low Engine Speeds                                          48

 32     Gaseous  Constituents of Crankcase Emissions, City  Bus
        6V-71,  Standard Speeds  (Average of Two  Runs)                  49

 33     Gaseous  Constituents of Crankcase Emissions, City  Bus
        6V-71,  Low Speeds (Average  of  Two Runs)                       49

 34     Crankcase  Mass  Emissions  from  City Bus  6V-71, Standard
        Speeds                                                        50

-------
                           LIST OF TABLES  (Cont'd)


Table                                                              Page

35     Crankcase Mass Emissions from City Bus 6V-71, Low Speeds     50

36     Crankcase Gas Individual Hydrocarbons, 13-mode Composite,
       6V-71 Bus Engine  (Standard Speeds)                           51

37     Crankcase Gas Aldehyde Concentrations for the 6V-71 Bus
       Engine, 13-mode Composite  (Standard Speeds)                  51

38     Crankcase Particulate Mass Emissions from a 6V-71 Bus
       Engine, Hi-Vol Sampler, 13-mode Composite  (Standard Speeds)  52

39     Crankcase Particulate Mass Emissions from a 6V-71 Bus
       Engine, 47 mm Sampler, 13-mode Composite                     '52

40     Crankcase Particulate Sizing Data for 6V-71 Bus Engine,
       13-mode Composite  (Standard Speeds)                          53

41     Amount and Boiling Range of Solubles in Crankcase Par-
       ticulate, 6V-71 Bus Engine, 13-mode Composite  (Standard
       Speeds)                                  '                    56

42     BaP Content of Crankcase Particulate, 6V-71 Bus Engine,
       13-mode Composite  (Standard Speeds)                          56
                                         •»
43     Elemental Composition of Crankcase Particulate, 6V-71
       Bus Engine, 13-mode Composite                                57

44     Sulfate in Crankcase Particulate-6V-71 Bus Engine, 13-
       mode Composite                ,                               58

45     Crankcase Gas DMNA Results for 13-mode Composite Tests
       by Engine, Impinger Samples/Hall Detector                    61

46     Summary of EPA-NEIC Nitrosamine Analysis of Mack ETAY
        (B)673-A Crankcase Gas Samples                               65

47     Data on Properties of New  and Used  Lubricating Oils from
       the Four Test Engines                                        72
                                    xi

-------
                         I.  SUMMARY AND CONCLUSIONS
      The objectives of this Task Order, which were to develop, demonstrate,
and utilize methods for sampling of crankcase emissions from heavy-duty die-
sel engines, have been achieved.  Methods for sampling were developed from
scratch where necessary, but were adapted from methods currently in use for
exhaust sampling if at all possible.  Analytical methods were mostly derived
from diesel exhaust analysis experience,' although the composition of the
crankcase emissions made necessary a number of refinements and modifications.
Development of methods was carried out on a dynamometer-mounted Detroit Diesel-
Allison 6V-71 test engine (referred to as the "6V-71 development" engine), and
characterization was performed on a truck-tractor powered by a Cummins NTC-350
engine and on a 6V-71 bus engine which had been removed from its chassis.  The
development engine and the NTC-350 were operated on a No. 2 fuel having 0.23
percent sulfur content, and the bus engine was operated on a No. 1 fuel con-
taining 0.10 percent sulfur.  The development engine was run on No. 2 fuel
because this fuel was necessary for some other research being conducted on
the engine at the time.  The No. 2 fuel was not necessarily considered "typ-
ical" for the 6V-71 development engine.

      Exhaust gaseous emissions (regulated pollutants, or HC, CO, and NOX)
were measured along with crankcase emissions, and it was found that crank-
case gaseous HC was emitted at the rate of 0.20 to 4.1 percent of exhaust HC.
Carbon monoxide and NOX emitted from the crankcases ranged from 0.005 to 0.43
percent of their counterparts in exhaust gases.  Individual low-molecular-
weight hydrocarbons (total)  were emitted from crankcases at concentrations
of 3.5 to 20 ppm C, and aldehydes at concentrations of 9.0 to 15.0 mg/m3  (2.8
to 4.7 ppm as butyraldehyde).  Total crankcase flowrates varied with engine
operating conditions, larger flows being emitted as engine speed and load
increased.  The range of these flows was from 2.4 m3/hr on the 6V-71 develop-
ment engine to 10.8 m3/hr on the turbocharged NTC-350.  Crankcase flows were
much smaller than engine intake air flows, ranging from about 0.25 to 1.0
percent of air flow when compared on the basis of 13-mode composites.

      Analysis for benzo(a)pyrene (BaP) was conducted on crankcase particulate.
The analysis included a step in which H2S04 was added to scrapings from the
TLC (thin-layer chromatograph) plate to form a fluorescent material for quan-
titative measurement.  Newer techniques do not include this step, and produce
lower BaP values which are considered more accurate than those reported here.
The BaP values are useful on a comparative basis, however.  With this quali-
fication in mind, mass flowrates computed from these crankcase studies ranged
from 10 to 77 yg per hour.  For comparison, a typical BaP rate in the exhaust
particulate of the 6V-71 development engine was 1670 ug per hour (all based
on 13-mode composite).

      Blowby particulate mass rates measured during this Task Order ranged
from about 0.80 to 2.4 g/hr, and the range of concentrations was approxi-
mately 93 to 628 mg/m3.  These concentrations were as high as or higher than
particulate concentrations found in the exhaust gases of heavy-duty diesels
(range approximately 50 to 100 mg/m3), but crankcase particulate mass emission
rates are much lower than those in exhaust due to the smaller crankcase flow-
rates.  Exhaust particulate mass emissions from engines similar to the test

-------
engines ranged from approximately 35 to 86 g/hr.  The participate matter
from the crankcases was predominantly organic soluble material, ranging
from about 88 to 98 percent solubles as determined by Soxhlet extraction.
Soluble content of crankcase particulate from the 4-stroke engine was vir-
tually identical to the average of that from the two 2-stroke engines.  Com-
parable figures for exhaust particulate, however, are 50 percent solubles
for 2-stroke engines and 10 to 20 percent solubles for 4-stroke engines.

      Sulfate emission rates from engine crankcases ran from 0.93 to 2.39
mg/hr in this study, and comparable exhaust emission rates using a fuel with
0.23 percent by weight sulfur content are around 1,300 mg/hr.  Exhaust sul-
fate rates using a low-sulfur fuel (0.04 percent) in a 6V-71 bus engine are
around 210 mg/hr at standard engine speeds and 330 mg/hr at low speeds (900
and 1500 rpm).  Sulfur in the sulfate emissions from diesel crankcases con-
stituted an average of about 15 percent of the total sulfur in the crankcase
particulate.

      Particulate from crankcases studied under this Task Order bore a great
similarity to lubricating oil according to a number of criteria.  Boiling
ranges of the soluble fractions of particulate were very similar to .those
of oils from the respective engines, although slightly wider (that is, sol-
ubles contained slightly more low-boiling and high-boiling materials than
oils).  Carbon and hydrogen content of solubles were only slightly lower than
that of oils in general, although two outlying data points tend to confuse
the results.  Small amounts of ash or other contaminants in particulate could
account for carbon/hydrogen differences between oils and particulate.  Percent-
ages of sulfur and phosphorus determined by X-ray were similar for oils and
solubles, but calcium was found only in the oils.  The remainder of the ele-
ments showed mixed results.

      Particulate size distributions from the three engines were somewhat
different, but bore a general similarity to one another.  Average data from
the impactor studies are presented in Summary Figures A and B, a log-proba-
bility and a semilog plot, respectively.  Crankcase particulate from the 6V-71
development engine was distributed more toward the larger sizes than that from
other two engines over most of the size range.

      Prior to this research Task, nitrosamines had not been found as an air
pollutant emitted by internal combustion engines.  No nitrosamines were
found in any of the intake air, crankcase particulate, or reagent blank
samples analyzed in this Task.  Three methods were employed to sample crank-
case gases for nitrosamine and nitrosamine precursor analysis, including
basic impingers (in two sizes), acidic impingers, and absorptive  (Tenax-GC)
columns.  A fourth engine was also used for some of the nitrosamine sampling.
It was a turbocharged 4-stroke Hack ETAY(B)673-A.

      No nitrosamine precursors were found in the acidic impinger samples,
although the ion chromatographic analysis used was very coarse.  No recognized
nitrosamine precursors or nitrosamines were found on the Tenax-GC cartridges
by GC-MS, although relatively high concentrations of N,N-dimethylacetamide
were found.  It is uncertain at this writing if  there is a possibility that
N,N-dimethylacetamide could be a nitrosamine precursor or a compound formed
by a reaction of a nitrosamine with some other material.

-------
SUMMARY FIGURE A.   LOG-PROBABILITY PLOT OF 13-MODE COMPOSITE CRANKCASE
         PARTICULATE SIZE DISTR1BUTIONTOR THREE TEST ENGINES

-------
                                                                            __ . 	 I

SUMMARY FIGURE B.  SEMILOG PLOT OF 13-MODE COMPOSITE CRANKCASE
     PARTICUIATE SIZE DISTRIBUTION FOR THREE TEST ENGINES

-------
      Indications of nitrosamines were found in basic impinger-collected
samples by both SwRI's Department of Environmental Sciences and EPA's National
Enforcement Investigations Center.  Since no qualification of the sampling
methodology was performed, these results cannot be construed as positive
identification.  GC-MS work performed on a relatively large sample gave strong
evidence that one of the compounds measured in the other samples was in fact
N-dimethylnitrosamine, but these results do not diminish the possibility that
nitrosamines are formed as artifacts in the impinger solution either during
or after sampling.

      Apparent DMNA concentrations in the crankcase gas samples taken from
one development engine and two characterization engines ranged from 27 to
170 ppt (parts per trillion).  These concentrations translate to a range of
apparent DMNA mass flowrates from 0.35 to 5.5 Vg/hr.

      A brief examination of available opinion and data on the status and
feasibility of crankcase emission controls for diesel engines has been made.
Some naturally-aspirated engines are equipped with control systems at the
factory, and at least one retrofit kit is offered for stationary and marine
engines.  Control systems for both naturally-aspirated and turbocharged en-
gines appear to be feasible with existing technology if the necessity for
such controls exists.

      Based on the above summary and the supporting data found throughout
the report, a number of conclusions have been drawn.  The most important con-
clusions are:

      1.  Although crankcase particulate matter seems very close to lubri-
cating oil in composition (boiling range, elemental analysis), enough dif-
ferences exist to indicate that such particulate is not simply droplets which
have been removed from the oil mechanically.  Reasons for this conclusion
are the slightly wider boiling range and lack of calcium in particulate.

      2.  Crankcase gas flows are much smaller than intake air flows, so
most gaseous and particulate crankcase emissions are very small as compared
to corresponding exhaust emissions.  Examples of such pollutants include the
following:

                                         % of Corresponding
                      Pollutant           Exhaust Emission
                  gaseous HC                0.2-4.1
                  CO, NOX, CO2              0.005 - 0.43
                  total particulate         0.9   - 2.9
                  sulfate (S04'a)            0.07  - 0.7
                  BaP                       0.6   - 4.6
                  aldehydes                 0.2   -1.0

      3.  Crankcase particulate matter can be sized with apparent accuracy
by an inertial impactor, but deposition on surfaces other than collection
discs seems to prohibit use of the device for particle mass flow computations.

-------
      4.  Analyses  conducted for nitrosamines and nitrosamine precursors in
crankcase gases were inconclusive.  No nitrosamines or precursors were found
in crankcase particulate or engine intake air samples.  Basic impinger-col-
lected samples appeared to contain DMNA at levels indicative of 27 to 170 ppt
(parts per trillion) in crankcase gases, and possibly other nitrosamines at
lower levels.  Cartridge samples appeared to contain no nitrosamines or known
precursors but did contain substantial amounts of N,N-dimethylacetamide.  It
is concluded that further research should be performed to qualify and employ
the most accurate possible nitrosamine methodology for both sampling and
analysis of diesel crankcase gases.

      5.  Systems for control,of diesel crankcase emissions are not yet in
widespread use.  Data should be acquired on performance and durability ef-
fects of such systems.  Design technology for adequate recirculation or other
control systems appears to be available, should it be shown necessary by fur-
ther characterization studies.

-------
                 II.  TEST PLAN AND METHODOLOGY DEVELOPMENT
      The original test plan for this Task Order essentially duplicated the
Scope of Work contained in the Task Specification Change Request for Task 4,
issued on January 29, 1976.  A copy of the aforementioned Scope of Work and
subsequent additions are given as Appendix A of this report.  The first part
of this section contains test plan details, and the second part contains de-
tails of sampling and analysis methodology development.  Methodology for the
sampling and analysis of nitrosamines is covered in Section VI.

A.    Development of Test Plan

      Criteria used in developing the Test Plan included test runs sufficient
to measure or collect samples for all exhaust and blowby constituents of
interest, constraints on sample size and acquisition rates imposed by ana-
lytical sensitivity and sampling hardware, and enough data points for assess-
ment of repeatability.  Many of the sampling and analytical methods used
were derived from experience gained under EPA Contract No. 68-02-1230^*
and (current) Contract No. 68-02-1777.  Previous work done on crankcase emis-
sions from diesels^) was found to be quite rudimentary, which is not sur-
prising since it was published in 1964.  It did provide some insight, however,
into expected blowby flowrates.

      The Test Plan  (or sampling and analysis plan) developed for this project
is outlined in Table I/ including contaminants sampled, methods, and numbers
of determinations.  Exhaust gas concentrations were measured during the same
modes as crankcase gas concentrations to provide a basis for comparison.  The
engine operating procedure for all test work was the EPA "13-mode" test*-*),
or some variation thereof with shorter times in each mode.

      It was intended that methodology be developed and demonstrated by per-
forming the analyses shown in Table 1 on a dynamometer-mounted Detroit Diesel-
Allison (DDAD) 6V-71 engine in use on another EPA contract  (68-02-1777).  This
plan was followed, and then attempts were made to obtain two other engines
(in trucks) for characterization.  By agreement with the Project Officer, orig-
inal criteria used for selection of the characterization engines were:

      - one engine should be a 4-stroke, naturally-aspirated engine and the
        other should be a 2-stroke, non-turbocharged engine

      - both engines should have accumulated approximately one half the ser-
        vice normally accumulated before overhaul  (125,000 to 200,000 miles,
        depending on the type of service)

      - both engines should have accumulated usage such that they are approxi-
        mately halfway through the normal oil change interval since an oil
        change, and they should be tested with the existing oil in the
        crankcase
* Superscript numbers in parentheses refer to List of References at end of
  this report.

-------
      - both engines should be mounted in vehicles suitable for chassis dyna-
        mometer operation.

In order to complete the test program, some deviations from these criteria
were necessary.  Although a great deal of effort was expended, no trucks
having the required engines were located in the area.  The primary difficul-
ties were that 2-stroke engines meeting the service criteria were unavailable,
and that all the naturally-aspirated, 4-strokes available had closed crank-
cases (blowby recirculation).  Two-stroke engines were found in buses, and
a bus was obtained for testing.  This vehicle proved unsuitable for chassis
dynamometer operation, however, in that it experienced two tire failures
before even one 13-mode sequence could be accomplished.
     TABLE 1.  TEST PLAN FOR DIESEL CRANKCASE EMISSIONS CHARACTERIZATION
Sample
Source
Crankcase
gases





Crankcase
parti culate











Exhaust
gases

Analysis for
HC, CO, C02, NOX
,
aldehydes
individual HC
nitrosamines


total particulate
sizing
appearance and
visual sizing
C, H, N, S
elements (metals)
sulfate
total organic
solubles
BaP in solubles
boiling range of
solubles
nitrosamines
HC, CO, CO,, NOV
* < *»

Collection and
Analysis Method (s)
bag samples, continuous
analyzers
DNPH
bag samples, G.C.
wet or column collection.
separation or elution,
G.C. and GC-MS
filtration, weighing
impactor, weighing

S.E.M. 'and photography
filtration , combustion
filtration, X-ray
filtration, BCA

filtration, extraction
T.L.C.

G.C.
extraction, G.C.
continuous analyzers


Determinations

3
3
3


5
30a
5

5b
£>
5
5

5
5

5
5
3

a performed on samples collected for other analyses
  S.E.M. work intended as a cross-check on impactor sizing study
      The vehicle and the engine finally characterized were a truck-tractor
with Cummins NTC-350 (turbocharged 4-stroke) engine and a city bus 6V-71 en-
gine removed from the bus and tested on a stationary dynamometer.  Identifi-
cation and service data on these vehicles/engines is given in Table 2, indi-
cating that they met the total service and oil service criteria quite well.

-------
                   TABLE 2.  IDENTIFICATION AND SERVICE DATA ON CHARACTERIZATION ENGINES
Engine type

Engine model

Engine serial number

Vehicle source

Vehicle identification

Vehicle model

Total engine service, km (mi)

Service since overhaul, km (mi)

Normal oil change interval, km (mi)

Service since oil change, km (mi)
Turbocharged 4-stroke

Cummins NTC-350

10370038 (mfd. 9/73)

ABC Truck Rental

No. 2200, Texas plate R32-988

1974 White Freightliner

337,067  (209,444)

no overhaul

24,140 (15,000)

17,038 (10,587)
Roots-scavenged 2-stroke

Detroit Diesel-Allison 6V-71

64A37135

San Antonio Transit Company

Original Bus No. 757 - now No. 695

GM Truck and Coach - City Coach

627,762 (390,074)

226,059 (140,467)

8,047 (5,000)

4,828 (3,000)
Note:  Short overhaul on city bus engine after testing found stuck rings on two cylinders,
       some wrist pin wear, no excessive bearing wear

-------
Both engines were tested with the existing oil in the crankcase, and all oil
added during the testing periods came from the normal sources (e.g., make-up
oil supplies of ABC and San Antonio Transit).

B.    Development of Sampling and Analysis Methodology

      Re-examining the test plan .in Tab3 ° 1, only the exhaust gas analysis
requires no explanation of either sampling or analysis technique.^3)   Details
of analysis are given in reference only for the following (summaries will be
given in text):

      DNPH method for aldehydes<4),
      G.C. method for individual HC(5), and
      BCA method for sulfate.(6>

All other sampling and analysis procedures require discussion, and they will
be approached in the order shown in Table 1.

      1.  Crankcase Gas Samples for HC, CO, CO?, and NOV

      Crankcase gas samples were taken direct for HC analysis and in bags for
CO, CO2, and NOX analysis.  Direct sampling is essential for HC from diesels
due to the need to measure at 190°C, and concentrations of the other gases
were low enough to permit bag sampling.  Direct sampling would have been
employed for all the gases, • but total continuous flowrates for all the ana-
lyzers combined exceeded the output from the crankcase of the development
engine in some operating modes.  Tedlar* sample bags were filled using a
Metal Bellows pump which pulled gases from the crankcase vent and through a
glass fiber filter.

      To document gas flowrates from the engines' crankcases and to permit
mass rate computations for gaseous constituents, measurements were taken on
both 6V-71 engines with the system shown schematically in Figure 1.  Using
the pump to scavenge the system prevented the crankcase vent from seeing any
increase in backpressure, thereby making the flow measurements as accurate
as possible.

      Crankcase flowrates from the NTC-350 engine were too high, under some
operating conditions, to use the dry gas meter system just described.  The
alternative adopted was to use the dilution system designed for impactor
studies as a measurement device.  This system is shown schematically in Fig-
ure 2 as set up for dilution/impactor work.  When used to measure blowby rate,
the dilution air inlet and the sample outlet were capped; and the blower speed
was adjusted to preserve atmospheric pressure at the crankcase outlet.  The
orifice downstream of. the blower provided the flow measurement, having been
calibrated against a laminar flow element.

      In addition to providing input data to pollutant mass flow calculations
where concentrations were measured by mode, the modal crankcase flowrates
also provided a basis for proportional sampling of other constituents during
* registered trademark of E. I. duPont De Nemours & Company.
                                      10

-------
       PRESSURE  (KEPT
      AT  ATMOSPHERIC)
               PRESSURE
                                            DRY
                                            GAS
                                           METER
        CRANKCASE
          TUBE
    MSA
RESPIRATOR
  FILTER
             FIGURE I,   CRANKCASE EMISSION FLOW MEASUREMENT SYSTEM
  FROM
CRANKCASE
  TUBE
                   MSA ULTRA-
                   AIRE FILTER
                             JOINT
                                           -REPLACEABLE
                                               PROBE
           FIGURE 2,  SAMPLING/DILUTION SYSTEM FOR  IMPACTOR STUDIES
                                       11

-------
composite runs.  The maximum ilowrate for a given sampling system was used
during the mode in which the engine emitted its maximum flow, and sampling
rates during the other modes were set proportionately lower.  "Composite"
flowrates were computed by time-weighting individual mode flowrates in the
same way as the Federal 13-mode test*3* is time-weighted, then taking an average.

       2.  Crankcase Gas Samples for Aldehydes and Individual Hydrocarbons

       Samples for aldehyde analysis by DNPH(4) were taken directly from the
blowby outlet using the system shown schematically in Figure 3.  The sample
line was heated as far as the impingers.  Maximum flowrate sampled was about
4 liters per minute, and flow through the system was held in constant propor-
tion to total crankcase flow.  The amount of DNPH reagent used in each impinger
was 40 ml.  In practice, samples for aldehyde analysis were usually taken con-
currently with those for individual hydrocarbons.  The system used to collect
samples for individual HC analysis(5) is shown in Figure 4, and it also oper-
ated with a maximum flowrate of 4 liters per minute.  Background air samples
were taken concurrently with selected runs.

       A gas chromatograph procedure was used to measure the individual hy-
drocarbon content of samples collected in the Tedlar bags.  The chromatograph
procedure employed a flame ionization detector and used a four-column arrange-
ment in conjunction with dual gas sampling valves.  The valves were pneumat-
ically operated and could be sequenced electrically to operate in either the
automatic or manual mode.  All blowby chromatograms were obtained automatically.

       Injection from the bags used for individual HC analysis to the chroma-
tograph was through an ice water (0°C) trap.  The trap was employed to retain
other crankcase gaseous species which had earlier been found to interfere
with and obscure the benzene and toluene chromatograms.

       3.  Crankcase Particulate Sampling Techniques

       Particulate in crankcase flow was measured by several methods, the
first of which was the sampler shown in Figure 5.  This sampler collected
all the crankcase particulate on a rectangular 203 mm x 254 mm (8 x 10 in)
glass fiber filter.  The filter was weighed before and after sampling under
controlled conditions to determine the amount of particulate emitted.  No at-
tempt was made during sampling to maintain the crankcase outlet pressure ex-
actly at atmospheric due to the "constant volume" procedure employed.  This
unit was called the "hi-vol" sampler due to its use of filters designed for
hi-volume ambient samplers, and the particulate collected was used for several
types of analysis discussed later.  Dilution air was cleaned up using a 99.99+
% OOP-efficient "Ultra-Aire" filter.

       Another measurement technique used was the dilution tunnel-impactor sys-
tem shown earlier in Figure 2.  In this system, the individual collection discs
and the final filter in the impactor constituted the collection apparatus, and
the other filter was used only to protect the blower.  The dilution system
was utilized so the impactor could be operated at a constanc flowrate, which
is necessary for obtaining an accurate size distribution.  Total particulate
collected with the impactor was determined by summing the amounts on each in-
dividual collection disc and the backup filter.  Glass fiber collection discs
                                       12

-------
               EXCESS
               GASES
                                                MOLECULAR
                                                SIEVE  (FILTER)
      FIGURE 3,   SYSTEM FOR COLLECTION OF CRANKCASE GAS
                SAMPLES FOR ALDEHYDE ANALYSIS
ENGINE
            FILTER
                               METAL
                              BELLOWS
                               PUMP
          EXCESS
           GASES
 ROTAMETER

VALVE
                EDLAR
               SAMPLE
                BAG
        FIGURE 4,  SYSTEM FOR COLLECTION OF CRANKCASE
           GAS SAMPLES FOR INDIVIDUAL HC ANALYSIS
                             13

-------
ENGINE

CRANKCASE TUBE

i
r
— 9
7\
                                                 AIR
                TEMP,
                       PRESSURE
                                FLOW
                                CHECK!
                                                            MSA ULTRA-
                                                            AIRE FILTER
                                                                FILTER
                                                                    SPEED
                                                                   CONTROL
                                                TO
                                           ATMOSPHERE
      FIGURE  5,   CRANKCASE EMISSION TOTAL PARTICULATE SAMPLER
were used for most of the testing, with stainless steel discs being used on
a few runs with the intent of examining the collected material using a scan-
ning electron microscope.

      Constant total  (diluted) flow was maintained by the hi-vol blower and
speed control, and total flow could be adjusted to suit each engine by using
an appropriate probe size and maintaining isokinetic conditions.  Dilution
air was cleaned by the "Ultra-Aire" filter shown.  Nominal  flow through the
impactor was maintained at 16.1 liters per minute (0.75 CFH), and probe dia-
meters available ranged from 6.35 to 12.7 mm (0.25 to 0.5 in).  Dilution tun-
nel dimensions were 50.8 mm (2 in) outside diameter, 47.6 mm  (1.875 in) in-
side diameter, and 0.56 m  (22 in) straight length upstream  of the probe tip.
The tube carrying undiluted sample into the tunnel was 25.4 mm  (1 in) O.D.
and was connected to the crankcase outlet with the minimum  possible length
of 12.7 mm (0.5 in) O.D. teflon tubing.
                                     14

-------
      The third and final particulate collection system was designed around
the 47 mm filters needed for several analyses (CHN, trace elements, and sul-
fate).  It is shown schematically in Figure 6.  When the 47 mm system was
tried on the development engine, it was found that the filters plugged too
quickly to permit completing an entire run.  Subsequent samples were, there-
fore, obtained by sampling from the dilution tunnel (described above) at the
point where the impactor was normally connected.  This technique proved quite
effective and permitted a constant flowrate through the filters instead of
modal adjustment to maintain proportional sampling.

      Photographs of the three particulate sampling systems described in
this subsection are given as Figures 7 through 9, Figure 7 shows the "hi-vol"
sampler connected to the 6V-71 development engine.  Figure 8 shows the dilu-
tion tunnel/impactor system, and Figure 9 shows the 47 mm system connected
to the dilution tunnel and operating with the 6V-71 bus engine.

      4.  Crankcase Particulate Appearance and Visual Sizing

      This analytical technique was proposed to serve as a "calibration" of
the impactor system.  The plan was to obtain photos via SEM of portions of
individual stage collection discs, and then to verify the range of particle
sizes on each disc by examination of the photos.  To permit good photos,
special stainless steel collection discs were designed to replace the com-
mercially-available glass fiber discs.  They were procured from a photo-
machining company (Microphoto, Inc.).  The discs were made of type 316 stain-
less steel and had a thickness of 0.051 mm (0.002 in).  Tare weight per stain-
less steel disc was approximately 0.925 gram for the "A" type disc and 0.853
gram for the "B" type disc, as compared to 0.151 gram and 0.148 gram (res-
pectively) for glass fiber.  Both types of discs  (A's were placed on odd
stages and B's on even stages) are shown with their glass fiber counterparts
in Figure 10.  The stainless discs shown are unused (before sampling), and
the glass fiber discs are used (they show different amounts of deposition).
One glass fiber backup filter (final stage) as used in the impactor is also
shown for comparison.

      It was recognized at the outset that the SEM's operation in vacuum  (1 x
10~5 mm Kg) might create problems with evaporation of some blowby particulate
constituents if the stage were maintained at room temperature.  It was de-
cided, therefore, to construct a "cold stage" to support sections of the
stainless discs in an attempt to keep the more volatile constituents of the
particulate from evaporating.  The cold stage is visible on the inside of a
mock-up chamber door in Figure 11, insulated thermally from the door by a
Teflon spacer.  The stage was connected electrically by a very thin wire to
ground, which is necessary for SEM operation.  Also visible in Figure 11 is
an insulated flask containing liquid N2, which provided the cooling medium
necessary to lower stage temperature to approximately -40°C.  Temperature of
the stage and sample were lowered by liquid N,, while the chamber pressure
was reduced.  Actual stage temperature was measured by a thermocouple and a
readout device while the SEM was in use.  A wider view of the ETEC Autoscan
Model U-l SEM is given in Figure 12.

      The SEM system as described proved unsuitable for crankcase particulate
matter due to unforeseen complications.  The first of these complications was
                                      15

-------
                       ^PRESSURE (KEPT)
                       /AT ATMOSPHERIC)
                                             SYSTEM CAN BE CONNECTED
                                             TO PROBE AND USED WITH
                                             DILUTION TUNNEL
                                                       47 MM FILTER
                                                       HOLDER UEA)
                                                        PUMP
                                                       (2 EA)
                                                         VALVE (4 EA)
                                                       GANGED OPERATION
                                                            DRY
                                                            GAS
                                                           METER
ROTAMETER
                                                                    TO
                                                                   ATM,
              FIGURE 6,  SAMPLING SYSTEM FOR 47 MM  FILTERS
                                    16

-------
FIGURE 7.   HI-VOL SAMPLER FOR TOTAL
 CRANKCASE PARTICULATE COLLECTION
FIGURE 8.   CRANKCASE PARTICULATE
  DILUTION TUNNEL FOR USE WITH
    IMPACTOR AND 47 ram SAMPLER
                         FIGURE 9.  47 nun SAMPLER FOR
                            PARTICULATE COLLECTION
                                      17

-------
FIGURE 10.  DISCS AND FILTER USED IN PARTICLE SIZING IMPACTOR
                              18

-------
FIGURE 11.  MOCK-UP SEM SAMPLE CHAM-
 BER SHOWING COLD STAGE AND LIQUID
     N02 SYSTEM MOUNTED ON DOOR
FIGURE 12.  GENERAL VIEW OF SEM
  that the individual collected particles had agglomerated into continuous
  films, presumably while being collected, as shown in Figure 13 (800X at 5kV,
  39°C).  This agglomeration meant that individual particles were not resolv-
  able.  In addition, the electron beam caused local volatilization of the de-
  posited material, leaving square "holes", with the surrounding material still
  at -408C and thus too viscous to flow immediately into the holes.  The re-
  sults of making several scans at relatively high magnification are shown in
  Figure 14 (20X at 5kV, -39°C) as the square "holes" in the deposited films.
  When the SEM was operated at room temperature, the holes were not apparent.
  This result was presumably due to the deposit's flowing back into the areas
  from which material had been volatilized.  Since the SEM technique was not
  developed further due to the problems mentioned above, it will not be dis-
  cussed further as a part of other report sections.

        5.  Analysis of Particulate for Carbon, Hydrogen, Nitrogen, and Sul-
        fur (CHNS)

        The CHNS analysis was performed on samples collected by 47 mm glass
  fiber filters.  Analytical work was done by Galbraith Laboratories, using
  automated thermal conductivity analysis of combustion products following com-
  bustion of whole filters.  Data were reported as percentages of total parti-
  culate on the filter  (particulate weight was supplied to the commercial lab)
  and were corrected for blank filter results.  All samples analyzed were "com-
  posites" of the 13-mode procedure.  This portion of the analysis was con-
  ducted using procedures developed under Contract No. 68-02-1230.
                                                                   (1)
                                       19

-------
FIGURE 13.  DIESEL CRANKCASE PARTICULATE DEPOSIT
AT 800X, 5kV, AND -39°C; INDIVIDUAL PARTICLES NOT
  RESOLVED DUE TO COALESCING OF LIQUID AEROSOL
FIGURE 14.  FILM OF DIESEL CRANKCASE PARTICULATE
AT 20X, 5kV, AND -39°C SHOWING EFFECTS OF HIGH-
MAGNIFICATION EXAMINATION OF SEVERAL SMALL AREAS

                       20


-------
      6.  Analysis of Particulate for Metals and Other Elements

      This elemental analysis was performed by the Institute's Mobile Energy
Division using X-ray fluorescence.  Filters used were 47 mm Fluoropore* discs,
which were selected for low background metal content.  Elements found in the
particulate matter included phosphorus, chlorine, iron, zinc, and sulfur.
The technique itself was capable of detecting and analyzing all the elements
having atomic number 11 or greater, but no others were detected.

      7.  Analysis of Particulate for Sulfate

      Sulfate samples collected on 47 mm fluorocarbon membrane  (Fluoropore)
filters were analyzed using the barium chloranilate procedure.^  This pro-
cedure calls for ammonia treatment of the filters to assure conversion of
sulfuric acid particulate into ammonium sulfate.  Conversion to the stable
ammonium salt improves the precision of the sulfate measurement by minimiz-
ing losses from the filter surface.

      The filter samples were placed face up in their opened plastic cases
in an ammoniation box.  The box was evacuated and ammonia, bubbled from con-
centrated ammonium hydroxide, was allowed to fill the box.  After a one-hour
ammoniation period, the soluble sulfates were extracted from the filters.
To begin this procedure, each filter was placed in a capped polypropylene
bottle to which a 10 ml aliquot of isopropyl alcohol (IPA) and 46 percent
deionized distilled water was added.  The bottle was then shaken vigorously
to permit proper leaching of the soluble sulfates.  A portion of the ex-
tracted sulfate sample was passed through a strong cation exchange resin
column, then through a column packed with barium chloranilate where the sul-
fates precipitated out as barium sulfate.  This reaction released colored
chloranilate ions which were measured colorimetrically with a UV photometer
at 310 nra.

      8.  Determination of Total Organic Solubles in Particulate

      Total organic solubles were determined by Soxhlet extraction of sam-
ples collected on hi-vol glass fiber filters.  The solvent used was benzene,
and refluxing occurred at about 20 cycles per hour for four hours.  This por-
tion of the analysis was conducted using procedures developed under Contract
No. 68-02-1230.U)

      9.  Analysis of Organic Solubles for Benzo (Ot) Pyrene  (BaP)

      Analysis for BaP was conducted using a method developed under previous
Contract No. 68-02-1230,^ except that a 19:1 mixture of pentane and ethyl
ether was used as the thin-layer chromatograph elutant.  The change to the
pentanerethyl ether mixture made the analysis more specific for BaP, reducing
interferences from other polynuclear aromatic compounds.  This BaP technique
has been supplanted by revised methods, however, since the subject work was
conducted.  Newer techniques generally produce lower values than those ob-
tained  in the subject tests, and these lower values are considered more accu-
rate than higher ones.
  registered trademark of Hillipore Corporation


                                      21

-------
      10.  Determination of Boiling Range for Organic Solubles

      Boiling range of organic solubles was determined using a gas chromato-
graph procedure, ASTM D-2887.  The column used was packed with Dexsil 300,
and the operating temperature range was approximately 100-600°C.  Internal
standards in the range of C8 to C]^ were used at about 20 percent concentra-
tion, and calibration? were performed using a 05 to 040 standard mixture.
This method was also used on the fuels and nils involved in the program for
purposes of comparison.
                                      22

-------
                     III.  DEMONSTRATION OF METHODOLOGY
      Sampling and analytical methods discussed in Section II were demon-
strated on a Detroit Diesel-Allison 6V-71 engine mounted on a 500 hp (373
kW) eddy current dynamometer stand.  This engine will be referred to from
this point on as the "6V-71 development engine".  To summarize the demon-
stration, all the methodology achieved some degree of success except the
SEM examination of crankcase particulate deposited on impactor discs.  The
problems experienced with the SEM work were discussed fully in Subsection
II.B.4.  Less serious problems occurred with some of the other analyses, and
they will be discussed along with corresponding results.

A.    Measurement of Crankcase Gas (Blowby) Flowrates

      Crankcase gas flowrates from the 6V-71 development engine are given in
Table 3, and they are in the expected range for such an engine which is in
good condition.  These crankcase flows represent about 0.26 to 0.39 percent
of engine airflow at corresponding conditions, and they were measured using
the system shown in Figure 1.
           TABLE 3.  CRANKCASE FLOWRATES, 6V-71 DEVELOPMENT ENGINE
Operating Condition
Engine rpm
440 (idle)
1260




2100




Load, %
	
0
25
50
75
100
0
25
50
75
100
13-mode composite0
Flowrate
Std. m3/hr*
0.68
1.8
1.7
1.9
1.9
1.9
3.6
3.7
3.8
4.2
4.3
SCFMa
0.40
1.0
1.0
1.1
1.1
1.1
2.1
2.2
2.2
2.5
2.5
kg/hr*
0.82
2.1
2.1
2.3
2.3
2.3
4.4
4.4
4.5
5.0
5.2
2.44
       * at p = 101.3 kPa and T = 21°C
         assuming gases have same density as air
       c time-weighting of individual modes as specified
         in Federal 13-mode procedure^

B.    Measurement of HC, CO, C02, NOX, and O2 in Crankcase Gases and Compari-
      son to Gaseous Exhaust Emissions

      Gaseous constituents of crankcase emissions from the 6V-71 test engine
were measured, yielding the results shown in Table 4.  These results can be
                                      23

-------
combined with the flowrates given in Table 3 to calculate mass emission rates
of gaseous constituents usi'ng the following equations:

      g/hr HC  - 0.000484  (ppm C) (M, kg/hr),
      g/hr CO  = 0.000970  (ppm CO)  (M, kg/hr),
      g/hr C02 = 15.2  (% CO2) (M, kg/hr). and
      g/hr N02 = 0.00159 (ppm N02)  (M, kg/hr).
            TABLE 4.  GASEOUS CONSTITUENTS OF CRANKCASE EMISSIONS ,
                          6V-71 DEVELOPMENT ENGINE
Operating Condition
Engine rpm
440 (idle)
1260




2100




Load, % '
	
0
25
50
75
100
0
25
50
75
100
HC,
ppm C
440
460
490
510
540
600
520
550
570
560
590
CO,
ppm
26
25
4
4
18
140
0
4
8
10
40
C02,
%
0.09
0.08
0.06
0.06
0.10
0.21
0.04
0.07
0.10
0.20
0.28
N0y,
ppm
7.6
4.5
4.7
4.6
10.
12.
2.7
3.8
6.1
13.
20.
o2,
%
20.9
20.9
20.9
20.9
20.8
20.8
20.9
20.9
20.8
20.8
20.8
The assumptions inherent in the equations are (1) that the average molecular
weight of gaseous crankcase emissions is the same as that of air, and  (2)
that the (hydrogen:carbon) ratio of hydrocarbons emitted is 2.0.  These as-
sumptions are essentially equivalent to those used in deriving similar equa-
tions for the Federal diesel engine emissions certification procedure.  Mass
emission rates thus calculated are given in Table 5.  A mathematical compo-
site has been computed which can be compared to 13-mode composite exhaust
emissions from this test engine (also given in Table 3).  This comparison
reveals that crankcase HC emissions (gaseous form only) amount to about 1.4
percent of exhaust HC and that other crankcase emissions range from 0.005 to
0.03 percent of their counterparts in exhaust gases.  Complete gaseous ex-
haust emissions data for this engine are given in Appendix B, pages B-2 and
B-3.

C.    Measurement of Individual Hydrocarbons in Crankcase Gases

      Individual hydrocarbons and aldehydes were sampled at rates proportional
to total crankcase flow in each mode.  These sampling rates are summarized in
Table 6, and the individual HC data are given in Table 7.  Methane, ethylene,
and propylene are seen to be present in the largest concentrations, but it
seems likely that ambient (background) methane contributed to levels for that
compound.  Smaller or trace quantities were noted for ethane, acetylene, pro-
pane, benzene, and toluene.
                                      24

-------
 TABLE 5.  CRANKCASE GASEOUS EMISSIONS FROM 6V-71 DEVELOPMENT ENGINE
Operating Condition
Engine rpm
440 (idle)
1260




2100




Load, %
	
0
25
50
75
100
0
25
50
75
100
13 -mode composite
of crankcase emissions
13-mode composite
of exhaust emissions
Emissions in grams per hour
HC
0.17
0.47
0.50
0.57
0.60
0.67
1.11
1.17
1.24
1.36
1.48

0.77

46.1
CO
0.021
0.051
0.008
0.009
0.040
0.31
0
0.017
0.035
0.048
0.20

0.062

425.
CO,
1.1
2.6
1.9
2.1
3.5
7.3
2.7
4.7
6.8
15.
22.

5.7

52,600.
NOX
0.010
0.015
0.016
0.017
0.037
0.044
0.019
0.027
0.044
0.10
0.17

0.041

665.
   TABLE 6.   COMPARISON OF INDIVIDUAL HC AND ALDEHYDE SAMPLING RATES
     TO CRANKCASE EMISSION FLOWRATES FROM 6V-71 DEVELOPMENT ENGINE
             DURING 13-MODE (3 min/mode) STEADY-STATE RUNS
Operating Condition
Engine rpm
440 (idle)
1260
1260
1260
1260
1260
440 (idle)
2100
2100
2100
2100
2100
440 (idle)
Load, %
»«
2
25
50
75
100
	
100
75
50
25
2
	
Crankcase Emission
Flow Rate, Std. m3/hra
0.68
1.8
1.7
1.9
1.9
1.9
0.68
4.3
4.2
3.8
3.7
3.6
0.68
Sample
Rate , S,/min
0.64
1.60
1.60
1.94
1.94
1.94
0.64
4.00
4.00
3.52
3.52
3.36
0.64
a at p = 101.3 kPa and T = 21°C
                                   25

-------
            TABLE 7.  DISTRIBUTION OF CRANKCASE GAS INDIVIDUAL HYDROCARBONS,
                       13-MODE COMPOSITE, 6V-71 DEVELOPMENT ENGINE
Constituent
Methane
Ethylene
Ethane
Acetylene
Propane
Propylene
Benzene
Toluene
Total non-Methane
ppm C
Run 1
4.3
2.4
0.3
0.2
0.2
2.2
0.3
tra
5.8
Run 2
3.4
1.5
0.2
tra
tr»
1.3
0.3
	
3.3
Run 3
3.7
^.4
0.2
tra
	
1.2
0.3
tra
3.1
Background
3.3
tra
0.1
	
	
	
	
tra
0.1
Avg. Net
0.5
1.8
0.1
tra
tra
1.6
0.3
	
3.8
      D.
           trace, less than 0.1 ppm C
Measurement of Aldehydes in Crankcase Gases
            Aldehydes were measured in the blowby of the 6V-71 development engine
      by the DNPH method.   The results of this analysis are summarized in Table 8,
      Run 1 having been discarded due to an apparent air leak in the sampling sys-
      tem.  Problems encountered with this analysis include interferences from
      fuel and oil constituents which were collected along with the sample and
      which eluted at the  same times as some of the DNPH derivatives.  The DNPH
      procedure was originally intended for use on dilute exhaust from gasoline-
      powered vehicles, and further development would probably be necessary before
      it could be considered entirely reliable for use with either diesel exhaust
      or diesel crankcase  flows.
                     TABLE 8.  CRANKCASE GAS ALDEHYDE CONCENTRATIONS
                   FOR THE 6V-71 DEVELOPMENT ENGINE, 13-MODE COMPOSITE
Date
5/4/76
5/4/76
Run
2
3
Concentration in mg/m^a
Form-
aldehyde
0.68
2.0
Acet-
aldehyde
0.93
1.8
Acetoneb
1.4
1.8
Isobutyr-
aldehyde0
1.9
2.4
Cro ton-
aldehyde
1.3
ND3
Hexanal
1.6
0.60
Benz-
aldehyde
1.3
0.24
Total
9.1
8.8
a at p = 101.3 kPa and T = 21eC
" includes acetone, acrolein, and propanal
c includes isobutyraldehyde ,  n-butyraldehyde , and methyl ethyl ketone
  none
           fieri
                                            26

-------
E.    Particulate Mass Emissions by Three Techniques

      Initial particulate sampling was performed using the hi-vol sampler,
permitting the engine's entire crankcase flow to be diluted and filtered.
These tests resulted in the data given in Table 9, showing good repeatabi-
lity between 13-mode runs.  The difference between particulate rates for the
11-mode and 13-mode procedures is due to the higher time-weighting factors
given to the high-power modes in the 11-mode sequence.
          TABLE 9.  CRANKCASE PARTICULATE  MASS  EMISSIONS  FROM 6V-71
                     DEVELOPMENT  ENGINE, HI-VOL SAMPLER

Engine
Operating Procedure
11-mode composite13

13-mode composite








Filter
No.
AR-6001
AR-6002
AR-6003
AR-6004
AR-6005
AR-6006
AR-6007
AR-6008
AR-6010
AR-6011
Average of 13-mode runs

Particulate
Weight, g
0.686
0.669
0.533
0.501
0.514
0.485
0.486
0.554
0.538
0.527
0.517
Particulate
Mass Rate,
g/hr
1.03
1.00
0.820
0.771
0.791
0.746
0.748
0.852
0.925
0.811
0.796
Concentration ,a
mg/m^
-.__—
	
336.
316.
324.
306.
307.
349.
379.
332.
331.
    at p  =  101.3 kPa and T  =  21°C
    initial runs conducted  during  exhaust particulate sampling tests
       Another  set of particulate  rate  data  was  obtained from test work  using
 the  47 mm sampler for  collection.   This  instrument collected samples on four
 filters simultaneously,  two  of  the  Fluoropore type and two of glass fiber.
 Data from these  runs are summarized in Table 10,  indicating reasonable  run-
 to-run agreement.   The average  mass rate for the  47 mm filters was 0.70 gram
 per  hour,  and  the average concentration  was 288 mg/m3.  These figures are
 about 12 percent lower than  those measured  by the hi-vol sampler, which is
 the  expected direction of change  given the  higher wall surface area per unit
 volume of sample which is characteristic of the 47 mm sampler.

       The final  set of particulate  rate  data was  acquired using the impactor
 system, although the primary goal of the impactor evaluation was to determine
 particle size  distribution.   These  data  are given in Table 11, and the  aver-
 age  mass rate  and concentration were about  0.48 gram per hour and 196 mg/m3,
 respectively.  These values, even lower  than those obtained using the 47 mm
 sampler, again  reflect  the  sampler wall surface  area per unit volume of sam-
 ple.  This unit  in  particular presents a number of opportunities for particles
 to adhere to surfaces  other  than  those on which particulate was measured directly.
                                      27

-------
     TABLE 10.  CRANKCASE PARTICULATE MASS EMISSIONS FROM 6V-71
        DEVELOPMENT ENGINE, 47 mm SAMPLER, 13-MODE COMPOSITE



Filter Type
Fluoropore




Filter Position
1
Filter
No.
FH4 7-6006
FH47-6008
FH4 7-6010
FH47-6012 ,
FH47-6014 '
Particulate
g/hr
0.66
0.62
0.70
0.81
0.68
mg/m3a
271.
253.
285.
334.
279.
2
Filter
No.
FH47-6007
FH47-6009
FH47-6011
FH47-6013
FH47-6015
Particulate
9/hr
0.67
0.68
0.69
0.81
0.67
mg/mja
274.
279.
284.
332.
273.



Filter Type
Glass fiber




Filter Position
3
Filter
No.
A4 7-6006 ,
A47-6008
A4 7-6010
A4 7-6012
A47-6014
Parti
g/hr
0.68
0.66
0.77
0.63
0.74
culate
mg/mja
280.
269.
315.
258.
301.
4
Filter
No.
A47-6007
A47-6009
A47-6011
A47-6013
A47-6015
Particulate
g/hr
0.73
0.76
0.81
0.59
• 0.68
mg/m1*3
299.
314.
333.
240.
280.
at p = 101.3 kPa and T = 21°C
     TABLE 11.  CRANKCASE PARTICIPATE MASS EMISSIONS FROM 6V-71
       DEVELOPMENT ENGINE, IMPACTOR SYSTEM, 13-MODE COMPOSITE
Disc Set Number
101
102
SS-106
Parti
g/hr
0.463
0.399
0.569
pulate
mg/m3a
190.
164.
233.
     a at p = 101.3 kPa and T = 21°C
                                  28

-------
F.
Size Distribution of Particulate
      As already mentioned, size distribution was determined using an iner-
tial impactor which sampled isokinetically from a diluted crankcase flow.
The flowrate through the impactor was held constant at 21.2 liters per minute
(0.75 CFM).  Data on particulate collected by impactor stage are given in
Table 12, showing good agreement between the runs using glass fiber collection
discs.  Data obtained using stainless steel collection discs are similar to
those just discussed, but different enough to lead to suspicion that the par-
ticle retention characteristics of the two types of discs are somewhat different.
           TABLE 12.  CRANKCASE PARTICULATE SIZING DATA FOR 6V-71
                    DEVELOPMENT ENGINE,  13-MODE COMPOSITE
Impactor
Stage
Number
filter
8
7
6
5
4
3
2
1
Min. Diam.a
Collected,
microns
under 0.43
0.43
0.62
1.05
2.1
3.3
4.6
7.0
11.1
Collected Weight
by Set, mg
101
1.082
1.519
3.671
4.865
0.839
0.753
0.274
0.0
0.0
102
1.051
1.445
3.420
3.904
1.380
0.612
0.251
0.0
0.0
SS106
3.892
2.638
4.017
3.835
1.196
0.546
0.340
0.077
0.048
Percent Recovered
by Set
101
8.3
11.7
28.2
37.4
6.5
5.8
2.1
0.0
0.0
102
8.7
12.0
28.4
32.4
11.4
5.1
2.1
0.0
0.0
SS106
23.5
15.9
2<..2
23.1
7.2
3.3
2.0
0.5
0.3
Cumulative Percent
by Set
101
8.3
20.0
48.2
85.6
92.1
97.8
100.0
100.0
100.0
102
8.7
20.7
49.0
81.4
92.8
97.8
100.0
100.0
100.0
SS106
23.5
39.4
63.6
86.7
93.9
97.2
99.2
99.7
100.0
'50 percent effective cutoff diameter"  (ECD)
      Visual inspection of the glass fiber discs after each run indicated ex-
tremely good definition of impact zones  (see Figure 10), especially as the
stage particulate size decreased.  Deposits on the stainless steel discs were
less well defined.  The fact that no weights were obtained on stages  1 and 2
of both runs is perhaps somewhat misleading, since some very light deposits
were noted on them.  Difficulty was experienced in weighing these two discs
in particular because the stage gaskets  cut them along their outer perimeter,
but the problem was not as pronounced with the other  stages.

      Graphical presentation of the impactor data is  given in Figures 15 and
16, the former being a log-probability plot and the latter a semilog  plot.
In order to make maximum use of the data, the dependent variable has  been
defined as "percent of particulate below stage effective cutoff diameter  (ECD)".
This definition means that filter-collected particles can be included on the
graph.  Using data for disc set 101 as an example, 8.3 percent is plotted
against 0.43 microns since that percentage was below  the 0.43 micron  (stage 8)
ECD.  Although the log-probability plot  is a more standard form for impactor
data, the semilog plot is also considered useful for  crankcase particulate
data.
                                      29

-------
10
99 99      99.9 99.8
                                  99
                                  H*f P»OB*BILITY X 2 LOG CYCLES
                                  •• C MCUFTCL • I SSC» CO  ntct l« » ^ •


                                   98       95      90        80      70    60    50
                                                                                                           468040

                                                                                                30    20        10       5

                                                                                                                                   2     1   0.5    0.2  0.1 0.05    0.01
     0.01    0.05 0.1  0.2    0.5    1
                                                                                                                                                        99.9      99.99

-------
SEMI-LOGARITHMIC       46 549O
•  -  CLi- X  .'C.  DIVI~IO*iJ   -lit I*, u  .*
       KLUF'tL  O CSSCP C^

-------
G.    Amount and Boiling Point Distribution of Organic Solubles in Particulate

      The amount of soluble organic material in the crankcase particulate mat-
ter was determined by Soxhlet extraction, and the boiling point distributions
were determined by gas chromatography.  Data on these properties are given in
Table 13 along with boiling ranges on new and used lubricating oils.  Approxi-
mately 88 percent of the parf'rulate, on the average, was soluble.  Solubles
had somewhat lower initial boiling points thvi either new or used oils but
were very similar to the oils in boiling range between 40 and 90 percent
points.  These data are certainly not conclusive, but they indicate a likeli-
hood that the particulate was mainly oil-derived.
        TABLE 13.  AMOUNT AND BOILING RANGE OF SOLUBLES IN CRANKCASE
          PARTICULATE, 6V-71 DEVELOPMENT ENGINE, 13-MODE COMPOSITE
Filter
No.
AR-6003
AR-6004
AR-6005
AR-6006
Particu-
late, g
0.5535
0.5006
0.5144
0.4850
Solubles,
9
0.4181
0.4628
0.4733
0.4360
Average
%
Solubles
78.3
92.4
92.0
. 90.0
88.2
New oil for comparison
Used oil for comparison
Boiling Range of Solubles
Temperature in °C it Weight % Off
1
266
274
277
278
274
311
316
10
336
331
348
344
340
436
428
20
428
401
434
428
423
465
460
40
483
467
485
482
479
491
489
60
506
501
513
515
509
515
5ll
80
528
533
537
542
535
545
537
90
552
576
558
566
563
578
557
95
576
577
603
585
627
574
97
591
596
594
660
580
H.    Benzo(a)Pyrene (or BaP) in Particulate Organic Solubles

      Determination of BaP was conducted to provide an indication of polynu-
clear organic matter in the soluble fraction of particulate.  These data are
given in Table 14 for the development engine, expressed in several units.
Computation of BaP in ug/hr was based on total BaP collected during each run
of 39 minutes duration.  Note that the analysis methodology used probably
produced higher values than newer methods would have produced.

I.    Elemental Composition of Particulate

      This subsection includes data from two sources.  Data on carbon, hydro-
gen, nitrogen, and sulfur by combustion were obtained by Galbraith Laboratories.
Data on elements by X-ray fluorescence were obtained by the Mobile Energy Di-
vision, assisting the Department of Emissions Research on this project.  The
resulting information is summarized in Table 15, indicating a predominantly
hydrocarbon composition with relatively small amounts of other elements.  Car-
bon/hydrogen ratio of particulate is similar to that of used engine lubricating
oil  (6.38 versus 6.63, respectively).  A measurable amount of sulfur was found
by both X-ray and combustion, but the X-ray data are considered more reliable
and accurate due to previous experience with both techniques.
                                      32

-------
       TABLE 14.   BaP CONTENT OF CRANKCASE PARTICULATE,  6V-71
               DEVELOPMENT ENGINE, 13-MODE COMPOSITE*
Filter
No.
AR-6003
AR-6004
AR-6005
AR-6006
Averages
Total
BaP, U9
22.7
10.2
11.4
20.3
16.2
Weight % BaP
in Particulate
0.0041
0.0020
0.0022
0.0042
0.0031
Weight % BaP
in Solubles
0.0054
0.0022
0.0024
0.0047
0.0037
BaP
pg/hr
35.
16.
18.
31.
25.
a technique used produced higher values than newer methods
     TABLE 15.  ELEMENTAL COMPOSITION OF CRANKCASE PARTICULATE,
             6V-71 DEVELOPMENT ENGINE,  13-MODE COMPOSITE

Filter No. (s)
A47-6006 & 07
A47-6008 & 09
A47-6010 & 11
A47-6012 & 13
A47-6014 & 15
Averages
Weight % of Particulate by Combustion
C
86.2
86.2
83.6
83.6
85.1
84.9
H
13.1
13.1
13.3
13.4
13.7
13.3
N
0.09
	 a
	 a
	 a
___a
	 a
S
0.4
0.4
0.7
0.5
0.8
0.6
E CHNS
99.8
99.7
97.6
97.5
99.6
98.8
Filter No. (s)
FP47-6007
FP47-6009
FP47-6011
FP47-6013
FP47-6015
Averages
Weight % of Particulate by X-raj,
P
0.04
0.02
0.03
0.04
0.04
0.03
S
0.43
0.38
0.37
0.40
0.39
0.39
Cl
0.02
0.03
0.02
	 b
	 b
0.01
Fe
b
	 b
0.0041
	 b
b

	 b

•
Zn
b
	 b
0.0044
	 b
b

b

   a  trace  -  less  than  0.01 percent
     none detected
                                    33

-------
J.    Sulfate (S04=)  in Participate

      Sulfate was measured by the BCA technique with the results shown in
Table 16.  Values determined were reasonably consistent compared with one
another, and it can be computed that the sulfur in the sulfate made up about
12 percent of the total sulfur in the particulate as determined by X-ray.
Data from the other engines tested will be examined on this point to see if
the trend is uniform.
             TABLE 16.  SULFATE IN CRANKCASE PARTICULATE, 6V-71
                    DEVELOPMENT ENGINE, 13-MODE COMPOSITE
Filter
No.
FH4 7-6006
FH4 7-6008
FH47-6010
FH47-6012
FH47-6014
Averages
ug so4=
on Filter
4.8
4.9
5.4
6.6
6.0
5.6
S in S04= as
Weight % Particulate
0.040
0.041
0.045
0.046
0.051
0.045
S04=,
mg/hr
0.80
0.75
0.93
1.12
1.03
0.93
                                      34

-------
         IV.  CHARACTERIZATION OF IN-SERVICE CUMMINS NTC-350 ENGINE
      Characterization work was conducted on the two in-service engines de-
scribed in Section II, and all the analytical methods were used except SEM
studies.  This section covers results achieved with the Cummins NTC-350 en-
gine alone to avoid confusion which would result if data from the two engines
were mixed.

A.    Measurement of Crankcase Gas (Blowby) Flowrates

      Measurement of crankcase emission flowrate on the truck NTC-350 diesel
engine was accomplished using the flow measurement systems shown schemati-
cally in Figures 1 and 2.  The system shown in Figure 2 waa used during those
modes for which blowby flows exceeded the flow capability of the system shown
in Figure 1.  The results, shown in Table 17, reflect inflow through the
blowby tube observed at idle, 1500 - 2 percent load, and 2100 - 2 percent
load, thereby obviating flow determinations at these modes.  The negative
pressures at the blowby tube ranged from 15 Pa at idle to 810 Pa at 2100 -
2 percent load.  The proportional sample rates shown were used when samples
were being taken for aldehydes and individual hydrocarbons.  The crankcase
flowrates from this engine were quite high as compared to those from the de-
velopment engine, but it is not known whether or not they were typical of
similar turbocharged engines.  No published data on crankcase flows from a
cross-section of engines have been located.
     TABLE 17.  CRANKCASE EMISSION FLOWRATES, TRUCK WITH NTC-350 ENGINE
Operating Condition
Engine rpm
600 (idle)
1500
1500
1500
1500
1500
2100
2100
2100
2100
2100
Load, %
	
2
25
50
75
100
2
25
50
75
100
Composite
Crankcase Flowrate
Std. m3/hra
	 c
	 C
8.7
13.0
16.0
21.3
	 C
12.8
17.2
21.1
24.6
SCFMa
	 c
	 C
5.1
7.6
9.4
12.5
	 C
7.5
10.1
12.4
14.5
kg/hrb
	 c
	 c
10.4
15.5
19.3
25.5
	 C
15.4
20.6
25.3
29.5
Proportional
Sample , i/min
0
0
1.42
2.11
2.61
3.46
0
2.08
2.79
3.44
4.00
10.8
     at p = 101.3 kPa and T = 21°C
     assuming gases have same density as air
     zero flow or inflow
                                      35

-------
 B.
Measurement of HC, CO, C02, NOX, and 02 in Crankcase Gases and Compari-
son to Gaseous Exhaust Emissions
       Gaseous constituents of crankcase emissions from the truck with NTC-350
 engine were measured in three separate runs,  and the average results are sum-
 marized in Table 18.   These concentrations have been combined with the flow-
 races given in Table n7 to calculate mass emission rates of gaseous consti-
 tuents, and the mass rates are given in Table 19.  All the^gaseous emissions
 were determined from 13-mode steady-state data.  The composite exhaust emis-
 sions tabulated at the bottom of Table 19 for comparison were taken concur-
 rently with the crankcase emissions.   Comprehensive exhaust emissions data
 are given  in Appendix B,  pages B-4 through B-6.  Comparison of crankcase and
 exhaust emissions reveals the crankcase (gaseous) HC amount as about 4.1 per-
 cent of exhaust HC and that other crankcase emissions range from about 0.08
 to 0.43 percent of their counterparts in exhaust gases.
            TABLE 18.   GASEOUS CONSTITUENTS OF CRANKCASE EMISSIONS
               TRUCK WITH NTC-350 ENGINE (AVERAGE OF THREE RUNS)
Operating Condition
Engine rpm
600 (idle)
1500




2100




Load, %
___a
2a
25
50
75
100
2a
25
50
75
100
HC
ppm C
	
««•
81
84
99
164
._.
70
73
91
105
CO
ppm
	
_«_
27
110
335
597
•.._
75
76
79
118
co2
%
	
____
0.30
0.29
0.35
0.38
_••»•
0.33
0.34
0.36
0.43
NOX
ppm
	
___
35.
46.
57.
61.
•»_«
37.
63.
44.
37.
°2
%
	
— .«
20.8
20.8
20.8
20.8
•.«_
20.8
20.8
20.8
20.8
   a zero flow or inflow
 C.     Measurement of Individual Hydrocarbons in Crankcase Gases

       Samples for individual HC analysis were taken in proportion to crank-
• case gas flows,  using the sampling rates given in Table 17.   The results of
 these tests are  summarized in Table 20.   Concentrations above 3.0 ppm C were
 found for methane and ethylene.  Propylene and benzene concentrations aver-
 aged just under  3.0 ppm C; while ethane, acetylene, and toluene averaged
 under 2.0 ppm C.  Propane was indicated in trace quantities  only.  These
 concentrations were generally somewhat higher than those found in the crank-
 case gases of the 6V-71 development engine.
                                      36

-------
  TABLE 19.   CRANKCASE MASS EMISSIONS FROM TRUCK WITH NTC-350 ENGINE
Operating Condition
Engine rpm
600 (idle)
1500


2100



Load, %
2
25
50
75
100
2
25
50
75
100
13 -mode composite
of crankcase emissions
13 -mode composite
of exhaust emissions
Emissions in grams per hour
HC


0.41
0.63
0.92
2.03
0.52
0.73
1.12
1.50
0.63
15.3
CO


0.27
1.66
6.26
14.77
1.12
1.52
1.94
3.38
2.48
582.
C02


47.6
68.5
102.
147.
77.1
106.
139.
193.
70.4
86,500
NOV
A


0.58
1.14
1.74
2.47
0.90
1.44
2.54
3.52
1.15
1170.
           TABLE 20.   CRANKCASE GAS INDIVIDUAL HYDROCARBONS,
                   13-MODE COMPOSITE, NTC-350 ENGINE
Constituent
Methane
Ethylene
Ethane
Acetylene
Propane
Propylene
Benzene
Toluene
Total non -Me thane
ppm C
Run 1
8.8
10.4
0.7
1.1
0.1
4.0
3.0
2.8
22.1
Run 2
5.1
5.6
0.4
0.5
tra
2.4
2.5
1.6
13.0
Run 3
4.2
5.0
0.3
0.3
tra
2.4
3.1
1.1
12.2
Background
2.7
tra
0.1
	
	
	
	
	
0.1
Avg. Net
3.3
7.0
0.4
0.6
tra
2.9
2.9
1.8
15.6
a trace - less than 0.1 ppm C
                                   37

-------
  D.    Measurement of Aldehydes in Crankcase Gases

        Aldehydes in the blowby gases of the NTC-350 engine were measured by
  the DNPH method, and the results are summarized in Table 21.  Run 1 was dis-
  carded due to a calibration error, but runs 2 and 3 were valid and the results
  appear to agree quite well.1, These samples also contain a greater fraction of
  simple aldehydes as compared to those taken from the 6V-71 development engine.
  Some reservations must be expressed about the aldehyde data, as discussed
  earlier, because it was originally developed for dilute exhaust from gasoline-
  fueled vehicles.
            TABLE 21.  CRANKCASE GAS ALDEHYDE CONCENTRATIONS FOR THE
                     NTC-350 TEST ENGINE, 13-MODE COMPOSITE
Date
7/14/76
7/14/76
Run
2
3
Concentration in mg/m3a
Form-
aldehyde
5.4
5.1
Acet-
aldehyde
2.3
1.5
Acetone**
1.2
0.72
Isobutyr-
aldehyde0
0.44
1.1
Croton-
aldehyde
0.41
0.40
Hex-
anal
0.15
0.047
Benz-
aldehyde
0.24
0.26
Total
10.1
9.1
at p = 101.3 kPa and T = 21°C
includes acetone, acrolein, and propanal
includes isobutyraldehyde, n-butyraldehyde, and methyl ethyl ketone
  E.
Crankcase Particulate Mass Emissions by Three Techniques
        The first type of particulate measuring system used on the.NTC-350 was
  the "hi-vol" system, which sampled the entire crankcase flow.  The results
  of the hi-vol tests are presented in Table 22,  showing good repeatability be-
  tween runs.   Concentration of particulate from this engine was lower than
  from the development engine, but crankcase flow was higher.  These effects
  compensated for each other to the point that particulate mass flowrate from
  the NTC-350 was only about 27 percent higher than that from the development
  engine.

        A second and independent set of particulate data was obtained from test
  results using the 47 mm sampler.  These data are given as Table 23, and they
  show reasonable consistency and agreement with the hi-vol data.  As averages,
  particulate rate was 1.03 grains per hour with the 47 mm sampler and particulate
  concentration was 95.8 mg/m3.  It was mentioned in the discussion of data on
  the development engine that particulate rate and concentration were expected
  to'be somewhat lower using the 47 mm system as compared to the hi-vol system.
  Data on the NTC-350 did not bear out this expectation, and no reasons have
  been discovered for this result.

        The final set of particulate rate data was acquired using the impactor
  system, even though the main purpose of the impactor tests was to define par-
  ticle size distribution.  Impactor rate data are given in Table 24, and they
  average about 0.76 grams per hour and 71 mg/m3.  These values are about 25
  percent lower than those obtained with either the hi-vol or the 47 mm systems,
  which is the expected direction of change.
                                        38

-------
       TABLE 22.  OKANKCASE  PARTU'IMJVPK MAS:- l-'MI:'.::ii»N:. I1-UI>M  AN
          NTC-350 ENGINE,  HI-VOL SAMPLER, 1J-MOLM-: iVMIMSLTK
Filter
No.
AR-6012
AR-6013
AR-6014
AR-6015
AR-6017
AR-6018
AR-6019
AR-6020
AR-6021
Averages
Participate
Weight, g
0.644
0.623
0.638
0.617
0.589
0.625
0.637
0.695
0.812
0.653
Particulate
Mass Rate, g/hr
0.991
0.958
0.982
0.949
0.906
0.962
0.980
1.07
1.25
1.01
Concentration , mg/m Ja
91.8
88.7
90.9
87.9
83.9
89.1
90.7
99.1
116.
93.1
 a at p = 101.3 kPa  and T = 21°C
    TABLE 23.   CRANKCASE PARTICULATE MASS EMISSIONS  FROM NTC-350
              ENGINE,  47 mm SAMPLER, 13-MODE COMPOSITE


Date
7/9/76
7/9/76
7/9/76
7/12/76
7/12/76

Run
No.
1
2
3
4
5
Particulate Rate by
Filter Position3, c
1
1.02
1.00
1.06
0.99
0.96
2
0.75
1.02
0.79
0.97
0.99
3
1.13
0.94
1.15
1.38
1.18
/hr
4
1.09
1.05
1.13
0.98
1.06
Particulate Concentration
by Filter Position3 'b, mg/m3
1
94.4
92.4
98.4
91.7
88.8
2
69.4
94.2
72.9
90.3
92.3
3
105.0
87.5
107.0
128.0
110.0
4
101.0
97.2
105.0
90.9
98.8
3 Fluoropore filters  at positions 1 and 2,
  at p = 101.3 kPa and T  =  21°C
glass fiber at  3 and  4
                                  39

-------
        TABU: 24.  CRANKCASE PARTICIPATE MASS EMISSIONS FROM NTC-SSO
                 ENGINE, IMPACTOR SYSTEM, 13-MODE COMPOSITE
Disc Set Number
103
104
105
108 ;
g/hr
0 886
0.663
0.5BO
0.920
mg/mja
82.0
61.4
53.7
85. 2
       a at p = 101.3 kPa and T = 21°C
F.    Size Distribution of Particulate

      Impactor samples for size distribution study were taken isokinetically
from a diluted total crankcase flow, maintaining a constant sample rate of
21.2 liters per minute (0.75 CFM).  Data on particulate collected by impactor
stage are given in Table 25, and they show good agreement between tests on
the basis of percentage collected.  Graphical presentation of these results
is given in Figures 17 and 18, the former being a log-probability plot and
the latter a semilog plot.  Again note that the dependent variable is "per-
cent of particulate below BCD".  Three of the runs yielded very similar plots
(disc sets 104, 105, and 108), but the first run had quite a bit less parti-
culate on discs 1 through 4 than did the other three.  It is possible that
cutting of discs by gaskets had some influence on the results.

G.    Amount and Boiling Point Distribution of Organic Solubles in Particulate

      The soluble fraction of the particulate was determined by Soxhlet ex-
traction, and the boiling range by gas chromatograph.  Data on these proper-
ties are given in Table 26 along with boiling ranges on new and used lubri-
cating oils.  About 93 percent of the particulate, on the average, was solu-
ble.  The solubles were very similar in boiling range; and as a group, they
were similar to the oils between 40 and 90 percent points.  Solubles were
somewhat lower-boiling than oils between IBP and 40 percent points, but the
opposite was true above the 90 percent point.

H.    Benzo(d)Pyrene (or BaP) in Particulate Organic Solubles

      BaP was determined as an indicator of total polynuclear organic matter
in the soluble fraction of particulate.  Data are given in Table 27 for the
NTC-350 engine, expressed in several sets of units.  Computation of BaP in
Mg/hr was based on total sample collected during each 39-minute run.  Note
that the analysis methodology used probably yielded BaP values which were
high as compared to those which would have been generated by newer methods.

I.    Elemental Composition of Particulate

      This subsection includes X-ray data determined by the Institute's Mo-
bile Energy Division and CHNS data obtained from Galbraith Laboratories.
Elemental data are presented as Table 28, and they indicate a predominantly
                                     40

-------
          TABLE 25.   CRANKCASE PARTICULATE SIZING DATA FOR NTC-350 ENGINE, 13-MODE COMPOSITE
Impactor
Stage
Number
filter
8
7
6
5
4
3
2
1
Minimum Diameter*
Collected
microns
under 0.43
0.43
0.62
1.05
2.1
3.3
4.6
7.0
11.1
Collected Weight
by Set, mg
103
2.364
2.483
4.044
2.518
0.312
0.072
0.0
0.0
0.004
104
2.050
2.024
2.814
1.578
0.304
0.099
0.081
0.0
0.126
105
1.824
2.074
2.081
1.121
0.199
0.063
0.128
0.055
0.086
108
2.447
3.359
4.001
1.671
0.551
0.181
0.187
0.124
0.069
Percent Recovered
by Set
103
20.0
21.1
34.3
21.4
2.6
0.6
0.0
0.0
0.0
104
22.6
22.3
31.0
17.4
3.3
1.1
0.9
0.0
1.4
105
23.9
27.2
27.3
14.7
2.6
0.8
1.7
0.7
1.1
108
19.4
26.7
21.8
13.3
4.4
1.4
1.5
1.0
0.5
Cumulative % by Set
103
20.0
41.1
75.4
96.8
99.4
100.0
100:0
100.0
100.0
104
22.6
44.9
75.9
93.3
96.6
97.7
98.6
98.6
100.0
105
23.9
51.1
78.4
93.1
95.7
96.5
98.2
98.9
100.0
108
19.4
46.1
77.9
91.2
95.6
97.0
98.5
99.5
100.0
"50 percent effective cutoff ciameter"  (ECD)

-------
0.01   0.05 0.1
                                                                                                              90
                                                                                                                      95       98    99
                                                                                                                                                 99.8 99.9       99 99

-------
I
SEMI-LOGARITHMIC     46 S49O
J CYCLES X 7O Divisions  MAPI is - i •
     KtuFrCL a e^scn co.

-------
         TABLE 26.   AMOUNT AND BOILING RANGE OF SOLUBLES IN CRANKCASE
                PARTICULATE,  NTC-350 ENGINE, 13-MODE COMPOSITE

Filter
No.
AR-6017
AR-6018
AR-6019
AR-6020
AR-6021

Particu-
late, g
0.5890
0.6251
0.6374
0.6947
0.8122

Solubles ,
9
0.5483
0.6016
0.6116
0.6132
0.7440
Averages

%
Solubles
93.1
- 95.2
96.0
88.3
, 91.6
93.0
New oil for comparison
Used oil for comparison
Boilinq R.-inqe of Solubles
Temperature in °C at Weight % off
1
304
313
303
303
298
304
312
311
10
390
385
3"1!
396
389
390
419
416
20
427
424
427
437
419
427
447
446
40
465
460
464
468
460
463
470
471
60
486
483
482
486
481
484
487
490
80
512
507
503
506
506
507
506
511
90
561
538
519
533
533
537
524
536
95
631
602
543
585
596
591
547
566
97
__a
622
568
636
— a
609
571
— a
out of range of analysis
               TABLE 27.  BaP CONTENT OF CRANKCASE PARTICULATE,
                      NTC-350 ENGINE, 13-MODE COMPOSITE3
Filter
No.
AR-6017
AR-6018
AR-6019
AR-6020
AR-6021
Averages
Total
BaP, yg
1.4
6.3
1.8
13.9
9.6
6.6
Height % BaP
in Particulate
0.00024
0.00101
0.00028
0.00200
0.00118
0.00094
Weight % BaP
in Solubles
0.00026
0.00104
0.00029
0.00227
0.00129
0.00103
BaP in
pg/hr
2.2
9.7
2.7
21.4
14.8
10.2
    a technique used produced high values as compared to newer methods
                                        44

-------
hydrocarbon material with relatively small amounts of other elements.  A
substantial amount of sulfur was found by both X-ray and combustion, but the
X-ray data are considered to be more reliable.  Carbon and hydrogen data are
highly variable, and the low average carbon values are suspect because the
resulting average C/H ratio by weight is very low (5.48 as compared to 6.49
for the engine's used lubricating oil).  The only probable explanation for
this result is that some contaminant in the solubles from participate created
a negative interference with the carbon analysis.
         TABLE 28.  ELEMENTAL COMPOSITION OF CRANKCASE PARTICULATE,
                      NTC-350 ENGINE, 13-MODE COMPOSITE
Filter Nos.
A47-6024 & 6025
A47-6026 & 6027
A47-6028 
-------
TABLE 29.   SULFATE IN CRANKCASE PARTICULATE,
      NTC-350 ENGINE, 13-MODE COMPOSITE
Filter
No.
FH47-6025
FH47-6027
FH4 7-6029
FH47-6033
FH47-6035
Averages
ug so4=
on Filter
1.82
2.63
1.62
2.14
2.22
2.09
S in S04= as
Weight % Particulate
0.028
0.040
0.023
0.034
0.036
0.032
so4-,
mg/hr
0.84
1.21
0.74
1.00
1.02
0.96
                        46

-------
     V.   CHARACTERIZATION OF IN-SERVICE DETROIT DIESEL 6V-71 BUS ENGINE
      Data on the 6V-71 development engine and the NTC-350 truck engine have
already been presented in Sections III and IV, respectively.  This section
deals with data on the 6V-71 bus engine only, to prevent confusing data on
the three engines.

A.    Measurement of Crankcase Gas (Blowby) Flowrates

      Where possible, within cost and time limitations, data were taken on
the 6V-71 bus engine during two distinct types of 13-mode tests.  One type
used "standard" engine speeds of 2100 and 1260 rpm (plus idle)/ which are the
speeds used to certify the 6V-71.  The other type used "low" speeds of 1500
and 900 rpm, which are considered more representative of bus operation.  All
testing on the 6V-71 bus engine was performed using a No. 1 diesel fuel, while
that on the NTC-350 and 6V-71 development engine was performed using a "national
average" No. 2 diesel fuel.

      Tables 30 and 31 give data on crankcase flowrates using both standard-
and low-speed procedures, in addition to the proportional sample rates for
individual hydrocarbon and aldehyde sampling  (standard jpeeds).  Although this
engine was similar to the 6V-71 development engine, it had higher crankcase
flows (by a factor of 1.6) and generally did not perform as well.

B.    Measurement of HC, CO, C02» NOX, and 02 in Crankcase Gases and Compari-
      son to Gaseous Exhaust Emissions

      Gaseous constituents of crankcase emissions from the 6V-71 bus engine
were measured in duplicate runs at both standard and low speeds.  The average
results are summarized in Tables 32 and 33.  Concentrations given have been
used with flowrates from Tables 30 and 31 to calculate mass emission rates
of gaseous constituents.  The results of these computations are given in
Tables 34 and 35.  Composite exhaust emissions tabulated at the bottoms of
Tables 34 and 35 were derived from data taken concurrently with the crankcase
gaseous emissions tests.  Comprehensive exhaust emissions data are included
in Appendix B, pages B-7 through B-10.  Comparison of emissions from crankcase
and exhaust shows that crankcase (gaseous) HC amounts to about 0.31 percent
of exhaust HC at standard speeds and about 0.20 percent at low speeds.  Other
crankcase emissions ranged from about 0.04 percent to 0.08 percent of their
counterparts in exhaust gases.

C.    Measurement of Individual Hydrocarbons in Crankcase Gases

      Samples for individual HC analysis were taken in proportion to crankcase
gas flows, using the sampling rates given in Table 30.  Results on these sam-
ples are summarized in Table 36, and they show slightly lower concentrations
than the other 6V-71 engine did  (the development engine).  Only ethylene and
propylene averaged over 1.0 ppm C, and all the other compounds are nearly in
the trace category (0.2 ppm C or less).
                                      47

-------
           TABLE 30.   CRANKCASE EMISSION FLOWRATES, 6V-71 CITY
                  BUS ENGINE AT STANDARD ENGINE SPEEDS
Operating Condition
Engine rpm
515 (idle)
1260
1260
1260
1260
1260
2100
2100
2100
2100
2100
Load, %
	
2
25
50
75
100
2
25
50
75
100
Composite

Std. m3/hra
1.2
-.7
2.9
3.1
3.1
3.2
5.6
5.8
6.2
6.4
6.8
se Flowra
SCFM3
0.7
1.6
1.7
1.8
1.8
1.9
3.3
3.4
3.6
3.8
4.0
te
kg/hr*
1.4
3.3
3.5
3.7
3.7
3.8
6.7
6.9
7.4
7.7
8.1
Proportional
Sample , i/min
0.71
1.61
1.71
1.80
1.83
1.86
3.27
3.40
3.63
3.78
4.00
3.9
a at p = 101.3 kPa and T = 21°C
  assuming gases have same density as air
           TABLE 31.   CRANKCASE EMISSION FLOWRATES, 6V-71 CITY
                     BUS ENGINE AT LOW ENGINE SPEEDS
Operating Condition
Engine rpm
515
900
900
900
900
900
1500
1500
1500
1500
1500
Load, %
	
2
25
50
75
100
2
25
50
75
100
Composite
Crankc
Std. m3/hra
1.2
2.2
1.8
1.6
1.5
1.2
3.6
3.7
3.9
4.1
4.3
ise Flowrat
SCFM*
0.7
1.3
1.0
1.0
0.9
0.7
2.1
2.2
2.3
2.4
2.5
a
kg/hrb
1.5
2.6
2.1
1.9
1.8
1.4
4.4
4.5
4.7
4.9
5.1
2.5
     a at p = 101.3 kPa and T = 21°C
       assuming gases have same density as air
                                    48

-------
TABLE 32.  GASEOUS CONSTITUENTS OF CRANKCASE EMISSIONS ,
 CITY BUS 6V-71, STANDARD SPEEDS  (AVERAGE OF TWO RUNS)
Operating Condition
Engine rpm
515 (idle)
1260
1260
1260
1260
1260
2100
2100
2100
2100
2100
Load, %
	
2
25
50
75
100
2
25
50
75
100
HC
ppm C
258.
229.
242.
241.
299.
321.
285.
330.
336.
377.
319.
CO
ppm
86.
37.
23.
39.
142.
695.
24.
21.
24.
41.
146.
co2
%
0.45
0.07
0.15
0.21
0.40
0.63
0.11
0.17
0.30
0.49
0.66
NOX
ppm
23.5
10.0
29.0
45.0
71.5
79.5
28.5
29.5
50.0
82.5
91.5
°2
%
20.7
20.8
20.7
20.5
20.2
19.5
20.7
20.6
20.4
20.1
20.0
TABLE 33.  GASEOUS CONSTITUENTS OF CRANKCASE EMISSIONS,
   CITY BUS 6V-71, LOW SPEEDS  (AVERAGE OF TWO RUNS)
Operating Condition
Engine rpm
515 (idle)
900
900
900
900
900
1500
1500
1500
1500
1500
Load, %
	
2
25
50
75
100
2
25
50
75
100
HC
ppm C
240.
196.
215.
226.
276.
330.
267.
299.
303.
336.
303.
CO
ppm
109.
33.
24.
123.
341.
1084.
32.
14.
33.
70.
416.
co2
0.12
0.07
0.14
0.41
0.50
0.66
0.07
0.14
0.29
0.45
0.67
NOX
ppm
12.5
3.0
23.0
59.5
94.0
80.0
8.5
32.0
52.5
70.5
91.5
°2
%
20.8
20.7
20.5
20.3
19.9
19.5
20.7
20.5
20.2
20.1
20.0
                           49

-------
TABLE 34.  CRANKCASE MASS EMISSIONS FROM CITY BUS 6V-71, STANDARD SPEEDS
Operating Condition
Engine rpm
515 (idle)
1260
1260
1260
1260
1260
2100
2100
2100
2100
2100
Load , %j
	
2
25
50
75
100
2
25
50
75
100
13-mode composite
of crankcase emissions
13-mode composite
of exhaust emissions
crankcase/exhaust
composite emissions, %
Emissions in grams per hour
HC
0.18
0.36
0.41
0.43
0.54
0.59
0.92
1.11
1.20
1.41
1.26

0.69

220.

0.31
CO
0.12
0.12
0.08
0.14
0.51
2.55
0.16
0.14
0.17
0.30
1.15

0.45

599.

0.08
co2
9.8
3.2
7.7
11.7
22.7
36.3
10.6
17.4
33.1
56.8
81.1

24.4

47,900.

0.05
NOX
0.05
0.05
0.16
0.26
0.42
0.48
0.30
0.32
0.59
1.01
1.18

0.39

809.

0.05
   TABLE 35.  CRANKCASE MASS EMISSIONS FROM CITY BUS 6V-71, LOW SPEEDS
Operating Condition
Engine rpm
515 (idle)
900
900
900
900
900
1500
1500
1500
1500
1500
Load, %
	
2
25
50
75
100
2
25
50
75
100
13-mode composite
of crankcase emissions
13-mode composite
of exhaust emissions
crankcase/exhaust
composite emissions, %
Emissions in grams per hour
HC
0.17
0.25
0.22
0.21
0.24
0.22
0.48
0.65
0.69
0.79
0.75

0.39

194.

0.20
CO
0.15
0.08
0.05
0.23
0.59
1.46
0.13
0.06
0.15
0.33
2.07

0.44

774.

0.06
C02
2.5
2.6
4.5
12.1
13.6
14.0
4.6
9.5
20.3
33.1
• 57.9

14.3

33,800.

0.04
NOX
0.03
0.01
0.08
0.18
0.27
0.18
0.06
0.23
0.39
0.55
0.75

0.22

625.

0.04
                                    50

-------
              TABLE  36.  CRANKCASE GAS  INDIVIDUAL HYDROCARBONS,
            13-MODE  COMPOSITE, 6V-71 BUS ENGINE  (STANDARD SPEEDS)
Constituent
Methane
Ethylene
Ethane
Acetylene
Propane
Propylene
Benzene
Toluene
Total non-Methane
ppm C
Run 1
2.5
1.6
0.1
0.2
tra
0.8
0.1
0.1
2.9
Run 2
3.4
2.0
0.2
tra
0.2
1.5
0.1
0.1
4.1
Run 3
3.4
1.9
0.3
-Jb
0.2
1.4
0.1
0.1
4.0
Background
3.0
tra
0.2
tra
tra
0.1
0.3
Avg. net
0.1
1.8
0.1
0.2
1.2
0.1
__b
3.4
      trace  -  less  than  0.1  ppm C
    b below  limit of  detection
 D.
Measurement of Aldehydes in Crankcase Gases
       Crankcase gases were analyzed for aldehydes by the DNPH method,  and the
 results are summarized in Table 37.  Run 1 was discarded due to anomalous re-
 sults, thought to have been caused by an oil or other hydrocarbon interference.
 In terms of general ranges, aldehydes from the 6V-71 crankcase were similar
 to those from the other two engines.   Consistency between runs, however,  was
 not as good as that observed with the other two engines.
             TABLE 37.  CRANKCASE GAS ALDEHYDE CONCENTRATIONS FOR
          • THE 6V-71 BUS ENGINE, 13-MODE COMPOSITE (STANDARD SPEEDS)

9/1/76
9/1/76
Run
2
3
Concentration in mg/m^3
Form-
aldehyde
3.2
3.8
Acet-
aldehyde
'2.1
3.3
Acetone*3
1.7
2.7
Isobutyr-
aldehydec
2.6
3.5
Croton-
aldehyde
0.68
1.4
Hex-
anal
0.77
3.4
Benz-
aldehyde
0.81
NDd
Total
11.9
18.1
at p = 101.3 kPa and T = 21°C
includes acetone, acrolein, and propanal
includes isobutyraldehyde, n-butyraldehyde,
none detected
                                      and methyl  ethyl  ketone
 E.     Crankcase Particulate Mass Emissions by Three Techniques

       Initial particulate samples from the 6V-71 bus engine were taken with
 the hi-vol system, which collects particulate from the entire (diluted) crank-
 case flow.  The results of these tests are summarized in Table 38, and they
 show good consistency.   This engine had crankcase particulate concentrations
 almost twice as high as the 6V-71 development engine, and 6.7 times as high
 as the NTC-350 test engine.  Consequently, the 6V-71 bus engine's crankcase
                                      51

-------
 particulate rate was about three times that of the 6V-71 development engine
 and about 2.5 times that of the NTC-350.   These particulate rates also re-
 flect the relatively high crankcase flowrates from the 6V-71 bus engine as
 compared to those from the 6V-71 development engine.
            TABLE 38.  CRANKCASE PARTICULATE MASS EMISSIONS FROM A
     6V-71 BUS ENGINE, HIrVOL SAMPLER, 13-MODE COMPOSITE  (STANDARD SPEEDS)
Filter
NO.
AR-6025
AR-6026
AR-6027
AR-6028
AR-6029
AR-6030
Averages
Particulate
Weight, g
1.541
1.558
1.573
1.641
1.637
1.581
1.588
Parl
Mass Rate, g/hr
2.37
2.40
2.42
2.52
2.52
2.43
2.44
ticulate
Concentration, mg/m33
609.
617.
622.
648.
648.
625.
628.
     at p = 101.3 kPa and'T = 21°C
       Additional particulate rate and concentration data were obtained from
 tests using the 47 mm sampler, and this information is presented as Table 39.
 Values obtained show good consistency and reasonable agreement with hi-vol
 data given the differences in the sampling systems.  Average particulate rate
 with the 47 mm system was 1.78 grams per hour, and average concentration was
 459 mg/m3.  These values are about 25 percent lower than those obtained using
 the hi-vol system.
            TABLE 39.  CRANKCASE PARTICULATE MASS EMISSIONS FROM A
              6V-71 BUS ENGINE, 47 mm SAMPLER, 13-MODE COMPOSITE
Date
8/26/76
8/27/76
8/27/76
8/27/76
8/27/76
Run
No.'
3
1
2
4C
5C
Particulate Rate by
Filter Position3, g/hr
1
1.59
1.82
1.33
1.32
2
1.64
1.81
1.40
1.37
3
1.63
1.94
1.85
1.33
1.37
4
1.65
2.08
1.83
1.30
1.36
Particulate Concentration
by Filter Position* 'b, mg/m^
1
409.
468.
540.
536.
2
422.
465.
566.
555.
3
419.
499.
476.
540.
555.
4
424.
535.
470.
526.
552.
a at p = 101.3 kPa and T = 21°C
b Fluoropore filters at positions 1 and 2,  glass fiber at 3 and 4
c low speeds (900 and 1500 rpm) - others at standard speeds
                                        52

-------
      In conjunction with particle sizing data, particulate rate information
was obtained using the diluter-impactor system.  Data from these tests indi-
cates that a sample line leak was present, making the particulate rate data
thus obtained of no value.  Size distributions obtained in these runs have
been scrutinized carefully for effects of sub-isokinetic sampling and erro-
neous impactor flowrates.
F.
Size Distribution of Particulate
      As mentioned in the previous subsection, data on particle size distri-
bution for the 6V-71 bus engine were somewhat suspect due to an apparent leak
in the sampling system.  For this reason, the actual particle weights collected
were meaningless; and  therefore, only the percentages are tabulated and dis-
cussed.  The percentage data are given in Table 40, and they show reasonable
agreement throughout the size range.  These sizing data are presented'graphi-
cally in Figures 19 and 20, a log-probability plot and a semilog plot, re-
spectively.  Note that the dependent variable is "percent of particulate
below BCD".  The plots for these two sets of discs are very similar, and
they compare quite well to corresponding data on the other two engines.  It
is concluded, therefore, that the sample line leak was upstream of the impac-
tor and that the size distribution data are valid on a percentage basis.
           TABLE 40.  CRANKCASE PARTICULATE  SIZING DATA FOR 6V-71
               BUS ENGINE,  13-MODE COMPOSITE (STANDARD SPEEDS)
Impactor
Stage
Number
filter
8
7
6
5
4
3
2
1
Minimum Diam.a
Collected,
microns
under 0.43
0.43
0.62
1.05
2.1
3.3
4.6
7.0
11.1
% Recovered
by Set
109
17.7
20.0
29.5
24.1
5.9
1.9
0.9
0.0
0.0
113
19.8
18.9
30.3
23.3
5.5
1.5
0.5
0.2
0.0
Cumulative %
by Set
109
17.7
37.7
67.2
91.3
97.2
99.1
100.0
100.0
100.0
113
19.8
38.7
69.0
92.3
97.8
99.3
99.8
100.0
100.0
G.
      "50 percent  effective  cutoff diameter"  (BCD)
Amount and Boiling Point Distribution of Organic Solubles in Particulate
      Total organic solubles in particulate were separated and determined by
Soxhlet extraction, and boiling range was obtained via gas chromatograph.
These properties are described in Table 41 along with boiling ranges on lu-
bricating oils  (new and used) for comparison.  As an average, about 98 per-
cent of the particulate matter was soluble.  The solubles had slightly lower
initial boiling points than new oil, while the used oil had a much higher IBP.
Most of the remaining boiling ranges were quite similar, especially between
                                      53

-------
                                         PROBABILITY X 2 LOG CYCLES
                                                          t  MMI m vt4
99.99       99.9 99.8
2     1    0.5   0.2  0.1 0.05    0.01    ,0
                             99   96       95      90        80706090403020
 0.01    0.05
                                                                                                                       95       98    99
                                                                                                                                                 99.8 99.9       99.99

-------
K-2
»tMJ LOuANlTNMIC   4tt 549O





    KCUFFEk ft »»C" ^n.

-------
   40 and 90 percent points.   It appears  likely that this engine's crankcase
   particulate was mostly oil-derived, but that the particulate may have been
   more  fuel-diluted than oil.
           TABLE  41.   AMOUNT AND BOILING RANGE OF SOLUBLES IN CRANKCASE
        PARTICULATE,  6V-71 BUS ENGINE,  13-MODE COMPOSITE (STANDARD SPEEDS)
Filter
NO.
AR-6025
AR-6026
AR-6027
Particu-
late, g
1.541
1.558
1.573
Solubles!
g
1.510
1.528
1.543 '
Averages
%
Solubles
98.0
98.1
98.1
98.1
New oil for comparison
Used oil for comparison
Boiling Range of Solubles
Temperature in °C at Weight % Off
1
264
290
293
282
307
366
10
415
415
420
417
462
465
20
457
471
477
468
491
493
40
477
503
506
495
511
513
60
499
512
524
512
529
528
80
528
522
551
534
552
550
90
565
586
585
579
576
575
95
619
625
	 a
622
617
595
97
	 «
	 e
	 t


a out of range of analysis
   H.
Benzo(a)Pyrene (BaP) in Particulate Organic Solubles
         This polynuclear organic compound was measured as an indicator of total
   PNA in the soluble fraction of particulate.  Data are given in Table 42 for
   the 6V-71 bus engine,  expressed in several types of units.  Emissions of
   BaP in crankcase particulate were computed on a mass rate (ug/hr)  basis using
   the total sample collected as representative of a 39-minute period.   BaP mass
   rate for this engine was higher than for either of the other two engines
   tested.   Concentration of BaP in crankcase particulate was about the same
   for the two 6V-71 engines, but the bus engine's higher particulate emissions
   made its BaP mass emissions higher.  Note that the BaP analysis method used
   probably resulted in high values as compared to newer methods.
              TABLE 42.   BaP CONTENT OF CRANKCASE PARTICULATE, 6V-71
                 BUS ENGINE, 13-MODE COMPOSITE (STANDARD SPEEDS)3
Filter
No.
AR-6025
AR-2027
Averages
Total
BaP, ug
40.3
. 60.2
50.2
Height % BaP in
Particulate
0.0026
0.0038
0.0032
Height % BaP
in Solubles
0.0027
0.0039
0.0033
BaP,
yg/hr
61.6
92.0
76.8
        technique used produced high values as compared to newer methods
                                        56

-------
I.    Elemental Composition of Particulate

      Included in this subsection are data determined by combustion analy-
sis at a commercial laboratory and X-ray data by the Institute's Mobile
Energy Division.  These data are given in Table 43, and they describe a pre-
dominantly hydrocarbon material with small amounts of other elements present.
Hydrogen values for solubles (standard speeds) seem anomalously low, leading
to a C/H ratio by weight of 7.09.  The C/H ratio for solubles of particulate
collected at low speeds is very similar to that for the engine's lubricating
oil (6.57 versus 6.42).  Data on particulate sulfur content by the two methods
are in some disagreement.  The X-ray data are considered to be more reliable.
         TABLE 43.  ELEMENTAL COMPOSITION OF CRANKCASE PARTICULATE,
                     6V-71 BUS ENGINE, 13-MODE COMPOSITE
Filter Nos.
A47-6042 & 6043
A47-6044 & 6045
A47-6046 & 6047
Averages
A47-6052 & 605 3a
A4 7-6054 & 605 5a
Averages
Weight % of Particulate by Combustion
C
86.8
83.4
78.7
83.0
82.4
84.4
83.4
H
11.7
12.1
11.3
11.7
12.6
12.8
12.7
N
0.7
1.3
1.3
1.1
0.2
0.1
0.2
S
0.7
2.0
0.9
1.2
1.7
0.6
1.2
E CHNS
99.9
98.8
92.2
97.0
96.9
97.9
97.4
Filter
No.
FH4 7-6044
FH47-6046
FH47-6048
Averages
FH47-6054*
FH47-60563
Averages
Weight % of Particulate by X-ray
P
0.06
0.07
0.09
0.07
0.08
0.07
0.08
S
0.25
0.26
0.27
0.26
0.21
0.22
0.22
Cl
	 b
	 b
0.01
	 b
0.01
0.03
0.02
Zn
0.44
0.22
	 b
0.22
0.28
0.41
0.34
     a samples taken at  low engine speeds  (900 and  1500  rpm)  - others
       at  standard  speeds
       none detected
                                       57

-------
J.    Sulfate (S04=) in Particulate

      Sulfate in cranJccase particulate was measured for the 6V-71 bus engine
with results as shown in .Table 44.  These data show reasonable consistency,
and it can be computed that the sulfur in the sulfate constituted about 18
percent of the X-ray sulfur at standaid speeds and about 27 percent of the
X-ray sulfur at low speeds.  These figures are somewhat higher than for the
other two engines.
             TABLE 44.  SULFATE IN CRANKCASE PARTICULATE, 6V-71
                        BUS ENGINE, 13-MODE COMPOSITE
Filter
NO.
FH47-6043
FH47-6045
FH47-6047
Averages
FH47-6053a
FH4 7-605 5a
Averages
ug so4=
on Filter
'4.61
3.82
3.01
3.81
6.06
4.42
5.24
S in S04= as
Weight % Particulate
0.054
0.040
0.044
0.046
0.069
0.050
0.060
S04=,
mg/hr
2.28
1.92
1.45
1.88
2.78
2.00
2.39
       a samples taken at low engine speeds  (900 and  1500  rpm)  -
         others at standard speeds
                                      58

-------
      VI.  ANALYSIS FOR NITROSAMINES IN CRANKCASE GASES AND PARTICULATE
      Part of the original test plan for this Task Order was sampling and
analysis for nitrosamines in crankcase gases and particulate.  This portion
of the project has not yet been discussed, because the nitrosamine work was
expanded beyond that called for in the original test plan.  Due to this ex-
pansion and the importance of the nitrosamine studies to the project, all
activity dealing with nitrosamines is collected in this section.

A.    Summary of Project Efforts on Nitrosamines

      The test plan for this project (Table 1) called for ten samples to be
taken from each engine for nitrosamine analysis, five from crankcase gases
and five from crankcase particulate.  These samples were collected from the
Detroit Diesel 6V-71 development engine, the Cummins NTC-350, and the Detroit
Diesel 6V-71 bus engine.  Collection and analysis systems will be described
in Subsection VI.B.

      The first additional nitrosamine activity was authorized under Task
Change No. 1, dated October 4, 1976.  Its purpose was to attempt confirmation
of N-dimethylnitrosamine by GC-MS (Gas Chromatograph - Mass Spectrometer) in
samples already obtained and analyzed as part of the original Task Scope of
Work.  Some samples were also taken for analysis by EPA and one of its other
contractors. Research Triangle Institute.

      Task Change No. 2, dated March 23, 1977, was added to collect one large
sample and a number of smaller ones from a fourth engine and to analyze the
samples by a variety of techniques.   The engine employed was a Mack ETAY(B)
673-A which was being tested under another EPA Contract (No. 68-03-2417).
Analyses were conducted by the Institute's Department of Environmental Sciences,
EPA's Research Triangle Park laboratories, Research Triangle Institute, and
EPA's National Enforcement Investigations Center in Denver.

B.    Initial Sampling Systems

      The system shown schematically in Figure 21 was used to collect gas
samples for nitrosamine analysis, the collection medium being 25 ml of water
in each impinger with a small amount of base  (KOH) or a greater amount of
buffer  (Na2C03) added to prevent the solution from becoming acidic.  Like
the two other impinger systems, it was designed to operate with a maximum
flowrate of 4 liters per minute.

      As reported by the Institute's Department of Environmental Sciences,
the specific recovery of DMNA was determined by spiking a blank KOH solution
with DMNA and subjecting it to the same procedures as the actual impinger
sample.  Blank glass fiber filters were similarly spiked and used for the
recovery experiments.  Quantitative determination of the DMNA was made by
injecting 20 \ii of the sample extract into a gas chromatograph operated
under the following conditions:
                                      59

-------
               EXCESS

               6ASESF,LTERS
                                                         MOLECULAR
                                                         SIEVE  (FILTER)
                       0.001  N  KOH
                          (Na2C03)
   FIGURE 21,   WINGER SYSTEM FOR COLLECTION  OF CRANKCASE  GAS
                  SAMPLES FOR NITROSAMINE  ANALYSIS
      column:       0.9 m (i) x 4 mm (i.d.) packed with Chrome-sorb  101
                   (80/100 mesh)

      temperature:  inlet - 220°C
                   column - 1408C
                   transfer line - 210°C

      detector:     Hall electrolytic conductivity, operated in the pyro-
                   lytic mode at 500°C, specific for nitrogen as ammonia.

      Analysis for DMNA in crankcase particulate was conducted on samples
collected by the hi-vol sampler on large glass fiber filters.  The  method-
ology used was essentially the same as that for gas samples, except that the
samples were extracted from the filters in 0.01 normal KOH and distilled
over into water.  This water solution was then treated in the same  way as
the aqueous samples obtained in the impingers.  More details of the analyt-
ical procedure are provided in the memoranda attached as pages C-2  and C-3
of Appendix C.
                                     60

-------
C.    Analysis of Filter and Impinger Samples taken Under Initial Scope of
      Work

      Data resulting from analysis of particulate samples taken during tests
on the 6V-71 development engine, the NTC-350 engine, and the 6V-71 bus engine
showed no evidence of nitrosamines.  Likewise, no indication of nitrosamines
was found in any of the impinger-collected engine intake air "blank" or "back-
ground" samples taken during these tests.  A tabulation of the filters and
impingers involved is included as Table C-l of Appendix C, page C-4.  Impin-
ger-collected samples of crankcase gases, however, did show indications of
N-dimethylnitrosamine using the Hall detector.  It must be noted that these
indications were not confirmed by GC-MS structural evidence.  Subject to
this qualification, data from impinger-collected gas samples are presented
in Table 45 for the three engines mentioned above.
              TABLE 45.  CRANKCASE GAS DMNA RESULTS FOR 13-MODE
           COMPOSITE TESTS BY ENGINE, IMPINGER SAMPLES/HALL DETECTOR

Engine
6V-71 Development





NTC-350






6V-71 Bus



Date
5/5/76
5/5/76
5/5/76
5/5/76
5/5/76
5/14/76
7/13/76
7/13/76
7/13/76
7/14/76
7/14/76
7/16/76
7/16/76
9/1/76
9/2/76
9/2/76
Im
Run No.
1
2
3
4
5
1
1
2
3
4
5
1
2
1
2
3
pinger
Solution pH
__b
— b
.Jo
— b
B^J3
B*J3
6.3
6.4
7.2
7.0
7.2
11. lc
11.0C
__b,c
__b,c
_Ja,c

DMNA3, yg/m3
0.58
0.99
0.27
0.54
0.58
0.45
0.95
0.71
0.24
0.49
0.42
0.56
0.39
0.10
0.085
0.085
     at p = 101.3 kPa and T = 21°C
     not determined
     contained Na2CO-j buffer
      Although a base (0.001 N KOH) was used in all the impingers, it was de-
termined while running the NTC-350 that some of the samples were turning slight-
ly acidic toward the end of the sampling period.  In order to avoid this situa-
tion, a buffer (Na2C03)  was added to subsequent impinger solutions to keep
them basic.  Comparing DMNA data for the first five samples from the NTC-350
to those for the last two, it appears that artifact formation in the unbuf-
fered solutions was not a problem.
                                      61

-------
      Concentrations of DMNA measured in gas samples from the first two en-
gines are quite similar, on '.-he average (0.57 yg/m3 for the 6V-71 development
engine versus 0.54 yg/m3 for the NTC-350).  Concentrations measured in gas
samples from the 6V-71 bus^engine are considerably lower, averaging 0.09 yg/m3.
No reasons are apparent which could account for this difference.  It has been
noted that the 6V-71 bus engine was operated on a No. 1 fuel, while the other
engines were operated on a No. 2 fuel; but there is no evidence thus far to
suggest that fuel type has any effect on measured concentrations of nitrosamines.

      Using the data in Table 45 as basis, average volumetric concentrations
and mass rates of DMNA measured in crankcase gases from the three engines are
as follows:  0.17 ppb and 1.4 yg/hr for the 6V-71 development engine; 0.15 ppb
and 5.5 yg/hr for the NTC-350; and 0.027 ppb and 0.35 yg/hr for the 6V-71 bus
engine.  These data must be considered tentative until confirmed by reprodu-
cible structural evidence,! which to date has not been obtained.

D.    First Attempt at GC-MS Confirmation of DMNA

      Prior to beginning work on this attempt at confirmation of the compound
thought to be DMNA, the plan was to run the individual imp nger-collected
samples by GC-MS (12 samples remained from the earlier work).  A problem
arose, however, in that the highest liquid phase concentration which could
be achieved with any of the individual samples was about 0.4 ng/y&.  Sensi-
tivity level of the GC-MS system was about 1.0 ng/y& at best, with higher
concentrations needed for better accuracy.

      To provide a more concentrated sample for the GC-MS, it was decided to
combine the existing samples and attempt to drive off the maximum amount of
solvent at low temperature.  Had no DMNA been lost in this procedure, its
liquid phase concentration would have been about 2.2 ng/y£.  Moderate losses
(about 25 percent)  were expected, but subsequent re-analysis by the Hall de-
tector showed DMNA loss during concentration to be about 50 percent.  There
was no indication of DMNA in the combined sample when run on the GC-MS, so
this test phase was inconclusive.

      Samples were also taken during this same time period for dimethylamine
analysis by EPA-RTF.  Dimethylamine is considered to be a possible precursor
of DMNA.  Results of the evaluations indicated no dimethylamine in any of the
samples taken.  The same impinger-collection technique was used for these
evaluations as was used for DMNA sample acquisition, except that the col-
lection reagent was acidic  (0.01 N H2S04).

E.    Collection and Analysis of Crankcase Gas Samples from Mack ETAY(B)
      673-A Engine

      Following the work described above on attempts to confirm the presence
of DMNA by GC-MS, it was decided that another series of collection runs would
be made in order to gather a larger sample.  The previously-used engines were
no longer on test stands or otherwise readily available, so the Mack ETAY(B)
673-A engine being used on another EPA contract was selected as a test engine.
The experimental plan for this phase of the project was made up of these items:
                                       62

-------
      - determine whether or not the Mack engine produces measurable amounts
        of DMNA in its crankcase gases,  as analyzed by the impinger-collection/
        Hall detector method

      - if the engine produces enough DMNA as characterized above, take a
        large sample of crankcase gases  suitable for analysis by both Hall
        detector and GC-MS

      - take small impinger samples, concurrent with large sample, for nitro-
        samine analysis by EPA-NEIC

      - take small impinger samples (in  acidic solution), concurrent with
        large sample, for amine (N-nitrosamine precursor) analysis by EPA-RTP

      - take small Tenax-GC* trap samples, concurrent with large sample, for
        nitrosamine analysis by Research Triangle Institute

      In order to simplify the sample acquisition process, it was decided to
operate the engine at a steady-state condition of 1900 rpm and 25 percent
load to gather the initial sample.  This procedure permitted sampling at a
constant rate and minimized engine fuel  consumption.  Fuel used was (code)
EM-239-F, "National Average" No. 2 diesel fuel, the same batch as that used
in the NTC-350 and 6V-71 development engines.  Data on the initial sample
(analysis by Hall detector) are summarized below:

      sample rate (nominal)                 4 liters/min  (0.004 m3/min)
      gas sampled                           0.108 std. m3 (101.3 kPa & 21°C)
      sample time                           30.0 minutes
      total DMNA recovered                  16.2 ng
      crankcase gas DMNA concentration      1.14 Ug/std. m3  (101.3 kPa & 21°C)

The DMNA standard analyzed at the same time as this sample shows only 13.2
percent recovery, whereas normal recoveries were 40 to 60 percent.  This low
apparent recovery contributed to the high (1.14 yg/m3) calculated concentration
of DMNA.

      Based on these data, it was decided to collect 5000 ng of DMNA by using
both a higher sample flowrate (with large impingers) and a longer sampling
time.  Using a projected sampling rate of 0.021 ra3/min  (0.75 ft3/min), time
to acquire approximately 5000 ng of DMNA was calculated to be 26.5 hours.

      The large-sample collection activity was performed on the Mack ETAY(B)
673-A engine, operating at 1900 rpm and 25 percent load.  Fuel coded EM-239-F
was used.  Sampling was conducted over a period of three days, with impingers
kept covered in an ice bath between (as well as during) sampling periods.
The sampling hardware used was identical to that shown schematically in Fig-
ure 21, except that larger impingers were used in place of the midget im-
pingers.  Data on the large sample are summarized below, with analysis by
Hall detector:
* registered trademark of Enka N.V., The Netherlands
                                       63

-------
      sample rate  (nominal)                 0.021 m3/min  (0.75 ft3/min)
      gas sampled                           41.3 std. m3  (101.3 kPa s. 21°C)
      sample time                           30.0 hr
      total DMNA recovered                  2,380. ng
      crankcase gas DMNA concentration      0.094 pg/std. m3  (101.3 kPa & 21°C)

Although a substantial amount of DMNA was recovered from the  large sample,
it was only about half the target amount.  Several factors may have combined
to produce this result, including losses of DMNA during the three-day sam-
pling period, recovery variation in the extraction and concentration steps,
and variability in the engine's production of DMNA.

      Having determined by the same methods used to analyze earlier samples
that the compound identified as DMNA was present in sufficient quantity, at-
tention was turned to GC-MS analysis of the large sample.  The complete re-
port filed by the Institute's Department of Environmental Sciences is inclu-
ded in Appendix C  (pages C-5 through C-19) for reference.  Summarizing their
findings, standards containing both DMNA and N-diethylnitmsamine (DENA) were
run by full-spectrum scanning and in the mass fragmentograt dc mode to deter-
mine GC-MS retention and response characteristics.  The full-spectrum chroma-
togram of standards is shown in Figure 2a of the above-mentioned report, page
C-ll of Appendix C.  A single-ion scan at 74 amu is shown in  Figure 2b  (page
C-12), indicating better resolution of the molecular ion of DMNA by computer
regeneration.  The mass fragment ion chromatograms of standards are shown in
Figure 3 of the same report, page C-13 of Appendix C.

      Chromatograms .of diesel crankcase gas concentrate are shown in Figures
4 and 5 of the report by the Department of Environmental Sciences, pages C-14
through C-18 and page C-19 of Appendix C, respectively.  The  full-spectrum
chromatogram in Figure 4a and the single-ion scans in Figures 4b through 4e
are complicated by the presence of interfering compounds.  This complication
meant that it was not possible to extract "clean" spectra, but rather good
indications of the presence of DMNA and DENA were obtained (at 30 and 74 amu
for DMNA and at 30 and 102 amu for DENA).  The mass fragmentograms of crank-
case gas samples shown in Figure 5 can be compared to fragmentograms for
standards in Figure 3 of the GC-MS report.  The sample fragmentograms show
strong evidence of the presence of DMNA and little visible evidence of the
presence of DENA.

      All the data and explanations thus far have dealt with  impinger-collected
gas samples from a diesel engine crankcase.  The indications  seem to be that
nitrosamine compounds are present in such samples, whereas they are not pre-
sent in impinger-collected samples of engine intake air or engine exhaust
or in samples of particulate from diesel engine crankcases or exhausts.  Test
results to this point cannot, however, rule out the possibility that nitro-
samines are formed as artifacts in the collection medium either during or
after collection.

      During each day of collection of the large sample, several smaller sam-
ples were acquired for independent analysis.  The first such  samples to be
discussed are those sent to EPA's National Enforcement Investigations Center
(NEIC) in Denver for nitrosamine analysis by gas chromatograph-thermal energy
analyzer.  Procedures for sample handling and analysis are detailed in EPA-
                                      64

-------
NEIC's report to the Project Officer and a copy of their nitrosamine analyt-
ical method, included as pages C-20 and C-21. respectively, of Appendix r.
The samples were collected using the apparatus shown in Figure 21.  Prior to
shipment, the aqueous samples were treated in the following steps:

      A.  Quantitatively transfer the impinge r solution to a 125 ml separatory
          funnel.
      B.  Add 13 g Nad per 50 ml solution.

      C.  Extract with dichlorome thane (3 x 10 ml), combine the extracts.

      D.  Place extract in light-tight glass bottles with Teflon* cap liners
          and pack carefully.

      The results of the EPA-NEIC analysis have been combined with pertinent
sampling data to yield the data shown in Table 46.  These data confirm the
presence of several nitrosamine compounds in the impinger-collected samples,
as well as two other nitrogen-containing compounds.  The amounts of DMNA
measured in these samples (average 0.28 yg/m ) fall between those found by
the Department of Environmental Sciences in the original small sample from
the ETAY(B)673-A engine (1.14 ug/m3) and in the large, sample subsequently
analyzed by GC-MS (0.094 ug/m3) •  The reasons for a lack of agreement are
not known, but they may involve time lag between sampling and analysis as
well as different collection/retention efficiencies for small (short-term)
samples as compared to the large sample.        '
              TABLE 46.  SUMMARY OF EPA-NEIC NITROSAMINE ANALYSIS
                 OF MACK ETAY(B)673-A CRANKCASE GAS SAMPLES
Date
1/4/77
1/5/77
1/6/77
Gas Sampled,
std.a m3
0.184'
0.190
0.187
Detection Limits
yg/std. m3a by Compound15
DMNA
0.24
0.32
0.29
0.027
DENA
0.14
0.12
0.12
0.027
DBNA
0.092
0.11
0.10
0.059
NPIP
0.060

-------
      Another type of snail sample was acquired during each day of large-sample
collection, this one in an.acidic impinger solution for nitrosamine precursor
analysis.  The particular constituent for which there was some reason to ana-
lyze was dimethylamine, since it was considered to be a possible precursor of
the N-dimethylnitrosamine already tentatively identified in diesel crankcase
gases.  The samples sent to EPA's Research Triangle Park laboratories were
analyzed by ion chro"
-------
volume placed in the bag initially and of total gas removed from the bag
yielded effective dilution ratio and amount of crankcase gas.  Nominal 1
liter and 5 liter (0.001 m^ and 0.005 m^) samples of crankcase gas were taken
and passed through traps.

      Results of analysis on these dilute samples by RTI are given in its re-
port to EPA, included as pages C-31 through O43 of Appendix C.  The single-
ion mass chromatograms in Figures 2 and  3 of this report showed no DMNA, to
detection limits of ^166 ppt and 'V-IOO ppt, respectively.  Translated into
units consistent with other studies reported thus far, these upper limits on
possible undetected DMNA were 0.51'and 0.31 yg/m3, respectively.  Note that
all the DMNA results on impinger-collected samples from the Mack engine,
from EPA-NEIC and from the Institute's Department of En/ironmental Sciences,
are lower than these detection limits with the exception of the initial small
sample by SwRI.

      Analysis of another trap-collected sample  (No. 6) was conducted by ob-
taining full mass spectra, producing identification of numerous organics as
listed in Table 2 of the RTZ report  (page C-41).  The compounds were not
quantitated, but the major components as observed by the total ion current
profile were phenol and N.N-dimethylacetamide.  None of the compounds RTI
considers N-nitrosamine precursors  (dimethylamine, aiethylamine, morpholine,
etc.) were found, nor were any other amines detected.  Detection limits for
nitrosamines in this sample were about the same as those already discussed
for sample No. 4.

      Although not included in the RTI report due to time requirements, sam-
ple No. 5 was also subjected to a full scan.  The (different) column used
for this analysis gave better resolution of the sample as a whole, but it
was not expected to yield good nitrosamine results.  Data on this sample are
given on pages C-44 through C-46.  The N,N-dimethylacetamide was also a major
constituent of this sample, as determined by the total ion current profile .

      Finding the N,N-dimethylacetamide  in the samples was not expected, and
thus  far no work has been  done to determine its  origin or possible  further
reactions.  According  to RTI, this  compound has  not been  found in other emis-
sions  from internal combustion engines,  but was  found at  trace levels in sam-
ples  from  a  forest  fire.

       Reservations  expressed by RTI  about their  results dealt mainly with  the
sampling procedure.  One concern was the possibility  that any  DMNA  present in
the crankcase gas might have condensed on the  walls of  the sample bag.  Ano-
ther  concern was that  the  sample  lines were not  heated, RTI  having  found
earlier that line temperatures of about  50°C were required to  minimize  DMNA
 losses.  Finally, the  lack of a specific recovery experiment on  an  authentic
standard created a  certain lack of  confidence  in the overall results.
                                       67

-------
        VII.  DISCUSSION OF CRANKCASE EMISSION CONTROL POSSIBILITIES
      A limited investigation has been made into possibilities for recircu-
lation or other controls on diesel crankcase emissions, as called for in the
Task Scope of Work.  Due to the efforts required to complete other phases of
the program, only a small effort could be devoted to this subject.  No testing
of crankcase emission control systems was performed.

      In order to reduce!oil emission from the crankcase, most engines employ
baffles and low flow velocities in the breather passages.  As evidenced by
the results of this study, however, the oil control systems employed on the
test engines were not adequate.  It may be that most of the oil or oil-derived
particles actually emitted were too small to be apprehended by passive mechan-
ical control systems (baffles, etc.), and such systems could probably have at
best a minor effect on gaseous crankcase emissions.

      At least one retrofit system is currently being marketed (by CECA En-
gineering, Division of Kar Products) for crankcase emission control of sta-
tionary and marine diesels, but it does not appear to be  sable on turbo-
charged or blown engines in its present state of development.  The system is
called "C.E.C.A." or "Clean Engine, Clear Air", and it operates by filtering
crankcase gases and recirculating them to the inlet manifold.  Such a system
might be adapted to turbocharged or blown engines using a pump downstream of
the filter and perhaps even a final charcoal filter element to trap hydrocarbon
material.  The information above was published on this system in Diesel and
Gas Turbine Progress, April 1976.^7'

      Some additional information on crankcase emission control systems used
by five manufacturers of marine diesel engines was submitted for this report
by the EPA Task Officer.  This information on marine diesels, dated December
7, 1976, is summarized as follows:

      Detroit Diesel-Allison Division, General Motors

            The crankcase breather tube is routed to the air cleaner  (or
      silencer) intake on all marine engines, turbocharged and naturally
      aspirated.

      Electro-Motive Division, General Motors

            The breather tube is connected to the exhaust system.

      Cummins

            Naturally aspirated - V903; internal breather connects to
                                  intake port
                                - all others; breather on valve cover
                                  connects to intake port

            Turbocharged - VT903; breather tube connected to the air
                           inlet
                         - all others; no control, vented to engine
                           compartment
                                       68

-------
      Deutz

            Individual contacted was not directly involved with controls/
      but said that his opinion was that there were no controls (the
      breather tube is routed outside the engine compartment).  Deutz
      is working on some systems in research.

      Alco

            No controls.  The breather tube is routed out of the engine
      compartment.

      Another manufacturer's viewpoint on crankcase emission controls for
diesels is presented in the letter and report from Caterpillar Tractor Com-
pany presented as Appendix D.  The Caterpillar viewpoint, in summary, is that
diesel crankcase emissions are an insignificant air pollution source.

      In this report section, we have primarily reported the work and opinions
of others, since no testing of diesel crankcase emission control systems was
conducted.  It appears, however, that crankcase controls for naturally aspi-
rated engines are feasible with available technology.  Indeed, some such sys-
tems are in use; and data could be gathered on their performance in the field
to determine whether they have positive or negative effects on engine perfor-
mance and durability.

      Adaptation of a recirculation system to a turbocharged engine is some-
what more complicated than adaptation of a system to a naturally aspirated
engine, if it is considered necessary to add the blowby gases to the inlet
tract downstream of the turbocharger.  Such a system would require a pump to
achieve flow in the desired direction.  If the blowby gases were routed to
the air cleaner, either directly or through some extra filter or trap system,
no such pumping requirements would exist.  Other alternatives might also be
considered for some applications, such as directing crankcase gases into the
engine's exhaust or simply cleansing them with filters plus hydrocarbon traps.
Even in the case of turbocharged engines, therefore, crankcase emission con-
trol does not appear to be a difficult technical problem.
                                      69

-------
               VIII.  ANALYSIS OF TEST '-.NGINE LUBRICATING OILS
      In addition to the demonstration and characterization data already pre-
sented, it was requested that information be obtained, on the engine oils.
These tests and measurements included the following:

      infrared spectral analysis
      quantitative mcLals by X-ray fluorescence
      boiling range by ASTM D-2887
      API gravity by ASTM D287
      viscosity in centistokes at 1008F and 212°F by ASTM D445
      pentane and benzene insolubles by ASTM D893
      total acid number by ASTM D664
      total base number by ASTM 02896
      sulfated ash by ASTM D874
      fuel dilution (of oil)
      carbon, hydrogen, and nitrogen fractions by combustion.
      benzo(a)pyrene

Boiling ranges have already been reported in conjunction with corresponding
data on particulate solubles, so they will not be repeated here.

      Copies of the infrared spectra of new and used oils for all four engines
are given in Appendix E, and their interpretations are given here.  Spectra
900 and 901 (pages E-2 and E-3) are for the 6V-71 development engine, and
they show that the used oil is essentially unchanged from new oil.  Oxidation
of the used oil is negligible, and no water or ethylene glycol is present.
Some soot is present in the used oil, and both contain a poly-isobutylene
type viscosity index improver.

      Spectra 902 and 903 (pages E-4 and E-5) show differences in new and used
oils for the NTC-350 engine.  The oil's additive package is partially consumed
in the used oil, and there are slight traces of oxidation and nitration.  The
additive package mentioned includes sulfonates (detergents and dispersants)
and zinc (ZDDP) anti-oxidants and anti-scuffing agents.  Soot is present in
the used oil, and its viscosity index improver (poly methyl methacrylate type)
is intact.   No ethylene glycol or water is present in either new or used oil.

      Spectra 930 and 931 (pages E-6 and E-7) are for the 6V-71 bus engine's
oils, and there are differences in new and used oils.  Soot appears in the
used oil, and the additive package in the used oil is partially consumed.
The used oil's viscosity index improver is intact (same type as in NTC-350
oils), and no water or ethylene glycol is present in either sample.  In gen-
eral, both used oils from the characterization engines show changes in pro-
perties which are normal for oils after considerable service.

      Spectra 1166 and 1167  (pages E-8 and E-9) are for the Mack ETAY(B)673-A
oils.  Very little soot appears in the used oil, and no water or ethylene
glycol are present in either new or used oil.  Part of the zinc additive pack-
age in the used oil has decomposed, as it is designed to do with use.  A small
amount of oxidation has taken place in the used oil, and its  (poly methyl
methacrylate type) viscosity index improver is intact.
                                       70

-------
      The remaining data on lubricating oils can be summarized in tabular
form, and Table 47 provides this information.  Most of these data show lit-
tle difference between new and used oils, and the oils from the four engines
are also quite similar to one another in most respects.  Differences which
are noticeable, however, show up in sulfur concentrations between engines.
Viscosity at 100°F is also considerably lower for the ETAY(B)673-A oil than
for the others.  Used oils seem more viscous excepting that from the 6V-71
development engine, and acid numbers of used oils were higher except for the
6V-71 bus oil.  Small amounts of fuel dilution were observed for all the used
oils except that from the 6V-71 development engine.  BaP values were quite scat-
tered, and are probably higher than those which would have been determined by
newer analytical techniques.
                                       71

-------
                                        TABLE 47.   DATA ON PROPERTIES OF HEW AND USED LUBRICATING OILS  FROM THE FOUR TEST ENGINES
10
Engine
Oil Description
Oil History
X-ray data. wt. *
Phosphorus (P)
Sulfur (S)
Chlorine (CD
Iron 
-------
                                 REFERENCES
1.  Hare, Charles T., "Methodology for Determining Fuel Effects on Diesel
    Participate Emissions."  Final Report No. EPA-650/2-75-056 to the U.S.
    Environmental Protection Agency under Contract No. 68-02-1230, March 1975.

2.  Chevalier, C. E., "Hydrocarbon Content of the Blowby Gases of Severely
    Worn Automotive Diesel Engines."  Report No. 15 prepared under Contract
    No. PHS 86-64-76 with the U. S. Department of Health, Education, and Wel-
    fare, October 1964.

3.  Federal Register, Vol. 37, No. 221, Part II, Subparts H and J (November
    15, 1972).

4.  Stump, Fred, "Oxygenated Compounds in Automobile Exhaust - Gas Chromato-
    graphic Procedure."  Unpublished Procedure developed by the U.S. Environ-
    mental Protection Agency.

5.  Black, F. M.; High, L. E.; and Sigsby, J. E.; "Methodology for Assignment
    of a Hydrocarbon Photochemical Reactivity Index for Emissions from Mobile
    Sources."  Final Report No. EPA-650/2-75-025 to the U.S. Environmental
    Protection Agency, March 1975.

6.  Tejada, Silvestre, "Determination of Soluble Sulfates."  Unpublished Pro-
    cedure developed by the U.S. Environmental Protection Agency, 1974.

7.  Diesel and Gas Turbine Progress, Vol. XLII, No. 4, April 1976.
                                      73

-------
                 APPENDIX A




TASK ORDER 4 SCOPE OF WORK AND MODIFICATIONS

-------
                              SCOPE OF WORK

TASK I

     Southwest Research Institute  (SwRI)  shall  adapt  the  appropriate
measurement procedures and methodology for sampling and analyzing
gases and particulates in diesel engine blowby  emissions.   The  following
compounds will be analyzed for:

Gaseous form:

     a)  Total hydrocarbons
     b)  Light hydrocarbons - individual  analysis
         of Cl through nC4
     c)  CO and C02
     d)  NOx
     e)  02
     f)  Aldehydes - Total polar aldehydes and  individual  aldehydes
         through C&.  03's will be analyzed as  a group.
     g)  Nitrosamines

Particulate:

     a)  Total particuiate
     b)  Distribution of particulars size.  This shall be  performed
         with an Anderson impactor.  Some of  the impactor  plates shall
         be analyzed b-r a scanning electron microscope to  determine
         the size distribution of the impacted  particules.
     c)  Nitrosamines
     d)  Total organic solubles
     e)  C,H,N, and S (particulate)
     f)  Boiling range of solubles
     g)  Sulfate
     h)  BaP in solubles

     SwRI shall also evaluate diesel blowby in  terms  of potential  emissions
of significant quantities of other compounds  that could represent  environmental
or health hazards.  Should such compounds be  judged a potential emission
product, SwRI shall report said compounds to  the Project Officer.  This
evaluation will be based on contractor knowledge and  experience in addition
to the contractor's analysis of the results of the above tests.  For  example,
if the contractor finds significant levels of BaP, he might report that
significant levels of ather POM comoouncs could be present in  -he  rlowbv.
r'lij is Because 3-ar is only one POM ana .s :r.e  one ;sea cc :.iarac_ariza
:-a PGM content of :r.e stream seing analyzed.   If significant  levels of
2aP vera :o be fcur.c, -hu csuld indicate ji^nificant levels jf :he  ;c.ier
POM'3.
                                  A-2

-------
TASK II

     SwRI shall employ the sampling and analysis methodology adapted  in
Task I to characterize the blowby emissions of two engines.  One of the
engines will be a heavy duty truck type 4-stroke and the other a city
bus type 2-stroke.  SwRI shall acquire the engines from local fleet
sources.  Both engines will be characteristic of the heavy duty engines
of their class that are currently in use.  Both engines shall have
accumulated sufficient hours of use to have reached about the half way
point between being a new engine and one that is in need of a major re-
building.  These hours of use shall be over roughly that type of use
expected of the engine type.

     The engines shall be tested with lubricating oil  that has accumulated
approximately one half of the use that the manufacturer recommends.   The
fuel to be used shall be Number 1 Diesel fuel for the  2-stroke and
Number 2 Diesel fuel for the 4-stroke.  Both fuels shall approximate
the national average composition for fuels of their type.  The test cycle
shall be the 13-mode cycle or some variant thereof which closely resembles
the 13-mode cycles.  Final choice of the 13-mode cycle will be dependent
on sample size and tire cooling problems, ana will be  made by mucal
agreement between SwRI and the Project Officer.

TASK III
     SwRI shall examine the possibilities for blowby  recirculaticn  or
other techniques for the control of diesel blowby emissions. . The results
of this analysis shall be included in the final  report.  This analysis
shall consist of a brief paper study that employs the contractor's  •
knowledge and experience in addition to his contacts  with  technical
people from the engine manufacturers.  If the contractor feels  that this
study is promising, he shall .do some limited engine experimentation such
as installing a PCV system and checking engine emissions.

Reports and Time Requirements

     The technical effort involved in performing this Scope of  Work
should require six months to perform.  During this time, SwRI shall
submit monthly letter reports briefly summarizing the status of the work
and the resource expenditures.  Three weeks after the end of this
Period of Performance SwRI shalL submit to the Task Officer five (5)
       of cne iraf- final raoor-.  Two vee&cs  if:ar r-scaioc  ::" :ai3  raocr-
    Project Ot'r'icar snaii racurn -.ie approve rqoon:  :o Swill for final
    ing ind raprcduc:i:n.  Thrae veeks if tar £v?.I racaivas  ine  r a via wed
   f- final raoorz, owP.I 3r..iil juomc :o :::a T.ISLC  }tfi^ar  30  lonies :c
                                  A-3

-------
CONTRACT TASK SPECIFICATION 'CHANGE REQUEST
NEGOTIATED CONTRACTS BRANCH //-VA
-------
         CONTRACT TASK SPECIFICATION'CHANCE REQUEST
                    NEGOTIATED CONTRACTS  BRANCH
                   CONTRACTS MANAGEMENT DIVISION
                  ENVIRONMENTAL PROTECTION  AGENCY
               NATIONAL ENVIRONMENTAL  RESEARCH  CENTER
                      CINCINNATI, OHIO 15268
                                                                'A LON I MALI NO
                                     CONTRACTOR
                                     Southwest Research  Institute
                                     TASK HO.
                                     TASK CHANCE NO.
                                                               DATE
           "Diesel Crankcase DHNA Emissions"
 ACCOUMIM5 ANO AfflOfllAllOM OAU
 Appropriation  DCN
Contract No.   Aect. No.
                               OC
                 Amt.
SFO
 TASK SPECIFICATIONS OR DESCRIPTION
           As per Scops of Work
 In addition  to the breakdown of costs required in the contract billing instructions,
 vouchers/ invoices must also identify the allocation of costs to each Task Order.
          ESTIMATE OP
                                       GOVERNMENT ESTIMATE
                                                                              ESTIMATE
  Task Level of Effort
          $5,000
210 hours
  DURATION Of TASK
                                    2 months after Initiation
•  TASK COMPLETION DATE
                                       -«nt-Vig affar Inl tlaUnn
i  PROJECT OFFICER -  MerTTll^W.^r^ECTD
                                                  OAu. CODE
                                                               TELEPHONE
                                                                           DA IE
                                            SIGNATURE
                                                                                DATE
  CONTRACTOR
    Southwest  Research Institute
  TITLE
                          S. H. Birgel, Manager
!  SIGNATURE
  Original signed by
    S H. Bireel
  CONTRACTING OFFICER
k
                                            A-5

-------
                             Scope of Work

Task I Engine Selection

Southwest Research Institute (SwRI) shall sample the crankcase blowby
emissions of the Mack ETAY(B) 673(A) diesel engine that is being used in
EPA Contract No. 68-03-2417 to establish whether or not sufficient: N-
dimethylnitrosamine (DMNA) is being emitted to provide a sample capable
of being analyzed by the SwRI gas chromatograph - mass spectrometer (GC-
MS).  The following decision line will then be pursed:

     a.  If insufficient DMNA flow is found - In this case, SwRI will
     perform the same test on other diesel engines that are currently on
     test stands.  If one is found that emits sufficient sample, proceed
     to Task II.  ,If not, then the DDAD 6V-71 sampling methodology
     development engine that was used in Task Order 04 of this contract
     shall be mounted on the test stand, followed by the, effort des-
     cribed in Task II.

     b.  If sufficient DMNA is found, proceed with Task II.

Task II  Sample Acquisition

The engine selected in Task I shall be operated over the appropriate
modes and the following samples drawn:

     1.  Sufficient sample for the SwRI GC-HEC and GC-MS systems to
     analyze.

     2.  Samples drawn through cartridges supplied by ORD/MSERC, to be
     returned to them for analysis.

     3.  Samples drawn in SwRI impengers, extracted and sent to EPA/NEIC,
     Denver.

Task III  Sample Analysis

SwRI shall analyze the samples drawn in Task II, by the GC-HEC and GC-MS
techniques.
                                    A-6

-------
                      APPENDIX B




GASEOUS EXHAUST EMISSIONS DATA ON THREE DIESEL ENGINES
                          8-'

-------
               TABLE B-l.  13-MODE FEDERAL DIESEL EXHAUST EMISSION CYCLE
                            6V-71 DEVELOPMENT ENGINE, TEST 1
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine
Speed
rpm
440
1260
1260
1260
1260
1260
440
2100
2100
2100
2100
2100
440
Torque
Ib-ft
3.5
10.5
134.8
271.4
406.2
542.7
5.3
479.7
358.9
239.8
119.0
8.8
3.5
Power
bhp
0.3
2.5
32.3
65.1
97.4
130.2
0.4
191.8
143.5
95.9
47.6
3.5
0.3
Fuel
Flow
Ib/min
0.04
0.13
0.27
0.45
0.63
0.92
0.05
1.33
1.04
0.77
0.54
0.31
0.05
Air
Flow
Ib/min
12.19
24.54
29.08
30.09
30.19
30.83
15.54
49.80
53.65
53.21
51. 71
47.87
9.93
Exhaust
Flow
Ib/min
12.23
24.67
29.35
30.54
30.82
31.76
15.59
51.14
54.70
53.99
51.80
48.18
9.98
Fuel
-Air
Ratio
0.003
0.005
0.009
0.015
0.021
0.030
0.003
0.027
0.019
0.015
0.010
0.007
0.005
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
HC
ppm
97
158
81
86
108
104
98
136
94
72
70
118
137
CO
ppm
110
218
151
108
155
4148
124
652
92
50
63
127
96
co2
0.81
1.07
2.06
3.31
4.75
6.52
0.82
6.39
5.00
3.71
2.56
1.51
0.81
NO**
ppm
96
54
175
337
669
879
91
992
627
361
163
59
88
Weighted
bhp
0.02
0.20
2.59
5.21
7.80
10.42
0.03
15.34
11.48
7.67
3.81
0.28
0.02
BSHC
g/hp hr


0.97
0.53
0.45
0.33

0.48
0.47
0.54
1.01


BSCO*
g/hp hr


3.62
1.33
1.29
26.61

4.57
0.92
0.73
1.79


BSCO2
g/hp hr


773
641
621
657

704
787
863
1151


BSN02**
g/hp hr


6.84
6.82
9.15
9.26

11.42
10.33
8.77
7.66


Hum.
gr/lb
72.6
72.6
72.6
72.6
72.6
72.6
76.0
76.0
76.0
76.0
76.0
76.0
74.3
                     Cycle Composite
                                BSHC
BSHC
BSCO*
BSN02**
BSNO-
BSCO-
    0.720 gram/bhp hr
    6.604 gram/bhp hr
    9.800 gram/bhp hr
=  10.521 gram/bhp hr
= 801.    gram/bhp hr
                                       BSFC    =   0.483 Ib/bhp hr
 * converted to wet basis
** converted to wet basis,  corrected to 75 grains of water per Ib of dry air
   and corrected to 85°F inlet temperature per Federal Register para. 85.974-18
                                           B-2

-------
               TABLE  B-2.   13-MODE  FEDERAL  DIESEL EXHAUST EMISSION CYCLE
                            6V-71 DEVELOPMENT ENGINE, TEST  1
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine
Speed
rpm
440
1260
1260
1260
1260
1260
400
2100
2100
2100
2100
2100
440
Torque
Ib-ft
1.8
10.5
134.8
271.4
406.2
542.7
1.8
479.7
358.9
239.8
119.0
8.8
1.8
Power
bhp
0.1
2.5
32.3
65.1
97.4
130.2
0.1
191.8
143.5
95.9
47.6
3.5
0.1
Fuel
Flow
Ib/min
0.04
0.13
0.28
0.45
0.64
0.92
0.04
1.33
1.03
0.78
0.53
0.32
0.05
Air
Flow
Ib/min
13.07
28.06
29.91
30.01
30.65
32.68
14.71
51.62
51.79
54.46
53.14
50.07
11.58
Exhaust
Flow
Ib/min
13.12
28.19
30.19
30.46
31.29
33.60
14.75
52.96
52.82
55.24
53.67
50.39
11.63
Fuel
-Air
Ratio
0.003
0.005
0.009
0.015
0.021
0.028
0.003
0.026
0.020
0.014
0.010
0.006
0.004
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
HC
ppm
81
122
80
74
96
106
94
136
96
75
72
94
117
CO
ppm
89
173
119
62
131
3839
120
627
99
62
83
132
109
"'
0.82
1.08
2.14
3.43
4.81
6.56
0.82
6.34
4.93
3.73
2.55
1.53
0.31
NO**
ppm
101
55
188
364
711
958
102
1047
671
374
174
64
100
Weighted
bhp
0.01
0.20
2.59
5.21
7.80
10.42
0.01
15.34
11.48
7.67
3.81
0.28
0.01
BSHC
g/hp hr


0.99
0.46
0.41
0.36

0.50
0.47
0.57
1.07


BSCO*
g/hp hr


2.91
0.76
1.10
26.06

4.55
0.96
0.94
2.45


BSC02
g/hp hr


826
663
638
699

723
749
887
1187


BSNO2**
g/hp hr


7.58
7.36
9.86
10.68

12.49
10.68
9.32
8.48


Hum.
gr/lb
79.3
79.3
79.3
79.3
79.3
79.3
77.6
77.6
77.6
77.6
77.6
77.6
77.7
                     Cycle Composite
BSHC
BSCO*
BSNO **
     **
                                BSHC + BSN0
                                       BSC02
                                       BSFC
  0.699 gram/bhp hr
  6.489 gram/bhp hr
 10.677 gram/bhp hr
 11.375 gram/bhp hr
822.     gram/bhp hr
  0.482 Ib/bhp hr
 * converted to wet basis
** converted to wet basis,  corrected to 75 grains of water per Ib of dry air
   and corrected to 85°F inlet temperature per Federal Register para. 85.974-18
                                           B-3

-------
              TABLE B-3.  13-MODE FEDERAL DIESEL EXHAUST EMISSION CYCLE
               OF TRUCK WITH CUMMINS NTC-350 ENGINE, TEST 1 (3 min/mode)
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine
Speed
rpm
600
1500
1500
1500
1500
1500
600
2100
2100
2100
2100
2100
600
Torque
Ib-fu
0.0
'20.3
252.1
507.7
759.8
1015.4
0.0
885.3
662.8
442.7
220.1
17.5
0.0
Power
bhp
0.0
5.8
72.0
145.0
217.0
290.0
0.0
354.0
265.0
177.0
88.0
7.0
0.0
Fuel
Flow
Ib/min
0.07
0.24
0.53
0.92
1.27
1.82
0.07
2.28
1.75
1.23
0.72
0.52
0.07
Air
Flow
Ib/min
7.89
18.09
21.43
23.34
25.53
32.54
8.29
52.61
44.24
36.93
28.44
25.10
8.29
Exhaust
Flow
Ib/min
7.96
18.33
21.97
24.25
26.79
34.36
8.36
54.89
45.99
38.16
29.16
25.63
8.35
Fuel
-Air
Ratio
0.009
0.013
0.025
0.039
0.050
0.056
0.008
0.043
0.040
0.033
0.025
0.021
0.008
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
HC
ppm
134
86
67
48
39
36
124
37
27
22
14
22
72
CO*
ppm
189
186
234
648
1761
2466
140
1099
754
448
289
271
188
C02
%
1.77
2.84
5.41
8.68
10.89
12.30
1.69
9.55
8.70
7.28
5.43
4.36
1.70
NO**
ppm
145
184
430
1113
1770
2045
'133
1419
1116
686
392
278
126
Weighted
bhp
0.00
0.46
5.76
11.60
17.36
23.20
0.00
28.32
21.20
14.16
7.04
0.56
0.00
BSHC
g/hp hr


0.27
0.11
0.06
0.06

0.08
0.06
0.06
0.06


BSCO*
g/hp hr


1.88
2.85
5.72
7.68

4.48
3.44
2.54
2.52


BSC02
g/hp hr


682
600
555
602

612
624
648
743


BSN02**
g/hp hr


5.67
8.04
9.44
10.47

9.51
8.37
6.39
5.61


Hum. ,
gr/lb
92.2
92.2
92.2
92.2
92.2
96.4
96.4
96.4
96.4
96.4
96.4
96.4
96.4
                     Cycle Composite
                                BSHC
BSHC
BSCO*
BSN02**
BSN02**
BSCO.,
= 0.113 gram/bhp hr
= 4.659 gram/bhp hr
= 8.910 gram/bhp hr
= 9.023 gram/bhp hr
=663   gram/bhp hr
 * converted to wet basis
** converted to wet basis and corrected to 75 grains water per Ib dry air
                                          B-4

-------
              TABLE  B-4.   13-MODE  FEDERAL DIESEL EXHAUST EMISSION CYCLE
               OF  TRUCK WITH CUMMINS  NTC-3SO  ENGINE,  TEST 2  (3 min/mode)
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine
Speed
rpm
600
1500
1500
1500
1500
1500
600
2100
2100
2100
- 2100
2100
600
Torque
Ib-ft
0.0
20.3
252.1
507.7
759.8
1015.4
0.0
885.3
662.8
442.7
220.1
17.5
0.0
Power
bhp
0.0
5.8
72.0
145.0
217.0
290.0
0.0
354.0
265.0
177.0
88.0
7.0
0.0
Fuel
Flow
Ib/min
0.07
0.25
0.53
0.92
1.27
1.82
0.06
2.28
1.75
1.23
0.72
0.45
0.07
Air
Flow
Ib/min
7.80
18.59
19.71
23.71
26.11
32.05
6.81
53.55
44.98
36.59
27.84
25.52
7.99
Exhaust
Flow
Ib/min
7.87
18.84
20.24
24.63
27.38
33.87
6.87
55.84
46.73
37.83
28.56
25.97
8.06
Fuel
-Air
Ratio
0.009
0.013
0.027
0.039
0.049
0.057
0.009
0.043
0.039
0.034
0.026
0.018
0.009
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
HC
ppm
118
90
54
31
10
1
100
41
33
37
32
68
106
CO*
ppm
137
160
260
672
1557
2340
146
996
644
416
288
248
167
co2
%
1.80
2.87
5.89
8.57
10.80
12.55
1.80
9.38
8.57
7.38
5.56
3.76
1.76
NO**
ppm
133
186
497
1067
1841
2046
152
1415
1137
694
419
257
138
Weighted
bhp
0.00
0.46
5.76
11.60
17.36
23.20
0.00
28.32
21.20
14.16
7.04
0.56
0.00
BSHC
g/hp hr


0.20
0.07
0.02
0.00

0.09
0.08
0.10
0.14


BSCO*
g/hp hr


1.92
3.00
5.17
7.19

4.13
2.99
2.34
2.46


BSC02
g/hp hr


684
601
563
605

611
624
651
745


BSN02**
g/hp hr


6.04
7.83
10.03
10.32

9.64
8.66
6.41
5.87


Hum. ,
gr/lb
108.7
108.7
108.7
108.7
108.7
99.1
99.1
99.1
99.1
99.1
100.8
100.8
100.8
                     Cycle Composite
       BSHC
       BSCO*
       BSN02**
BSHC + BSN02**
       BSC02
= 0.113 gram/bhp hr
= 4.312 gram/bhp hr
= 9.041 gram/bhp hr
= 9.154 gram/bhp hr
=662   gram/bhp hr
 * converted to wet basis
** converted to wet basis and corrected to 75 grains water per Ib dry air
                                          3-5

-------
              TABLE B-5.  13-MODE FEDERAL DIESEL EXHAUST EMISSION CYCLE
               OF TRUCK WITH CUMMINS NTC-350 ENGINE, TEST 3  (3 min/mode)
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine
Speed
rpm
600
1500
1500
1500
1500
1500
600
2100
2100
2100
2100
2100
600
Torque
Ih-ft
0.0
20.3
252.1
507.7
759.8
1015.4
0.0
885.3
662.8
442.7
220.1
17.5
0.0
Power
bhp
0.0
5.8
72.0
145.0
217.0
290.0
0.0
354.0
265.0
177.0
88.0
7.0
0.0
Fuel
Flow
Ib/min
0.07
0.26
0.53
0.92
1.28
1.82
0.07
2.28
1.75
1.23
0.72
0.45
0.07
Air
Flow
Ib/min
7.78
19.57
19.37
24.55
26.26
35.20
7.92
53.32
44.73
35.88
28.59
25.41
8.21
Exhaust
Flow
Ib/min
7.85
19.83
19.91
25.47
27.54
37.02
7.98
55.60
46.48
37.11
29.31
25.86
8.28
Fuel
-Air
Ratio
0.009
0.013
0.027
0.037
0.049
0.052
0.009
0.043
0.039
0.034
0.025
0.018
0.009
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
HC
ppm
112
81
44
30
13
1
101
47
44
35
54
79
125
CO*
ppm
184
211
293
565
1684
2587
183
901
614
451
133
299
208
C02
%
1.80
2.80
5.95
8.30
10.89
12.85
1.78
9.42
8.59
7.51
5.41
3.77
1.70
NO**
ppm
131
188
522
992
1820
2120
154
1411
1072
741
399
266
155
Weighted
bhp
0.00
0.46
5.76
11.60
17.36
23.20
0.00
28.32
21.20
14.16
7.04
0.56
0.00
BSHC
g/hp hr


0.16
0.07
0.02
0.00

0.10
0.10
0.10
0.24


BSCO*
g/hp hr


2.13
2.61
5.62
8.69

3.72
2.83
2.49
1.17


BSO>2
g/hp hr


680
602
571
677

611
622
650
744


BSN02**
g/hp hr


6.24
7.53
9. 98
11.69

9.57
8.12
6.71
5.74


Hum. ,
gr/lb
92.2
92.2
92.2
92.2
92.2 ,
98.1
98.1
98.1
98.1
98.1
98.1
96.4
96.4
                     Cycle Composite   BSHC    = 0.127 gram/bhp hr
                                       BSCO*   = 4.502 gram/bhp hr
                                       BSN02** = 9.202 gram/bhp hr
                                BSHC + BSN02** = 9.329 gram/bhp hr
                                       BSCO2   =676   gram/bhp hr

 * converted to wet basis
** converted to wet basis and corrected to 75 grains water per Ib dry air
                                          B-6

-------
                TABLE B-6.   13-MODE FEDERAL DIESEL EXHAUST EMISSION CYCLE
            OF DETROIT DIESEL 6V-71 BUS ENGINE, TEST  1 HIGH SPEED  (3 min/mode)
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine
Speed
rpm
515
1260
1260
1260
1260
1260
515
2100
2100
2100
2100
2100
515
Torque
Ib-ft
0.0
12.3
119.0
238.1
357.1
476.2
0.0
388.6
290.6
194.3
96.3
8.8
0.0
Power
bhp
0.0
2.9
28.6
57.1
85.7
114.2
0.0
155.4
116.2
77.7
38.5
3.5
0.0
Fuel
Flow
Ib/min
0.05
0.17
0.25
0.44
0.61
0.89
0.06
1.30
1.00
0.77
0.54
0.37
0.05
Air
Flow
Ib/min
11.79
30.96
30.11
30.21
30.39
30.21
11.77
48.70
48.38
49.28
48.54
49.12
11.36
Exhaust
Flow
Ib/min
11.84
31.12
30.36
30.65
31.01
31.10
11.83
50.00
49.38
50.05
49.08
49.48
11.71
Fuel
-Air
Ratio
0.004
0.005
0.008
0.015
0.020
0.030
0.005
0.027
0.021
0.016
0.011
0.007
0.004
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
HC
ppm
1024
915
464
572
460
504
780
380
288
308
432
416
872
CO*
ppm
861
752
328
133
158
4107
724
1127
125
73
103
227
568
"2
0.71
1.07
2.06
3.21
4.51
6.21
0.65
5.99
4.71
3.65
2.45
1.61
0.66
NO**
ppm
39
67
325
658
928
897
47
973
825
559
353
178
68
Weighted
bhp
0.00
0.24
2.28
4.57
6.85
9.14
0.00
12.43
9.30
6.22
3.08
0.28
0.00
BSHC
g/hp hr


6.51
4.05
2.20
1.81

1.61
1.62
2.62
7.27


BSCO*
g/hp hr


9.18
1.87
1.51
29.41

9.53
1.40
1.24
3.45


BSC02
g/hp hr


903
712
674
698

796
827
971
1290


BSN02**
g/hp hr


14.92
15.24
14.51
10.55

13.52
15.14
15.54
19.43


Hum.
gr/lb
77.7
77.7
77.7
77.7
77.7
71.1
71.1
71.1
71.1
68.1
68.1
68.1
68.1
                       Cycle Composite
                                  BSHC
BSHC
BSCO*
BSN02**
BSN02**
BSCOo
  4.017 gram/bhp hr
 10.595 gram/bhp hr
 14.852 gram/bhp hr
 18.869 gram/bhp hr
888.    gram/bhp hr
 * converted to wet basis
** converted to wet basis and corrected to 75 grains water per Ib dry air

-------
                 TABLE B-7.  13-MOD£ FEDERAL DIESEL EXHAUST EMISSION CYCLE
              OF DETROIT DIESEL 6V-71 BUS ENGINE, TEST 2 HIGH SPEED (3 min/mode)
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine '
Speed
rpm
515
1260
1260
1260
1260
1260
515
2100
2100
2100
2100
2100
515
Torque
Ib-ft
0.0
14.0
119.0
239.8
358.9
479.7
0.0
392.1
294.1
196.1
98.0
14.0
0.0
Power
bhp
0.0
3.4
28.6
57.5
86.1
115.1
0.0
156.8
117.6
78.4
39.2
5.6
0.0
Fuel
Flow
Ib/min
0.05
0.16
0.28
0.45
0.64
0.87
0.06
1.31
1.00
0.77
0.56
0.38
0.06
Air
Flow
Ib/min
11.69
31.05
30.22
30.87
30.03
29.74
11.59
48.23
48.23
48.23
48.23
49.28
11.76
Exhaust
Flow
Ib/min
11.74
31.21
30.50
31.31
30.66
30.62
11.65
49.54
49.23
49.00
48.79
49.66
11.82
Fuel
-Air
Ratio
1 0.004
0.005
0.009
0.014
0.021
0.029
0.005
0.027
0.021
0.016
0.012
0.008
0.005
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
HC
ppm
936
941
458
520
476
580
792
416
268
352
440
416
800
CO*
ppm
754
843
340
137
184
4477
601
1311
128:
89
112
237
615
C02
%
0.68
1.07
2.06
3.22
4.56
6.20
0.66
6.02
4.68
3.32
2.34
1.68
0.64
NO**
ppm
55
67
321
674
923
852
59
958
850
583
371
190
56
Weighted
bhp
0.00
0.27
2.28
4.60
6.89
9.21
0.00
12.54
9.41
6.27
3.14
0.45
0.00
BSHC
g/hp hr

,
6.46
3.74
2.24
2.04

1.73
1.48
2.90
7.23


BSCO*
g/hp hr


9.55
1.96
1.72
31.33

10.89
1.40
1.47
3.68


BSC02
g/hp hr


907
724
671
681

786
809
857
1203


BSN02**
g/hp hr


14.79
15.84
14.20
9.79

13.07
15.38
15.75
19.96


Hum.
gr/lb
66.4
66.4
66.4
66.4
66.4
63.0
63.0
63.0
63.0
75.9
75.9
75.9
75.9
                       Cycle Composite
                                  BSHC
BSHC
BSCO*
BSN02**
BSN02**
BSC02
  4.028 gram/bhp hr
 11.307 gram/bhp hr
 14.717 gram/bhp hr
 18.744 gram/bhp hr
861.    gram/bhp hr
 * converted to wet basis
** converted to wet basis and corrected to 75 grains water per Ib dry air
                                             B-8

-------
                 TABLE B-8.   13-MODE FEDERAL DIESEL EXHAUST EMISSION CYCLE
              OF DETROIT DIESEL 6V-71 BUS ENGINE,  TEST 1 LOW SPEED (3 min/mode)
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine
Speed
rpm
515
900
900
900
900
900
515
1 1500
1500
1500
1500
1500
515
Torque
Ib-ft
0.0
12.3
124.3
248.6
372.9
498.9
0.0
467.4
348.4
232.8
115.5
8.8
0.0
Power
bhp
0.0
2.1
21.3
42.6
63.9
85.5
0.0
133.5
99.5
66.5
33.0
2.5
0.0
Fuel
Flow
Ib/min
0.04
0.11
0.20
0.34
0.48
0.69
0.05
1.01
0.75
0.54
0.37
0.20
0.05
Air
Flow
Ib/min
11.70
22.75
22.29
22.92
23.26
22.40
11.50
35.42
35.49
35.95
35.55
36.31
11.69
Exhaust
Flow
Ib/min
11.74
22.86
22.49
23.26
23.74
23.09
11.55
36.43
36.24
36.48
35.92'
36.52
11.74
Fuel
-Air
Ratio
0.003
0.005
0.009
0.015
0.021
0.031
0.004
0.029
0.021
0.015
0.010
0.006
0.004
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
HC
ppm
1096
1016
797
444
516
668
960
396
312
358
346
578
880
CO*
ppm
558
864
349
178
336
8266
617
2277
122
109
215
667
703
C°2
,0.75
0.92
1.99
3.26
4.65
5.96
0.69
6.12
4.32
2,95
1.94
1.09
0.68
NO**
ppm
.68
60
339
680
929
837
69
1030
915
659
343
67
66
Weighted
bhp
0.00
0.17
1.70
3.41
5.11
6.84
0.00
10.68
7.96
5.32
2.64
0.20
0.00
BSHC
g/hp hr


11.11
3.20
2.53
2.38

1.43
1.50
2.59
4.97


BSCO*
g/hp hr


9.69
2.56
3.28
58.71

16.34
1.17
1.57
6.15


BSC02
g/hp hr.


868
735
713
665

690
650
668
872


BSN02**
g/hp hr


15.45.
16.05
14.91
9.77
i
12.14
14.40
15.62
16.11


Hum.
gr/lb
69.7
69.7
69.7
69.7
69.7
69.7
'69.4
69.4
69.4
69.4
72.6
72.6
75.8
                       Cycle Composite
                                  BSHC
BSHC
BSCO*
BSN02**
BSN02**
BSC02
  4.322 gram/bhp hr
 17.793 gram/bhp hr
 13.943 gram/bhp hr
 18.265 gram/bhp hr
756.    gram/bhp hr
 * converted to wet basis
** converted to wet basis and corrected to 75 grains water per Ib dry air

-------
                 TABLE B-9.   13-MODE FEDERAL DIESEL EXHAUST EMISSION CYCLE
              OF DETROIT DIESEL 6V-71 BUS ENGINE, TEST 2 LOW SPEED (3 min/mode)
Mode
1
2
3
4

5
6
7
8
9
10
11
12
13
Engine
Speed
rpm
515
900
900
900

900
900
515
1500
1500
1500
1500
1500
515
Torque
lb-ft.
0.0
10.5
124.3
250.3
i
374.6
500.7
0.0
465.7
348.4
232.8
115.5
10.5:
0.0
Power
bhp
0.0
1.8
21.3
42.9

64.2
85.8
0.0
133.0
99.5
66.5
33.0
3.0
0.0
Fuel
Flow
Ib/min
0.05
0.10
0.19
0.31

0.47
0.70
0.05
1.02
0.75
0.54
0.37
0.21
0.06
Air
Flow
Ib/min
11.69
23.38
22.60
22.60

22.60
22.69
11.67
35.72
35.91
36.16
35.79
35.72
11.77
Exhaust
Flow
Ib/min
11.74
23.48
22.79
22.91

23.07
23.39
11.72
36.74
36.66
36.70
36.16 '
35.93
11.82
Fuel
-Air
Ratio
0.004
0.004
0.009
0.014

0.021
0.031
0.004
0.029
0.021
0.015
0.010
0.006
0.005
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
HC
ppm
835
984
780
452
544
732
1016
440
364
400
386
612
756
CO*
ppm
728
824
372
145
276
8260
502
2126
88
70
213
639
647
C02
%
0.74
0.91
1..97
3.20
4.57
6.04 ,
0.65
6.33
4.56
3. '24
2.06
1.13
0.65
NO**
ppm
64
61
352
721
1018
875
56
1033
973
635
345
86
58
Weighted
bhp
0.00
0.14
1.70
3.43
5.14
6.86
0.00
10.64
7.96
5.32
2.64
0.24
0.00
BSHC
g/hp hr


11.02
3.19
2.58
2.63

1.60
1.77
2.91
5.58


BSCO*
g/hp hr


10.47
2.04
2.61
59.22

15.44
0.85
1.01
6.14


BSCO2
g/hp hr


871
706
678
680

722
694
738
932


BSN02**
g/hp hr


16.27
16.64
15.80
10.30

12.33
15.49
15.13
16.33


Hum.
gr/lb
74.2
74.2
74.2
74.2
74.2
74.2
70.9
70.9
70.9
70.9
69.4
69.4
69.4
                       Cycle Composite
                                  BSHC
BSHC
BSCO*
BSN02**
BSN02**
BSCO-,
  4.473 gram/bhp hr
 17.361 gram/bhp hr
 14.440 gram/bhp hr
 18.913 gram/bhp hr
780.     gram/bhp hr
 * converted to wet basis
** converted to wet basis and corrected to 75 grains water per Ib dry air
                                             B-10

-------
                    APPENDIX C





SUPPLEMENTAL NITROSAMINE DATA AND ANALYSIS REPORTS
                           C'\

-------
M e m o r andum

May 27, 1976


To:       Daniel Montalvo - Division 11

From:    Charles of. RocL'iguez - Division 01   (/If^

Subject:  Determination of nitrosamine in diesel blowby.
          Project 11-4291-006
          We have completed analyses of the impinge r solutions and
glass fiber filters which you used to sample diesel blowby exhaust.
The methods we found that best extract dimethylnitrosamine (DMNA)
from the sample medium follow.

I.         Extraction from glass fiber filters

          A.  .   Cut the filter into small pieces and place in a 200 ml
                 round bottom flask,

          B.     Add 75 ml 0. 01 N KOH solution to the flask,

          C.     Distill approximately 45 ml into a 100 ml beaker
                 containing 5 ml H^O,

          D.     Wash the distillation tube with a small volume deionized
                 water into the beaker,

          F.     Proceed as below, steps A - E.

II.        Extraction from alkaline impinger solution

          A.     Quantitatively transfer the impinger solution to a
                 125 ml separatory funnel,

          B.     Add 10 ml hexane and shake for 1 minute, withdraw
                 and discard the hexane,

          C.     Extract with dichloromethane (2 x 10 ml), combine
                 the extracts,

          D.     Evaporate the extract to less  than 500 fj.1 with a Kuderna-
                 Danish evaporative concentrator,

          E .     Make up  the extract to a final volume of 500  ul
                 with dichloromethane.
                                  02

-------
Memorandum
October 20, 1976
To:          Charles Hare - 11

From:       Charles F.  Rodriguez - 01

Subject;      Method for  the determination of Dimethylnitrosamine in
             dieselblowby 11-4291-006
             I am sending you a copy of the method we have used to
extract samples taken from diesel blowby gases and determine the
dimethylnitrosamine levels.  The method is essentially the same
except after step II. B-NaCl is added to the impinger solution in the
amount of 13g per 100 ml solution.  This serves to enhance the partition
ratio of the DMNA in our favor and provides a better phase separation
between the aqueous and organic liquids.
CFR: nn
File:   11-4291-006
                               C-3

-------
    TABLE C-l.  SUPPLEMENTARY DATA ON CRANKCASE PARTICULATE AND
        ENGINE INTAKE AIR SAMPLES ANALYZED FOR NITROSAMINES
Engine
6V-71 Development
NTC-350
6V- 71 Bus
Date
3/23/76
4/22/76
4/26/76
4/26/76
4/26/76
5/14/76
7/06/76
7/06/76
7/07/76
7/07/76
7/07/76
7/13/76
7/16/76
8/25/76
8/25/76
8/25/76
9/01/76
Impinger
Run No.
a
a
a
a
a
1
a
a
a
a
a
3
2
a
a
a
1
Solution pH
_____a
a
a
a
a
b

a
a
a
a
a
8.7
11.0
a
a
a
b

Filter No.
AR-6001
AR-6007
AR-6008
AR-6010
AR-6011
a

AR-6012
AR-6013
AR-6014
AR-6015
AR-6016
a
a

AR-6028
AR-6029
AR-6030
3i

a does not apply
b not measured
                                C-4

-------
           SWRI  DEPARTMENT OF  ENVIRONMENTAL SCIENCES
        CONFIRMATION OF THE PRESENCE OF DIMETHYLNITROSAMINE IN
                         DIESEL BLOWBY GASES

                 GAS CHROMATOGRAPHY/MASS SPECTROMETRY
     For some time now we have been quantitatively determining one
nitrogen-containing compound in diesel crankcase gas samples and re-
porting it as n-dimethylnitrosamine.  This compound is among those
sampled in aqueous KOH and subsequently removed by extraction into
dichlorom'ethane, back extracted with acid to remove interfering amines,
and concentrated by evaporation in a micro Kuderna-Danish apparatus.
Quantitative determination of the compound called n-dimethylnitro-
samine is then done by injection onto a gas chromatographic column
which separates compounds of a similar nature.  They are then sensed
by a Hall Electrolytic Conductivity Detector operated in the pyrolytic
mode (which makes it selective for amines and nitrosamines)il»2).
Regardless of the specificity of this detector, there is no molecular
information provided.  Consequently, identification of n-dimethyl-
nitrosamine cannot be considered absolute.  This program then was
initiated to attempt confirmation of the identity of the compound
heretofore called n-dimethylnitrosamine.

EXPERIMENTAL

     Quantitative Determination of DMNA

     The sample which was received dissolved in approximately 500 ml
0.01N  KOH>was transferred to a large separately funnel, and NaCL
 (26 g/100 ml) was added.  Extraction of amines, nitrosamines, and
any  other basic compounds was effected with three 50 ml volumes
of dichloromethane  (DCM).   The combined extracts were transferred
to a Kuderna-Danish evaporative concentrator and placed in a water
bath maintained at 58-60°C.  Volume was reduced, by evaporation of
the DCM,  to less than 5 ml.  The concentrate was then made up to
5 ml with DCM and back extracted with mild ^804 solution to remove
free amines.  A quantitative determination was made by comparing
chromatographic response of an aliquot with those of standard solutions
of DMNA which had been subjected to the same extraction and concen-
tration procedures.  The extract was then transferred to a micro-
Kuderna-Danish concentrator, 0.25 ml methanol was added, and the so-'
lution was evaporated to a final volume of approximately 300 Ul.
Methanol was then added  to make the final concentration of DMNA 10 ng/pl
as determined by the quantitative results above.

     Quantitative determinations were made with a Model 220 gas chro-
ma tograph; equipped with a Hall electrolytic conductivity detector
operated in the pyrolytic mode.  The instrument was operated under
the following conditions:
                                   C-5

-------
           SWR1  DEPARTMENT 0^  ENVIRONMENTAL  SCIENCES
     Column:  0.9 m(l) x 4 nun (i.d.) packed with Chromosorb 101
              (80/100 mesh)

     Carrier Gas:  Helium, 60 cc/mir @ 50 psig

     Temperature:  Inlet - 220°C
                   Column - 140°C
                   Transfer line - 210°C

     Detector:  Quartz tube reaction zone
                Inert helium atmosphere
                Furnace - 500°C
     Gas chromatographic/mass spectrometric determinations were made with
a Finnigan Model 3300F instrument and Model 6100 data system operated under
the following conditions:

     Column:  1.5 m(l) x 2 mm (i.d.)  packed with Chromosorb 101
              (80/100 mesh)

     Carrier Gas:  Helium, 15 cc/min @ 14 psig •

     Temperature:  Inlet - 220°C
                   Column - 150° - 250°C @ 6c/min
                   Glass Jet separator - 220°C
                   Column effluent transfer - 200°.C
                   Separator transfer - 200°C
                   Ion Source: 70C-80°C

     Mass Spectrometer
                   Electron impact ionization
                   Ionizing voltage - 70 ev @ 250 y amp
                   Accelerating voltage - programmable, set for
                                          optimum resolution and
                                          sensitivity
                   Mass range:  29-105 amu
                   Integration time:   11 msec/amu
                   Scan Time: 1 sec

     Results and Discussion

     Gas chromatography with the Hall detector-equipped instrument in-
dicated there was a total of 2380 ng in the concentrated blowby extract.
The smallest volume we felt the extract could be concentrated to was
350 ul, which provided a concentration of 6.8 ng/yl DMNA.   The chroma-
togram of 20 ul extract (136 ng DMNA) is shown here as Figure 1.  Two
peaks are shown as coinciding with the retention times of n-dimethyl-
nitrosamine (DMNA) and n-diethylnitrosamine (DENA).  These are the only
two peaks normally observed when smaller samples of blowby exhaust were
taken.  In several of those samples,  only the peak labeled DMNA was seen.

-------
           SWRI  DEPARTMENT OF  ENVIRONMENTAL SCIENCES
There are at least four and possibly six other compounds shown here which
respond as nitrosamines, although it is by no means certain that they are
not some other type of nitrogen-containing compounds.

     Once it was determined that there was a workable amount of DMNA in the
extract, GC-MS  studies were begun by determining retention and response
characteristics of authentic standards of DMNA and DENA.  The mass spec-
trometer was set to scan in either the continuous ,  full spectrum mode or
the mass fragmentographic mode at 30, 42, 74 and 102 amu.  The chromato-
gram produced by full-spectrum scanning is shown as Figure 2, and that for
the mass fragmentographic mode as Figure 3.

     The chromatogram in Figure 2 is 1 ng of each compound, and it is ap-
parent that at this level the nitrosamines are barely discernible.  Thus
for the full-spectrum node the limit of detection is approximately 10 ng.
Figure 2 shows, however, that an identifiable chromatogram can be extrac-
ted with the computer by having it regenerate the chromatogram of response
at 74 amu only, the molecular ion of DMNA.

     The mass spectra and the fragment ions that give rise to these spectra
have been described^'4).  The use of these characteristic fragment ions
to provide'useful chromatographic data is shown i.i Figure 3.  There is a
response at 30, 42, and 74 amu for DMNA; and at 30, 42, and 102 amu for
DENA.  There is a difference noted in the relative abundances of these
fragments when compared with those in the literature (which indicate the
molecular ion is the most abundant).  Here the ion at 42 amu is the most
abundant.  This situation is not uncommon and arises from differences in
instruments and operating conditions.  The operating conditions for chroma-
tographing the blowby concentrate were established by running these stan-
dards, and were used throughout this study.

     The full-spectrum (total ion) chromatogram of the diesel blowby con-
centrate is shown in Figure 4a. It is apparent that there is a large
number of compounds in the extract, many more than had been expected.
This complication made the situation much more difficult.  The single-
ion responses at 30, 42, 74 and 102 amu were computer-extracted from this
chromatogram, and are shown as Figures 4b - 4e, respectively.  The re-
sponse at 42 amu has to be ignored because it saturates at an early point
in the chromatogram.  The combination of responses at 30 and 74 amu for
DMNA, and at 30 and 102 amu for DENA indicate the presence of these com-
pounds.  It was not possible, however, to extract a "clean" spectrum
from either of these regions of the chromatogram.

     A mass fragmentogram was recorded for the diesel blowby concentrate
and is shown in Figure 5.  The response shown for DMNA very clearly indi-
cates the presence of the three characteristic fragment ions at 30, 42,
and 74 amu.  They also occurred in the expected order of abundance, i.e.,
42 amu the most abundant  and 30 amu the least abundant.  The presence
of DENA is also indicated, but in a much less definitive fashion due to
the overlapping  of peaks at 42 amu.  The oscilloscope output monitored
by the GC-MS operator showed a definite peak indicating DENA at 102 amu,
but the computer calculates on a relative response basis and plots from
these results.  Therefore, the presence of a large peak at 42 amu tended
to make the plotted peak at 102 amu difficult to see.


                                   C-7

-------
           SWRI  DEPARTMENT  OF ENVIRONMENTAL SCIENCES
     The evidence presented here  is  a  very  strong indication that the
compound previously reported as dimethylnitrosamine in diesel blowby
concentrates is in fact that compound.  To  make identification of DMNA
absolute, it would be required to isolate the DMNA from interfering
compounds and thereby be able to  produce a  "clean" mass spectrum of
DMNA.  The identification of DENA in ~he concentrate is less clear-cut,
as pointed out previously, due to the  smaller amount present.  Never-
theless, there is some evidence to support  the identification of this
compound as DENA.  The problem of determining the origin of these species;
i.e., "Are the nitrosamines produced in the diesel crankcase system or
are they artifacts of sampling and/or  analysis?", was not addressed
here because it was beyond the scope of this project.
                                   C-8

-------
           SWRI  DEPARTMENT OF ENVIRONMENTAL  SCIENCES
                             REFERENCES
1.   "Gas Chromatography  and Selective Detection of N-nitrosamines."
    J.  W. Rhoades  and D. E. Johnson, J. Chromato. Sci., 8, 616 (1970)

2.   "Method for the  Determination of N-nitrosamines in Tobacco-Smoke
    Condensate."  J.  W.  Rhoades and D. E. Johnson, J. Natl. Cancer
    Inst., 48,  1841  (1972)

3.   "Mass Spectrometry of Pesticides and Pollutants."  S. Safe and 0.
    Hutzinger,  CRC Press, 1973 p. 204

4.   "N-nitrosodimethylamine in Fish Meal."  N. P. Sen, L. A. Schwinghamer,
    B.  A. Donaldson  and  W. F. Miles, J.' Agr. Food Chem., 20, 1280 (1972)
                                  C-9

-------
2 NJ
m o
c:

m
      FIGURE C-I,  CHRDMATOGRAM OF BLOWBY EXHAUST, HALL NITROGEN DETECTOR

-------
?
              223L60  DJMETHYLNITROSHMINE   GC/MS/EI
              100
                                                                        DMNA
                                                                      7.76 MIN
                 456
                   TIME, MINUTES
                           DENA
                        12.15 MIN
1
               MiiT'lTTTYV I i I i I i I ' I ' pfi lYiTii
               JO     550      6bO     650
                   11      12      13
                   TIME, MINUTES
                                                                                          10
                   FIGURE  C-2A,   CHROMATOGRAM OF STANDARDS,  FULL SPECTRUM SCANNING

-------
   22 3L60  DI HE TH YLN .1 TROSflM INE
              R6C RT M/E 74
   100
GC/MS/EI
                                                             DMNA
                                                           7.76 WIN
                                                        -t-
                                                        7
   iUO
  456

   TIME,  MINUTES
8
—i—
 10
                 DENA
              12.15 MIN
                  eb'ti
        11      12      13
        TIME^ MINUTES
FIGURE C-2B,  SINGLE  ION  CHROMATOGRAM COMPUTER-EXTRACTED FROM FULL SPECTRUM  SCAN

-------
                 223L20 GC/MS/EI
                                                           DENA
                                                        12.21 MIN

                                                                           30
100     200     300    400
100     200     300     400
                                             100    200    300     400
                      78
                    ^  MINUTES
                                10     11      12     13
                                                                                                14
FIGURE C-3,  MASS  FRAGI^NTOGRAM OF STANDARDS

-------
n
              :?3L50   DIHETHYLNITROSHHINF:  PC/MS/EI
                         TIC  OF BLOWBY
              100
              100
                                                                        DMNA
                                                                      7.64 MIN
                              100
'I'""""!1"
150     200
T	1	
3       4
 "I"
  250
300
350
400
 5       6
^ MINUTES
                                                                            8
                             DENA
                          12.27 MIN
              •3JO     550      6DO      (ibO
                    11     12      13
                    TIME, MINUTES
'|'l' T'TTTTI

450
                              10
                   FIGURE  MA.   TOTAL  ION  CHROMATOGRAM Of DIF.SFI.  BLOWBY CONCECTRATF

-------
              223L50   DIMETHYLNITROSnMINE  GC/MS/EI
                         lOflHU
o
(-•
Ln
100
                       50
                                             456
                                              TIME, MINUTES
                                                            8
                              DEN A
                           12.27 MIN
                                                    \J
                                          i
                                                             V
                                                                     450
                        10
                ' I » I ' I I I Ml I I I » I I I I I I I I I < I I I I I l| I | I | I | II I | . 1 t | 1 | I I I | I | I | l | i I I | I | I | I | > I I | I | I | I | . I i | I | i | i | i • i | i | i |
               5DO      550     000     650     TOO      750     8DO     850     9BO      950
                     11     12     13      14       15     16
                                              TIME, MINUTES
17
                                                          18
19
20
             FIGURE MB.  SINGLE ION SCANS  COMPUTER-EXTRACTED  FROM TOTAL ION CHROMATOGRAM

-------
   223L50   DIMETHYLNITR03HMINE  GC/MS/EI
              42BHU

   100
    r
    0
   100
                                                             t
                                                           DMNA
                                                         7.64 MIN
50
                  100
""I1"
   150
200
250
300
                                             350
                                                                 400
          456
           TIME, MINUTES
                   t
                 DENA
              12.27 MIN
450
                                                                  10
' 1 1 1 1 1 ' 1 1 1 1 1 ' 1 1 1 1 1 1 1 1
30 550 ffl
— i i
11 12
51 1 » i » i
•
13
650 ' "A
14
M ' 1 ' 1 ' 1
30
15
750

800
i
16

17
850

9
18
bo""
19
950


20
                                  TIME, MINUTES
FIGURE C-4c,   SINGLE ION SCANS COMPUTER-EXTRACTED FROM TOTAL  ION CHROMATOGRAM

-------
  223L50   OIMETHYLNITROSflMINE  GC/MS/EI
             74RMU
   100
                                                        350
 400     450
   100
                                 456
                                  TIME,  MINUTES
8
                 DENA
              12.27 MIN
 10
  500     550     600      650     TOO     750     800     850     900     950
        11      12      13     14       15      16     17     18
                                 TIME,  MINUTES
      19
20
FIGURE C-4D,   SINGLE ION SCANS COMPUTER-EXTRACTED FROM TOTAL ION CHROMATOGRAM

-------
OD
              223L50   DIMETHYLNITROSflMINE  GC/MS/EI
                         RGC HT M/E 102
              100
                                                                      DMNA
                                                                    7.64 MIN
                      50
100
150
200
""I"
  250
' I' I ' I ''I ' I ' I ' I ' I ' I ' I ' I' I I I I I I I I I ' I I I I I I I 11
300      350      400      450
                            DENA
                         12.27 MIN
               456
                TIME, MINUTES
                                                                           8
                                                    10
              500     550    ' eb'd'     650
                   11      12     13
                   TIME., MINUTES
            FIGURE  ME.   SINGLE ION SCANS  COMPUTER-EXTRACTED FROM TOTAL ION CHROMATOGRAM

-------
                          223L20 GC/MS/EI
                                                                                                 102
                                                                                   74
100     200
300
100
ZOO     300
400
100     200
                        8    9    10   11
                        TIME, MINUTES
                                                           12    13   14    15   16    17   18
 FIGURE 05,   MASS FRAGMENTOGRAM OF DIESEL BLOWBY CONCENTRATE

-------
                            ENVIRONMENTAL PROTECTION AGENCY
                                  OFFICE OF ENFORCEMENT
                       NATIONAL ENFORCEMENT INVESTIGATIONS CENTER
                           BUILDING 53, BOX 25227, DENVER FEDERAL CENTER
                                  DENVER, COLORADO  80225


       Mr. Tom Baines                                      DATE  February 1.  1977
       EPA - ECTD
.KO«    chief
       Chemistry Branch
SUBJ1< I
       Results of Nitrosamine Analysis of Diesel Exhaust Samples

       As discussed by phone 1/31/77, we have analyzed the methylene chloride
       extracts of diesel exhaust samples which you had sent to us  from the
       Southwest Research Institute and have obtained the following results.

                                                 Nitrosamines (ng/sample)
       Sample Description                  DMN   DEN   DBN   NPIP   NMORPH   DAN

       Bottle blank (1/6/77)                ND    ND    ND    ND      ND      ND
       Reagent blank (1/4/77)               ND    ND    ND    ND      ND      ND
       Background (1/4/77)                  ND    ND    ND    ND      ND      ND
       Background (1/5/77)                  ND    ND    ND    ND      ND      ND
       Background (1/6/77)                  ND    ND    ND    ND      ND      ND
       Sample (1/4/77)                      45    25    17    11      51      43
       Sample (1/5/77)                      61    23    21   

-------
                NEIC ANALYTICAL PROCEDURE FOR NITROSAMINES
Air  Samples
Air  samples are  taken  in a foil-covered impinger (SGA Catalogue #JV8550)
filled with 60 ml of 1 N KOH.  Sample air is drawn through the impinger
by a  vacuum pump and metered through a calibrated stainless steel
hypodermic needle (B&D #21).  Normal collection volumes range from
100  to 200 liters.

The  KOH  is then  extracted in a 125-ml separatory funnel with three 8-ml
portions of dichloromethane (Burdick and Jackson, "distilled in glass").
The  combined extract is taken down to 0.5 ml on a Kuderna-Danish type
assembly consisting of a three ball Snyder column attached to a
specially made 50-ml concentrator flask which in turn is attached to
a 4-ml calibrated receiving tube.  The column and receiving tube are
available through Kontes Glass Company; the 50-ml flask was custom made
locally.  Before taking the solution down, 0.5 ml of 2,2,4 trimethylpentane
is added as a keeper.  A hot water bath maintained at 58 to 60°C 1s used
as the heat source.

All  laboratory work is completed under low UV lighting; "bag" lights are
used  for this purpose.  Typical extraction recoveries are about 40%.

Microliter aliquots of the concentrates are then injected into a gas
chromatograph attached to a Thermo Electron Corporation Thermal Energy
Analyzer.  A dry ice/acetone mixture is used in the cold trap on the TEA.
The  GC column used is 20 ft. 1/8 inch 0.0. stainless steel with 15% FFAP
on 60/80 mesh    Chromsorb WAW DCMS.  Typical injection volumes are 5 jil •


Water Samples

Grab  samples are collected in amber glass quart bottles with teflon liners.
Two  50-ml portions of dichloromethane are used to extract the sample in
•a 2000-ml separatory funnel.  The combined extract is taken down as
above, only using a 250-ml concentrator flask and 1 ml of 2,2,4 trimethylpentane
is added as a keeper before concentration.  In some cases it may not be
necessary to concentrate to 1 ml in order to obtain measurable amounts
of nitrosamines.  Similarly, it is advisable to keep volumes injected
into  the GC/TEA  as small as possible.

'Granular anhydrous sodium sulfate (Mallinkrodt AR) may be used to treat
emulsions during the extractions.  Its freedom from interfering materials
should be established by running several blanks.


                                                 Richard C. Ross
                                                 Chemistry Branch
                                      C-21

-------
           ENVIRONMENTAL PROTECTION AGENCY
                RESEARCH TRIANGLE PARK
TO:      Charles T. Hare, SwRI
FROM:    Roy Zweidinger, EPA
DATE:    May 26, 1977
SUBJECT: Diesel Blow-by Samples

     Enclosed are the spectra  obtained from the ion
chromatographic analysis of the diesel blow-by samples
sent by SwRI.  Spectra of several amines am3 other cations
are included.  Retention times are written above each peak.
     Na+, NH.+ and K+ appear to be the only detectable
ions, however methylamine and ethanolamine could be present
and obscured  by the NH4+ and K+ peaks.
                           C-22

-------
o
to
                    FIGURE  C-6.   ION CHROMATOGRAM OF MACK ETAY(B)673-A CRANKCASE
                               GAS SAMPLES,  BLANK, AND BACKGROUND AIR

-------
1.J-*
          FIGURE C-7,   ION CHROMATOGRAM OF MACK  ETAY(B)673-A CRANKCASE
                 GAS SAMPLES, BACKGROUND AIR, AND NA+  STANDARD

-------
?
U1
                                                                                          L\
                           FIGURE C-8.  ION CHROMATOGRAM OF AMINE  STANDARDS

-------
FIGURE C-9,   ION CHROMATOGRAM OF NH^+ STANDARDS

-------
o
i
-J
           1 I
                        FIGURE C-IO,   ION  CHROMATOGRAM OF NH/|+ AND NA+  STANDARDS

-------
SOUTHWEST  RESEARCH   INSTITUTE
8500 CULEBRA ROAD  •  POST OFFICE  DRAWER 28510  •  SAN ANTONIO. TEXAS 7828«

                          February 14, 1977
Dr. Edo Pellizzari
Research Triangle Institute
Cornwallis Road
PO Box 12194
Research Triangle Park, NC 27709

Dear Edo:

    This will confirm our telephone conversation regarding diesel
crankcase gaseous samples taken on February 11, 1977.  The following
lists the quantity of crankcase vapor diluted in a premeasured amount
of zero-grade dry N2-  In the case of the "1 liter" s<-nples, the pre-
measured amount was 0.44 cu ft (12.46 liter) of N2, while the "5 liter"
samples were added to a premeasured 2.2 cu ft (62.3 liter) of l?2.  In
both cases, assuming 1 liter and 5 liter samples were taken, the tar-
get dilution level was 13.5.

    The 13.5 dilution level would then result in the nominal 60 ppm
NOX raw vent gas concentration being reduced to about 4.8 ppm.  We
wanted to get below 5 ppm without excessive dilution.  As mentioned,
the sample rate from the engine was a nominal 1 liter/min monitored
and timed by a Brooks flowmeter and stop watch.  The engine crankcase
was backpressured slightly  (approximately 0.4 in H20) to force the
sample through a 3/8 inch OD stainless probe that was connected to the
flowmeter.  The sample probe was 8 inches long and was unheated.  It
was inside a larger pipe fitting which connected to the crankcase vent
through which the remainder (most) of the vent gases passed around the
probe and then to atmosphere via a gate (backpressure) valve.

    No external heating was employed and only the flowmeter was open
to the cell ambient.  No water or oily condensate was noted.  I esti-
mate the sample gas temperature at the inlet to the flowmeter to be
on the order of 100-125 °F.

    New Tedlar sample bags were used for each sample.  First, about 0.5
cu ft N2 was added to the evacuated bag, and then the bag was evacuated
to a vacuum of 10 inches water.  Then the desired amount of N2 diluent
was added, metered by a wet test meter.  This bag was then connected to
the sample flowmeter outlet and the crankcase gases were added at a 1
liter/min rate into the bag.  The contents of the bag were thoroughly
mixed for about 1 minute.  The bag was then connected to the items you
furnished;  namely, the fiberglass prefilter, the cartridge holder and
then our pump, a flowmeter, and finally the same wet test meter.  The
amount of sample was calculated by subtracting the N2 from the N2 +
sample column:  A new prefilter was used for each sample.
        SAN ANTONIO. HOUSTON. CORPUS CHRIST). TEXAS. AND WASHINGTON O C
                                  C-28

-------
Dr. Edo Pellizzari
February 14, 1^77
page 2

    Prior to taking samples for record, the sampling and cartridge
loading procedures were thoroughly developed to assure that what was
added to the bag could be accurately metered to give a close value
for sample amount.  The system was leak-checked and precalibrated.

    When loading the cartridge, a flow of 7.7 liter/minute was main-
tained constant and continued until the bag was evacuated to 10 in Hj
vacuum.  The withdrawal time was approximately 1.7 minutes for bags
containing 1 liter and 8.7 minutes for bags containing 5 liter of
sample.

    By copy of this letter, I am informing Tom Baines and Ran Dradow
of the technique we used and the basic sampling data.  We are very
interested in your findings with these cartridges.

                                              Very truly yours,
                                              {Carl J. i/Spr
                                              Director
                                              Department of Emissions Research
KJS:Cp
cc:  Tom Baines
     Ron Bradow
                                 C-29

-------
Date:  2/11/77
                TABLE 1.  CRANKCASE GAS SAMPLING DATA - TENAX GC CARTRIDGES
                                    MACK ETAY(B) 673A

                 Fuel:  EM-239-F
                        Gulf 2D @ 0.23% S
               N2 Diluent
                                     Cartridge Load
Crankcase Sample
Cart
No.
Volume,
tt3
obs
£
Temp
Op
pBar
in Hg
Volume,
ft3
obs Temj
£ °F
Nominal 1 liter
1
2
3
0
CJ
o 4
5
6
0.44
0.44
0.44
12.46
12.46
12.46
74
74
76
29.105
29.088
29.083
0.472
0.479
0.469
13.365 74
13.563 75
13.280 76
Nominal 5 liter
2.2
2.2
2.2
62.3
62.3
62.3
76
77
77
29.071
29.071
29.072
2.368
2.386
2.382
67.053 77
67.562 77
67.449 77
? PBar
in Hg
Sample
29.105
29.088
29.083
Sample
29.071
29.071
29.072
Volume ,
ft3

0.032
0.029
0.029

0.168
0.186
o.n2
obs
£

0.906
0.821
0.830

4.765
5.275
5.154
£

0.789
0.7 3
0.718

4.109
4.549
4.444
Dilution
Ratio *2'

14.75
16.52
16.0

14.07
12.81
13.09
mCorrected
                  *m'
                  S = std T & P of 32 °F & 29.92 in Hg
                  m = Meter Conditions
Note manometer reading of Wet Test Meter = 0.20 in H2O

               460 +32   /29.105 + (0.20)  (.074) - 0.838) c
        '*     460 +74   V              29.92              /
        where the 0.838 is water vapor pressure, in Hg 
-------
                   Table 1.  CRANKCASE GAS SAMPLING DATA - TENAX GC CARTRIDGES
                                  MACK ETAY(B) 673A (SRI DATA)
Date:  2/11/77
                   Fuel:  EM-239-F
                          Gulf 2D @ 0.23% S
               N? Diluent
                                       Cartridge Load
Crankcase Sample
Cart
No.

1
2
3
0
Ul
4
5
6
Volume
tt-»

0.44
0.44
0.44
, obs
I

12.46
12.46
12.46
Temp
°F

74
74
76
pBar
in Hg
29.105
29.088
29.083
Volume,
ft3
obs Temj
i °F
Nominal 1 liter
0.472
0.479
0.469
13.365 74
13.563 75
13.280 76
Nominal 5 liter
2.2
2.2
2.2
62.3
62.3
62.3
76
77
77
29.071
29.071
29.072
2.368
2.386
2.382
67.053 77
67.562 77
67.449 77
> pBar
in Hg
Sample
29.105
29.088
29.003
Sample
29.071
29.071
29.072
Volume,
ft3

0.032
0.029
0.029

0.168
0.186
0.182
obs
a

0.906
0.821
0.830

4.765
5.275
5.154
Corrli|
£

0.789
0.713
0.718

4.109
4.549
4.444
Dilution
Ratio (2)

14.75
16.52
16.0

14.07
12.81
13.09
(1)Corrected I
          Tm PS
               Vm,  S a std T & P of 32 °F & 29.92 in Hg
                    m = Meter Conditions
  Note manometer reading of Wet Test Meter « 0.20 in H20
                 46° * 32   /29.105 » (0.20)  (.074) - 0.838 \
Example:  V  =
                              /29.
                              \              fc .7 • "*             /
             where the 0.838 is water vapor pressure,  in Hg @ 74°F and 29.0 in Hg Bar.

^Dilution Ratio = Dilution Vol +  Sample Vol  « Cartridge Load (observed)
                           Sample Vol              Sample Vol (observed)

-------
N-nitrosodimethylamine in the inlet manifold for recoverying vapors trapped on


Tenax GC cartridge samplers have been described.  ' ~    The operating


conditions for the inlet-manifold were as follows:  desorption chamber


2$5°C; six-port valve, 275°C; cryogenic trapping, -195°C; injection,


180°C; thermal desorption time, 4 min.  Gas-liquid chromatography was on


a 48 m glass SCOT capillary coated with DECS in 0.1% of benzyltri-


phenylphosnonium chloride.  The capillary column was programmed from 70-


205°C at 4°/min and the carrier gas flow (helium) was 1.5 ml/ min.  A


single stage glass jet separator (220°C) interfaced the SCOT capillary


column to the mass spectrometer.  The mass spectrometer (Varian MAT CH-7


620L computer) conditions were filament current, 300 pg; ion source


vacuum, 2 x 10   torr; magnetic current to monitor tn/e 74.  The inten-


sity of m/e 74 was recorded vs time.


     Standards of DMN were used for determining the response of the mass


spectrometer (area 4 m/e 74 peak at retention time for DMN) V£ the con-


centration of DMN was obtained.  Air DMN vapor was synthesized and


specific quantities collected on Tenax GC cartridges.  The cartridges


were then desorbed and the m/e 74 ion monitor to obtain detector res-


ponse vs_ concentration.  Based on the volume of air sampled, the concen-

                                                        3
tration of N-nitrosodimethylamine was calculated in ng/m  of ambient


air.
                                C-36

-------
                                6.
3.0  Results and Discussion



     The Tenax GC sampling cartridges which were used to collect vola-



tile organic vapors including N-nitrosoamines were subjected to gc/ns



analysis.  Two cartridges corresponding to samples No. 1 and 4 (1 and 5



£ of blow-by gas) were analyzed by single ion monitoring (m/e 74) for N-



nitrosodimethylamine.  Figure 1 depicts a mass chromatogram for 25 ng of



N-nitcosodimethylamine.  The retention time was 19.5 min.  The maximum



limit of detection was observed to be ~0.5 ng.



     Figures 2 and 3 present the mass chromatograms for samples Nos. 1



and 4 of diesel blow-by gas.  As indicated in each figure by the arrow



at 19.5 min, no N-nitrosodimethylamine was detected.  Sample No. 1



(Table 1, Fig. 2) was run at an attenuation of 0.1 V which represents



the limit of detection for the instrumentation.  Thus, if DUN was pre-



sent on the sampling cartridge it was below 0.5 ng or ~166 ppt in the



blow-by gas.  Sample No. 4 (Table 1, Fig. 3) represented ~5 £ of blow-by



gas and in this case the detection limit was ~100 ppt.



     Since no DUN was detected in either the 1 or 5 £ samples of diesel



blow-by gas, we proceeded to examine sample No. 6 under acquisition of



full mass spectra.  Sample No. 6 was subjected to gas chromatography/



mass spectrometry/computer analysis utilizing the same chromatographic



column (48 ra SCOT DECS) that was used in the previous analyses.  Table 2



presents the results of the volatile organic vapors which were iden-



tified in sample No. 6.  Although these constituents were not quanti-



tated, the major components as observed by the total ion current profile



were phenol and N,N-dimethylacetamide.  It was estimated that these two



components comprised -30% of all of the observed constituents.  The



primary reason for the analysis of sample No. 6 was to determine whether
                                 037

-------
KJU
90

60
P
|TO
£
£ 60
V)
1U
£ 50
2 | 40
o> 6
20
to
n
25 ng DMN
48 m DECS SCOT
80-205°C, 4°C/min
•

i
»
1

1
. ^ 1 11 _
i i i ii i i
      6
9
15
18
21
                   TIME  (MIN.)
Figure 1.  Mass chromatogram of N-nitrosodimethylamlne.

-------
f?
              90
              80
           p70
              60
              50
              40
            i so
              20
               10
                0
                                                         12
is         ia
   TIME (MN.)
21
24
27
30
                                        Figure 2.   Mass  cltromatogram of  sample  No.  1  of  diesel blow-by  gas.

-------
       9          12         IS         18        21         24



                                 TIME  (MIN)




Figure 3.  Mass  chromatogram of sample No. 4 of dlesel blow-by  gas.
27        3O
33

-------
                                  10
Table 2.  VOLATILE ORGANIC VAPORS
                          (SAMPLE
IDENTIFIED IN DIESEL BLOW-BY GAS
NO. 6)
Chromaco-
graphlc
Peak No.
1
2
2A
2B
3
3A
4
4A
5
SA
SB
6
6A
7
8
8A
8B
9
9A
9B
9C
10
11
11A
11B
12
13
13A
13B
14
14A
14B
14C
IS
ISA
1SB
16
16A
17
18
19
20
21
22
23
24
Elutlon
Temp.
CO
80
81
82
82
83
84
85
85
87
88
88
91
92
94
96
97
97
101
102
103
104
105
107
108
108
111
114
115
117
118
119
121
123
124
125
126
128
130
133
133
140
142
144
146
149
.151
Compound
C02
ii-hexane
C?Hj6 laomer
C7H14 lsomor
n.-ociane
C9H20 l80Bar
ji-nonana
C10H22 l80Bar
n^deeane
C11H24 180B"
C10H20 l80mar
£-undecane
C11H22 * C12H26 1"°»ara
C12H26 i80"r
toluene
C12H24 i80Bar
CUHJO isomer
nrtridecane
ethylbensene + Cip26 l8oner
2-xylene
C14H28 I80mar
2-hexanol
o-xylene
C.-alkyl benzene laoaar
C13H24 I80"r
C14H30 taoBar
ti-tecradecane
C.-alkyl benzene laomer
C15H30 18CB"
C16H34 I80mar
C^-alkyl benzene Isomer
C15H30 l80Ber
alkyl alcohol laomer (?)
ji-pentadecane
C^-alkyl benzene laomer
C15H28 lsonar
C16H32 I8™""
C.-alkyl benzene Isomer
C16H34 iaoB1"r
C16H34 l8CBar
C16H30 180B"
C.H-4 iaomar
C17H36 I80"ar
C18H36 1<10""
CpH^ Isomer
C1»H38 I80n"r
Chromaco-
graphlc
Peak No.
25

27
28
29
31
31A
32
33
34
35
37
38
39
40
40A
41
42
43
44
45
45A
46
47
48
51
52
S3


















Eluclon
Temp.
CO
153

161
165
174
180
180
184
184
190
192
19S
198
200
203
203
205
205
205
205
205
205
20S
205
205
205
205
205


















Compound
acetic acid (tent.) *
C18H38 1»OBar
C..H-- iaomer (tent.)
14 20
H,H-dlmathylacetaalde
phenyl acetate
aliphatic acid (?)
acetophenone
naphthalene
N-methylacetanlde
aliphatic acid (?)
C..H..O laemer (tent.)
methylnaphthalene Isomer
C6HIQ0 laomer (tent.)
mathylnaphthalene Isomer
dlechylphthalate
Cj-alkyl naphthalene Isomar
dlaethylnaphthalene laomer
dlaethylnaphthalene Isomer
dlmethylnaphthalene Isomer
phenol
dlmathylphenol laomer
C.-alkyl naphthalene isomer
crasol Isomer
cresol isomer
Cj-alkyl naphthalene Isomar
dime thy Iphenol isomer
ethylohenol Isomer
Cj-alkyl phenol Isoner


















                                   C-41

-------
                                11
precursors for the formation of N-nitrosoamines were present such as



dimethylamine, diethylamine, dipropylaraine, etc. and/or morpholine.



There was no evidence of the presence of amines in this sample, however



two nitrogen containing compounds, N,N-dimethylacetamide and N-methyl-



acetamide were observed.
                                 C-42

-------
                                12
4.0  References



1.   Research in RTI Laboratories under EPA Contract No.  68-02-1228,



     Development of Analytical Techniques for Measuring Ambient




     Atmospheric Carcinogenic Vapors, 1976 (in preparation).



2.   Cohen, J., EPA-NFIC, Denver, CO, private communication.



3.   Pellizzari, E. D.  Development of Method for Carcinogenic Vapor



     Analysis in Ambient Atmospheres.  Publication No.  EPA-650/2-74-



     121, Contract No. 68-02-1228, 148 pp., July, 1974.



4.   Pellizzari, E. D.  Development of Analytical Techniques  for



     Measuring Ambient Atmospheric Carcinogenic Vapors.  Publication



     No. EPA-600/2-75-076, Contract No. 68-02-1228, 187 pp.,



     November, 1975.



5.   Pellizzari, E., J. E. Bunch, R. E. Berkley and J.  T. Bursey,



     Biomed. Mass Spec., 3, 196 (1976).



6.   Pellizzari, E. D., J. E. Bunch, J. T. Bursey, R. E. Berkley,



     E. Sawicki and K. Krost, Anal. Lett., 9, 579 (1976).
                                C-43

-------
Table 3.  VOLATILE ORGANICS  IN DIESEL  BLOW-BY GAS'
                        SAMPLE 5
Chroma to-
graphic
Pejk So.
1
2
3
3A
4A
1
SA
6

78
8
8A
9
9A
10
IDA

12

12A

12C
13

138
13C
14
14A
IS
ISA

16
16A
168
16C
16D
17
18
18A
19
20
20A
21
22
22A
Elution
Top.
t'C)
40
43
47
47
48
49
SO
SO

S3
56
57
57
58
59
60

64

66

68
69

70
71
73
74
75
76

77
78
79
79
81
82
84
8S
88
89
90
92
9)
94
Compound
co2
1-butene
acetaldehyde
isopentane
CjH10 isonor
2-pencane
C^K.-O laonet
acetone

CS, (tent.)
2-mathylpencane
C4HjO isoner
3-aethylpantane
CgH., isomer
hexafluorobenxena (el)
n-hexane

n«ehylcyclapencan« *
pecfluortoluen* (el)
methyl ethyl ketcne (cent)

C7»l* "o""
banzena

cyclohoxana
2-mechylhexaDa
3-me thy Ihoxana
CjH.. isomer
CgH,. isoner
+ C.H. . Laomer

o-hcptana
CjH14 isaner
n-pentanal
CjH^j laooer
2-pencanone
nathy Icyc lohenana
C.H,- iaoner
B 10
CgH., laonat
4-nethyl-2-pencanone
toluene
4-aec hy 1- 2-pancanone
C.H,. Isomer
O 19
C.H,, isomer
C.H,, iaoner (cent)
S 14
Chroma to-
graphic
Peak No.
23
23A
25
23A
26
27
28
28A
29
30
30A
31
31A
32
33
33A
338
33C
34
35
37
38
39
40
41
41A
42
43
44

45
4SA
4SB
46

47
47A
47B
48

49
SO
SI
52
S3
Elucion
Temp.
Co
95
97
98
100
101
103
104
106
107
108
109
110
110
112
113
113
114
114
11 S
117
120
122
123
124
125
125
126
127
129

130
131
132
133

136
137
138
139

141
142
143
144
14S
Compound
CBH16
C8H14 1SOMC
cetrachloroe chy lene
C6H12° isonec
C7K_4 isooer (cane)
C9H20 lsOBet
methyl ethylcyclopancane
Cgli^g Isoner
C.H.g Isoner
ethylbenzene Isoner
C9H20 is°net
S-jcylane
C9K20 iSOner
C9H20 isoner
diacetone alcohol (tent.)
styrene
o-xylena
C,Hlfl isooer
N^S-dlnethylaeetanide
2-nonane
laopropylbaniane
C10H22 i'oam







isoner















C^-alkyl eyclohexane isomer
CiOH22 1BOB"
C10H22 i80Mr
n-propylbenzene
unknown
ethyltoluane isooer
trinethylbenzena + e]_oH72

C10H22 lsoner
C10H20 1"oer
C10H20 l"Mr
1,2,4-trlBethylbeozenc +
C10H20 lsoaer
phenol •*• n-dec»n«
C4-alkyl benzene laoaer
C^alkyl benaane iaener
1.2.3-triaethylbenzane *
C11H22 i30ate
C11H24 lsoBer





Lsomers












indan (tent) * C11H20 laoaer
C^-alkyl cyclohaxane isooer
C11H22 iso*"
C,-alkyl benzene * C,,.H.,

isoner
                                C-44

-------
TABLE 3 (Cont'd).
VOLATILE ORGANICS IN DIESEL BLOW-BY GASa
        SAMPLE 5
grapnic
Pejk. So
54

53
56
57
58
58A '
39

39A
60
60A

61
62
62A
628
63
63A
64
64A
648
63
65A
66
47
67A
63
69

70
71
7 LA
72
73
74
74A
73
76

77
77A
78
79
80
81
j- Elutidn '
Temp . Compound
CC)
143

146
147
- 148
149
130
130

131
132
133

154
136
136
137
137
13B
139
139
160
160
161
162
163
163
164
163

166
167
168
169
171
173
173
174
173

176
176
177
178
170
130
C4-elkyl benzene + ciiH22
laooers
C11Hj4 Isoaer
C10H18 lsooer
CjjHjj liomer
C4-alkyl benzene Isoaer
C..H.J laoaer
C4-alkyl benzene * ci2Hi2
laoaers
'C..H,, Isoaer
C-.Hj, laomer
C.-alkyl benzene + ereaol
Isomers
n-undecane
C,,H,- laoaer
C^-alkyl benzene Isomer
C.,H,fi laomer
cnHin * ci/"in Isomer* (tent.)
11 20 14 20
C12H20 lsooar
Cl2H2fi isooer
Cj-alkyl benzene Isomer
CUH20 I800er
Cj-alXyl cyclohexane Isomer
C12H,& laomer
C^-alkyl benzene laomer
C12H,6 laomer
Cj-alkyl phenol Isooer
C4.-alkyl benzene Isomer
C.-H,. laomer
12 26
C.-alkyl phenol isomer
C12H22 l*oner
naphchalene
C12H24 itaa"
'n-dodacane
C13H28 ison*r
C13H26 l*""r
C.-alkyl benzene laomer
Cj-alkyl benzene + Cji^fi
• iaomera
C12H22 tjon'r
C13H16 iaaa"
C.-alkyl cyclohexane Isomer
'0
C^Hjj iaoner
C..H,, laomer
C. , H., laoaer
1 J fcO
Chronuco-
graphlc
Peak No.
82A
82B
83
84
83
86
86A
86B
87
SB
89
90
91
92
92A
93
94
94A
93
96
96A
96B
97
98
99
100
101
101A
101B

102

103
104
103
106
107

108
109
110
111
112
113
114
113
Eluclon
Temp . Compound
CO
182
183
184
183
187
188
188
190
191
192
194
194
196
197
199
200
201
202
203
204
204
206
206
208
210
211
213
214
213

217

219
220
221
222
224

223
230
232
233
234
236
238
239
C13H24 lsomar
C13HZ6 I88mer
C14HJ8 lsoner



3-mechylnaphcnalene + n_-trldecane
C12H22 UoMr
a-mechylnaphchaleae
C14H30 l"B8r
C14H26 I38mar
C,4H^8 Isomer
C7-«lkyl cyclohexane Isoner
C14H26 lsonmr
C14K30 lSOBar
C1SH30 tsooer
C15H32 lsooer
C..H28 laomer
n-cecradecane
Cj-alkyl naphchalene Isomer
C13H30 I80mer
C2-alkyl naphchalene lsomar
CjjHj,, Homer
C13H28 tsoaer
C1SK28 lMBBr
Cg-»lkyl cyclohexane Isoaer
C..H.4 laomer
C13H28 isoa*T
C16H32 l80Mr
n,-pencaddcane
C.-alkyl naphchalene
C..H.. Isomer
IS 30
C.-alkyl naphchalene + ^\(f
laoaera
C16H30 " C17H34 1»°°«"
C^-alkyl cyclohexane laomar
C16H30 llolMr


























32




alkyl phchalace Isomer (BKC)
2.2.4-crlmechyl penca-1.3-dlol
dl-laobucyrace (BKC)
ii-hexadecane
C17HJ4 laomer
C H isoBsr
C.Q-alkyl cyclohaxan* ijeraa
C17H34 * C17H32 I80mt"
C.gH.. laomer
n-hepcadecane
C17H]4 laomer (tent.)




r




                                C-45

-------
    TABLE  3  (Cont'd).
VOLATILE ORGRNICS IN DIESEL BLOW-BY  GAS3
        SAMPLE 5
C.-iromjco-
.jrjphic
Peak Nu.
116
116A
117
118
119
ilution
T«np.
CO
240
240
240
240
240
Compound
C.gH.. iioraer
C19H3S lsomer
C,.-ilkyl cyclohexane liomar
ja-octadccane
C.nH.A ISOBBT
19 40
Chroma co- Eluclon
jrjphlc Temp. Compound
Peak No. CO





Analysis was performed on a 100 m glass SCOT column coated with OV-101
 programmed from 20-240'BC @ 4°/min.  Carrier flow  (He) was 2.25 ml/min.
                                     C-46

-------
               APPENDIX D

CATERPILLAR TRACTOR COMPANY REPORT TO EPA
      ON DIESEL CRANKCASE EMISSIONS

-------
                                   CATERPILLAR TRACTOR CO.
                                   TECHNICAL CENTER
                                   Peoria. Illinois 61629
                                  February 10, 1976
Mr. C. T. Hare, Manager
Advanced Tech.
Department' of Emissions Research
Southwest Research Institute
8500 Culebra Road
San Antonio, TX   78228

Dear Mr. Hare:

During my February 4 visit to Southwest Research Institute you indicated
that EPA remains curious about crankcase blowby gas from diesel engines
as an atmospheric pollutant.  Enclosed is a copy of my June 19, 1974,
letter to John McFadden and a copy of a Caterpillar report dated
June 6, 1974.

I hope this report will contribute to your investigation.  We have done
no further checking since this report was written since we are convinced
that diesel blowby is an insignificant air pollution source.  If you
have questions, let me know.

                                  Very truly yours,
R. D. Henderson
Telephone:  (309) 578-6112
jcc

Encs.
                                  Emissions Control Manager
                                  Engineering G.O.
                                 D-2

-------
               REPORT BY






       CATERPILLAR TRACTOR CO.






                  to






U. S. ENVIRONMENTAL PROTECTION AGENCY






       OFFICE OF AIR PROGRAMS






             TO SUPPLY






    INFORMATION ON DIESEL ENGINE






     CRANKCASE BLOWBY ANALYSIS








                 AND






          DISPOSAL SYSTEMS
           JUNE 6, 1974
                   D-3

-------
Introduction

The purpose of this report is to provide the U. S. Environmental Protection

Agency with diesel engine crankcase blowby information.  This is necessary

because diesel blowby gas may be mistakenly considered an air pollution

source of sufficient magnitude to justify future regulatory control.  Currently

the blowby gas from most di'esels is discharged to the atmosphere.



Our measurements show that Idiesel engine crankcase blowby gas contributes

very little to total engine emissions so control is not justified from

that standpoint.  Blowby gas disposal systems are used in some instances

by Caterpillar Tractor Co. for commercial reasons but are not readily

adapted to all engines, particularly those that are turbocharged.



Discussion

In the gasoline engine, crankcase blowby gas is a significant source of

pollution, particularly hydrocarbons, while in the diesel it is not.  This

is primarily because of the fundamental difference in the method of mixing

air and fuel.   In spark ignition engines the mixing takes place in the

carburetor resulting in a nearly homogenous fuel-air mixture flowing into

the cylinders during the intake stroke.  Thus, all  cavities in the cylinder,

including the volume above the top ring between the piston and the cylinder
                   i
wall, are filled with the fuel mixture which isn't completely burned

during the power stroke.  Some of this unburned fuel escapes into the

crankcase and is contained in the blowby gas.
                                     D-4

-------
In the diesel engine, fuel is not injected into the cylinder until  late
in the compression stroke so cylinder cavities such as the one around the
piston above the top ring contain air only.  The diesel also operates with
excess combustion air so the fuel tends to be more completely consumed.
Thus, the diesel engine tends to have lower amounts of HC in the exhaust
and insignificant HC in the crankcase blowby gas.


For better understanding of this diesel characteristic, the blowhy gas
from a precombustion chamber, turbocharged V8 engine was analyzed and
compared with the exhaust emissions using the EPA 13 mode diesel engi.ie
test as a basis.  The following table shows the results:


          Exhaust - g/hp hr   Crankcase Blowby - g/hp hr   Blowby/Exhaust (Ratio)
CO                2.74                  .0063                    1/435
HC                0.41                  .0125                    1/33
N02               4.75                  .0036                    1/1320


In this engine the mass flow of the exhaust was approximately 200 times that
of crankcase blowby.  In order for the blowby to add significantly to  total
diesel engine emissions,  the concentrations of CO, HC, and NO  would  have to
                                                              A
be extremely high.  We have concluded that the control of blowby gas is
unnecessary from an emissions standpoint.

However, there are instances where diesel blowby gas is controlled in  ways other
than discharging directly to the atmosphere.  The reasons are primarily
commercial.  Caterpillar Tractor Co. produces a medium duty naturally  aspirated,
V8 truck engine which has had a PCV system since its introduction.  This was
done at the request of a major customer.

                                       D-5

-------
In future low emission models of this engine we may not be able to use the




PCV system which directs the blowby gas into the inlet manifold.  This is



because the blowby may cause soot particles from recriculated exhaust gas to



deposit in the inlet manifold and ports to an unacceptable degree.








In a 500 hour test with EGR, excessive build up of deposits was experienced



in the inlet manifold and inlet ports as shown in the attached photographs.



As a result of this experience, EPA and Bureau of Mines (Bartlesville1) people



decided to discharge the blowby gases to the atmosphere when testing the




durability of a Caterpillar 1160 model engine: modified to reduce emissions.



EGR was one of the emission control methods utilized.   The test setup and



results are reported in EPA-460/3-73-010,  "Dmability of Advanced Emission



Controls for Heavy Duty Diesel and Gasoline Fueled Engines" by Fleming and




French.








Return of crankcase blowby gas to the inlet manifold of a turbocharged engine



presents significantly greater problems and may be completely impractical.



Routing the gas through the turbo compressor wheel is not a good practice



because dust may.be caused to collect and reducs the compressor efficiency



with resulting deterioration of engine performance.  In the most extreme



case, balance of the very high speed compressor wheel may be affected.








If blowby gas is introduced into the intake manifold after the compressor, a



pump is necessary because of the higher pressure in the manifold.  Even if



a pump is used, there is still some likelihood of fouling the surfaces of the
                                     D-6

-------
after-cooler core when one is used.  Caterpillar currently does not consider



return of blowby gases to the intake manifold in turbocharged engines a



satisfactory (or practical control method.








Although crankcase blowby gas is discharged to the atmosphere in most



Caterpillar engines, this is done only after careful attention to breather



design.  Oil mist in the breather discharge is avoided by careful location



of the outlet from the engine, by an effective system of baffles, and by



maintaining suitably low gas velocity.  The essentially oil free gas contains



insignificant amounts of pollutants which escape from the combustion chamber



into the crankcase as shown earlier in this report.








Conclusion



Preventing discharge of diesel crankcase blowby gas to the atmosphere is not



justified from an emission control standpoint.  When a PCV control system



returning the gas to the intake manifold is used for commercial reasons, it



should not be considered an emission control device.
                                      D-7

-------
mENGINEERIIMG
LEiS.   RESEARCH
                                                   LEFT BANK CYLINDER
                                                   HliAl) PLENUM CHAMBER
                                                   DEPOSITS AFTER 500
                                                   CYCLE ENDURANCE HPS
     RIGHT  BANK CYLINDER
     HEAD PLENUM CHAMBER
     DEPOSITS AFTER 500
     CYCLE  ENDURANCE HRS
                                             VIEW OF CROSSOVER  (INTAKE)  PIPE
                                             FROM BOTTOM -  500  IKDURS
                                             BLOWBY UNDUCED  HERE


                                             NOTE:  BLOWBY FLOWED MOSTLY TO
                                                   LEFT BANK OF CYLINDERS.
                                                   ECR INLET PORT 'iS PARTIALLY
                                                   BLOCKED  CAUSING UNEVEN BANK
                                                   TO BANK  RECIRCULATION
                                  D-8

-------
             APPENDIX E




INFRARED SPECTRA OF LUBRICATING OILS

-------
                                             WAVELENGTH
KJ
I
to
                                                                                                15   20  30 50
               4000 3800 3600 3400  3200 3000 2800 2600  2400 2200 2000 1800 1600 1400  1200   00  800  600   400

                                                   FREQUENCY (CM'1)
20
                                                                                                             0
SPECTRUM NO._£fc£L_
SAMPLE (a\)-7(
/V^FW /£>/<-


ORIGIN

PURITY
PHASE
THICKNESS -,92- s-
LFfiFND
1.
2.
DATE S-3-7L
OPERATOR L>.&
RFMARKS




[OCO
>-o
71 ?2
c
•y
p


-------
                                 WAVELENGTH
                                                          LM)
  25
100
                                                                                        15  20  30 50
                                                                                         J
Q|iiiiiimii!iiimiiim|MiiiimiMimii'MMl|liiHilMi"''"'i'i!iip'"'''"'''"
4000 3800 3600  3400  3200  3000 2800 2600  2400  2200 2000  1800  1600  1400 1200  1000  800   600   400  200

                                        FREQUENCY  (CM'1)
SPECTRUM NO. 
-o
Q
~2_
0


-------
I
                 25
               100
                80
             $40
                20
                 0
WAVELENGTH        (MM)
       4             5
itmiiiiiiini: Minimi inn mi
                              lju  m
                          Ml          it
                      HiiiUII  i I    I I 111 I'    II!I!!1!!
                            lilllmllllmliliitlilimilll!!!!!!!!!!!

                                                         6     7   8  9 10
                                                                i    T i. I • i
                                I!!!!!!         Juluiii iiiiiiii  !!ji|
15  20  3050
' Itit-Htnttil ^i tilth 00
               80
                 4000  3800  3600  3400 3200  3000 2800  2600 2400  2200  2000  1800 1600  1400  12(X   1000
                                                            FREQUENCY (CM"1)
                                                                                      600   400   200
SPECTRUM NO. 90 3L
SAMPLE /WCL-3S7)



ORIGIN

PURITY
PHASF
THICKNESS .^>7<-
LEGEND
1.

DATE g-3-7^
OPFRATOR J>.A
REMARKS




coco
II
m ^f
c
•yr
O


-------
 2.5
WAVELENGTH
      4
15  20  3050
                                                                                              100
4000  3800  3600 3400 3200  3000  2800 2600 2400 2200 2000  1800  1600  1400  1200 1000 800  600  400  200
                                     FREQUENCY (CM*1)
SPECTRUM NO-J^^i.
SAMPLE /V/^-J5»
Llsi=D OtL.


ORIGIN

PURITY
PHASE
THICKNESS 0L*-
LEGEND
1.
9
DATE f-J-7^
OPERATOR D./?
REMARKS




SAMPLE
SPEQRUM NO.

-------
                              WAVELENGTH
                                                                              15  20  3050
                                                                                           20
4000  3800 3600 3400 3200 3000  2800 2600 2400  2200 2000  1800  1600 1400 1200  1000  800  600  400  200

                                    FREQUENCY (CM1)
SPECTRUM NO. ?3o
SAMPLE &V/-7/
A/fvs £>tt-

•
ORIGIN

PURITY
PHASE
THICKNESS ^JC_
LEGEND
1.
9
DATE 9-i£-7(,
OPERATOR ^A
REMARKS
i



                                                                                          J*~ • I»
                                                                                          iq

                                                                                          "c



                                                                                            O

-------
M
  2.5

100
                                             WAVELENGTH

                                                 .  f          . ?
6     7   8  9 10      15  20  30 50
      i      • i.i       •i    i.i.i
                                                                                                           100
                                                                                                          0
               4000 3800  3600 3400  3200 3000 2800  2600 2400 2200 2000  1800 1600 1400  1200  1000 800  600  400   200

                                                   FREQUENCY (CMM
SPEaRUM NO._Z^1_
SAMPLE ^,l/-7/
2/S
-o
m
i/i
-o
m
Q
PD
c
p

-------
                                        WAVELENGTH
                                                                           6       7    8   9  10        15   20  30  50
                                                                                             './..I....!..;..'  '*	L  •  '   •'
                                                                                                           miiiiiiiiinuniiiit
                                                                                                                           100
Q|MIMIIlM'MIIIIIIIMil|inil!l|IIMII|iinil'M|T:"M1MMI!IIM'll!!'|'MIIMiMMIIllMlllMI|i!l|li;il||ll|lllinil.M|l»IIIMII|!|llll»liini[IUI|ll!!ltnillllllMII|ll!lllll!ll'|t|Q

4000  3800  3600  3400  3200  3000  2800  2600  2400  2200  2000  1800  1600  1400  1200  1000  800   600   400   200


                                                FREQUENCY (CM'1)
SPECTRUM NO. //&6
SAMPLE t-vin \



ORIGIN

PURITY v^o.-V
PHASE .icp'.A
t.
THICKNESS .^*".
LEGEND
1.
7
DATE T/PO/ ''1
OPERATOR L>'
REMARKS




1
?
m
^
>
X.
SPECTRUM NO..1
i

-------
  25
loofe
*&
3
WAVELENGTH
       4            5
6     7   8   ? 10
                                                                                           15  20  3050
                                     i     iluimiililjiiiimiitilMimiiiiiiitniiiiiitiiiiiHii
                                      I II ill

    4000  3800 3600 3400 3200 3000  2800  2600  2400 2200 2000  1800  1600 1400  1200  1000

                                           FREQUENCY  (CM'1)
                                                                                          600   400  200
SPECTRUM NO. /H*1
SAMPLF e^ X



ORIGIN

PURITY wccV
PHASE l.au.-.rl
R
THICKNESS .oJT»,v¥s
LEGEND
1
?,
DATE S/^ <•'//'/
OPERATOR l.>*
RFMARKS





SPECTRUM NO..
SAMPLE CT/HD


^.
tT
-\

-------