APTD-156-
        OPTIMUM WORKING FLUIDS
                    FOR AUTOMOTIVE
                    RANKINE  ENGINES
VOLUME  I  - EXECUTIVE  SUMMARY
           U.S. ENVIRONMENTAL PROTECTION AGENCY
                Office of Air and Water Programs
          Office of Mobile Source Air Pollution Control
       Advanced Automotive Power Systems Development Division
                 Ann Arbor, Michigan 48105

-------
                                                    APTD-1563
                    Prepared By

     D. R. Miller, H.  R.  Null, Q.  E.  Thompson

           Monsanto Research Corporation
             800 North Lindbergh  Blvd.
            St. Louis, Missouri    63166
              Contract No.  68-04-0030
               EPA Project Officer:

                   K.  F.  Barber



                   Prepared For

       U.S.  ENVIRONMENTAL PROTECTION AGENCY
         Office of Air and Water  Programs
   Office of Mobile Source Air Pollution Control
Advanced Automotive Power Systems  Development Division
              Ann Arbor,  Michigan   48105

                    June  1973

-------
The APTD (Air Pollution Technical Data) series of reports is issued by
the Office of Air Quality Planning and Standards, Office of Air and
Water Programs, Environmental Protection Agency, to report technical
data of interest to a limited number of readers.  Copies of APTD reports
are available free of charge to Federal employees, current contractors
and grantees, arid non-profit organizations - as supplies permit - from
the Air Pollution technical Information Center, Environmental Protection
Agency, Research Triangle Park, North Carolina 27711 or may be obtained,
for a nominal cost, from the National Technical Information Service,
5285 Port Royal Road, Springfield, Virginia  22151.
This report was furnished to the U.S. Environmental Protection Agency
by Monsanto Research Corporation in fulfillment of Contract No. 68-04-0030
and has been reviewed and approved for publication by the Environmental
Protection Agency.  Approval does not signify that the contents necessarily
reflect the views and policies of the agency.  The material presented in
this report may be based on an extrapolation of the "State-of-the-art."
Each assumption must be carefully analyzed by the reader to assure that it
is acceptable for his purpose.  Results and conclusions should be viewed
correspondingly.  Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
                           Publication No. APTD - 1563
                                       ii

-------
                         ACKNOWLEDGEMENT
The authors acknowledge with gratitude the helpful counsel of
Messrs. K. Barber, S. Luchter and F. P. Hutchins, all of the
Environmental Protection Agency's Advanced Automotive Power Systems
Development Division.  Many persons within Monsanto Research Corp-
oration and Monsanto Company contributed substantially to the
research program reported here.   These included R. C. Binning,
J. A. Conover, L. L. Fellinger,  R. L. Green, R. L. Koch, R. J.
Larson, G. J. Levinskas, R. A. Luebke, N.  F. May, J. T. Miller,
L. Parts, A. C. Pauls, P. F. Pellegrin, J. V. Pustinger, W. R.
Richard, A. D. Snyder, W. N. Trump, W. M.  Underwood, J. A. Webster,
E. P. Wheeler and F. J. Winslow.  Their manifold inputs are
gratefully recognized, as also are those of Professor P. A. S.
Smith of the Department of Chemistry, University of Michigan, who
served as a consultant.

Also recognized are the contributions made by D. B. Wigmore and
R. E. Niggemann of Sundstrand Aviation, Division of Sundstrand
Corporation, both in the Research Program and in the Engine
Optimization Studies performed for this report.
                               iii

-------
                            CONTENTS





Volume I.   Executive Summary                                 Page



     Introduction                                               1



     Summary                                                    2



     Conclusions                                               27



     Recommendations                                           30





Volume  II.   Technical Section



Volume III.   Technical Section - Appendices



Volume  IV.   Engine Optimization
                                iv

-------
                          INTRODUCTION
This is the final report of a research program performed by
Monsanto Research Corporation for the Environmental Protection
Agency, Advanced Automotive Power Systems Development Division
(AAPSDD) under Contract 68-04-0030.  The objective of the work
was to determine the best working fluids for Rankine cycle
automotive power plants.

The Rankine Engine has been identified by EPA as a potential
alternative to the internal combustion engine for the automotive
application in the event the internal combustion engine cannot
satisfy the 19.75/76 Federal Emission Standards.  Development work
has been proceeding on the Rankine engine, as well as other alter-
native power plants, under the direction of AAPSDD since initiation
of the Clean Air Act Amendments of 1970.

In a Rankine cycle power plant, the working fluid, which is
repeatedly vaporized, expanded and recondensed, plays a key role
in determining the competitiveness of this system.  The thermo-
dyriamic, chemical and physical properties of the fluid have a
strong influence on the power plant efficiency, materials of con-
struction, component weight and size, the degree of hazard, if
any, and to some extent the power plant cost.  Therefore, to make
certain that the Rankine engine could be evaluated at its greatest
potential, this study program was initiated to search out the best
fluids currently available for the automotive Rankine engine appli-
cation.

To accomplish this objective, specific guidelines were established
for the fluid screening process.  These guidelines, contained within
the contract work statement, covered all significant fluid char-
acteristics: availability, compatibility, thermal stability, cycle
efficiency, cold start ability, applicability in principal engine
types, owner cost, toxicity, flammability and environmental hazard.
Quantitative targets set for these characteristics were intended to
approximate the upper limit of what could be achieved with presently
known materials and current technology.  The results indeed reveal
that the criteria were well chosen, as only two possible fluids from
over 100 candidates survived the screening process, and these did
not totally satisfy all of .the stated requirements.  In reviewing
the results it should be kept in mind that less stringent or
different selection criteria could very well have led to the
selection of different candidates.

The volume of reportable material resulting from this program has
dictated the physical subdivision of the detailed technical matter
into three separate volumes (II through IV) with this Volume I
serving as an executive introduction and summary.

-------
                            SUMMARY


A translation of system and working fluid contractual goals into
fluid screening criteria was the first step toward meeting con-
tract objectives.  Contractual goals are listed in Table 1, while
the derived screening criteria are given in Table 2.  A detailed
discussion of the derivation of the criteria is given in Volume II.
It may be noted that the screening criteria are mainly of the go/
no-gO) variety useful in the early elimination of unsuitable
materials.  Additional requirements on cost, hazards, compatibility
and system operability were treated on an ad hoc basis involving
only a few of the more .promising candidate fluids.

On the basis of predicted thermal stability and melting/boiling
points, 112 candidate fluid constituents were identified and,
where possible, procured.  Included in the candidate list
(Table 3) for comparison are several organic fluids previously
used or suggested as Rankine working fluids.  Except for water,
all identified candidates are organic chemicals.  Although
inorganic candidates were sought and a list of 32 possibilities
was similarly compiled, none was judged to be compatible with
low-cost metals at the 712°F minimum peak fluid temperature
needed to meet cycle efficiency requirements.

Binary and ternary mixtures of several of the 112 constituent
candidates were also identified with the expectation that such
mixtures offer wider latitude in matching diverse fluid re-
quirements .

Identified fluids were next subjected to screening tests under
the first five criteria.  Criteria 1 through 3 required mainly
literature work.  A special steel ampoule thermal exposure test
was employed to screen fluids for 720°F thermal stability under
Criterion 4.  Criterion 5, designed to distinguish between fluids
requiring "reasonable" and "too-much" regeneration, drew on
published thermophysical properties or, when these data were not
available, on recognized property prediction procedures to com-
pute the I-factors.

Candidate screening results under Criteria 1 through. 5 are
summarized in Table 3-  Here it is seen that the majority of
initial candidates failed to survive 200 hours at 712°P in the
Criterion 4 steel ampoule test.  The few which did are listed in
Table 4 with their corresponding capsule survival times, I-factors
(at normal boiling points), and Criterion 3 (normal boiling point)
results.  Brackets in the I-Factor column delineate the fluids
satisfactory per se (innermost bracket) from those suitable only
as components of a mixed (or complex) working fluid (outermost
bracket) and from those satisfactory as neither, all based upon
the tolerable regenerator size, Criterion 5-

-------
       Table 1.  PROPULSION SYSTEM AND WORKING FLUID GOALS
Propulsion System Characteristic
  Type Expander
  Approx. Net2 Full Power
  Max. Weight
  Max. Volume
  Regenerator Size
  Min. WF6 Condensing Temp.
  Max. Low Ambient Temp.
  Min. High Ambient Temp.
  Max. Starting Time (idle at -20°F.
  Max. Starting Time (65% power) @ 60°F,
  Average Annual Usage
  Life Expectancy
Working Fluid Characteristic
  Presently available?
  Min. Carnot Cycle Efficiency
  Min. Ideal Cycle Efficiency
  Max. Cost5 to Owner, 5 yr. period
  Compatible with Low Cost Materials?
  Compatible Lubricant Needed?

  Health Hazard
  Fire and Explosion Hazard
  Environmental Hazard
     Goal1
1-stage recip. and
1-stage turbine
1453 HP
1600 Ib
35 cu ft
Nil or small
220°F
-40°F
125°P
25 sec
*J5 sec
350 hr
3500 hr
Yes
     Goal1
30%
$100
Yes
Yes - recip.
Maybe - turbine
Nil
Nil
Nil
NOTES:
1.  Prime sources:  1971 Vehicle Design Goals (EPA) as amplified in
    consultation with EPA
2.  Gross expander shaft less feed pump shaft
3-  Sundstrand calculation based on performance requirements of
    1971 Vehicle Design Goals (EPA)
4.  Prime source:  contract work statement
5-  Initial factory fill plus materials only for any subsequent
    refills
6.  "Working Fluid"

-------
          Table. 2.  WORKING FLUID SCREENING CRITERIA


1.  Fluid component(s) known and available at reasonable cost.

2.  Flow point1 below:
    a)  20°C (68°F) for mixture component

    b)  -29°C (-20°F) for final working fluid

3.  Normal boiling point 65 to 120°C (150 to 250°F)

4.  Stable in mild steel ampoule test for at least 200 hours
    at 720°F.

5.  I-Factor2 (at normal dew point) between:

    a)  0.65 arid °° for mixture component

    b)  0.75 and 1.5 for final working fluid

6.  Reference ideal cycle efficiency at least 30% (UAk < »)

7.  Equivalent real cycle efficiency - report only - (UAk =
    125 Btu/HP-hr-F)

8.  Isentropic expansion enthalpy drop no greater than 200 Btu
    Ib  (turbine expander only, equivalent real cycle)

9.  Isentropic expansion density ratio no greater than 25
    (reciprocating expander only, equivalent real cycle).
NOTES:

1.  Lowest temperature at which fluid will freely flow (melting
    po.int in case of pure compound).

2.  Parameter related to slope of saturated vapor curve on T-S
    diagram (1 = vertical)

-------
  Table 3.  SUMMARY OF SCREENING TEST RESULTS, CRITERIA 1 to 5

                                                           «
                                                    Result,
   Defined Organic Component                     Criterion No.
FSN           Name                              I  1  3  1  5
 1   benzotrifluoride                           0  0  0  X  X
 2   o-bromobenzotrifluoride                             3
 3   m-bromobenzotrifluoride                    0  0  X  X
 4   p-bromobenzotrifluoride                    0  0  X  X
 5   2-bromo-l,4-difluorobenzene                00X6
 6   o-bromofluorobenzene                       0  0  X  X
 7   m-bromofluorobenzene                       0  0  X  X
 8   bromopentafluorobenzene                    0  0  X  X  X
 9   p-bromofluorobenzene                       0  0  X  X
10   p-bromophenyl trifluoromethyl ether        0  0  X  X
11   o-chlorobenzotrifluoride                   0  0  X  X
12   p-chlorobenzotrifluoride                   0  0  X  X
13   m-chlorobenzotrifluoride                   0  0  X  X  X
14   o-chlorofluorobenzene                      0  0  X  X  0
15   m-chlorofluorobenzene                      0  0  X  X  0
16   p-chlorofluorobenzene                      0  0  X  X  0
17   chloropentafluorobenzene                   0  0  0  X  X
18   1,2-dichlorohexafluorocyclobutane          0  0  X  X  X
19   2,3-dichlorooctafluorobutane                       18
20   1,2-dichlorooctafluorocyclohexene-1                20
21   1,2-dichlorotetrafluorocyclobutene-l       0  0  0  X
22   o-difluorobenzene                          00000
23   p-difluorobenzene                          00000
24   m-difluorobenzene                          00000
25   2,4-difluorotoluene                        0  0  0 " 0  X
26   2,5-difluorotoluene                        0  0  0  0  X
27   2,6-dimethylfluorobenzene                  0  0  X  0  X
28   2,3-dimethylfluorobenzene                  0  0  X  0  X
29   3,4-dimethyIfluorobenzene                  X
30   l,3-di(trifluoromethyl)benzene             0  0  0  0  X
31   l,4-di(trifluoromethyl)benzene             0  0  0  X  X

        0  means  pass
        X  means  fail
       42  means  failed by implication by FSN-42

-------
Table 3.  SUMMARY OF SCREENING TEST RESULTS. CRITERIA 1 to 5 (Cont'd.)
   Defined Organic Component
 FSN          Name
  32  o-fluoroanlsole
  33  m-fluoroanlsole
  34  p-fluoroanisole
  35  fluorobenzene
  36  p-fluorobenzotrifluoride
  37  m-fluorobenzotrifluoride
  38  o-fluorobenzotrifluoride,
  39  fluoroether No. 3
  40  2-fluoro-3-methylpyridine
  41  p-fluorophenyl trlfluoromethyl ether
  42  2-fluoropyridine
  43  o-fluorotoluene
  44  p-f luorotoluene
  4 5  m-f luorotoluene
  46  hexafluorobenzene
  47  methylpentafluorobenzene
  48  octafluorotoluene
  49  pentafluorobenzene
  50  pentafluoropropanol-1
  51  perfluoro(methylcyclohexane)
  52  perfluoroalkane-70
  53  perfluorotributylamine
  54  perfluoro-n-hexane
  55  perfluoro-2-butyltetrahydrofuran
  56  perfluoro(dimethylcyclohexanes)
  57  phenyltrlfluoromethyl ether
  58  phenyltrlfluoromethyl sulflde
  59  1,2,4,5-tetrafluorobenzene
  60  1,2,3,5-tetrafluorobenzene
  61  1,2,3,4-tetrafluorobenzene
  62  2,3,5j6-tetrafluorotoluene
    Result,
 Criterion No.

I  2  1  4  5_
0  0  X  X  X
0  0  X  X  X
0  0  X  X
00000
        37
0  0  0  X  X
0
X

0
0
0
0
0
0
0
0
0
0
0
0


0
0
0
0
0
0
0
0
0
0
0
0


0
X
0
0
0
0
0
0
0
0
0
X

42

X
X
X
X
0
X
0
0
X
0
X


X

X
X
X
0
X
X
0

X
0  0  0  X  X
0  0  X  X  X
0  0
0
0
0
X
0
0  0
0  0
0  0  OX  X
0  0  X  X
00000
0000
0  X  X
         0 means pass
         X means fail
        42 means failed by implication by FSN-42

-------
Table 3.  SUMMARY OF SCREENING TEST RESULTS. CRITERIA 1 to 5 (Cont'd.)
   Defined Organic Component
 FSN          Name
  63  trifluoroacetophenone
  64  1,2,4-trifluorobenzene
  65  1,3>5-trifluorobenzene
  66  2,2,2-trifluoroethanol
  67  tris(trifluoromethyl)-s-triazine
  68  resorcinol bis(trifluoromethyl)  ether
  69  3-chloropyridine
  70  3-methylpyridine
  71  pyrazine
  72  pyrlmldine
  73  toluene
  7*1  methanol
  75  1-propanol
  76  benzene
  77  monochlorobenzene
  78  pyridine
  79  tetrachloroethylene
  80  trichloroethylene
  81  trichloroethane
  82  2,2,3,3-tetrafluoropropanol
  83  perfluorosuccinic anhydride
  84  perfluoroglutaric anhydride
  85  undecafluoropropoxy-2-propanol-l
  86  2,4,6-trifluoropyrimidine
  87  hexafluoroacetone trihydrate
  88  hexamethyldisilazane
  89  perfluoroether P-ID (Allied)
  90  water
  91  acetic acid
  92  morpholine
  93   ethanol
        0 means pass
        X means fail
        2 means failed by implication by FSN-42
   Result,
Criterion No.
1
0
0
0
0
0
X
0
0
0
0
0
0
0
0
0
0
0
0

0
X
0
X
0
2
0
0
0
0
0

0
0
X
X
0
0
0
0
0
0
0
0

0

0

0
3
X
0
0
0
0

X
X
0

0
0
0
0
X
0
X
0

0

0

0
4_
X
0
0
X
0

X
0


0
X
X
0
X
0
X
X
80
X

X

X
5.

0
0

X

0
0
0

0
0
0
0
0
0
0


0

X

0
Corrosive
0
0
0
0
0
0
0
0
0
0
0
0
X
X
0
0
X
0
X
0
0
X
X
X
X
X
0
X
0


-------
Table 3.  SUMMARY OP SCREENING TEST RESULTS. CRITERIA 1 to 5 (Cont'd.)
   Defined Organic Component
 FSN          Name
  94  neopentanol
  95  ethylene glycol monomethyl ether
  96  anisole
  97  thiazole
  98  N-methylmorphollne
  99  thlophene
 100  1,1,1,2,1,1,1-heptafluoro-2,3,3-tri-
      chlorobutane
 101  perfluoroheptene-1
 102  2,3,5,6-tetrafluorophenol
 103  pentafluorophenol
 101  N-nonafluorobutyloctafluoromorphollne
 105  2-methylpyridine
 106  1-methylpyrldine
 107  pentafluoropyridlne
 108  2-chlorothiophene
 109  2,6-dimethylpyridine
 110  2-methylpyrazine
 111  hexamethyldisiloxane
 112  trlmethyl borate
    Result,
 Criterion No.

0
0  0
0  0
         I  5_
0  X  0  X  X
0  0  X  X  0
0  0  X  X
X
0  0  0  X  X
000X0

0  0  0  X
0  0  0  X
0  X  X
0  X  X
0  0  0  X
0  0  X  0  X
00X00
   0  0
   0  0
X  X
X  0  0
0  0  X  X
0  0  0  0  X
0  0  0  X
   Defined Mixtures
 MFSN        Components
  1   FSN-73 and PSN-78
  2   FSN-71 and PSN-90
  3   PSN-90 and FSN-93
  1   FSN-78 and FSN-91
  5   FSN-75, FSN-76 and FSN-90
  6   FSN-76, PSN-90 and FSN-93
  7   FSN-73 and FSN-75
  8   FSN-75 and FSN-76
000
0  0  0  X  X
000X0
0  0  0  X  X
0  0  0  X
0  0  0  X
0  0  0  X
0  0  0  X
         0 means pass
         X means fail
        142 means failed by implication by FSN-12

-------
Table 3.  SUMMARY OF SCREENING TEST RESULTS. CRITERIA 1 to 5 (Cont'd.)
Defined Mixtures
MFSN
9
10
11
12
13-
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
FSN-73
FSN-74
FSN-75
FSN-74,
FSN-73
FSN-71
FSN-78
FSN-23
FSN-30
FSN-61
FSN-23
FSN-66
FSN-70
FSN-71
FSN-90
FSN-90
FSN-90
FSN-90
FSN-46
FSN-46
FSN-70,
FSN-90,
FSN-70,
Components
and FSN-74
and
and
FSN-76
FSN-90
FSN-78 and FSN-90
and
and
and
and
and
and
and
and
and
and
and
and
and
and
and
and
FSN-76
FSN-90
FSN-90
FSN-78
FSN-78
FSN-78
FSN-96
FSN-90 ("Fluorinol")
FSN-90
FSN-90
FSN-105
FSN-106
FSN-109
FSN-110
FSN-107
FSN-49
FSN-90 and FSN-105
FSN-105 and FSN-106
FSN-90 and FSN-106
Result ,
Criterion
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
0
0
0
X
0
0

0
0
0
0
X
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
X
0
0

0
0
0
0

0
0
X
0
0
0
0
0
0
n
X
X
X
X
0
X
0
X

X
X
X
0
X
0
0
0
X
0
0
0
0
0
*
No.
5


0

0

0

X

X

0

0
0


0
0
0
0
0
         0 means pass
         X means fail
        42 means failed by implication by FSN-42

-------
     Table 4.  THERMALLY STABLE COMPONENTS AND I-PACTORS
PSN              Name
 56  perfluoro(dimethylcyclohexanes)
 89  perfluoroether P-ID (Allied)
111  hexamethyldisiloxane
 67  trls(trifluoromethyl)-s-trlazlne
 51  perfluoro(methylcyclohexane)
 30  1,3-di(trifluoromethyl)benzene
 28  2,3-dimethylfluorobenzene
 27  2,6-dimethylfluorobenzene
 48  octafluorotoluene
105  2-methylpyridlne
 25  2,4-difluorotoluene
 26  2,5-difluorotoluene
 73  toluene
 46  hexafluorobenzene
 1|9  pentaf luorobenzene
 59  lf2,1,5-tetrafluorobenzene
 60  1,2,3,5-tetrafluorobenzene
107  pentafluoropyridine
109  2,6-dimethylpyridine
 64  1,2,4-trifluorobenzene
 65  1,3,5-trlfluorobenzene
 35  fluorobenzene
 22  o-difluorobenzene
 23  p-difluorobenzene
 24  m-difluorobenzene
 70  3-methylpyrldlne
106  4-methylpyridine
 78  pyridine
 76  benzene
 90  water

        tt 0 = pass, X = fail
Stable
Hrs . at
720°P
>336
>336
312
>312
>336
>336
>288
>336
>336
>288
>336
>336
>336
>336
>336
>336
>336
>336
>288
>336
>336
>336
>336
>336
>336
288
240
>336
>336
NBPt
I- Result^
Factor Grit. 3
0.27 0
0.28 X
0.35 0
0.36 0
0.40 0
0.42 0
0.50 X
0.50 X
0.55 0
0.56 X
0.60 0
0.60
0.66
0
0
0.68 0
0.71 o
0.74
0.75
0.75
0.75
0.76
0.76
0.77
0.78
0.79
0.81
0.82
0.82
0.87
0.89
2.81
0
0
0
X
0
0
0
0
0
0
X
X
0
0
0
                             10

-------
  Chemicals in the lower part  of Table  4,  i.e.  the primary  auto-
  motive candidates,  all fall  in several  chemical  categories  with
  the characteristics shown in Table  5.
         Table 5-  ADVANCED FLUID COMPONENT CLASSIFICATIONS


  I.  Aromatics (benzene, toluene, pyridine and
                 methylpyridines)

      A.  inexpensive (articles of present commerce)

      B.  flammable

      C.  water solubility

          1.   Benzene, toluene - insoluble

          2.   Pyridines - soluble in all proportions

      D.  recognized toxicity

 II.  Polyfluorinated aromatics (fluorinated benzene, toluene,
                                 pyridine and methylpyridines)

      A.  inherently expensive, several in limited production

      B.  Flammability

          1.   fluorobenzene,  difluorobenzenes, trifluoro-
              benzenes - flammable in normal air (by flash
              point test)

          2.   tetra-, penta-, hexa-fluorobenzenes, penta-
              fluoropyridine, octafluorotoluene - non-
              flammable in normal air (by flash point test)

      C.  water insoluble

      D.  toxicity not well established, thought to be of
          same order as parent aromatics

III.  Water
                                  11

-------
Of  the simple  fluids  within the innermost I-factor.bracket in
Table 4, not all meet the Criterion 2 requirement for final
working fluids (flow  point below -29°C).  Those which do are
listed in Table 6 with their melting points  (same as flow point)
and an indication of  flammability in normal  air.


                Table 6.  SIMPLE FLUID CONDIDATES


                                  Melting      Flam.
FSN                               Ft. (°C)     Test*

 60  1,2,3,5-tetrafluorobenzene     -48          0

107  pentafluoropyridine            -44          0

 64  1,2,4-trifluorobenzene         -32          X

 35  fluorobenzene                  -42          X

 22  o-difluorobenzene             -34          X

 24  m-difluorobenzene             -78          X

 78  pyridine                       -42          X
     •flashpoint in normal air; 0 = pass, X = fail


When a requirement for non-flammability is added to Criteria 1 to
5, then, the only simple candidates to emerge are:

     l,2,3»5-tetrafluorobenzene, and pentafluoropyridine

While test amounts of these materials are available, their prices
are so high as to suggest that even high-volume production would
not lower them enough to meet owner cost requirements of $100/5 y
supply.  This added impetus to a search for mixed (or complex)
fluids to take advantage not only of the wider choice of physical
parameters so afforded, but also of the possibility of lowering
the cost below that of the most expensive component.  Water is an
especially attractive component of such a mixture since it can be
expected to reduce flammability and toxicity as well as cost.

Some 31 mixtures were defined at various stages of the search
(Table 3).  Most of these were eliminated from consideration when
they failed the Criterion 4 test, or were found to contain com-
ponents which had failed.  Two more (MFSN's 1 and 13) are quite
flammable and were eliminated on that basis.  The remaining group
is constituted as in Table 7.


                                12

-------
                Table 7.  COMPLEX FLUID CANDIDATES


               1.  hexafluorobenzene with

                   a.  pentafluorobenzene (MFSN-28)

                   b.  pentafluoropyridine (MFSN-27)

               2.  water with

                   a.  pyridine (MFSN-15)

                   b.  2-methylpyridine (MFSN-23)

                   c.  3-methylpyridine (MFSN-21)

                   d.  4-methylpyridine (MFSN-24)

                   e.  binary combinations of b,  c and d
                       (MFSN-29,  30, 3D

Each of the three methylpyridines violates either the I-factor
or boiling point criterion.   Retention of these as candidate
components is justified primarily on the grounds  that they are
the only water-soluble organics (other than pyridine itself)
found to have reasonable thermal  stability.

Predictive computer routines were used to enable  cycle evaluations
of the advanced candidates of Tables 5, 6 and 7.   Their functions
are listed in Table 8.

These routines were used extensively to analyze the behavior of
advanced candidates in the "Reference Ideal Cycle", defined as
follows:

     Reference Ideal Cycle - An ideal Rankine cycle (clockwise
     rectangle on a pressure-entropy diagram) operating between
     a top (bulk fluid) temperature of 712°F and  a bottom tem-
     perature of 220°F (saturated liquid at condenser outlet)
     and with the peak fluid pressure chosen as the lesser of:

     a.  that just preventing expansion penetration into the
         2-phase region;

     b.  1000 psia
                                13

-------
                Table 8.  COMPUTER ROUTINES EMPLOYED


1.  Estimation of pure compound

    a.  critical constants (P, T, V)

    b.  perfect vapor heat capacity coefficients

    c.  vapor pressures

    d.  liquid densities

    e.  enthalpies of vaporization

2.  Calculation of liquid activity coefficients from vapor
    pressure data on particular complex fluids.

3.  Calculation of vapor phase properties by a special general-
    ization of the Redlich-Kwong equation of state to cover com-
    plex as well as simple vapor mixtures.

4.  Prediction of principle PVT/thermodynamic functions of either
    simple or complex fluids, using as input, data found in the
    literature, or generated as in 1 to 3 above, and having as an
    option (when treating simple fluids) the use of the Hirschfelder,
    Buehler, McGee and Sutton formulation.

5-  Calculation of specified ideal or real Rankine cycles (only
    with the Hirschfelder et al. formulation).

6.  Plotting of temperature-entropy diagrams for any fluid processed
    in i».

-------
Reference ideal cycles calculated for many of the Table 5-7
candidates are compiled in Table 9.  The first column is a test
of Criterion 6, which requires an ideal cycle efficiency of at
least 30% at any level of regenerator effectiveness.  All straight
organic fluids pass, whereas the pyridine-water mixtures fail.
While water itself appears disqualified, it is capable of higher
efficiencies by raising its upper operating temperature.

Other entries in the table demonstrate compliance with certain
interim (non-criterion) goals; namely,

     1.  condenser pressure from 5 to 50 psia

     2.  ideal expansion density ratio less than about 40
         (to serve in a single expansion reciprocator)

     3.  ideal expansion enthalpy drop less than about 200
         Btu/pound (to serve in a single-stage impulse turbine)

     4.  I-factor in the approximate range of 0.65 to » com-
         ponents) or 0.75 to 1.5 (final fluids)


The ideal Rankine cycle with a fully effective regenerator is
only of limited applicability in evaluating working fluids,
especially as concerns cycle efficiency.  A 100% effective re-
generator is by definition infinite in size.  Since the space
available for an automotive regenerator is severely restricted
in the engine compartment, and since the cycle efficiency is
highly sensitive to extent of regeneration (as well as to ex-
pander efficiency), a more realistic "Equivalent Real Cycle"
was defined as follows:

     Equivalent Real Cycle - A real Rankine cycle which matches
     the Reference Ideal Cycle at the two extreme corners (ex-
     pander and pump inlets),  has engine and pump efficiencies
     of 75% each, has pressure losses in the regenerator, vapor
     generator and condenser representative of those predicted
     by system contractors (in January 1972),  and has an adiabatic
     pump and expander.  The "size" of the regenerator may be
     limited by restricting its q/At (log mean) ratio (or equiv-
     alently its UAk product)  to a "reasonable value".

Equivalent  Real Cycle calculations employing many of the same
simple fluids recorded in Table 9 are compiled in Table 10.   Com-
parison of  the entries with the reciprocating/turbine suitability
Criteria 8  and 9 reveals all fluoroaromatics and benzene suitable
for either  engine type and all aromatics but benzene suited only
for turbine use.
                                15

-------
Table 9.  SUMMARY OF REFERENCE IDEAL CYCLE CALCULATIONS ON THERMALLY STABLE CANDIDATES
    Aromatic3
      Benzene
      Toluene
      Pyridine
      3-Methylpyrldine
    Fluoro Aromatlcs
      Fluorobenzene
      m-Dlfluorobenzene
      p-Dlfluorobenzene
      1.2,3,5-Tetrafluoro
        benzene
      1,2,4,5-Tetrafluoro
        benzene
      Pentafluorobenzene
      Octafluorotoluene
      Pentafluoropyrldine
      Hexafluorobenzene

    Water
Pyridine/Water Solutions

  6056
  30$
  60%
  30%
  60%
           Pyridine

           2-Methylpyridine

           3-Methylpyrldine
      30%
      60% f4-Methylpyridine
                               Ideal
                               Cycle
                              Efflc.E*
31.8
31.7
31.5
32.1
31-9
31.6
31-7

30.6

30.8
30.3
30.2
30.3
31.1

20.5
27.4
28.3
27.9
27-9
27-3
27.7
27.6
28.6
          Condenser
          Pressure
           (psia)
                                       29.4
                                       12.3
                                       10.5
                                        4.2
                                       25.9
                                       28.5
                                       23.6

                                       28.2

                                       22.8
                                       26.6
                                       11.9
                                       28.2
                                       30.5

                                       17.3
            21.3
            19.6
            22.8
            21.6
            19-3
            15-9

            13.9
            14.3
                                                 Expansion
                                                  Density
                                                   Ratio
                         40.3
                        145-9
                        145.1
                        338.1
                         42.4
                          5.8
                                  Expansion
                                  Enthalpy
                                    Drop
                                  (Btu/lb)
 81.3
 79-3
 93-9
 97.3
 13-0
226.0
           I-Pactor
              at
            220°F
0.85
0-75
0.98
0.89
46.6
43-7
54.6
70.1
56.9
59-1
0.65
0.70
0.71
0.76
54.7
15-2
86.9
41.6
40.3
44.8
35.3
29-7
38.2
35.2
0.77
0.68
0.55
0.75
0.54
3.34
36.3
47.1
33.8
49.4
39.1
65-4
40.1
76.4
154.9
125.9
160.8
107.3
167-1
112.3
165-5
114.8
1.66
1.33
1.48
0.89
1.75
1.21
1.73
1.20
           *  Full regeneration on all simple fluids: full regeneration or regeneration
              only to the dew line, whichever occurs first, on all solutions.
          **  Concentrations shown are mol percents.
                                         16

-------
Table 10.  SUMMARY OF EQUIVALENT REAL CYCLE CALCULATIONS ON
                    SIMPLE THERMALLY STABLE FLUIDS
Aromatics
benzene
toluene
pyrldine
3-methylpyridlne
Fluoro Aromatics
f luorobenzene
m-dif luorobenzene
p-d if luorobenzene
1 j 2, 3 > 5- tetraf luorobenzene
pentaf luoropyridine
hexaf luorobenzene
Notes: 1. Regenerator Q/AT(log
Real
Cycle
Effic.(
22.8
22.7
23-7
24.8
22.1
21.9
22.3
21.2
20.9
20.7
mean)
Expansion
Density
%1} Ratio2
15-1
46.4
13-5
83.6
17.0
16.2
19-5
15.7
15.4
15-2
= UAk = 125
Enthalpy
Drop,( Btu/lb)
57.4
56.5
67.3
70.2
49-5
40.2
41.8
30.4
27-0
24.8
Btu/HP-hr-°F
2. Isentropic, from engine inlet to exhaust enthalpy
3. Condenser pressures
Table 9
and 220
°F I-factors
same as in
                              17

-------
A searching examination of the complex fluid candidates listed in
Table 7 was carried out to further limit the number of final can-
didates.  This included experimental studies of flammability,
toxicity and freeze (flow) point, in addition to the development
of criteria data discussed above.

The first choice was obvious.  Of the nonflammable fluids, only
the highly fluorinated benzenes stand a reasonable chance of
approaching the cost goal in large-scale production.  Hexa-
fluorobenzene is too high melting to be used alone, but mixtures
with tri-, tetra- and pentafluorobenzenes possess the requisite
low flow points.  Such mixtures actually result from the fluor-
ination of benzene, so processing economies could be expected.
As a particular model representing this class, the first final
Rankine candidate fluid, designated RC-1, was fixed as follows:

                              RC-1

                                      Concentration
        Constituent        Mol Wt     Mol '%    Wt %

    pentafluorobenzene     168.1      60.0     57-5

    hexafluorobenzene      186.1      40.0     42.5


The only obvious way to avoid using the expensive fluoro compounds
and still secure some measure of fire hazard protection is to
employ an organic-water solution.  Only the pyridines, with their
water solubility, permit this approach.  Mixtures involving two
or more of the three methylpyridines and pyridine itself, all con-
templated in this category, offer some advantage in processing
economics because of the simultaneous production of all four in
practice.  In consideration of freeze (flow) point depression,
cycle utility, and flash/fire points, it appeared that a mixture
of 2-methylpyridine in water was the best binary choice, so as
the second final candidate the representative model defined as
follows was chosen:

                              RC-2

                                      Concentration
        Constituent        Mol Wt     Mol %

    water                   18.0      65.0

    2-methylpyridine        93-1      35-0
                                18

-------
Thorough experimental and computational characterization of the
two final candidates followed their definition.  Physical property
measurements included flow points, vapor pressures, liquid densities3
specific heats, viscosities and thermal conductivities.  Selected
values appear in Table 11.  Computations using these and pure com-
ponent literature data then provided such derived properties as
vapor viscosities and thermal conductivities, ideal vapor heat
capacity coefficients, and liquid activity coefficients.

Thermodynamic tables and T-S diagrams were next prepared, and the
two principal Rankine cycles were calculated, as summarized in
Table 12.  It is significant to compare the cycle efficiencies of
the two fluids.  While RC-2 fails the 30$ ideal cycle efficiency
stricture, it actually shows a slightly higher real cycle effi-
ciency than does RC-1.  The main reason is that the fixed "size"
regenerator is more effective in recovering the lesser superheat
of the RC-1 fluid.  This supports the original use of the I-factor
screening criterion.

Application-oriented test work was also undertaken to further
explore candidate utility.  Included were stability-compatibility
tests in dynamic pumping loops and steel ampoules, animal acute
toxicity experiments, flammability tests, lubrication studies, and
manufacturing cost estimations.  Highlights are listed in Table 13-

Fluid RC-2 was found to be incompatible with SAE 4130 steel boiler
tubing at 720°F.  Failure of the steel to passivate was evidenced
by formation of hydrogen gas and magnetic iron oxide.  An attempt
to prepassivate the steel with water/steam reduced, but did not
stop, the water-iron reaction.

An idea of the inhalation toxicity hazards of the two final can-
didates was garnered from comparisons of the rat LC50 results with
(a) expected inhalation doses in a hypothetical "worst case"
Rankine automobile accident involving driver entrapment and fluid
release, and (b) known inhalation accidents involving common vol-
atile fluids.  The inhalation hazards of both candidates are con-
sidered comparable to those of gasoline.  Products of flame de-
composition of each fluid were identified.

Attempts to find lubricants with all characteristics desired in
the reciprocating engine application were largely unsuccessful.
Desired characteristics were (a)  stable, compatible, and a liquid
at 712°F; (b)  incompletely soluble in the working fluid at ambient
temperature; and (c)  solidification point (in suspension in working
fluid) below -20°F.  All lubricant candidates known to be stable
above 700°F were completely miscible in RC-1 and most were highly
soluble in RC-2.  A number of highly silylated benzenes were syn-
thesized which had minimal solubility in RC-2 and thermal stability
comparable to the RC-2.   These compounds suffered from inherent
expense and too-high solidification points.
                                19

-------
        Table 11.  SELECTED PROPERTIES OF FINAL CANDIDATES
                                        Final Candidate
        Property
Average molecular weight
Flow point, °F
Normal boiling point, °F
Condensing pressure
  @ 220°F, psia
Heat of vaporization
  6 NBPt, Btu/lb
I-Factor at 220°F
Liquid density at
  77°F, g/ml
Specific heat at
  77°F, Btu/lb
 RC-1
175.3

172

 30

 79-1
  0.72

  1.549
  0.290
 RC-2
200

 21
378
  1.38

  0.985
  0.668
                                20

-------
Table 12.   FINAL CANDIDATE CYCLE CALCULATION SUMMARY
Final Candidate:
Cycle1 :
Cycle Efficiency
Per 100 Net Cycle HP
Fluid rate, Ib/hr
Eng. exhaust, cfm
Pump rate, gpm
Pump HP
Heat flows, kBtu/hr
Vapor gen.., kBtu/hr
Regen., kBtu/hr
Conden. , kBtu/hr
Engine, % Ef f ic . 2
Pressure ratio
Density ratio
Isentropic 3
Exhaust quality, %
Enthalpy drop, Btu/lb
Pump, % Effic.2
Pressure ratio
Regenerator, % Effect.
UAK, Btu/HP-hr-°F
Log mean temp, diff., °F
Notes: 1. "Reference ideal"
earlier
2. Input specified
3. From engine inlet
RC-1
Ideal Real
30.0 21.0

7470 11500
259 3^0
12.2 18.8
7.34 16.8 -
848 1220
391 692
409 95.0
100 75
36.6 29.8
44.6 37.9
44.6 15.4
176 190
36.6 25-8
100 75
36.6 42.1
95-3 .91.8
12.0 12. 52
RC-2
Ideal Real
28.8 22.2

1760 2520
233 310
3.93 5-66
2.43 5-04
883 1150
64.8 213
629 890
100 75
43.6- 35.4
35 -.6 32.9
35.6 12.1
108 118
149 106
100 75
43.6 50.1
100 98.4
12. 52
32.6 55-6 0 23.0
and "equivalent real" as defined


condition to exhaust enthalpy
                        21

-------
      Table 13-  PERFORMANCE PARAMETERS OF FINAL CANDIDATES
                                             Candidate
Stability. Compatibility

1000-hour dynamic loop test

  Failure mode

Toxic Hazards

Inhalation
  4-hr rat LC50, ppm

  30-min rat LC50, ppm

Oral, rat, LD50, mg/kg

Dermal, rabbit, LD50, mg/kg

Irritation score
  Rabbit skin (8 max.)
  Rabbit eye (110 max.)

Flammability

Flash point, modified COC, °F

Fire point, modified COC, °F

Auto ignition temp., °F

Hot compartment ignition
  rating (benzene3!*!)

Cost (to owner, 5 yr supply), est.$
                                        RC-1
                            Pass
    RC-2



    Fail

Decomposition
16,000
40,0002
12,000
>8,000
3
10
8,000*
-
8101
2501
5.6'
711
e
e
e
0
112
130
145
1,060
8
<100
Notes:
Experiments conducted with a 40/60 mol % mixture of
2-methylpyridine/water

Experiment conducted with a 50/50 mol % mixture of
penta-/hexafluorobenzene by Imperial Smelting Corp-
oration
                                22

-------
A low cost synthetic fluid - biphenyl/terphenyl eutectic - was
evaluated as a miscible lubricant in friction-wear experiments
involving cast iron rubbing against various typical mating
materials.  This material has promise as a miscible lubricant in
either RC-1 or RC-2.  No petroleum-based lubricants can be ex-
pected to survive the high temperatures demanded in this appli-
cation.

Fairly detailed economic studies were carried out to project the
ultimate cost to the driving public of the RC-1 fluid.  Present
manufacturing techniques are not suited to high volume low-cost
production.  A considerable technological investment would be
necessary to bring the cost to the level indicated in Table 13-
The RC-2 fluid is inherently cheaper and is currently abundant.

Ancillary experimental studies were carried out to probe the
ultimate thermal stability of "Fluorinol 85" with and without
added refrigeration oil.  The upper temperature limit of these
fluids was placed near 650°F.  Another study of a class of semi-
organic heat stable fluids called "carboranes" revealed a
flammability deficiency.

The two final candidate fluids were subjected to a Rankine Cycle
Engine Optimization Analysis where the size, weight, and efficiency
of an engine for each of these fluids was determined.  Fluid RC-1
was used to optimize an engine with a turbine expander and fluid
RC-2 was used to optimize an engine with a reciprocating expander.
An optimum automotive engine was determined by using a criterion
that weighed engine size and weight against engine overall efficien
(miles per gallon) over a time-weighted average of five operating
load points representing country, suburban, and city driving con-
ditions .

The Sundstrand Rankine Cycle Design Optimization Program used is
a comprehensive analytical design procedure in which detailed
design point and off-design calculations are made, and components
meeting cycle requirements are calculated.  In the design procedure
a design point cycle is first calculated, from which engine com-
ponents are designed.  The performance of the system using these
components is then evaluated by means of off-design cycle calcula-
tions.  The overall system and its performance is then evaluated
against an optimizing criterion, the results of which are used to
vary the inputs to the design point cycle analysis.  A new design
point cycle is then calculated, components designed, and off-design
performance evaluated.  This procedure is repeated until an optimum
system, based on the optimizing criterion, is found.
                                23

-------
The first step in defining an optimum engine is to define the load
points for which the engine is to be optimized, and to define the
parameters of the optimizing criterion.  For the purposes of this
study only steady state load points were to be used, and the
optimizing criterion was a trade-off between system efficiency
and system weight.  The optimizing criterion was based on the
fact that high-efficiency systems tend to be high-weight systems,
and reduced-weight systems result in lower efficiencies.  As
system weight is directly related to system volume, system size
is also being optimized by this criteria.

The load points used in the analysis consisted of one design point
load and four off-design load conditions.  These load points were
derived from the Federal driving cycle and EPA's prototype vehicle
performance specification.

The load points selected were:

                                          Speed    Power    % Time

Design Point  maximum power acceleration  70 mph   108 HP     0.5
Off-design 1  steady speed                60 mph    32 HP    16.5
Off-design 2  steady speed                30 mph   8.4 HP    33-0
Off-design 3  acceleration              25-1 mph  9-92 HP    28.0
Off-design 4  acceleration              l.?4 mph 1.143 HP    22.0

The power is the power output from the transmission to the drive
shaft, the engine power required being determined by the mechanical
losses, accessory requirements and the cycle calculations.  These
load points were also used to calculate the fuel load required for
a 200 mile range.

The optimizing criterion was constructed using miles-per-gallon as
representative of engine efficiency, weighed against system weight
in pounds.  In the optimizing process a weighting factor was varied
to place more emphasis on system efficiency or system weight to
see how this affected the system configuration.  The optimum con-
ditions found in each calculation are listed in Tables 14 and 15.  "

The turbine engine was optimized as a supercritical engine assuming
a turbine inlet temperature of 712°F (required for a 200°F con-   v
denser in order to meet the EPA specifications of 42% carnot cycle
efficiency), and a 425 psia turbine inlet pressure.  This pressure
was selected to keep pump work to a minimum since the fluid prop-
erties did not indicate any significant benefit in increased
adiabatic head or reduced regenerator size by going to a higher
pressure.  The turbine tip speed was limited to 1200 ft/sec to
allow economical materials, and the condenser assumed was the
AiResearch design for the EPA Thermo Electron engine.
                                24

-------
Table 14.  PRINCIPAL RESULTS_OF__ENGINE OPTIMIZATION CALCULATIONS
     Engine Type Working Fluid
Weighting Factor, Effic. (Vs Weight)

Overall Miles/Gallon

Weights, Pounds
  Expander and Gear Box
  Condenser and Pan
  Regenerator
  Burner and Vapor Generator
  Transmission and Drive Train
  All Other Components and Fuel

       Total System

Volumes, Cubic Feet
  Expander and Gear Box
  Condenser and Fan
  Regenerator
  Burner and Vapor Generator
  Transmission
  Unoccupied Space

       Total System
Turb.
RC-1
 0.5
 173
 195
 380
 355
 322
1571
Turb.
RC-1
0.97
 178
 172
 234
 368
 355
 319
1626
Recip.
RC-2
 0.5
13.33   13.81   18.89*
 225
 142
  36
 281
 355
 322
1361
0.23
7-55
4.13
9.58
1.50
15.^1
38.3
0.28
7.71
5-01
8.83
1.50
15.57
38.9
2.25
5-56
0.82
9.79
1.50
13-3
33-2
     Based on 3% of stroke clearance volume and square wave
     operation of valves porting 20$ of piston area
                                25

-------
Table 15.  OPTIMIZED DESIGN POINT PARAMETERS
Engine Type Working Fluid
Weighting Factor, Ef f Ic . (Vs Weight)
Condensing Temperature, °F
Regenerator Press. Drop/Press. In, %
Regenerator Effectiveness, %
Condenser Effectiveness, %
Condenser Air Power, kW
Expander Speed
Tip Speed, ft/sec
Pitch Diam. , in.
RPM (3 in. stroke)
Admission
Arc/Total Circum. , %
Intake Angle, Degrees
Flue Gas Press. Split (Vap. Gen/
Total) %
Turb.
RC-1
0.5
179
1.27
93
81
5-48

677
4.96
-

7.7-1
-
80.4
Turb.
RC-1
0.97
179
1.27
95
81
4.93

677
6.00
-

100.
-
89.4
                                               Recip
                                               RC-2
                                                0.5
                                                242
                                               1.43
                                                 87
                                                 80
                                                5.0
                                               1900


                                                 54

                                               72.6
                      26

-------
The optimization analysis for the turbine engine was performed
over six values of the optimization weighting factor (which can
vary from 0 to 1; the higher the value, the greater the emphasis
on high efficiency).  The result of this analysis showed that there
was no appreciable variation in the weight of the engine or in
miles per gallon over the.range of the weighting factor.  It was
also seen that there was little variation in the design point
condenser temperature, which varied from 179°P to 200°F.  It
appeared from an analysis of the results that this situation was
the result of the restriction of condenser size to one fitting
the frontal region of a normal engine compartment.  The size
restriction on the condenser so severely restricts the condenser
size for the power level required that essentially the same size
condenser results with all values of the optimization weighting
factor.  As a result, the condenser influence on the rest of the
power system is such that there is very little variation in the
optimum engine weight and efficiency throughout the range of the
optimizing weighting factor.

The reciprocating engine was analyzed as a subcritical engine
using RC-2.  The design assumed an expander inlet temperature of
712°F and a 700 psia inlet pressure for the same reasons as in
the turbine engine, and to keep cylinder bore diameter a reasonable^
size for an automotive engine.  The reciprocator was a single
acting, single expansion engine whose speed was limited to 2000
rpm to keep valve train forces within reasonable limits.  The
condenser core was that of the AiResearch design developed for
the Thermo Electron engine.

As can be seen by the miles  per gallon, the reciprocating engine
shows very encouraging results.  However, it must be taken into
consideration that the simplified reciprocator model used assumed
a three percent of stroke clearance volume and "square wave"
valve operation that had flow areas equal to one-fifth the piston
area.  These rather optimistic assumptions result in low valve
losses and high volumetric efficiency which may be difficult to
attain in an actual engine.   The result of these assumptions is a
high efficiency expander which results in the high miles per
gallon.figures.   It should be borne in mind when comparing these
figures to that of the turbine, that the turbine expander model
used was a well developed turbine analysis that gave realistic
results based on current state-of-the-art technology.
                           CONCLUSIONS
1.  No organic or aqueous organic fluids were found to completely
    satisfy all the requirements established in the contract for
    Rankine cycle automotive power plant working fluids.
                                27

-------
Fluid RC-1 comes nearest to satisfying the automotive organic
working fluid requirements and goals set by AAPSDD.  It meets
all of the established performance criteria for both turbine
and reciprocating engines, including efficiencies, temperature,
pressure and density limitations, and regenerator size.
Additionally, it is considered fire-safe and unlikely to cause
serious injury on accidental human exposure (subject to
verification in extended toxicity testing recommended as a
follow on).  The fluid is quite stable at temperatures to at
least 720°F in low-cost system materials, and it is expected
to last for at least the 5-year span of initial vehicle owner-
ship.  It possesses a distinct advantage over water in not
freezing at temperatures down to -4o°F.  A low-cost, soluble
lubricant is available if needed.  The main contractual
deficiency of the fluid is its projected cost of no less than
$112/over its expected five year life, as compared to a $100
contract limit.  To arrive at the $112 cost it was necessary
to assume (a)  successful advancement of the state of fluor-
ination technology beyond that now practiced, and (b)  the
existence of a 50 million pound per year benzene fluorination
plant.  Current prices of the higher fluorinated benzenes are
many times the projected price.

Turning to cycle practicality, the comparatively high flow
rate (Table 12) of the RC-1 fluid per net cycle horsepower
(related to its higher molecular weight) may be considered a
disadvantage despite the lack of a contractual limit.  All in
all, this fluid is worthy of further development.

Fluid RC-2 is a back-up candidate for automotive use, either
in a turbine or reciprocating engine.  It satisfies all per-
formance criteria except the 30% minimum ideal cycle efficiency.
Even so, the fluid is capable of high efficiency usage in a
real cycle with a compact regenerator.  Although the fluid is
flammable, it is less so than common fuels.  In burning it
releases considerably less energy on account of its water con-
tent, and its products of combustion are not hazardous to
humans.  The fluid itself is unlikely to cause serious injury.
It is marginally stable at ?20°F in steel.  SAE 4130 steel is
not a compatible material for boiler tube construction with
RC-2.  The fluid acts chemically to corrode this steel, forming
F6304 and hydrogen.  A more suitable steel alloy needs to be
found before this fluid can be considered a contender with
RC-1.  Assuming a satisfactory steel were found, the RC-2 fluid
would cost the vehicle owner about $100 in the first five years
of vehicle life.  This presumes three system refills.  Tradi-
tional lubricants cannot be used with this fluid above ?00°F,
but low cost synthetics offer promise when mutual solubility
is permissible.  Synthetic silane lubricants may be considered
if immiscibility is mandatory, but further development is
needed.  Fluidity above -40°F is another feature of importance.
The prime ingredient of the fluid is commercially available at

                             28

-------
    low cost now, but in lesser quantity than would be commanded
    by a million cars-per-year market.  RC-2 is worthy of further
    development only in the unlikely event that candidate RC-1
    must be abandoned.

^.  Many organic fluids possess the state and thermodynamic prop-
    erties needed to attain engine performance goals.

5.  Very few fluids possess the requisite thermochemical stability
    in steel at 712°F (and higher).  Those that do fall in these
    categories:

                       Aromatics

                       Fluoroaromatics
                       Aromatic tertiary amines

                       Perfluoroaliphatics
                       Water

6.  Flammability eliminates the aromatics from contention, and
    places the aromatic tertiary amines in a contingency category.
    Since the aromatic tertiary amines (or pyridines) are quite
    flammable and water soluble, their safe usage in automobiles
    is contingent upon a suitable reduction in flammability
    effected by mixing with water.

7.  An excessive demand for regenerator space to maintain tol-
    erable cycle efficiency rules out the perfluoroaliphatics.
    This is clearly shown by the I-factor criterion and the
    supporting cycle calculations.

8.   Additional restrictions on vapor pressure,  melting point
    (or "flow" point), cost, and toxicity further narrow the
    choice to
                       mixed fluorobenzenes and

                       aqueous pyridines,

    from which categories the final candidates  RC-1 and 2 were
    chosen.

9-   The search rationale adopted at the outset  was basically
    sound, although some adjustments were necessary.   Particu-
    larly valuable features were:

         a.   an efficient thermal stability screening program

         b.   development and effective use of the I-factor con-
             cept for thermodynamic screening - including con-
             firmation via I-factor - efficiency correlations

         c.   maximal use of computation throughout, but especially
             in these areas:

                                29

-------
                  prediction of thermodynamics of non-ideal/
                    ideal mixtures

                  I-factor and physical property predictions
                  ideal and real cycle calculations

     d.  the concept of progressive advancement of candidates,
         exposing each to increasingly rigorous examination.

10.  As shown by the optimization studies, both the RC-1 and RC-2
     final candidate fluids are appropriate fluids for automotive-
     size engines and are conducive to good efficiencies for
     automotive application.


                         RECOMMENDATIONS
Relative to fluid RC-1, it is recommended that

     1.  it be considered an acceptable organic fluid for a
         prototype engine development program.

     2.  static thermochemical stability tests be performed
         to learn how far above 720°F the fluid might be
         driven

     3.  providing prototype experience is favorable, the
         fluid be subjected to a carefully planned set of
         exposure tests in the areas of
                        chronic inhalation

                        chronic dermal contact

                        environmental fate

         as necessary to qualify the fluid for widespread use

     4.  encouragement be given to prospective manufacturers to
         develop the necessary low-cost processes for large scale
         production.

Relative to fluid RC-2 and assuming abandonment of RC-1, it is
recommended that:

     5.  additional work be undertaken to find a suitable steel
         and/or steel passivation technique to avoid the iron-
         water reaction at high temperature

     6.  if appropriate, additional synthesis work be done to
         finalize an immiscible silane lubricant
                                30

-------
 7.  a fire hazard simulation be performed - preferably in-
     volving a mock-up burner-vapor generator and a simulated
     high pressure fluid leak

 8.  the first three recommendations under RC-1 be followed

 9-  additional engine optimization studies be carried out
     using computer models of components which more closely
     represent that for automotive type advanced state-of-the-
     art designs.   These models would give a better descrip-
     tion of component weight, size, effectiveness, and
     efficiency.

10.  further engine optimization studies be performed that
     optimize more parameters than were done for this pro-
     gram.  A more detailed optimizing criteria would also
     be beneficial in determining an optimum engine.
                            31

-------
                                  TECHNICAL REPORT DAYA
                           (Please rrcJ IniLn.ctiuns o;i //(..• ri true beJ
1. REPORT NO.
    APTD  -
            1563
                                                          3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
  Optimum Working Fluids for Automotive Rankine Engines
  :Volume  I-  Executive Summary
                                                          5. REPORT DATE
                                                          6. PERFORMING ORGANIZATION CODE
7. AUTHORlS)

  D.R." Miller,  H.  R.  Null, Q. E  Thompson
                                                          8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 Monsanto  Research Corporation, SOONorth Lindbergh Blvd
  St. Louis,  Missouri  63166   and Sundstrand Aviation
  Rockford,  Illinois
                                                          10. PROGRAM ELEMENT NO.
                                                          11. CONTRACT/GRANT NO.
                                                             68 -  04 -  0030
12. SPONSORING AGENCY NAME AND ADDRESS
 U. S. ENVIRONMENTAL PROTECTION AGENCY
 Office  of  Mobile  Source Air Pollution Control
 Advanced Automotive Power Systems Development Division
 Ann Arbor,  Michigan.    *~48105	
                                                          13. TYPE OF REPORT AND PERIOD COVERED
                                                          14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
   The objective was to determine  the  best working fluids for Rankine cycle  automotive
 ppwer plants. This study program  was  initiated to search out the best  fluids  currently
 available for the automotive  Rankine  engine application. Specific guidelines  were
 established for the fluid  screening process. These guidelines covered  all significant
 fluid characteristics: availability,  compatibility, thermal stability, cycle  efficiency
 cold start ability, applicability in  principal engine types, owner cost,  toxicity,
 flammability and environmental  hazard.  This Volume i serves as an executive introductidjr
 and summary. The best Rankine candidate fluid, designated RC-1, was fixed as  follows.:
 pentafluorobenzene 60 Mol%, hexaflurorbenzene 40 Mol%. The best binary choice,  RC-2
 was fixed as follows: water 65  Mol%,  2-methylpyridine 35Mol%. Fluid RC-1 was  used  to
 optimize an engine with a  turbine expander and fluid RC-2 was used to  optimize  an
 engine with a reciprocating expander. The optimization studies indicate that  both  the
 RC--1 and RC-2 final candidate fluids  are appropriate fluids for automotive-size engine
 and are conducive to good  efficiencies  for autonotive application. Additional conclu-
 sions and recommendations  are given.  The results of the study are summarized  and
 presented in fifteen tables.
17.
                               KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                             b.lOENTIFIERS/OPEN ENDED TERMS  C. COSATI 1'iclU/GfoUp
 Air Pollution
 Rankine cycle
 Engines
 Design
 Turbines
 Reciprocating Engines
                                               Working Fluid
                                               Mixed Fluorobenzenes
                                               Aqueous Pyridines
    13 B
    21 G
    21 E
    14 B
     7 C
13. DISTRIBUTION STATEMENT

      Release  Unlimited
                                             19. SECURITY CLASS (This Report)
                                               Unclassified	
21. NO. OF HAGtS
       36
                                             20. SECURITY CLASS (This page)
                                             «-s Unclassified
                                                                        22. PRICE
EPA Form 2220-1 ,-V-73)
                                            32

-------