EPA-600/R-97-065
July 1997
NEW CHEMICAL ALTERNATIVE FOR OZONE-DEPLETING SUBSTANCES:
HFC-236fa
By:
N. Dean Smith, Theodore G. Brna, Cynthia L. Gage, and Robert V. Hendriks
U. S. Environmental Protection Agency
National Risk Management Research Laboratory
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
Prepared for:
U. S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
-------
TECHNICAL REPORT DATA , , . ....,.
(Please read Instructions on the reverse before comp \ ||| | ||| || | | |||| 11
1. REPORT NO. 2.
EPA-600/R-97-065
PB97-1
86308
4. TITLE AND SUBTITLE
New Chemical Alternative for Ozone-Depleting
Substances: HFC-236fa
5. REPORT DATE
July 1997
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
N. Dean Smith, Theodore G.Brna, Cynthia L.Gage,
and Robert V. Hendriks
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
See Block 12
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
NA (Inhouse)
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 12/96 - 4/97
14. SPONSORING AGENCY CODE
EPA/600/13
15.supplementary notes proiect officer is N. Dean Smith, Mail Drop 63, 919/541-
2708.
i6. abstract rep0rt gives results of a preliminary evaluation of a new hydrofluoro-
carbon (HFC)—HFC-236fa or 1,1,1, 3, 3, 3-hexafluoropropane--as a possible alterna-
tive for chlorofluorocarbon (CFC)-114 (1, 2-dichloro-l, 1, 2, 2-tetrafluoroethane) re-
frigerant for chillers and as a possible fire suppressant replacement for halon-1301
(bromotrifluoromethane). (NOTE: HFCs form a class of chemicals having the poten-
tial to replace stratospheric ozone depleting substances such as CFCs and hydro-
chlorofluorocarbons.) Evaluation tests included an examination of flammability,
stability, atmospheric lifetime, thermophysical properties, lubricant miscibility
and solubility, materials compatibility, inhalation toxicity, refrigeration perfor-
mance, heat transfer characteristics, and flame suppression. Results of these eval-
uations indicate that HFC-236fa is a viable alternative for CFC-114 refrigerant and
for halon-1301 or -1211 fire extinguishing agent. Its relatively long atmospheric life-
time may be a concern from a global warming perspective.
PROTECTED UNDER INTERNATIONAL COPYRIGHT
ALL RIGHTS RESERVED.
NATIONAL TECHNICAL INFORMATION SERVICE
U.S. DEPARTMENT OF COMMERCE
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. cosati Field/Group
Pollution Halohydrocarbons
Refrigerants Propane
Fire Extinguishers Hydrofluorination
Stratosphere
Ozone
Greenhouse Effect
Pollution Control
Stationary Sources
Hydrofluorocarbons
Global Warming
Hex af luor opr op ane
13B 07C
13 A
13L
04A
07B
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
- 62
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policyand
approved for publication. Mention of trade names
or commercial products does not constitute endorse
ment or recommendation for use.
-------
FOREWORD
The U. S. Environmental Protection Agency is charged by Congress with pro-
tecting the Nation's land, air, and water resources. Under a mandate of national
environmental laws, the Agency strives to formulate and implement actions lead-
ing to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet this mandate, EPA's research
program is providing data and technical support for solving environmental pro-
blems today and building a science knowledge base necessary to manage our eco-
logical resources wisely, understand how pollutants affect our health, and pre-
vent or reduce environmental risks in the future.
The National Risk Management Research Laboratory is the Agency's center for
investigation of technological and management approaches for reducing risks
from threats to human health and the environment. The focus of the Laboratory's
research program is on methods for the prevention and control of pollution to air,
land, water, and subsurface resources; protection of water quality in public water
systems; remediation of contaminated sites and groundwater; and prevention and
control of indoor air pollution. The goal of this research effort is to catalyze
development and implementation of innovative, cost-effective environmental
technologies; develop scientific and engineering information needed by EPA to
support regulatory and policy decisions; and provide technical support and infor-
mation transfer to ensure effective implementation of environmental regulations
and strategies.
This publication has been produced as part of the Laboratory's strategic long-
term research plan. It is published and made available by EPA's Office of Re-
search and Development to assist the user community and to link researchers
with their clients.
E. Timothy Oppelt, Director
National Risk Management Research Laboratory
i i i
-------
ABSTRACT
Hydrofluorocarbons (HFCs) form a class of chemicals having the potential to replace
stratospheric ozone depleting substances such as chlorofluorocarbons (CFCs) and
hydrochlorofluorocarbons (HCFCs). This report gives results of a preliminary evaluation of a new
HFC (HFC-236fa or 1,1,1,3,3,3-hexafluoropropane) as a possible alternative for CFC-114 (1,2-.
dichloro-l,l,2,2-tetrafluoroethane) refrigerant for chillers and as a possible fire suppressant
replacement for halon-1301 (bromotrifluoromethane). Evaluation tests included examinations of
flammability, stability, atmospheric lifetime, thermophysical properties, lubricant miscibility and
solubility, materials compatibility, inhalation toxicity, refrigeration performance, heat transfer
characteristics, and flame suppression. Results of these examinations indicate that HFC-236fa is a
viable alternative for CFC-114 refrigerant and for halon-1301 or halon-1211 fire extinguishing
agents. Its relatively long atmospheric lifetime may be a concern from a global warming perspective.
iv
-------
TABLE OF CONTENTS
Page
ABSTRACT iv
FIGURES V1
TABLES vii
SCIENTIFIC SYMBOLS AND ABBREVIATIONS v.i i i
ACKNOWLEDGMENTS ix
1. INTRODUCTION 1
2. SUMMARY AND CONCLUSIONS 6
3. THERMOPHYSICAL AND APPLICATION PROPERTIES OF HFC-236fa 9
SYNTHESIS OF HFC-236fa 9
THERMOPHYSICAL PROPERTIES 9
ATMOSPHERIC LIFETIME 17
INHALATION TOXICITY 18
FLAMMABILITY AND FIRE SUPPRESSION CHARACTERISTICS 26
HFC-236fa/POE LUBRICITY 27
HEAT TRANSFER CHARACTERISTICS 30
STABILITY AND MATERIALS COMPATIBILITY 31
4. REFRIGERANT PERFORMANCE 40
HFC-236fa AS A CFC-114 REPLACEMENT IN CHILLERS 40
THEORETICAL PERFORMANCE EVALUATION 40
CENTRIFUGAL COMPRESSOR EVALUATIONS 42
COMPRESSOR CALORIMETRY RESULTS OF CENTRIFUGAL
CHILLER SIMULATION 45
5. REFERENCES 51
v
-------
FIGURES
No. Page
1. Fourier-transform infrared spectrum of HFC-236fa 10
2. Mass spectrum of HFC-236fa 11
3. Volume change of elastomers in HFC-236fa 34
4. Weight change of elastomers in HFC-236fa 35
5. Linear swell of elastomers in HFC-236fa 36
6. Hardness change of elastomers in HFC-236fa 37
7. Coefficient of performance (theoretical) ratio of HFC-236fa relative to CFC-114 43
8. Volumetric capacity ratio of HFC-236fa relative to CFC-114 44
9. Experimental cooling capacity ratio of HFC-236fa relative to CFC-114 47
10. Volumetric efficiency ratio of HFC-236fa relative to CFC-114 48
11. Coefficient of performance (experimental) ratio of HFC-236fa relative to CFC-114 49
12. Compressor isentropic efficiency ratio of HFC-236fa relative to CFC-114 50
VI
-------
TABLES
No. Page
1. Chemical codes, formulas, and boiling points of the 37 new chemicals synthesized 3
2. Chemicals selected for further characterization 4
3. Thermophysical properties of liquid and vapor HFC-236fa 12
4. Summary of experimental data used in mBWR correlation of HFC-236fa 14
5. Coefficients for ancillary equations (1) - (4) 15
6. Wear tests results (unit is teeth of wear) 29
7. Extreme pressure (step test) results (lb) 30
8. Compatibility test matrix 32
9. Tested elastomers and plastics 33
10. Desiccant results with HFC-236fa 39
11. Properties of HFC-236fa and other CFC-114 alternatives 41
12. Centrifugal compressor characteristics at 4°C evaporating and 40°C condensing
temperature 45
vii
-------
SCIENTIFIC SYMBOLS AND ABBREVIATIONS
cP
centipoise
cSt
centistokes
cP
ideal gas heat capacity at constant pressure
Cv
ideal gas heat capacity at constant volume
COP
coefficient of performance
fpi
fins per inch
Ma2
Mach number
N
compressor speed (revolutions per minute)
NOAEL no observable adverse effect level
Pe
evaporating pressure
Pc
condensing pressure
PPm
parts per million
Pc
critical pressure
Pa
vapor pressure
Qe
compressor refrigerating capacity
R
universal gas constant
RMS
root mean square
rpm
revolutions per minute
Tc
condensing temperature
Tc
critical temperature
Tb
boiling point
compressor impeller tip speed
vs
volumetric flow rate, suction
Vd
volumetric flow rate, discharge
Pc
critical density
Pl
saturated liquid density
Pv
saturated vapor density
v i i i
-------
ACKNOWLEDGMENTS
Initial synthesis of HFC-236fa and preliminary determination of its thermophysical properties
were performed by Darryl D. DesMarteau and Adolph L. Beyerlein of Clemson University with the
joint sponsorship of the U. S. Environmental Protection Agency (EPA Cooperative Agreement
CR-815134) and the Electric Power Research Institute (EPRI). Extended thermophysical property
measurements and determination of the rate constant for reaction of HFC-236fa with hydroxyl radical
were contributed by the National Institute of Standards and Technology (NIST) under EPA
sponsorship (Interagency Agreement DW13935432). Krich Ratanaphruks, Michael W. Tufts,
Angelita S. Ng, and Richard Snoddy of Acurex Environmental Corporation performed the
flammability, thermal/chemical stability, materials compatibility, lubricant miscibility, and lubricity
evaluations under EPA Contracts 68-D0-0141 and 68-D4-0005. Georgi Kazachki, Evren Bayoglu,
and Rob Delafield of Acurex Environmental Corporation were responsible for the compressor
calorimeter tests under EPA Contracts 68-D0-0141 and 68-D4-0005. The New Mexico Engineering
Research Institute (NMERI) performed the flame extinguishment tests under EPA Cooperative
Agreement CR-817774. Toxicity tests were supported in part by funding from the Strategic
Environmental Research and Development Program (SERDP) and were conducted by DuPont
Haskell Laboratory and Zeneca Central Toxicology Laboratory under EPA Contracts 68-D2-0063
and 68-D4-0005, respectively.
ix
-------
1. INTRODUCTION
Fully halogenated chlorofluorocarbons (CFCs) and their bromine-contaixiing relatives (halons)
are recognized as primary contributors to depletion of the Earth's stratospheric ozone layer. As early as
1978, the United States, on the basis of what was then largely theoretical evidence for depletion of
stratospheric ozone by CFCs, responded by promulgating restrictions on the use of CFCs as propellants
in aerosol products. At that time, aerosol propellants represented the single largest commercial use of
CFCs. Only Sweden, Norway, and Canada joined the United States in enacting measures to reduce
CFC emissions at that time.
In the mid-1980's, as scientific evidence confirming stratospheric ozone depletion mounted, the U.
S. Environmental Protection Agency (EPA) began considering additional regulatory restrictions on the
use of CFCs and halons by weighing the environmental benefits against the economic and societal
consequences of such action. In the course of this study, it became apparent that few, if any, alternative
chemicals were readily available or had been proven applicable to the numerous CFC and halon uses
which had grown dramatically in the short time span following the 1978 CFC aerosol propellant ban.
In the Spring of 1987, the EPA convened a panel of international experts to assess the likelihood
of finding suitable alternative chemicals. Compounds in which one or more of the chlorine or bromine
atoms of the CFCs and halons were replaced by hydrogen atoms appeared to offer a high probability of
serving as successful alternatives. It was noted that the presence of hydrogen in the molecules would
increase the probability that the molecules would degrade in the lower atmosphere and be dissipated
before reaching the ozone layer. However, it was also noted that replacing halogen atoms with
hydrogen atoms could lead to undesirable flammability, toxicity, and/or reduced efficacy of the
compounds in certain applications. Among the findings of the expert panel was the conclusion that
governments should sponsor research to enhance knowledge of chemical substitutes, particularly
research into those chemicals that were not already under active consideration by chemical companies
(Nelson, 1988).
1
-------
Shortly after the landmark Montreal Protocol on Substances that Deplete the Ozone Layer was
negotiated in September 1987 (UNEP, 1987), the EPA's Air and Energy Engineering Research
Laboratory (AEERL)* initiated a search for new chemical alternatives. This effort was undertaken to
try to provide an expanded set of candidate alternatives in the event that the few chemicals which had
then been proposed as alternatives fell short of expectations.
In 1988, AEERL and the Electric Power Research Institute (EPRI) formed a cooperative project
with the chemistry departments of Clemson University and The University of Tennessee to synthesize a
number of partially fluorinated propanes and butanes and fluoroethers which, on the basis of molecular
structure and anticipated boiling points, were thought to be possible CFC or halon alternatives. These
compounds also possessed structural features believed to enhance degradation of the compounds in the
troposphere. Over a 3-year period, 37 compounds were prepared of sufficient stability and in sufficient
yield and purity to obtain a limited set of relevant property measurements. Of the 37 chemicals
synthesized, 15 were hydrofluoropropanes and butanes (HFCs), 8 were hydrofluoroethers (HFEs), 5
were fully fluorinated ethers (FEs), and 9 were hydrochlorofluoropropanes (HCFCs). These
compounds and their normal boiling points are listed in Table 1.
AEERL selected for further study 12 of the 37 synthesized chemicals whose measured boiling
points and critical temperatures most closely matched those of CFCs-11, -12, -114, and -115 or which
might form mixtures having desirable properties. Alternatives for CFC-113, widely used as a solvent,
were not emphasized since a number of chemicals and technologies appeared to already exist or be
emerging to replace this compound. Also, compounds which possessed chlorine (HCFCs) were ruled
out for further evaluation since such compounds had non-zero ozone depletion potentials and would also
be subject to future phaseout. Table 2 presents the 12 compounds selected by AEERL for further
examination.
* Recently redesignated the Air Pollution Prevention and Control Division of the National Risk
Management Research Laboratory, Research Triangle Park (NRMRL - RTP), NC.
2
-------
Table 1. Chemical codes, formulas, and boiling points of the 37 new chemicals synthesized.
Chemical Code
Chemical Formula
Tb (°C)
HFC-227ca
CF3-CF2-CF2H
-16.3
HFC-227ea
CF3-CHF-CF3
-18.3
HFC-236ca
CHF2-CF2-CHF2
12.6
HFC-236cb
CF3-CF2-CFH2
-1.4
HFC-236ea
CF3-CHF-CF2H
6.5
HFC-236fa
CF3-CH2-CF3
-1.1
HFC-245ca
CF2H-CF2-CFH2
25.0
HFC-245cb
CF3-CF2-CH3
-18.3
HFC-245fa
CF3-CH2-CF2H
15.3
HFC-254cb
CF2H-CF2-CH3
-0.8
HFC-329mccp
CF3-CF2-CF2-CF2H
15.1
HFC-338pccq
CHF2-CF2-CF2-CH2F
42.5
HFC-338mccq
CF3-CF2-CF2-CFH2
27.8
HFC-338meem
CF3-CHF-CHF-CF3
25.4
HFC-347mccs
CF3-CF2-CF2-CH3
15.1
HCFC-225ba
CF3-CFCI-CFHCI
51.9
HCFC-225da
CF3-CHCI-CF2CI
50.8
HCFC-226da
CF3-CHCI-CF3
14.1
HCFC-226ea
CF3-CHF-CF2C1
17.1
HCFC-234da
CF3-CHCI-CFHCI
70.1
HCFC-235ca
CF3-CF2-CH2CI
28.1
HCFC-243da
CF3-CHCI-CH2CI
76.7
HCFC-244ca
CF2H-CF2-CH2CI
54.8
cy-HCFC-326
cy-
-------
Table 1. Continued
Chemical Code
Chemical Formula
Th (°C)
HFE-125
cf3-o-cf2h
-34.6
HFE-134
cf2h-o-cf2h
4.7
HFE-143
cf2h-o-cfh2
29.9d
HFE-143a
CF3-0-CH3
-24.1
HFE-227ca
cf3-o-cf2-cf2h
-3.1
HFE-124B1
CF2H-0-CF2Br
24.5
FE-115B1
CF3-0-CF2Br
-5.4
FE-116
CF3-O-CF3
-58.7
FEE-218
CF3-0-CF2-0-CF3
-9.8.
cy-HFE-225
cy-CHF-CF2-0-CF2-
3.4
cy-HFE-234
cy-CH2-CF2-0-CF2-
21.2
cy-FE-216
cy-CF2-CF2-0-CF2-
-28.2
cy-FEE-216
cy-CF2-0-CF2-0-CF2-
-22.1
Note: Chemical codes for the ether compounds have not been standardized. The listed codes are
AEERL designations, "d" = decomposes.
Table 2. Chemicals selected for further characterization.
Alternatives for:
Chemical Code
Chemical Formula
Chemical Name
CFC-1 l/HCFC-123
HFC-245ca
CF?H-CF9-CFH?
1,1,2,2,3-pentafluoropropane
HFC-245fa
cf3-ch2-cf2h
1,1,1,3,3-pentafluoropropane
HFC-338mccq
cf3-cf2-cf2-cfh2
1,1,1,2,2,3,3,4-octafluorobutane
CFC-12
HFC-227ca
CFvCF9-CFoH
1,1,1,2,2,3,3 -heptafluoropropane
HFC-227ea
cf3-chf-cf3
1,1,1,2,3,3,3-heptafluoropropane
HFC-245cb
cf3-cf2-ch3
1,1,1,2,2-pentafluoropropane
HFE-143a
cf3-o-ch3
1,1,1-trifluorodimethyl ether
CFC-114
HFC-236cb
CF^-CF?-CFH9
1,1,1,2,2,3-hexafluoropropane
HFC-236ea
cf3-cfh-cf2h
1,1,1,2,3,3-hexafluoropropane
HFC-236fa
cf3-ch2-cf3
1,1,1,3,3,3-hexafluoropropane
HFC-254cb
cf2h-cf2-ch3
1,1,2,2-tetrafluoropropane
CFC-115
HFE-125
CF3-0-CF2H
pentafluorodimethyl ether
Extended evaluation of these 12 candidates was undertaken by AEERL with emphasis on their
potential use as refrigerants and as blowing agents for insulation foams. Expanded evaluation included
4
-------
determination of atmospheric lifetimes, inhalation toxicities, chemical stabilities, material
compatibilities, vapor thermal conductivities, lubricant miscibilities, and refrigeration performance.
This report summarizes results obtained for one of the candidates, HFC-236fa (1,1,1,3,3,3-
hexafluoropropane), as a potential alternative refrigerant for CFC-114 in chillers and as a possible
replacement for the halon fire extinguishing agents halon-1301 (bromotrifluoromethane) and halon-1211
(bromochlorodifluoromethane).
5
-------
2. SUMMARY AND CONCLUSIONS
Theoretical analysis and experimental evaluations in a semihermetic compressor confirm that
HFC-236fa should be considered as a replacement for CFC-114 in shipboard chillers. HFC-236fa has
a higher refrigerating capacity and coefficient of performance (COP) than CFC-114 at shipboard chiller
conditions because of its higher volumetric efficiency in the semihermetic compressor and is, therefore,
not an exact drop-in replacement. The HFC-236fa volumetric capacity and the measured compressor
capacity with HFC-236fa are higher than with CFC-114 at condensing temperatures below about 80°C.
Although the theoretical HFC-236fa COP is lower than that of CFC-114, the experimental COP is
higher than or equal to that of CFC-114 at condensing temperatures up to 70°C. There is some drop-off
in performance of HFC-236fa at higher condensing temperatures.
HFC-236fa contains no chlorine or bromine and therefore has zero potential to deplete
stratospheric ozone. Its measured reaction rate at 298 K with hydroxyl (OH) radical was 1.1 x 10"14
cm3 molecule"1 sec"1 as determined by the National Institute of Standards and Technology (NIST), and
0.034 x 10"14 cm3 molecule"1 sec"1 as determined by the Jet Propulsion Laboratory (JPL) of the
National Aeronautics and Space Administration (NASA) (JPL, 1994). The favored value is that of JPL
due to small amounts of impurities present in the material examined by NIST which may have
contributed to a higher than expected rate constant. Accepting the JPL measured rate constant gives an
atmospheric lifetime due solely to reaction with an OH radical of 192 years relative to an OH
atmospheric lifetime for methyl chloroform of 6.6 years. A lifetime of this duration gives cause for
concern from a global warming perspective. The Intergovernmental Panel on Climate Change (EPCC,
1995) has assigned a global warming potential (GWP) of 6300 to HFC-236fa (100-year horizon, GWP
of C02 = 1).
Considerable toxicity testing has been completed for HFC-236fa due in large part to the U. S.
Navy's interest in retrofitting its shipboard CFC-114 chillers with HFC-236fa. Toxicity tests included
6
-------
acute inhalation, cardiac sensitization, genetic toxicity, developmental inhalation, and 90-day
subchronic inhalation. Maximum concentration of the chemical administered to rats and rabbits for the
inhalation toxicity evaluations was 50000 ppm. The one notable compound-related effect was a
diminished response or lack of response by the test animals to an alerting stimulus during exposure.
Although rats exposed to 50000 ppm were generally non-responsive and rats exposed to 20000 ppm
have a diminished response during the first week of exposure of the 2-week study, most animals
exhibited normal alerting responses during the second week. Any diminished response effect was
completely reversible upon cessation of exposure. A maximum concentration of 200000 ppm of HFC-
236fa in air was used for the cardiac sensitization tests using male beagle dogs as the subjects where
half of the dogs indicated positive responses. There were no positive responses when the concentration
was at 100000 ppm or less. Based on all toxicity tests performed, it is concluded that HFC-236fa
should pose no significant toxicity problems.
HFC-236fa was included in a matrix of four HFCs (HFCs-236fa, -236ea, -245ca, and -245fa),
and one HFE (HFE-125) examined for thermal and hydrolytic stability and materials compatibility.
These tests showed that, with and without a polyolester (POE) lubricant present, Buna-N™, Buna-S™,
Geolast®, Hypalon®, silicone rubber, and EPDM elastomers gave acceptable overall performance.
Fluoropolymers such as Viton®, Kalrez®, and Teflon® were especially susceptible to absorption of
HFCs, including HFC-236fa, resulting in unacceptable swelling. Hydrogenated nitrile butyl rubber and
natural rubber showed excessive swelling when the POE lubricant was present. Neoprene was deemed
unsuitable due to shrinkage and embrittlement, with and without the lubricant present.
Aluminum, steel, cast iron, copper, brass, and bronze were compatible with HFC-236fa and the
POE lubricant. Of the four molecular sieve bead desiccants tested, three were of 0.3 nm average pore
diameter and exhibited no apparent reactivity with HFC-236fa. The fourth desiccant had an average
pore diameter of 0.4 nm and showed some evidence of degradation of HFC-236fa based on the
increased fluoride ion content of the desiccant after the aging test.
7
-------
HFC-236fa was found to be completely miscible with ISO-68 polyolester lubricant over the
temperature range of -30.1 to + 125°C. Lubricity tests indicated that the chemical is compatible with
this type of lubricant and that the presence of the refrigerant in the oil improved the wear performance
of the oil.
Heat transfer coefficients of HFC-236fa were determined in test rigs configured to investigate
refrigerant-side coefficients in centrifugal chiller systems. Coefficients were measured for two
conventional finned tubes and three performance-enhanced tubes during shell-side condensation and
pool boiling on the outside of a single horizontal tube. As expected, the high-performance enhanced
tubes were found to increase heat transfer and produce higher heat transfer coefficients than the
conventional finned tubes. Comparison of shell-side heat transfer coefficients obtained for HFC-236fa
with those obtained for CFC-114 under identical conditions showed HFC-236fa to have better heat
transfer during condensation with a maximum increase of 40 percent relative to CFC-114. For pool
boiling, HFC-236fa provided a maximum heat transfer increase of 80 percent relative to CFC-114.
For pool boiling tests, a miscible POE lubricant was added to the HFC-236fa up to a lubricant
concentration of 3 weight percent. With the exception of one of the high performance tubes, the oil
caused the heat transfer coefficients to decrease by less than 10 percent from those obtained for the
HFC-236fa alone.
HFC-236fa has received a safety classification of "Al" (low toxicity, nonflammable) by
standards set forth by the American Society of Heating, Refrigerating and Air-Conditioning Engineers
(ASHRAE, 1992). A concentration of 5.6 volume percent HFC-236fa in air was found to extinguish an
n-heptane flame in a standard cup burner test. This extinguishing concentration is equivalent (within
experimental error) to that of a commercially available fire extinguishing agent (HFC-227ea,
1,1,1,2,3,3,3-heptafluoropropane).
8
-------
3. THERMOPHYSICAL AND APPLICATION PROPERTIES OF HFC-236fa
SYNTHESIS OF HFC-236fa
HFC-236fa was prepared in laboratory quantities (ca. 100 g) by the reaction of 0.028 moles of
1,1,1,3,3-pentafluoropropene (CF3CH=CF2) with a mixture of 3 g of potassium fluoride (KF) and 25
mL of formamide (HCONH2) at 28°C. Yield from this reaction was 86 percent.
KF
CF3-CH = CF2 > CF3-CH2-CF3
formamide
28°C
The product obtained from this reaction was characterized by gas chromatography (GC), infrared
19 1
(IR) spectrophotometry, mass spectrometry (MS), and F and H nuclear magnetic resonance (NMR)
spectrometry. Purity of the HFC-236fa thus prepared was 99.5 percent. Initial laboratory synthesis
and preliminary thermophysical property measurements were performed by Clemson University under
joint sponsorship of the EPA and EPRI.
Larger (kilogram) quantities of the compound needed for performance and stability/compatibility
tests were procured from PCR, Inc. in Gainesville, FL. Individual lots of the compound delivered to
AEERL for testing were subjected upon receipt to purity assay by combined GC/MS, and Fourier
transform infrared spectroscopy (FUR) and found to be 98.9 percent pure. Figures 1 and 2 present
FT1R and mass spectra for HFC-236fa, respectively.
Quantities of several thousand kilograms of HFC-236fa required for extended duration toxicity
tests were supplied by the DuPont and 3M companies.
THERMOPHYSICAL PROPERTIES
Table 3 gives thermophysical property data for HFC-236fa derived by the National Institute of
Standards and Technology (NIST) from measured data correlated by the modified Benedict-Webb-
Rubin (mBWR) equation of state.
9
-------
120
a
eg
100
80
m
G
c*
u
s
u
u
Ph
60
40
20
0 1 1 1 1 1 L
4000 3800 3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600
Wavenumber (cm"1)
Figure 1. Fourier-transform infrared spectrum of HFC-236fa
-------
69
31
45
I I I I I I'l I I'I'I M|l*
64
i i I'm i'i i I
113
133
I I'l I | I'l II I i'l II | I II II I'l II I I I I II
I I I I I I I I I I I
140 160
20
40
60 80 100
Fragment molecular weight
120
Figure 2. Mass spectrum of HFC-236fa
-------
Table 3. Thermophysical properties of liquid and vapor HFC-236fa
T(°C)
P(kPa)
Density
(kg/m3)
Volume
(m3 /kg)
Entropy
(kJ/kaK)
Enthalpy
(kJ/ksK)
Cv
(kJ/kg K)
Cp
(kJ/kgK)
vapor
liquid
vapor
liquid
vapor
liquid
vapor
liquid
vapor
liquid
vapor
liquid
-50
7.623
0.6292
1590
1.589
6.29E-04
0.5924
-0.2274
126.7
-56.2
0.6469
0.7998
0.7041
1.113
-45
10.60
0.8571
1575
1.167
6.35E-04
0.5896
-0.2027
130.1
-50.6
0.6558
0.8031
0.7137
1.122
-40
14.48
1.149
1560
0.8705
6.41E-04
0.5876
-0.1782
133.5
-45.0
0.6648
0.8072
0.7235
1.130
-35
19.47
1.517
1546
0.6594
6.47E-04
0.5862
-0.1542
137.0
-39.3
0.6740
0.8120
0.7336
1.139
-30
25.80
1.974
1531
0.5065
6.53E-04
0.5854
-0.1304
140.4
-33.6
0.6833
0.8174
0.7440
1.148
-25
33.71
2.537
1516
0.3942
6.60E-04
0.5852
-0.1070
143.9
-27.8
0.6928
0.8233
0.7547
1.158
-20
43.48
3.222
1501
0.3104
6.66E-04
0.5854
-0.0838
147.4
-22.0
0.7025
0.8296
0.7658
1.167
-15
55.42
4.046
1486
0.2472
6.73E-04
0.5861
-0.0609
150.9
-16.2
0.7123
0.8363
0.7772
1.177
-10
69.84
5.029
1471
0.1988
6.80E-04
0.5873
-0.0382
154.3
-10.2
0.7222
0.8434
0.7890
1.187
-5
87.10
6.193
1456
0.1615
6.87E-04
0.5888
-0.0158
157.8
-4.30
0.7323
0.8507
0.8012
1.197
0
107.6
7.560
1440
0.1323
6.94E-04
0.5906
0.0064
161.3
1.70
0.7426
0.8581
0.8138
1.208
5
131.6
9.155
1425
0.1092
7.02E-04
0.5928
0.0284
164.8
7.80
0.7529
0.8657
0.8269
1.218
10
159.7
11.00
1409
0.0909
7.10E-04
0.5952
0.0501
168.3
13.9
0.7635
0.8734
0.8405
1.229
15
192.2
13.14
1393
0.0761
7.18E-04
0.5978
0.0717
171.7
20.1
0.7741
0.8812
0.8546
1.240
20
229.6
15.58
1376
0.0642
7.27E-04
0.6007
0.0931
175.2
26.4
0.7849
0.8890
0.8694
1.252
25
272.4
18.38
1360
0.0544
7.35E-04
0.6037
0.1143
178.6
32.7
0.7959
0.8968
0.8849
1.264
30
321.0
21.57
1343
0.0464
7.45E-04
0.6068
0.1353
182
39.0
0.8070
0.9047
0.9013
1.277
35
376.0
25.19
1325
0.0397
7.55E-04
0.6101
0.1562
185.3
45.4
0.8182
0.9125
0.9186
1.291
40
437.8
29.30
1308
0.0341
7.65E-04
0.6135
0.1770
188.6
51.9
0.8297
0.9203
0.9371
1.305
45
507.0
33.95
1289
0.0295
7.76E-04
0.6169
0.1976
191.9
58.5
0.8413
0.9282
0.9569
1.321
50
584.2
39.21
1270
0.0255
7.87E-04
0.6203
0.2181
195.1
65.2
0.8531
0.9360
0.9785
1.338
55
669.9
45.15
1251
0.0222
8.00E-04
0.6237
0.2386
198.3
71.9
0.8652
0.9439
1.0020
1.356
60
764.7
51.87
1230
0.0193
8.13E-04
0.6270
0.2590
201.3
78.7
0.8775
0.9519
1.0280
1.377
65
869.4
59.49
1209
0.0168
8.27E-04
0.6303
0.2794
204.3
85.6
0.8902
0.9600
1.0580
1.400
70
984.5
68.15
1187
0.0147
8.43E-04
0.6333
0.2997
207.1
92.7
0.9033
0.9682
1.0920
1.427
(continued)
-------
Table 3. Continued
T(°C)
P(kPa)
Density
Volume
Entropy
Enthalpy
Cv
Cp
(kg/m3)
(m3/kg)
(kT/keK)
(fcj/kg)
(kJ/kaK)
(kJ/ksK)
vapor
liquid
vapor
liquid
vapor
liquid
vapor
liquid
vapor
liquid
vapor
liquid
75
1111
78.02
1163
0.0128
8.60E-04
0.6362
0.3201
209.9
99.80
0.9169
0.9767
1.132
1.458
80
1249
89.33
1138
0.0112
8.79E-04
0.6387
0.3406
212.4
107.1
0.9311
0.9856
1.181
1.494
85
1400
102.4
1111
0.0098
9.00E-04
0.6409
0.3612
214.7
114.6
0.946
0.9949
1.241
1.538
90
1565
117.6
1081
0.0085
9.25E-04
0.6425
0.3820
216.8
122.2
0.9619
1.005
1.319
1.593
95
1744
135.6
1049
0.0074
9.53E-04
0.6434
0.4030
218.6
130.1
0.9791
1.016
1.424
1.664
100
1940
157.1
1013
0.0064
9.87E-04
0.6434
0.4245
219.9
138.2
0.9981
1.028
1.576
1.760
105
2152
183.7
972.7
0.0054
1.03E-03
0.6421
0.4465
220.7
146.7
1.019
1.042
1.814
1.902
110
2383
117.5
924.6
0.0046
1.08E-03
0.6388
0.4695
220.6
155.7
1.044
1.058
2.229
2.136
115
2634
263.2
865.1
0.0038
1.16E-03
0.6326
0.4941
219.2
165.5
1.074
1.079
3.107
2.615
120
2907
332.7
782.4
0.0030
1.28E-03
0.6206
0.5221
215.5
176.8
1.111
1.107
5.896
4.199
-------
Measurements were made using HFC-236fa which had been repurified by preparative gas
chromatography to a purity of 99.98 percent. The normal boiling point of the repurified chemical was
-1.1°C. Table 4 lists the types and ranges of experimental data used in the mBWR correlation. Due to
a lack of data in some regions, predictions from an extended corresponding states (ECS) model for
HFC-236fa were used. These "data" were used to ensure that extrapolation of the mBWR equation
beyond the range of the experimental data would not result in physically unreasonable behavior.
Table 4. Summary of experimental data used in mBWR correlation of HFC-236fa
Source
No. of Data
Temp.
Pressure
Density
Points
Range (K)
Range
(MPa)
(mol/L)
Vapor pressure
28
306 - 358
0.35 -1.4
—
18
266 - 334
0.08-0.8
—
7
180 - 240
2E-04 - 0.02
—
Sat'd liquid density
28
248 - 372
6.7-9.8
3
370 - 390
5.4-6.7
Sat'd vapor density
11
180-310
1E-04 - 0.17
5
350 -390
0.54-1.8
Ideal gas heat capacity
6
276-380
—
Pressure-Volume-
325
248 - 372
1.0-6.5
7.1-10.1
Temperature
842
274 - 345
1.0-35
8.2-9.5
202
276-400
0.01 -0.36
0.003-0.156
41
370-450
0.01 -10
0.003 - 7.8
Isochoric heat capacity
117
274 - 343
—
—
78
183 -327
—
—
40
210-450
—
—
Speed of sound
108
276-400
—
—
Second & third virials
7
276 - 400
—
—
As part of the fitting routine, vapor pressure, saturated liquid and saturated vapor density, and
ideal gas heat capacity data were fit to ancillary equations. These equations were then used by the
mBWR fitting routine to calculate other derived thermodynamic properties. Ancillary equations used
for vapor pressure (PCT), saturated liquid density (pL), saturated vapor density (pv), and ideal gas heat
capacity (CP) are:
14
-------
In
PL
Ei
Pc
c^r+a,r15 +a1r2 +a'3ta +a4z65
1 ~T
pc[l +doTP + dii2/3 + d2x + d3T4/3 + (I4T2 + d5x3]
p =P
rv o
1+/y+fr2f>+f2*+A*2+/y
Pa{Zc-\)
+1
1 + fsT
1
00
-1
Cp° - Co + CjT + C2T2
where: T = temperature (in kelvin)
Tc = critical temperature
Tr = T/Tc
t = 1 - Tr
Pc = critical pressure
pc = critical density
P = 0.325
Zc = Pc/(pcRTc)
Coefficients for the ancillary equations are listed in Table 5.
Table 5. Coefficients for ancillary equations (1) - (4)
(1)
(2)
(3)
(4)
i
0
1
2
3
4
5
RMS (%)
CXi
-8.034525
2.784457
-2.548348
-4.719707
0.586451
0
0.041
di
0.782372
6.233671
-9.692730
5.808706
0
0
0.160
fi
-0.284461
-2.986825
2.361121
1.588164
-1.577292
0
0.251
Ci
53.466256
0.228092
0.000353
0
0
-0.663588
0.111
Critical constants used in the fit are:
Tc = 398.07 K
Pc = 3.200 MPa
pc = 3.626 mol/L (551 kg/m3)
15
-------
and R = 0.008314 L« MPa/(mol«K)
The mBWR equation used was that proposed by Jacobsen [Jacobsen and Stewart, 1973] and i
of the form:
is
15
i-17
P = j=12 ai(T)p' + exp(-52) W0S bi(T)p2i"
where 8 = p/pc, the temperature dependence of the ^ coefficients
ai = RT
a2 = b,T + b2T05 + b3 + b/T + bs/T2
a3 = b6T + b7 + b8/T + bs/T2
a4 = bi0T + bn + bi2/T
as = bi3
a6 = bi/T + bis/T2
a7 = b)6/T
a8 = bn/T + bi8/T2
and the values for the b; coefficients are:
bi = - 0.6611218748E-01 bn = -0.1112078439E+04
b2 = 0.8617639027E+01 bH = -0.2637100515E+00
b3 = -0.2337322560E+03 bM= 0.4775211631E+02
b4 = 0.4374862328E+05 bis = 0.1978040351E+04
b5 = -0.5396777615E+07 bus - -0.4857108989E+01
b6 = -0.7575885520E-02 b17= 0.1448211964E+00
b7 = 0.1073795635E+02 bis =-0.2210593229E+02
b8 = -0.1066265886E+05 b»= 0.9262701699E+00
b9 = -0.1030474554E+06 b20= 0.5779206662E+07
(5)
are:
ag — big/T2
s-io — b20/T2 + b2i/T3
an = b22/T2 + b23/T4
an = b24/T2 + b25/T3
ai3 = b26/T2 + b27/T4
aw = b28/T2 + b29/T3
ais = b30/T2 + b3i/T3 + bB^T4
b23 = 0.3194201231E+10
bM = 0.7929461073E+04
b25 = -0.6936062956E+06
b26= 0.8498362591E+02
b27= 0.2097020511E+07
b28 = 0.1106003692E+01
b29= 0.9537147118E+02
b30 = -0.8818152066E-02
b3i = 0.9731949088E+01
16
-------
bio = -0.1948680916E-02 b21 = -0.9855110656E+09 b32 = -0.9355169222E+03
b,i = 0.4383652281E+01 b22= 0.1971998080E+06
The mBWR equation for HFC-236fa was shown to exhibit reasonable behavior upon
extrapolation to temperatures down to the triple point (180 K) and up to 500 K and pressures up to 40
MPa.
Thermophysical properties for HFC-236fa are also available from the NIST database
REFPROP Version 5.0 (NIST, 1996).
ATMOSPHERIC LIFETIME
Compounds such as HFC-236fa which do not contain chlorine or bromine atoms are not capable
of destroying stratospheric ozone. However, the presence of carbon-fluorine (C-F) bonds in the
molecule can render the chemical a strong absorber of infrared radiation thereby posing global warming
concerns if the atmospheric lifetime of the compound is sufficiently long. The presence of hydrogen (H)
atoms in the molecule subjects the chemical to degradation by reaction with atmospheric hydroxyl (OH)
radicals. Therefore, atmospheric stability of a chemical containing C-H bonds is usually determined by
measuring the reaction rate of the chemical with the OH radical. Atmospheric lifetimes are then
calculated by comparing the measured rate constant with that of methyl chloroform (9. 9 x 10"15 cm3
molecule"1 sec"1) whose atmospheric lifetime due to OH removal alone (6.6 years) is independently
known.
Kinetic experiments for the HFC-236fa/OH reaction were performed for the EPA by NIST using
the technique of resonance fluorescence spectrometry. This technique is based on the rate of
disappearance of OH radical in the presence of a great excess of the stable substrate. While the method
has some advantages over other approaches, it is very sensitive to the presence of even trace amounts of
reactive impurities. By this method, the reaction rate constant for abstraction of a H atom from HFC-
-------
236fa with OH was determined to be 1.1 x 10"15 cm3 molecule-1 sec-1 at 298 K leading to an OH
atmospheric lifetime of 59 years. However, a plot of the logarithm of the rate constant vs. 1/T was non-
linear which is not consistent with the well-established temperature behavior of HFCs and which
suggested the presence of a small amount of a highly reactive impurity. A GC analysis of the material
indicated that the sample vapor was 99.6 percent pure. It is therefore possible that this measured rate
constant is too large and the corresponding atmospheric lifetime too short.
Subsequent to the NIST determination of the HFC-236fa/OH rate constant, researchers at the Jet
Propulsion Laboratory (JPL) of the National Aeronautics and Space Administration (NASA) measured
the rate constant to be 0.34 x 10"15 cm3 molecule"1 sec"1 at 298 K (JPL, 1994). The JPL method differed
from the NIST method by measuring the simultaneous rate of disappearance of the unknown and a
known reference chemical. By obtaining the ratio of the two rate constants determined simultaneously,
any effect due to any impurities cancels out. The JPL rate constant is more in line with what is
predicted based on the molecular structure of the chemical. Adopting this rate constant gives an OH
atmospheric lifetime of 192 years for HFC-236fa.
An atmospheric lifetime of 192 years raises concerns about the global warming potential of the
HFC-236fa. The Intergovernmental Panel on Climate Change (IPCC) has assigned a Global Warming
Potential (GWP) of 6300 to HFC-236fa (100-year horizon, GWP of C02 = 1) as compared to a GWP
of 9300 for CFC-114 (IPCC, 1995).
INHALATION TOXICITY
With U. S. Navy co-sponsorship, the EPA undertook a rigorous evaluation of the inhalation
toxicity of HFC-236fa. Included in this battery of toxicity tests were a 5-day acute inhalation test, a
genetic toxicity screening, a cardiac sensitization test, a 2-week inhalation test, a 90-day subchronic
test, and developmental toxicity tests. A brief synopsis of test protocols and results follows.
18
-------
Mutagenicity Testing of HFC-236ea in the Salmonella tvphimurium and Eschirichia coli Plate
Incorporation Assay
HFC-236fa was evaluated for mutagenicity in Salmonella tvphimurium strains TA100, TA1535,
TA97, and TA98 and in Escherichia coli WP2uvrA (pKMlOl) with and without an exogenous
metabolic activation system (S9). The maximum concentration tested was an atmosphere of 100
percent of HFC-236fa. Additional lower concentrations of 0, 20, 40, 60, 70, and 80 volume percent
HFC-236fa in air were also evaluated. No evidence of mutagenic activity was detected in two
independent trials.
In-vitro Assay of HFC-236fa for Chromosome Aberrations in Human Lymphocytes
HFC-236fa was evaluated for clastogenic (chromosome damaging) activity in human
lymphocytes in-vitro following 3-hr treatments with and without metabolic activation (S9). Two
independent trials were conducted. In Trials 1 and 2, concentrations of approximately 0, 20, 30, 40,
and 100 volume percent and 0, 40, 60, 80 and 100 volume percent HFC-236fa in air were evaluated,
respectively. No statistically significant increases in the percent of chromosomally abnormal cells
occurred at any HFC-236fa concentration evaluated, and no concentration-related trends in chromosome
aberration induction were observed. HFC-236fa was not clastogenic in this assay.
Mouse Bone Marrow Micronucleus Assay of HFC-236ea by Inhalation
This test was conducted to determine whether HFC-236fa induces an increase in the frequency of
micronucleated polychromatic erythrocytes in bone marrow. In this study, groups of male and female
Crl:CD®-l(ICR)BR mice were placed in inhalation exposure chambers and exposed whole body to
HFC-236fa at target concentrations of 0, 5000, 20000, and 50000 ppm for 6 hours/day for 2
consecutive days. Bone marrow smears were prepared approximately 24 and 48 hours after the second
exposure, and 2000 polychromatic erythrocytes per animal were evaluated for the presence of
micronuclei.
19
-------
No statistically significant increases in micronucleated polychromatic erythrocytes were observed
in the animals at any HFC-236fa concentration tested. In addition, no significant decreases in the ratio
of young polychromatic erythrocytes to mature normochromatic erythrocytes were observed. Under the
conditions of this study, HFC-236fa was negative; i. e., HFC-236fa did not induce micronuclei in bone
marrow cells.
Acute Inhalation Toxicity of HFC-236fa in Rats
Acute inhalation toxicity of HFC-236fa in male rats was investigated and found to be very low.
Lethal concentrations in air were not determined since no rats died during or after exposure. The
highest concentration tested for HFC-236fa was approximately 189000 ppm for 4 hours. HFC-236fa
caused symptoms of narcosis (sleep induction) at a concentration of 189000 ppm, but not at 134000
ppm. Rats recovered quickly (within an hour) after cessation of exposure, and no delayed effects were
seen in a 1 to 4 day recovery period.
2-Week Inhalation Toxicity Study with HFC-236fa in Rats
Four groups of five male and five female Crl:CD®BR rats were exposed by inhalation for 6
hours a day, 5 days a week for 2 weeks to concentrations targeted at 0, 5000, 20000, or 50000 ppm
HFC-236fa (one concentration per group). The rats were weighed prior to exposure each day and were
observed for clinical signs both prior to and following each exposure. The response to an alerting
stimulus (response to sound stimulus during exposure) was determined during exposures. Clinical
pathology evaluations were conducted prior to the end of the study. Immediately following the last
exposure, rats were sacrificed and necropsied. All rats were examined for gross and microscopic
pathological changes, and liver tissue was evaluated for peroxisomal p-oxidation activity.
Rats exposed to HFC-236fa had no body weight effects and no abnormal clinical observations
either prior to or following exposures. The one notable compound-related effect was a diminished
response or lack of response to an alerting stimulus during exposure. Although rats in the 50000 ppm
group were generally non-responsive and rats in the 20000 ppm group had a diminished response during
20
-------
the first week of exposure, most of these rats exhibited normal alerting responses during the second
week. All rats exposed to 5000 ppm behaved normally during exposure and were responsive to a
stimulus following exposure.
No compound-related effects were noted during clinical pathology evaluations. No organ weight
differences or gross abnormalities were seen during necropsy and no histopathologic changes were
observed upon microscopic evaluation of tissues.
Although HFC-236fa did not induce peroxisome proliferation, a slight but statistically significant
decrease in hepatic P-oxidation activity occurred in the 50000 ppm rats that were sacrificed after the
tenth exposure. This decrease was considered compound-related but not biologically adverse since the
change was not accompanied by relevant clinical pathologic, organ weight, or liver histopathologic
changes.
Under the conditions of this study, the no-observable-adverse-effect level (NOAEL) was 5000
ppm, based on the transient, reduced responsiveness to an alerting stimulus during exposure to
concentrations of 20000 ppm or greater. This effect was completely reversible upon cessation of
exposure.
90-Dav Inhalation Toxicity Study with HFC-236fa in Rats
Four groups of 10 male and 10 female Crl:CD®BR rats were exposed by inhalation for 6 hours a
day, 5 days a week, over a 14-week period (65 exposures) to concentrations targeted at 0, 5000, 20000,
and 50000 ppm (one concentration per group). All rats were weighed weekly and were also individually
observed for clinical signs of toxicity following each exposure. The response to an alerting stimulus
was determined during the exposures. Ophthalmological examinations were performed prior to
commencement of the study and at the end of the study. Clinical pathology evaluations were conducted
near the midpoint and at the end of the study. The day following the last exposure, the rats were
sacrificed and necropsied. All rats were examined for gross and microscopic pathological changes, and
liver tissue was evaluated for peroxisomal p-oxidation activity.
21
-------
HFC-236fa had no effect on body weights, no abnormal clinical observations were seen in rats
either prior to or following exposures, and no effect on food consumption or food efficiency was
observed. The one notable compound-related effect was a diminished response to an alerting stimulus
during exposure in the 50000 ppm male and female rats during the first 2 weeks of the study. A
diminished alerting response was noted in some rats, primarily during the last 2 hours of the exposure.
The number of rats affected generally decreased with successive exposures such that by study day 18 all
rats from this group had a normal alerting response. All rats exhibited normal alerting response during
the evaluation period immediately after exposure.
No compound-related effects were detected during the ophthalmological and clinical pathology
evaluations. No organ weight differences or gross abnormalities were seen during necropsy and no
histopathologic changes were observed upon microscopic evaluation of tissues. No biologically
significant alterations in hepatic peroxisomal p-oxidation activity were seen in any group.
The assigned NOAEL was 20000 ppm, based on the transient, reduced responsiveness to an
alerting stimulus during exposure to 50000 ppm. This effect was not seen in all rats from the 50000
ppm group and was completely reversible upon cessation of exposure.
Inhalation Developmental Toxicity Study of HFC-236fa in Rats
A pilot study was conducted at 0, 5000, 20000, and 50000 ppm HFC-236fa in rats. Results
indicated a diminished to absent alerting response (response to sound stimulus during exposure at 50000
ppm). Some of the rats exposed to 20000 ppm also had diminished responses. No changes in alerting
response were seen after test day 4 in the 20000 ppm exposed rats. Normal reactions occurred in rats in
the 5000 ppm groups each day and in rats from all groups when they were tested 10-20 minutes after
the generation of the test material ceased. The abnormal alerting responses were reversible, compound
related effects, since they occurred only during exposures. Under the conditions of this pilot study,
maternal toxicity characterized by diminished alerting response during exposure, was detected at
concentrations of 20000 and 50000 ppm. There was no other evidence of maternal toxicity. No
22
-------
fetotoxicity was observed at any concentration level. Based on these results, exposure levels of 0, 5000,
20000, and 50000 ppm HFC-236fa were chosen for the main developmental study in rats.
HFC-236fa was administered by inhalation to four groups of 25 Crl:CD®BR female rats on days
7 - 16 of gestation at a daily concentration of 0, 5000, 20000, or 50000 ppm (one concentration per
group). At 20000 and 50000 ppm, there were significant dose related decreases in maternal body
weight gain over the first 2 days of inhalation exposures. At 50000 ppm, this reduction in weight gain
was accompanied by a significant reduction in maternal food consumption and occasional instances of
diminished alerting responses during the inhalation exposures. No evidence of maternal toxicity was
detected at 5000 ppm. There was no evidence of developmental toxicity at any level tested.
Fetal findings showed that there were no compound-related effects on the incidence of early, late,
or total resorptions detected. There were no dead fetuses. No compound-related effects for mean fetal
weight were detected. The values for mean fetal weight were comparable across the control and
exposure.
Under the conditions of this study, significant maternal toxicity was demonstrated at daily
exposure levels of 20000 and 50000 ppm. The maternal NOAEL was 5000 ppm. Developmental
toxicity was not detected at any exposure level. The developmental NOAEL was 50000 ppm. Thus,
HFC-236fa is not considered to be uniquely toxic to the rat conceptus.
Extended Acute Inhalation Toxicity of HFC-236fa in Rats
This study was carried out to assess the toxicity of HFC-236fa following the extension of the
normal single 4-hour exposure period employed in acute inhalation studies to exposure for 5 consecutive
days. The target concentration used in this study was 50000 ppm as selected by the EPA.
A group of five male and five female Alpk:APFSD® rats (approximately 7 weeks old) supplied
from a colony maintained at Alderly Park, Cheshire, UK, was exposed nose-only to a target
atmospheric concentration of 50000 ppm, for 6 hours per day, for 5 consecutive days. These rats were
then retained for a period of 14 days for observation following the last exposure.
23
-------
Test animals were observed for gross clinical abnormalities during each exposure period and
were checked daily thereafter. In addition, animals were given a detailed examination after exposure on
days 1-5 and then daily during the 14-day observation period up to and including day 19.
HFC-236fa was generated close to target concentration. The only finding possibly attributable to
exposure of HFC-236fa was associated with slight respiratory tract irritation. However, in the absence
of any effects on body weight, and lung appearance and weight, this is considered to be of little
significance at the concentration tested.
On the basis of this test, HFC-236fa is considered to have low acute inhalation toxicity at
concentrations of 50000 ppm and lower.
Inhalation Developmental Toxicity Study of HFC-236fa in Rabbits
HFC-236fa was administered to groups of 20 Hra:(NZW)SPF female time-mated rabbits on
days 7-19 of gestation at daily exposure levels of 0, 5000, 20000, and 50000 ppm. The highest level,
50000 ppm, was the highest concentration which could be attained without supplementing chamber
oxygen.
There was no evidence of any maternal or developmental toxicity at any exposure concentration
tested. There were no compound-related effects on maternal body weights, weight changes, food
consumption, clinical observations, or post-mortem findings. There were no compound-related
developmental or reproductive effects; the endpoints evaluated included mean fetal weight, mean litter
size, measures of pre- and post-implantation embryo lethality, and the incidences of fetal malformations
and variations. Thus, the maternal and developmental no-observed-effect-level (NOEL) was 50000
ppm. Therefore, the results of this study indicate that HFC-236fa was not uniquely toxic to the rabbit
conceptus.
Cardiac Sensitization
The potential of HFC-236fa to cause cardiac sensitization by inhalation in beagle dogs was
determined. Cardiac sensitization to adrenaline is a phenomenon associated with the inhalation of a
24
-------
number of unsubstituted and halogenated hydrocarbons. After inhalation of the sensitizing agent,
challenge with adrenaline causes cardiac arrhythmias. The technique used in this study involved the
intravenous injection of adrenaline before and during the inhalation of the test gas. The effect of
adrenaline on an electrocardiogram was examined in both cases and compared to assess any positive
response to the test gas.
Six male beagle dogs were selected for the test. These dogs had been acclimatized to laboratory
conditions and handling procedures and trained to accept restraint procedures during the course of
previous studies. The cardiac response of the dogs to adrenaline alone was first monitored. The dogs
were then exposed to the HFC-236fa by a snout-only delivery system. Concentrations of HFC-236fa
examined were 5, 10, 15, 25, and 30 volume percent in air. At least 1 calendar day was allowed
between exposure sessions to allow the dogs to recover.
The criterion for a positive effect was the appearance of a burst of multifocal ventricular ectopic
activity (MVEA) or ventricular fibrillation (VF). Ventricular tachycardia alone was not necessarily
considered definitive evidence of a positive response. At 5 and 10 percent (50,000 and 100,000 ppm)
HFC-236fa all responses were negative. There were no clinical signs of any toxicological significance.
At 15 percent (150,000 ppm) HFC-236fa, 33 percent of responses were positive. One dog of the
six responded with MVEA and one dog responded with VF: both responses were fatal. A third dog was
observed to be unsteady in gait after removal from the restraint.
At 20 percent (200,000 ppm) HFC-236fa, 50 percent of responses were positive; i.e., two of the
remaining four dogs responded with MVEA but survived. All dogs exposed at this concentration were
observed to be unsteady in gait after removal from the restraints.
Exposures at 25 and 30 percent HFC-236fa were prematurely terminated because the dogs either
struggled significantly during exposure or were anaesthetized. Overall, half of the dogs gave positive
responses at 20 percent HFC-236fa in air. There were no positive responses at 10 percent or less HFC-
236fa.
25
-------
FLAMMABILITY AND FIRE SUPPRESSION CHARACTERISTICS
Flammability of HFC-236fa was evaluated by means of the American Society for Testing and
Materials E681-94 standard method for determination of flammability limits (ASTM, 1994). Tests
were conducted at room temperature (27°C or 300 K) and at 100°C (373K) at atmospheric pressure
(ca. 754 torr or 100.5 kPa) using a 5-L glass test vessel and a 0.1 second alternating current (ac) spark
generated by a 15 kV, 30 mA power supply as the ignition source. Air mixed with the HFC-236fa had
a relative humidity of 50 percent. HFC-236fa/air mixtures were prepared by individually metering the
HFC and air into the evacuated (ca. 1.7 torr or 0.23 kPa) flask equipped with a pressure transducer
accurate to ±1.5 torr or 0.2 kPa. At room temperature, tested concentrations of HFC-236fa in air were
5.1, 8.1, 11.0, and 15.0 volume percent. At 100°C, tested concentrations were 8.0, 12.0, 15.9, and 20.0
volume percent. No flame was generated or propagated in these tests, indicating that the material was
nonflammable. This is in agreement with the axiom that fluorocarbons having a C-F/(C-H +C-C) bond
ratio >1 are not flammable. Calculated standard enthalpies of combustion (298.15 K, 101.3 kPa) for
the two possible overall combustion reactions for HFC-236fa (with all reactants and products in the
vapor state) are (Smith and Tufts, 1996):
C3H2F6 + 202 + 2H20 3C02 + 6HF AH0 =-941.2 kJ/mol"1
C3H2F6 + 202 -» C02 + 2HF + 2COF2 AH0 = -781.7 kJ/mol"1
Because of its nonreactivity, HFC-236fa was among several chemicals examined by the EPA as
possible alternatives for ozone depleting fire extinguishing agents such as the halons. This work was
conducted by the New Mexico Engineering Research Institute (NMERI) under EPA Cooperative
Agreement No. CR817774. Laboratory scale suppression of a n-heptane flame was measured using the
cup-burner technique (Skaggs et al., 1995). The concentration of HFC-236fa in air found to just
extinguish the flame was 5.6 volume percent. This is nearly twice the concentration of halon-1301
(CF3Br) to extinguish the same flame under the same conditions (3.0 volume percent). Taking into
account the difference in the liquid densities of HFC-236fa and halon-1301, the volume of liquefied
26
-------
HFC-236fa which would have to be available to extinguish a fire is 2.16 times the volume of liquefied
halon-1301. The greater fire extinguishment effectiveness of halon-1301 is due to the presence of
bromine atoms in the molecule. Bromine atoms have been found to be very effective scavengers of
reactive combustion radicals, thereby allowing combustion chain reactions to be terminated. It would
appear, however, that HFC-236fa would be as effective in fire suppression as HFC-227ea
(1,1,1,2,2,3,3-heptafluoropropane), a commercial streaming fire extinguishing agent.
HFC-236fa/POE LUBRICITY
A wear test based on ASTM Method D2670-88, "Standard Test Method for Measuring Wear
Properties of Fluid Lubricants (Falex Pin and Vee Block Method)" (ASTM, 1988) was carried out.
This ASTM method was modified to accommodate testing the lubricant/refrigerant combinations by
substituting the standard oil test cup with a glass cup with an internal glass frit in the bottom and a side
arm. Lubricant was added to this glass cup and the test gas added through the side arm. The test gas
would then bubble up through the oil.
The Falex test machine rotates a metal journal between two metal stationary vee blocks immersed
in an oil bath. Load is applied to the system through lever arms (load arms) to the vee blocks. An
analog, dial-type gauge measures the applied load, and an electronic load cell measures the torque
transferred by the friction between the rotating pin and the vee blocks. The load is generated by a
screw-drive clamp mounted across the ends of the load arms. The amount of wear of the test pieces is
measured as the amount of rotation of a toothed wheel attached to the screw-drive (200 teeth equal one
full rotation).
The wear test uses a 5 minute break-in period at fixed load. Following the break-in period, the
wear test consists of operating the machine at a fixed load for 15 minutes; the total number of teeth
passing a reference point is reported as the result.
The extreme pressure test used the step test procedure described in ASTM D3233-92, "Standard
Test Methods for Measurement of Extreme Pressure Properties of Fluid Lubricants (Falex Pin and Vee
-------
Block Methods)" (ASTM, 1992). The pressure test also involves a 5 minute break-in period at fixed
load. Following the break-in period, the load is increased in 250-lb increments and maintained for 1
minute at each increment. The extreme pressure test is complete when a brass shear pin breaks (at
approximately 100 in-lb torque) or when the maximum system load (2885 lb) is reached without failure
of the shear pin. The load at failure (or maximum load) is reported as the result. Results of the step
tests are in "pounds" of load applied to the test pieces at failure. In this case, a larger number indicates
better performance. A value "2885+" indicates that the test piece did not seize or break after the
completion of the maximum load of the test machine.
A combined test matrix was developed using a mineral oil-sulfur mixture (ASTM D 2670-88
Blend B) and argon gas as controls. The order of the tests was randomized to remove possible bias as a
result of test order. Argon was chosen as an inert control gas since nitrogen might react with the hot
test pieces during testing. A gas flow rate of 0.1 L per minute was maintained in all tests. The cylinder,
valve, flow meter, and lines were kept at ambient temperature. The mass flow rate for HFC-236fa was
approximately 1 g per minute.
Results for the wear tests given in Table 6 show the number of teeth advanced on the screw drive
(higher number equals more wear) or that the test pieces seized during the test. Using the POE oil in the
absence of argon or HFC-236fa resulted in seizure of the test pieces in six of the eight trials, while the
other two trials gave less wear than did operation with Blend B. This binodal distribution of results
(either seizure or superior performance compared to Blend B) was observed in all tests with HFC-
236fa. It was observed that fine metal particulate generated during the wear tests remained suspended
in the Blend B oil but agglomerated in the POE oil and frequently accumulated in the test piece
interfaces causing seizure. Agglomeration of the metal particulate in the POE oil may be related to an
additive in the oil which was not present in Blend B. Overall, addition of argon to the POE oil gave
significant improvement in wear resistance, as did addition of HFC-236fa. Still, even with the addition
of these gases, seizure of the test pieces occasionally occurred. It is postulated that the decrease in the
28
-------
amount of wear upon adding argon or HFC-236fa may be due to a reduction in the oil temperature
caused by the flow of gas through the oil and/or to gas bubbles flushing metal particles from the
journal/vee-block interfaces. For some tests, it was observed that all of the wear occurred in the last 5
minutes of the test when the oil had reached a temperature of 90 to 100°C.
Table 6. Wear tests results (unit is teeth of wear)
Blend B
POE Lubricant
w/o Argon
w/Argon
w/o Argon
w/Argon
w/HFC-236fa
114
162
seized
56
0
117
140
seized
135
9
109
165
seized
90
42
111
seized
45
74
128
seized
43
118
131
seized
seized
seized
117
41
106
118
73
Results for the extreme pressure (step) tests are given in Table 7. Results are tabulated as either
the load (in pounds) at which the rotating journal seized or "2885+" to indicate that the lubricant was
tested to the limits of the machine without failing. Again, the standard Blend B oil outperformed the
POE oil in all tests except for those in which HFC-236fa was added to the POE oil. Addition of argon
slightly enhanced the lubricity of both the Blend B and POE oils in the step test, possibly due to a
cooling effect. Addition of HFC-236fa to the POE oil improved its performance by a factor of about
2.5 in the step test. In general, it was found that the presence of HFC-236fa in the POE oil significantly
improved the lubricity of the oil as measured in the ASTM wear and extreme pressure tests.
29
-------
Table 7. Extreme pressure (step test) results (lb)
Blend B
POE Lubricant
w/o Argon
w/Argon
w/o Argon
w/Argon
w/HFC-236fa
2600
2885+
765
1100
2885+
2600
2885
930
1100
2885+
2450
2885+
765
1100
2885+
2450
765
2885+
2885+
930
765
2885+
1100
765
2600
2600
HEAT TRANSFER CHARACTERISTICS
The shell-side heat transfer performance of HFC-236fa was evaluated for both conventional
finned tubes (26 and 40 fins per inch [fpi] tubes) and high performance-enhanced Turbo-CII, Turbo-B,
and Turbo-BII tubes.
Condensation of pure HFC-236fa was conducted on a 26 fpi, a 40 fpi, and a Turbo-CII tube.
Pool boiling on four tube types (26 fpi, 40 fpi, Turbo-B, and Turbo-BII) was tested for pure HFC-
236fa and for HFC-236fa mixed with 1 and 3 weight percent IS068 POE lubricant. These tubes had
nominal outer diameters of 19.1 mm (1/4 in) and were evaluated at a saturation temperature of 40°C for
condensation and 2°C for pool boiling over the heat flux range of 15 to 40 kW m"2.
Heat transfer was improved for HFC-236fa by using the high-performance enhanced tubes.
Specifically, the Turbo-CII tube performed better than the two conventional finned tubes in the
condensation tests, while the performances of the Turbo-B and Turbo-BII tubes were superior to those
of the two conventional finned tubes in the pool boiling tests.
The maximum increase in heat transfer coefficients for the Turbo-CII tube was 80 percent
relative to the 26 fpi tube and 70 percent relative to the 40 fpi tube, while for the Turbo-B it was 1.7
and 2.2 times greater than for the 26 and 40 fpi tubes, respectively. In addition, the Turbo-BII tube
gave boiling heat transfer coefficients up to 80 percent larger than those of the Turbo-B tube.
The heat transfer performance of HFC-236fa was compared with that of CFC-114 determined in
the same facility under the same conditions. For all the tubes tested, except the Turbo-CII tube, the heat
30
-------
transfer results showed that HFC-236fa outperformed CFC-114 for both shell-side condensation and
pool boiling. The heat transfer coefficients for HFC-236fa during condensation were up to 40 percent
larger than those for CFC-114 and during pool boiling were up to 80 percent larger.
Effects of compressor oil on heat transfer performance during pool boiling were also investigated.
The presence of up to 3 weight percent IS068 POE lubricant in HFC-236fa decreased the boiling
performance of pure HFC-236fa by less than 10 percent for all but one of the tubes tested. The Turbo-
BII tube, the only exception, showed an increase in boiling coefficients of up to 30 percent over the pure
refrigerant values for tests with 1 percent lubricant, and up to 15 percent with 3 percent lubricant
present.
STABILITY AND MATERIALS COMPATIBILITY
Thermal and hydrolytic stability of HFC-236fa in the presence of metal catalysts was evaluated
by placing the chemical (0.7 g) in evacuated sealed glass tubes in contact with copper, steel, and
aluminum coupons and heating the tubes at 175°C for 14 days in accordance with the methods
described in ANSI/ASHRAE Standard 97-1989 (American Society of Heating, Refrigerating and Air-
Conditioning Engineers) (ASHRAE, 1989). These tests were repeated with added POE lubricant
(0.7 g) and with added moisture (0.07 |j.L). Chemical to lubricant content was based on the mass rather
than the volume ratio. At the end of the 14-day aging period, the contents of the tubes were visually
inspected and the vapor and liquid phases analyzed by a combination of GC and FTIR. There was no
evidence of copper plating on the steel and no evidence of either refrigerant or lubricant breakdown
indicating excellent thermal and hydrolytic stability of the HFC-236fa both alone and in combination
with lubricant.
Another matrix of sealed tube samples was prepared combining HFC-236fa with various desiccants,
plastics, and elastomers with and without lubricant to test for compatibility of the HFC-236fa with these
31
-------
materials. These samples were aged in an oven at 125°C for 14 days.
Table 8 gives the sample matrix for the compatibility tests. Table
plastics.
All tests were run in duplicate.
9 describes the elastomers and
Table 8. Compatibility test matrix
Sample
Number
Material(s)
Sample
Number
Material(s)
1 a,b
HFC-236fa/Buna™-N
22 a,b
HFC-236fa/Geolast®/POE
2 a,b
HFC-236fa/E-70
23 a,b
HFC-236fa/Mylar®™
3 a,b
HFC-236fa/HNBR
24 a,b
HFC-236fa/Nomex®
4 a,b
HFC-236fayHypalon®
25 a,b
HFC-236fa/Nylon 6,6
5 a,b
HFC-236fa/Kalrez® C
26 a,b
HFC-236fa/Mylar®/POE
6 a,b
HFC-236fa/Natural rubber
27 a,b
HFC-236fa/Nomex®/POE
7 a,b
HFC-236fa/NBRS
28 a,b
HF C-236fa/Nylon 6,6/POE
8 a,b
HFC-236fa/Neoprene 3229
29 a,b
HFC-236fa/Brass, Bronze
9 a,b
HFC-236fa/S-70
30 a,b
HFC-236fa/Cast iron
10 a,b
HFC-236fa/Teflon®
31 a,b
HFC-236fa/Brass, Bronze/POE
11 a,b
HFC-236fa/Geolast®
32 a,b
HFC-236fa/Cast iron/POE
12 a,b
HFC-236fa/Buna™-N/POE
33 a,b
HFC-236fa/Activated desiccant H-5
13 a,b
HFC-236fa/E-70/POE
34 a,b
HFC-236fa/Activated desiccant H-6
14 a,b
HFC-236fa/HNBR/POE
35 a,b
HFC-236fa/Activated desiccant H-7
15 a,b
HFC-236fa/Hypalon®/POE
36 a,b
HFC-236fa/Activated desiccant H-9
16 a,b
HFC-236fa/Kalrez®-C/POE
37 a,b
HFC-236fa/Activated desiccant H-5/POE
17 a,b
HFC-236fa/Natural rubber/POE
38 a,b
HFC-236fa/Activated desiccant H-6/POE
18 a,b
HFC-236fa/NBRS/POE
39 a,b
HFC-236fa/Activated desiccant H-7/POE
19 a,b
HFC-236fa/Neoprene 3229™/POE
40 a,b
HFC-236fa/Activated desiccant H-9/POE
20 a,b
HFC-236fa/S-70/POE
41 a,b
HFC-236fa/Viton® A
21 a,b
HFC-236fa/Teflon®/POE
42 a,b
HFC-236fa/Viton® A/POE
32
-------
Table 9. Tested elastomers and plastics
Polymers
Descriptions
Buna™-N
Copolymer of 1,3-butadiene (70 percent) and acrylonitrile (30 percent)
HNBR
Hydrogenated nitrile butyl rubber
Hypalon®
Chlorosulfonated high density polyethylene
Geolast®
Nitrile polypropylene
NBRS or Buna™-S
Copolymer of 1,3-butadiene (70-75 percent) and styrene (25-30
percent)
Neoprene
Polychloroprene
Kalrez®-C
Perfluoropolymer of tetrafluoroethylene and perfluoromethyl vinyl
ether
E-70 or EPDM
Ethylene propylene diene methylene rubber
S-70 or SI
Silicone rubber
Viton®-A
Copolymer of vinylidene fluoride and hexafluoropropylene
Teflon®
Polymer of tetrafluoroethylene
Nomex®
Polymer of m-phenylenediamine and isophthalic acid chloride
Mylar®
Polyethylene teraphthalate
Nylon 6,6
Polymer of adipic acid and hexamethylenediamine
Figures 3 through 6 graphically display the results of the changes in volume, weight, linear swell,
and hardness for the elastomeric materials with HFC-236fa with and without the lubricant. Values
represent averages of the duplicate samples for each material. Some swelling of elastomeric materials is
desired for gaskets and O-rings to form a good seal in equipment. However, volume increases of greater
than 20 percent or linear swell of greater than 5 percent may be considered excessive and detrimental.
Also, any shrinkage of the material is not desired. A change of hardness of ±10 percent may indicate
excessive softening or embrittlement and may be considered unacceptable. Depending on where in the
equipment the O-ring and gasket materials are placed, they may experience contact primarily with HFC-
236fa or with a combination of HFC-236fa and lubricant. Therefore, a given elastomer or plastic may
be suitable for use in one section of the equipment and not in another.
In common with three other HFCs and one HFE tested for material compatibility, results of the
HFC-236fa tests showed that with and without a polyolester (POE) lubricant present, Buna-N™, Buna-
S™, Geolast®, Hypalon®, silicone rubber, and EPDM elastomers gave acceptable overall performance.
33
-------
30
M
•S
o
25
20 +
15
10
a
« -
9. 5 +
•P".
-10
-15
[J
ffl HFC-236fa without POE
~ HFC-236fa with POE
55
J
o
r-
W
§
a
J2
ffl
(D
1
"8
£
s
d>
(-H
&
53
o
r-
oo
d
Q
d
t§
d>
H
£
X
§
J!
vo
VO
d
O
5
Elastomers
Figure 3. Volume change of elastomers with HFC-236fa
-------
30
U)
Ul
4>
A
O
"S
a
if
VO
VO
a
o
5
Figure 4. Weight change of elastomers in HFC-236fa
-------
m HFC-236fa without POE
~ HFC-236fa with POE
Elastomers
Figure 5. Linear swell of elastomers in HFC-236fa
-------
10
u>
-J
-------
Fluoropolymers such as Viton®, Kalrez®, and Teflon® were especially susceptible to absorption
of HFCs, including HFC-236fa, resulting in unacceptable swelling. Hydrogenated butyl nitrile rubber
and natural rubber showed excessive swelling in the presence of the HFC/lubricant mixtures. Neoprene
was deemed unsuitable due to shrinkage and embrittlement with the HFCs, with and without the
lubricant present.
Evidence for degradation of HFC-236fa was sought by comparison of the infrared spectra and
gas chromatograms of the vapor phase from each of the 42 aged samples against the corresponding
spectrum and chromatogram of imaged HFC-236fa. Degradation of the lubricant in aged samples was
also checked by infrared spectral comparison with the imaged lubricant.
Neither HFC-236fa nor the lubricant showed any evidence of degradation in the presence of the
various metals after the 2-week heating period. Some IR spectral changes were observed in the liquid
(lubricant) phase for some of the samples containing the elastomers/plastics, but these features could
not be attributed unambiguously to degradation of the elastomers/plastics or the lubricant or both. The
most likely source of the new DR. absorption features seemed to be leaching of some components of the
elastomers such as fillers, accelerators, or plasticizers included in the formulations.
Desiccant compatibility with the refrigerant was checked by fluoride analysis of the condensate
collected by passing steam over a bed of desiccant mixed with vanadium pentoxide (V205) in a nickel
tube furnace at 975°C. Fluoride analysis in desiccants was based on industry procedures (UOP, 1978;
Association of Florida Phosphate Chemists, 1991). Four molecular sieve desiccants in the
aluminosilicate family were aged along with HFC-236fa and the HFC-236fa/POE mixture. All types of
unaged and aged desiccants were analyzed for fluoride. The original (unaged) desiccants did not show
any fluoride content. However, one desiccant type in the aged group showed a small amount of fluoride
deposition (<2 percent); see Table 10.
38
-------
Table 10. Desiccant results with HFC-236fa
Desiccant and Refrigerant
Combinations
Percent Fluoride
without POE with POE
HX-51 & HFC-236fa
1.66
0.12
HX-62 & HFC-236fa
0.30
0.00
HX-72 & HFC-236fa
0.00
0.00
HX-92&HFC-236fa
0.00
0.00
1 pore size = 0.4 nm
2 pore size = 0.3 nm
39
-------
4. REFRIGERANT PERFORMANCE
HFC-236fa AS A CFC-114 REPLACEMENT IN CHILLERS
CFC-114 has been used as the refrigerant of choice in centrifugal chillers with capacities
ranging from 440 to more than 1200 kW. The U. S. Navy has been a major user of this type of
equipment to supply comfort air conditioning and critical cooling needs on board its ships. Another
application for this refrigerant has been high temperature heat pumps employed in waste heat
recovery and utilization. HFC-236fa is one of several alternatives with normal boiling points near
that of CFC-114. Table 11 presents a list of some CFC-114 alternatives in the order of their normal
boiling points along with their critical properties.
THEORETICAL PERFORMANCE EVALUATION
Both theoretical thermodynamic and experimental analyses were performed comparing the
performance of HFC-236fa with CFC-114 under shipboard chiller conditions (2 to 13°C evaporating
temperature and 40 to 65°C condensing temperature) and high temperature heat pump conditions (0
to 35°C evaporating and 40 to 110°C condensing temperatures) (Kazachki et al. 1995 and 1994).
Important parameters for centrifugal compressors were compared for HFC-236fa and CFC-114
(Kazachki and Gage, 1993).
40
-------
Table 11. Properties of HFC-236fa and other CFC-114 alternatives
Refrigerant
Name
Chemical
Formula
TC
(°C)
PC
(kPa)
Tb
(°C)
HCFC-124
2-chloro-1,1,1,2-tetrafluoroethane
CF^CFCIH
122.5
3660
-13.2
HC-600a
isobutane
135.0
3648
-11.7
HCFC-142b
1 -chloro-1,1 ,-difluoroethane
CClFoCH,
137.2
4120
-9.2
HFC-236cb
1,1,2,2,3-pentaQuoropropane
CF,CF9CFH?
130.2
3118
-1.4
HFC-236fa
1,1,1,3,3,3-hexafluoropropane
CF,CH<>CF,
130.7
3177
-1.1
HFC-254cb
1,1,2,2-tetrafluoropropane
HCF,CF0CH,
146.2
3753
-0.8
HC-600
n-butane
c4h10
152.0
3797
-0.4
CFC-114
Dichlorotetrafluoroethane
CClFoCClF-)
145.7
3248
3.6
HFE-134
1,1,2,2-tetrafluorodimethylether
HCF9OCF9H
147.1
4228
4.7
HFC-236ea
1,1,1,2,3,3-hexafluoropropane
CF,CHFCF0H
141.2
3533
6.5
HFC-236ca
1,1,2,2,3,3-hexafluoropropane
HCF^CFoCFoH
155.2
3405
12.6
A theoretical thermodynamic analysis was performed using a computer model based on the
thermodynamic properties of HFC-236fa predicted by REFPROP (NIST, 1996). Volumetric
refrigerating capacity and coefficient of performance (COP) of the vapor compression cycle were the
two main criteria used to evaluate performance. Since both CFC-114 and HFC-236fa have wet
isentropic vapor compression, a vapor compression cycle with throttling and dry compression was
used. Dry compression is defined as isentropic compression of refrigerant vapor with sufficient
superheat for the compression process to end on the vapor saturation curve. Since the goal was to
evaluate the performance of HFC-236fa relative to CFC-114, the results are presented as ratios of the
particular characteristics of HFC-236fato CFC-114.
41
-------
The COP ratio of HFC-236fa to CFC-114 is shown in Figure 7. At all conditions, the ratio is
<1, and as much as 30 percent lower at the lowest evaporating temperature and highest condensing
temperature for which the COP is calculated. The lower critical temperature of HFC-236fa
compared to CFC-114 affects its performance in a negative way.
The volumetric capacity ratio of HFC-236fa to CFC-114 is presented in Figure 8 for five
condensing temperatures. For up to 80°C the HFC-236fa volumetric capacity is 0 to 20 percent
higher than CFC-114, probably because of the lower normal boiling point of HFC-236fa (-1.10 vs.
3.68°C). Due to the lower critical temperature of HFC-236fa (130.65 vs. 145.88°C), its volumetric
capacity at 93.3°C condensing temperature is lower than that of CFC-114.
CENTRIFUGAL COMPRESSOR EVALUATIONS
The majority of compressors used in air-conditioning chillers are centrifugal. Important
parameters for centrifugal compressors have been evaluated for HFC-236fa and compared to CFC-
114. Table 12 presents some of these parameters for a compressor with an impeller diameter of
0.5 m and a tip width of 0.025 m for these refrigerants. In developing this table, the following values
were used for the compressor coefficients: a compressor head coefficient of 0.6, a tip flow coefficient
of 0.3, and a compressor volume flow coefficient of 0.053. This analysis shows that HFC-236fa has
slightly higher discharge pressures. The compressor's refrigerating capacity, QE, with HFC-236fa is
about 26 percent higher, indicating that a smaller impeller (or compressor) can be used.
42
-------
0.95 --
0.9 --
0.85
o
1
0.8 --
0.75
0.7 -
0.65 -
Tc=40.60
Tc=51.70
Tc=65.60 —X—Tc=79.4
Tc=93.3
0.6
0
5
10
15
20
25
30
35
Evaporating temperature (°C)
Figure 7. Coefficient of performance (theoretical) ratio of HFC-236fa relative to CFC-114
-------
1.2 -
o
»¦«
2
&
• P4
u
"S
s
3 0.9 --
o
>
0.8 -
Tc=40.6
Tc=51.7
Tc=65.60
Tc=79.40
Tc=93.30
0.7
0
5
10
15
20
25
30
35
Evaporating temperature (°C)
Figure 8. Volumetric capacity ratio of HFC-236fa relative to CFC-114
-------
Table 12. Centrifugal compressor characteristics at 4°C evaporating and 40°C condensing
temperatures
(From Kazachki and Gage, 1993)
Refrigerant
P
e
P
C
U2
Ma2
N
Vs
VD
Qe
(kPa)
(kPa)
(m/s)
-
(rpm)
(m /s)
(m /s)
(kW)
HFC-236fa
123.1
416.2
174
1.43
6654
2.30
2.05
2076
CFC-114
102.7
336.3
163
1.41
6214
2.15
1.92
1653
COMPRESSOR CALORIMETRY RESULTS OF CENTRIFUGAL CHILLER SIMULATION
One technique for additional evaluation of an alternative refrigerant is a calorimeter test of
both the original refrigerant and the proposed alternative refrigerant. This calorimeter test suggests
how the alternative refrigerant will perform in a certain compressor under actual operating conditions.
Information on capacity and efficiency characteristics of the tested compressor is the output of such a
test. A calorimeter evaluation was made of HFC-236fa as a potential alternative to CFC-114 under
chiller operating conditions.
Comparative evaluations of CFC-114 and HFC-236fa were conducted in a compressor
calorimeter test rig with a semi-hermetic compressor. The compressor was designed for use with HFC
refrigerants, was lubricated with IS068 POE lubricant, and included an air-cooled 0.56 kW (3/4 hp)
3
motor. Its volumetric flow rate was 1.329 L/sec (169.2 ft/hr) at a 1750 rpm motor speed. The
evaluations were made over a wide range of evaporating and condensing temperatures representing
both shipboard chiller and high temperature heat pump conditions. Some limitations on these
conditions were imposed by the calorimeter's capacity rather than by the refrigerant's properties or
compressor characteristics. The tests were performed according to ASHRAE Standard 23-1993
(ASHRAE, 1993).
45
-------
Experimental cooling capacities are presented in Figure 9 as the ratio of HFC-236fa to CFC-
114. The cooling capacities of HFC-236fa are higher than those of CFC-114, except at the higher
condensing temperatures and lower evaporating temperatures. This is consistent with the theoretical
results. The compressor volumetric efficiency with HFC-236fa is essentially the same as with CFC-
114 (Figure 10).
The ratios of measured coefficients of performance (COP) for HFC-236fa relative to CFC-114
are presented in Figure 11. At condensing temperatures up to about 65°C, the HFC-236fa COP is
greater than the CFC-114 COP. At higher condensing temperatures, it drops somewhat below that of
CFC-114 but generally increases with increasing evaporating temperatures. The compressor
isentropic efficiency with HFC-236fa (Figure 12) is about 10 percent better than with CFC-114,
resulting in 10 percent higher COP compared to theoretically predicted values.
With the HFC-236fa refrigerant, the compressor ran for almost 1,800 hours without failure,
excessive noise, vibration, or any other indication of abnormal operation. These test results indicate
that HFC-236fa is a viable alternative refrigerant for CFC-114 in chillers.
46
-------
1.2 --
o
2
*
u
« - „
a 0.9 —
03
u
0.8 --
0.7 --
Tc=40.60
Tc=51.70
Tc=65.60
Tc=79.40
Tc=93.3
0.6
10
15
5
20
25
30
0
35
Evaporating temperature (°C)
Figure 9. Experimental cooling capacity ratio of HFC-236fa relative to CFC-114
-------
1.06
1.04 —
o
V>
rt
u
1.02 -
£
g
o
u
*C 0.98
"5
s
s
O 0.96 —
>
0.94 --
Tc=40.6
Tc=65.6
Tc=51.7
Tc=79.4
Tc=93.30
0.92
10
15
0
5
20
25
30
35
Evaporating temperature (°C)
Figure 10. Volumetric efficiency ratio of HFC-236fa relative to CFC-114
-------
1.2
o
0.8 -
0.7 --
Tc=40.6
Tc=51.70
Tc=65.60
Tc=79.4
Tc=93.3
0.6
5
10
20
25
0
15
30
35
Evaporating temperature (°C)
Figure 11. Coefficient of performance (experimental) ratio of HFC-236fa relative to CFC-114
-------
0.9
¦ Tc=40.60 -0-Tc=51.70 —A— Tc=65.6 Tc=79.4 Tc=93.30
0.8
10 15 20 25 30 35
Evaporating temperature (°C)
Figure 12. Compressor isentropic efficiency ratio of HFC-236fa relative to CFC-114
-------
5. REFERENCES
ANSI/ASHRAE 97-1989, Sealed Glass Tube Method to Test the Chemical Stability of
Material for Use Within Refrigerant Systems, American Society of Heating, Refrigerating and Air-
Conditioning Engineers, 1791 Tullie Circle, NE, Atlanta, GA, 1989.
ASHRAE Standard 34-1992, Number Designation and Safety Classification of Refrigerants,
American Society of Heating, Refrigerating and Air-Conditioning Engineers 1791 Tullie Circle, NE,
Atlanta, GA, 1992.
ASHRAE Standard 23-1993, Methods of Testing for Rating Positive Displacement
Refrigerant Compressors, American Society of Heating, Refrigerating and Air-Conditioning
Engineers, 1791 Tullie Circle, NE, Atlanta, GA, 1993.
Association of Florida Phosphate Chemists, Methods of Analysis for Phosphate Rock; No. 14
Fluoride-F in Methods Used and Adopted by the Association of Florida Phosphate Chemists, Seventh
Edition, P. 0. Box 1645, Bartow, FL, 1991.
ASTM (American Society for Testing and Materials), Standard Test Methods for Measuring
Wear Properties of Fluid Lubricants (Falex Pin and Vee Block Methods), ASTM Designation:
D2670-88, 1916 Race St., Philadelphia, PA, 1988.
ASTM (American Society for Testing and Materials), Standard Test Methods for
Measurement of Extreme Pressure Properties of Fluid Lubricants (Falex Pin and Vee Block
Methods), ASTM Designation: D3233-92, 1916 Race St., Philadelphia, PA, 1992.
ASTM (American Society for Testing and Materials), Standard Test Method for
Concentration Limits of Flammability of Chemicals, ASTM Designation: E681-94, 1916 Race St.,
Philadelphia, PA, 1994.
51
-------
EPA (U. S. Environmental Protection Agency) Health Effects Test Guidelines, Office of Toxic
Substances, Federal Register, 50, No. 188, Part 798.1150, September 1985.
IPCC (Intergovernmental Panel on Climate Change), Climate Change 1995, Cambridge
University Press, UK, 1995.
Jacobsen, R. T. and R. B. Stewart, Thermodynamic Properties of Nitrogen Including Liquid
and Vapor Phases from 63 K to 2000 K with Pressures to 10,000 Bar, J. Phys. Chem. Reference
Data, 2, 757, 1973.
JPL (Jet Propulsion Laboratory), Chemical Kinetics and Photochemical Data for Use in
Stratospheric Modeling, Evaluation Number 11, Jet Propulsion Laboratory, California Institute of
Technology, Pasadena, CA December 15, 1994.
Kazachki, G. S., and C. L. Gage, Thermodynamic Evaluation and Compressor Characteristics
of HFC-236ea and HFC-245ca as CFC-114 and CFC-11 Replacements in Chillers, Presented at the
1993 International CFC and Halon Alternative Conference, Washington, DC, October 20-22,1993.
Kazachki, G. S., C. L. Gage, and R. V. Hendriks, Chlorine-Free Alternatives for CFC-114:
Theoretical and Experimental Investigations, Presented at the 1995 International Congress of
Refrigeration, the Netherlands, August 20, 1995.
Kazachki, G. S., C. L. Gage, R. V. Hendriks, and W. J. Rhodes, Investigation of HFC-236ea
and HFC-236fa as CFC-114 Replacements in High Temperature Heat Pumps, Presented at
52
-------
International Institute of Refrigeration Conference: CFCs, the Day After, Padua, Italy, September
21 -23, 1994.
Nelson, T. P., Findings of the Chlorofluorocarbon Chemical Substitutes International
Committee, EPA-600/9-88-009 (NTIS PB88-195664), April 1988.
NIST Standard Reference Database 23, Thermodynamic Properties of Refrigerants and
Refrigerant Mixtures (REFPROP) Version 5.0, National Institute of Standards and Technology,
Boulder, CO, 1996.
Skaggs, S. R., E. W. Heinonen, T. A. Moore, and J. A. Kirst, Low Ozone-Depleting
Halocarbons as Total-Flood Agents: Volume 2 - Laboratory Scale Fire Suppression and Explosion
Prevention Testing, EPA-600/R-95-150b (NTIS PB96-109079), September 1995.
Smith, N. D. and M. W. Tufts, Thermodynamic Properties of Selected HFC-Refrigerants,
International Journal of Heating, Ventilating, Air-Conditioning and Refrigerating Research, 2:3, 257-
261, July 1996.
UNEP (United Nations Environment Programme), Montreal Protocol on Substances that
Deplete the Ozone Layer - Final Act, 1987.
UOP Tarrytown Analytical Laboratory Procedure, Determination of Fluoride in
Aluminosilicates by Pyrohydrolysis/Ion Selective Electrode, UOP, 25 East Algonquin Road, Des
Plaines, IL, June 1978.
53
------- |