lhm ||III;|I|H1 Irnm \ \l
`
`l11IrI‘11;:li-m.:l In l’-Lin-.nl|;nm1lluu.\\ m|1uMl;u_\..l;m1I;u'\ 20. 201!-
`
`“ E The Engineering Society
`For Advancing Mobility
`“Land Sea Air and Space®
`I N T E R N A T I 0 N A L
`
`400 COMMONWEALTH DRIVE, WARRENDALE. PA 15096-0001 U.S.A.
`
`SAE Technical
`Paper Series
`
`Current Development Status of HFC-134a
`for Automotive Air Conditioning
`
`900213
`
`D. J. Bateman
`
`Du Pom, Wilmington, DE
`
`MAR
`
`71990
`
`SAE
`LIBRARY
`
`International Congress and Exposition
`Detroit, Michigan
`February 26 — March 2, 1990
`
`Page 1 of 14
`
`Arkema Exhibit 1113
`
`

`
`Dim nlnzult-(l l'rum S \l~Z lIlIl'l'll2llil|ll:ll h_\ Riulical llauniltun. \\'t-(lm-~(|:t_\. .I;|nu;u‘_\ 20. 2(II(-
`
`The appearance of the ISSN code at the bottom of this page indicates SAE‘s consent that
`copies of the paper may be made for personal or internal use of specific clients. This consent
`is given on the condition, however, that the copier pay a $5.00 per article copy fee through the
`Copyright Clearance Center, lnc., Operations Center, 27 Congress St., Salem, MA 01 970 for
`copying beyond that permitted by Sections 107or 108 of the U.S. Copyright Law. This consent
`does not extend to other kinds of copying such as copying tor general distribution, for
`advertising or promotional purposes, for creating new collective works, or for resale.
`
`SAE routinely stocks printed papers for a period of three years following date of publication.
`Direct your orders to SAE Customer Service Department.
`
`To obtain quantity reprint rates, permission to reprint a technical paper or permission to use
`copyrighted SAE publications in other works, contact the SAE Publications Group.
`
`the SAE Global Mdwllry Dalabsai
`
`All SAE papers are
`abstracted and indexed in
`
`SAE GLOBAL MOBIJTY DATABASE 0
`
`No part of this publication may be reproduced in any form, in an
`electronic retrieval system or otherwise, without the prior writ-
`ten permission of the publisher.
`
`ISSN 0148-7191
`
`Copyright 1990 Society of Automotive Engineers, Inc.
`
`Positions and opinions advanced in this paper are those of the
`author(s) and not necessarilythose of SAE. The author is solely
`responsible forthe content of the paper. A process is available
`by which discussions will be printed with the paper if
`it
`is
`published in SAE Transactions. For permission to publish this
`paper in full or in part, contact the SAE Publications Division.
`
`Page 2 of 14
`
`Persons wishing to submit papers to be considered for presen-
`tation or publication through SAE should send the manuscript
`or a 300 word abstract of a proposed manuscript to: Secretary,
`Engineering Activity Board, SAE.
`
`Printed in USA
`
`

`
`“H\\ll]H;l(l(‘(| [rum \ \li lllIl‘l'll:IliHll;lI h_\
`
`Iii-.inr-.1 llulnillnn. \\ (‘(lll(‘\(|:l}. .l;inu:ir_\ Ill. lllllu
`
`900213
`
`Current Development Status of HFC-134a
`
`for Automotive Air Conditioning
`
`D. J. Bateman
`
`Du Pont, Wilmington, DE
`
`Abstract
`
`HFC-134a is the refrigerant of choice to replace
`CFC-12 in automotive air conditioning. The
`purpose of this bulletin is to communicate the
`current test status of HFC-134a for use in new
`
`automotive air conditioning systems. It expands
`on the SAE Technical Paper (#890305)—“Per-
`formance Comparison of HFC-134a and CFC-12
`in an Automotive Air Conditioning System" pre-
`sented by Du Pont at the 1989 SAE International
`Congress and Exposition. Topics covered in this
`paper include:
`
`- Environmental Impact
`- Refrigerant Properties
`- Heat Transfer
`
`- Refrigeration Performance
`- Materials Compatibility
`— Hoses
`— Elastomers
`— Desiccants
`— Metals
`- Lubricants
`
`— Solubility
`— Lubricity
`
`Based on the test work completed, HFC-134a is
`not a “drop-in" replacement for CFC-12 in most
`existing systems. Depending on the system,
`certain design modifications may be required to
`achieve optimum performance and durability.
`The significant conclusions from this study are
`listed on page 12.
`
`Introduction
`
`Chlorofluorocarbon 12 (CCIZFZ) is included
`among the compounds regulated under the
`Montreal Protocol, with the possibility of nearly
`
`Page 3 of 14
`
`complete production phase-out by the turn of the
`century. Extensive searches in the 1970s and
`again in 1986-1988 for alternative compounds
`having lower ozone depletion and necessary
`product properties have resulted in industry con-
`sensus that HFC-134a (CH2FCF3) is the most
`suitable replacement for CFC-12 in new automo-
`tive air conditioning applications.
`
`Environmental Impact
`
`Because of its chlorine content and atmospheric
`lifetime of approximately 120 years, CFC-12 has
`been assigned a relative ozone depletion poten-
`tial of 1.0. In addition, due to its ability to absorb
`light rays of certain wavelengths, CFC-12 has an
`estimated global warming potential of 3.0.
`
`In contrast, since HFC-134a contains no chlorine
`
`it has an ozone depletion potential of zero. Due
`to its different chemical structure and much
`
`shorter atmospheric lifetime (about 16 years), it
`also has a much lower global warming potential
`of 0.3. Because of the large volume of CFC-12
`currently used by the automotive industry, replac-
`ing it with HFC-134a provides a significant envi-
`ronmental improvement.
`
`Refrigerant Properties
`
`Thermodynamic properties for HFC-134a have
`been developed by a number of organizations.
`Recently, the National Institute of Standards and
`Technology (NIST) has conducted extensive
`experimental work‘ in developing what many
`consider to be the most accurate set of thermo-
`
`dynamic properties for HFC-134a. Copies of this
`information are available through NIST.
`
`

`
`l)U\\llll|:I(l(‘(l [rum \ \l-L llll(‘l'Il:Ilil|ll;ll h_\ Hizinczi Humillnn. \\ (‘(lll(‘\(|:I). .l;inu:ir_\ III. III]!-
`
`Transport properties (liquid and vapor viscosity,
`heat capacity, and thermal conductivity) are still
`being developed for HFC—134a. Heat transfer co-
`efficients for evaporation and condensation are
`also being determined. These data, along with
`the thermodynamic properties are required by
`the industry to help in redesigning systems to
`optimize the performance of HFC-134a. These
`properties will impact many of the system operat-
`ing parameters such as:
`
`— Coefficient of performance (energy
`efficiency)
`
`— Cooling capacity
`
`— Heat exchanger capacities
`
`— System temperatures/pressures
`
`— Pressure drop in hoses, tubing, heat
`exchangers, etc.
`
`Heat Transfer
`
`As mentioned previously, heat transfer coeffi-
`cients for HFC-134a are required by the industry
`to help understand what changes will be required
`in heat exchanger design to optimize the per-
`formance of HFC—134a. Du Pont has sponsored
`a research project at Iowa State University to
`develop certain heat transfer data. A portion of
`this work has been completed and will be pub-
`lished? in the International Journal of Refrigera-
`tion later this year.
`
`In-tube heat transfer coefficients have been
`
`experimentally determined for HFC—134a and
`CFC-12 at Iowa State. For evaporation at similar
`mass fluxes, HFC-134a heat transfer coefficients
`
`were 35% to 45% higher than those for CFC-12;
`see Figure 1 for the comparison at 10°C. For
`condensation at similar mass fluxes, HFC-134a
`
`heat transfer coefficients were 25 to 35% higher
`than those of CFC-12. Figure 2 is a plot of this
`data at 40°C. In both cases, the increase in heat
`transfer coefficient for HFC-134a is probably due
`to its higher liquid thermal conductivity. Tests
`were also conducted at other temperatures; this
`data will be included in the published paper.
`
`Comparisons were also made for equivalent
`cooling and heating capacities by multiplying
`mass flow rates and enthalpies of vaporization.
`This comparison is based on the fact that the
`enthalpy of vaporization is higher for HFC-134a
`compared to CFC-12, and therefore lower mass
`flow rates are required for HFC-134a to provide
`
`equivalent heat transfer rates. As the mass flow
`rate of HFC-134a is reduced, the in-tube heat
`transfer coefficient is also reduced. Even when
`the heat transfer coefficient is reduced because
`
`of lower mass flow rates, it is still higher for
`HFC-134a. The preceding discussion is appli-
`cable when considering using HFC-134a in a
`heat exchanger designed for CFC-12. For
`equivalent cooling, HFC—134a heat transfer
`coefficients were 5 to 15% higher than CFC-12.
`At equivalent heating capacity, HFC-134a heat
`transfer coefficients were 10 to 20% higher.
`
`Refrigeration Performance
`
`Performance comparison of HFC-134a and
`CFC-12 in automotive air conditioning has been
`reported in several papers“ 5 presented at
`various industry association meetings. Most of
`the tests have been conducted in vehicles de-
`
`signed for CFC-12. As more knowledge is gained
`about HFC-134a and designs are modified, per-
`formance deficiencies noted in these tests
`
`will most probably be reduced and possibly
`eliminated.
`
`The theoretical performance of CFC-12 and
`HFC—134a in a typical automotive system is
`compared in Figures 3 and 4. In this example,
`the data indicate that CFC-12 is about 6% more
`
`energy efficient and has about 4% greater capac-
`ity. Another key system operating parameter is
`compressor discharge pressure. The pressure
`with HFC-134a is about 25 psi higher than with
`CFC-12. Although the compressor discharge
`pressure is higher, the discharge temperature for
`HFC-134a is about 14°F lower due to its higher
`specific heat.
`
`Based on theoretical calculations, performance
`tests and information gained from other tests,
`such as those reported in this paper, HFC—134a
`is not a “drop-in” replacement for CFC-12 in most
`automotive systems. As stated earlier, it is the
`replacement of choice but design modifications
`will be required to optimize performance and
`durability. Development work is well underway
`throughout the industry in the following areas:
`compressors, condensers, evaporators, desic-
`cants, hoses and elastomers. There is also work
`
`underway to determine if the cycling clutch orifice
`tube system is acceptable, or if the use of a
`thermal expansion valve will provide better
`performance.
`
`Page 4 of 14
`
`

`
`l)n\\ l]Il|;I(ll‘(I Irum .\ \E IIlI('l'll2llIIIll:lI h_\ ltiaulcu llumillun. \\ ('(IIl(‘\(I1l). .l:u1u:rr_\
`
`lII_ llllfu
`
`FIGURE ‘I.
`
`FIGURE 2.
`
`MEASURED EVAPORATION HEAT TRANSFER
`COEFFICIENTS FOR HFC-1348 AND CFC-12
`
`MEASURED CONDENSATION HEAT TRANSFER
`COEFFICIENTS FOR HFC-1343 AND CFC-12
`
`X103
`5
`
`X103
`4
`
`4
`
`2
`
`‘Te
`E
`5
`q)
`
`8
`.5
`
`I:
`16
`OJ
`I
`
`N
`? 3
`g
`E
`::'
`(1)
`
`8 2
`9‘:
`
`+-
`6
`0)
`I
`
`1
`
` 2
`
`Temp - 40°C
`
`0
`
`0
`
`100
`
`200
`
`300
`
`400
`
`500
`
`o ' I
`0
`100
`200
`300
`400
`500
`
`Mass Flux (kg/m2-Sec)
`
`Mass Flux (kg/m2-Sec)
`
`HFC-134a
`
`— — CFC-12
`
`HFC-134a
`
`— — CFC-12
`
`FIGURE 3.
`
`FIGURE 4.
`
`CFC-12: THEORETICAL PERFORMANCE
`IN AUTOMOTIVE AIR CONDITIONER
`
`HFC-134a: THEORETICAL PERFORMANCE
`IN AUTOMOTIVE AIR CONDITIONER
`
`Operating Conditions:
`T (condenser exit) = 150°F:
`Liquid sub-cooled to 140'°F:
`Compressor (isentropic)
`efficiency = 0.7
`Pressure drop at compressor inlet = 5 psi
`
`T (evaporator exit) = 40°F
`Suction superheated to 50°F
`
`Operating Conditions:
`T (condenser exit) = 150°F:
`Liquid sub-cooled to 140°F:
`Compressor (isentropic)
`efficiency = 0.7
`Pressure drop at compressor inlet = 5 psi
`
`T (evaporator exit) = 40°F
`Suction superheated to 50°F
`
`Enthalpy
`_V____ __(Etu/l_I_3_)_ _ ____P_(psla)
`V
`Condenser inlet
`92.32
`249.48
`Condenser exit
`45.08
`249.48
`Evaporator inlet
`42.18
`51.61
`Evaporator exit
`82.81
`51 .61
`Compressor inlet
`84.46
`46.61
`Compressor exit
`103.93
`249.48
`
`V
`
`Refrigeration Performance
`Compression Ratio
`Net Refrigeration (Btu/lb)
`Net Condenser Heat (Btu/lb)
`Refrigerant Circulated (lb/min)
`Compressor Heat (Btu/lb)
`Compressor Power (Btu/min)
`(Watt)
`Compressor Displacement (cu ft/min)
`Refrigeration Load (Btu/min)
`C.O.P. for Cooling
`
`T_(°F)
`150.00
`150.00
`40.00
`40.00
`50.00
`213.10
`
`4.83
`40.62
`47.24
`3.88
`19.47
`75.55
`1 ,327.21
`3.50
`157.62
`2.09
`
`_7__
`Condenser inlet
`Condenser exit
`Evaporator inlet
`Evaporator exit
`Compressor inlet
`Compressor exit
`
`Enthalpy
`(Btu/lb)
`122.05
`64.96
`60.46
`108.83
`111.11
`135.61
`
`P (psia)
`274.93
`274.93
`49.70
`49.70
`44.70
`274.93
`
`Refrigeration Performance
`Compression Ratio
`Net Refrigeration (Btu./lb)
`Net Condenser Heat (Btu/lb)
`Refrigerant Circulated (lb/min)
`Compressor Heat (Btu/lb)
`Compressor Power (Btu/min)
`(Watt)
`Compressor Displacement (cu ft/min)
`Refrigeration Load (Btu/min)
`C.O.P. for Cooling
`
`T (°F)
`150.00
`150.00
`40.00
`40.00
`50.00
`198.60
`
`5.53
`48.38
`57.09
`3.13
`24.50
`76.76
`1,348.60
`3.50
`151.56
`1.97
`
`Based on a fixed compressor displacement of 3.50 cu ft/min
`
`Based on a fixed compressor displacement of 3.50 cu ft-‘min
`
`Page 5 of 14
`
`

`
`I)n\\l1ln2l(|(‘(| lruin .\ \l-L lllll‘l‘Il;llilIlliIl h_\ “i:lll(':l llumiltun. \\('(lll(‘\(|:l_\..lilllll1lI'} lit, Juli‘.
`
`Hose Permeation
`
`Since HFC-134a will replace CFC-12 for automo-
`tive air conditioning, it was necessary to deter-
`mine if permeation rates through existing hoses
`are acceptable.
`
`Permeation Rate
`
`The studies were run with an initial 80 volume 0/0
`
`liquid loading of refrigerant at 80°C in 5/8" l.D. x
`about 30" length air-conditioning hoses. The tests
`included: CFC-12 as a base line, and HFC-134a.
`
`The hoses were:
`
`- Nylon-lined
`- HYPALONE 48
`
`- “Nitrile-1"
`- “Nitrile-2"
`
`and nylon hoses would have permeation loss
`rates of 0.05-0.08 pounds per year.
`
`Experimental:
`
`After leak testing the hoses by pressurizing with
`400 psig of nitrogen and immersing in water, the
`hoses were dried, evacuated, and charged with
`refrigerant to give an 80 volume % liquid fill at
`80°C. These hoses were laid horizontally in an
`80°C oven along with control hoses with no
`refrigerant contained in them so that the weight
`losses of these control hoses could be subtracted
`
`from the hoses with refrigerant to get the true
`weight loss due to permeation. These weight
`losses are shown graphically in Figure 5.
`
`Hose construction was:
`
`Nitrile
`
`Elastomer/Plastic Compatibility
`
`Nylon
`
`HYPALON-‘A 48 (1 & 2)
`
`Inner liner
`Second layer
`Reinforcement
`Outer cover
`
`Nylon
`—
`Nylon
`Chlorobutyl
`
`HYPALON *1 48 Nitrile
`Rayon
`Rayon
`2 braids
`2 braids
`EPDM
`EPDM
`
`Elastomers and plastics are used in all automo-
`tive systems as seals. gaskets. O-rings and for
`construction of hoses. Tests of several commonly
`used elastomers and plastics with CFC-12/
`mineral oil and HFC-134a/polyalkylene glycol
`
`FIGURE 5.
`
`HOSE PERMEAT|0N—HFC-134a VS. CFC-12
`
`120
`
`100
`
`20
`
`The permeation rates were calculated in units of
`lbs/ft — yr for the 5/8" l.D. hose. The results were
`as follows:
`
`Permeation Rate (lbs/ft — yr)
`Nylon HYPALON“ 48 “NitriIe-1" “Nitrile-2”
`
`CFC-12
`HFC-134a
`
`0.3
`0.2
`
`1.0
`0.2
`
`1.5
`1.8
`
`1.9
`2.7
`
`At the end of the tests, the hoses were sectioned
`
`to permit internal examination. All of the inner
`surfaces appeared and felt acceptable for con-
`tinued use. Based on permeation rate tests,
`nylon-lined and HYPALON-‘~ 48 hoses appear
`to be suitable for use with HFC-134a.
`
`Please note that these permeation rate tests
`provide a comparison of the various hoses in a
`laboratory oven at 80°C and are not an indication
`of the actual permeation losses from an operat-
`ing automotive air conditioning system. The
`actual permeation losses will be much lower
`because the system does not operate continu-
`ously at 80°C. It has been estimated (based on
`typical operating time and temperatures) that
`permeation losses from a system containing
`CFC-12 and nitrile hoses is about 0.2 pounds per
`year per vehicle. For further comparison, it has
`been estimated that systems containing CFC-12
`
`Page 6 of 14
`
`O)O
`
`
`
`PermeationLoss(grams) A0':OO
`
`0
`
`100
`
`200
`
`300
`
`400
`
`500
`
`600
`
`700
`
`800
`
`Time (hours)
`
`j 12/Nylon
`—o— 12/HYPALON?'48
`—E}— 12/Nitrile-1
`—A— 12/Nitrile-2
`
`—)(~ 134a/Nylon
`
`-9- 134a/HYPALor\F’48
`—I-— 134a/Nitrile-1
`—.A.—_ 134a/Nitrile—2
`
`

`
`”H\\llllI:I(l(‘(l [rum \ \l-L I111;-rnulimml h_\ Ilium-;i llumilfnn. \\ 1-(lm-~(|:r\. _l:inu:ir_\ III. 101(-
`
`(PAG) have been performed to determine relative
`compatibility. The following materials were
`tested:
`
`with CFC-12. This is not consistent with
`
`other results for this material and may be
`incorrect.
`
`Elastomers:
`
`Plastics:
`
`VITONFT A
`Epichlorohydrin
`HYPALON's 48
`Neoprene W
`
`Lubricants:
`
`Nylon
`Graphite-filled
`TEFLON-'-‘ (~15%)
`
`Paraffinic mineral oil
`PAG
`
`Refrigerants:
`
`CFC-12
`HFC-134a
`
`Each material was exposed to CFC-12/mineral
`oil and HFC—134a/PAG at two temperatures
`(176°F and 268°F) for a period of 14 days in
`sealed Pyrexii compatibility tubes. At the end of
`the 14 days, the test specimens were removed
`from the tube and immediately checked for
`change in length and weight (temporary change).
`The materials were then stored at room tem-
`
`perature for an additional 14 days and the final
`changes in length and weight were determined.
`
`Although additional testing will have to be done
`for specific application requirements, these tests
`provide a general understanding of the relative
`compatibility with HFC-134a/PAG compared to
`CFC-12/mineral oil. Figure 6 graphically shows
`the temporary and final length change of the
`materials when exposed to the two refrigerant/oil
`mixtures at 176°F and 268°F. For example: the
`temporary length change for Neoprene W (CFC-
`12 at 176°F) was 2.7%, and the final length
`change was -1.4%. In the case of nylon, the
`temporary and final length changes were the
`same for CFC-12 at 176°F and HFC-134a at both
`
`temperatures. Figure 7 provides similar informa-
`tion for change in weight. Some basic conclu-
`sions can be drawn from this work:
`
`- VITON9 A shows excessive temporary
`weight and length change with both CFC-12
`and HFC-134a, and is probably unaccept-
`able in this application with CFC-12 or
`HFC-134a.
`
`- The temporary and final length changes for
`the other five materials are within plus or
`minus 5% of original, and are similar for
`CFC-12 and HFC-134a. The one exception
`is the graphite-filled TEFLON~‘i' at 176°F
`
`- The temporary and final weight changes for
`the nylon and graphite-filled TEFLONE is
`very similar for CFC-12 and HFC-134a.
`These materials should be suitable for use
`
`with either refrigerant/oil mixture.
`
`- The temporary and final weight changes for
`HYPALON5 48 and Neoprene W are within
`plus or minus 10% of original. However,
`the changes are in opposite directions and
`these materials might require further evalu-
`ation. Neoprene W is currently being used in
`many HFC-134a test systems and does not
`appear to present any problems.
`
`- The temporary weight change for epich|o-
`rohydrin and HFC—134a was higher than the
`other materials (except VlTON-’“- A), espe-
`cially at 268°F. Further testing will be re-
`quired for specific applications to determine
`if this is a valid concern.
`
`Desiccant Compatibility
`
`UOP (formerly Union Carbide Molecular Sieve)
`is a major supplier of molecular sieve deslccants
`for automotive air conditioning. They have been
`doing a great deal of work to develop suitable
`deslccants for use with HFC-134a. The data in
`
`this section have been supplied by UOP. Figure
`8 summarizes compatibility tests of HFC-134a/
`PAG oil with two commercially available desic-
`cants, 4A-XH-5 and 4A-XH-6; and a develop-
`mental molecular sieve (XH-7). Compatibility of
`CFC-12/Suniso 3G8 with 4A-XH-5 is also pro-
`vided for comparison. The water capacity, crush
`strength, and attrition data are reported as per-
`cent or multiple of fresh values so that a compari-
`son can be made. Chloride, fluoride, and water
`
`content are also given.
`
`The water content was determined by Karl Fisher
`titration. In the tests with HFC-134a the initial
`
`water level on the activated samples was less
`than 1.0%. Any additional water comes mainly
`from refrigerant/oil/desiccant reactions. In smaller
`amounts it also comes from handling in the
`atmosphere and from water contained in the
`refrigerant and oil charged. The molecular sieve
`contained 0.02 to 1.7% fluoride and 0.05 chloride
`
`after conditioning at 550°F in nitrogen purge to
`remove residual refrigerant, water, and oil. The
`
`Page 7 of 14
`
`

`
`I)u\\IlIn;u|(~«I lrum \ \I, Illit-I’Il;iliul1;1I|I\
`
`I’-inn: :1 Hzllllillull. \\ l‘(lIll‘\lI;l_\. ,I.IIllI;II’\ I0. 301(-
`
`FIGURE 6
`
`CHANGE IN LENGTH OF ELASTOMERS/PLASTICS
`
`(°/o Length Change vs. initial)
`
`-10
`
`-5
`
`0
`
`%
`
`5
`
`10
`
`15
`
`20
`
`Nylon
`
`V|TON®A
`
`ECO
`
`HYPALON®48
`
`Neoprene W
`
`Graphite-
`Filled TFE
`
`14-DayExposure }
`
`14-Day Drying
`
`CFC-12/M.O. -7”; _
`
`HFC-134a/PAG $71111 %i23!32§
`Temp.
`Final
`Temp.
`Final
`
`1 76°
`
`268°
`
`Page 8 of 14
`
`

`
`I):-u l||II;II|(‘ll [rum \ \l
`
`lIIIvI‘II;IlinII;II I-\
`
`I’-i;IIII;I l|;IInilluII. \\ l‘||ll(’\|I£L‘. .I;IIIII;II’\ 20. 20!!-
`
`FIGURE 7
`
`WEIGHT CHANGE OF ELAST0MERS/PLAST|CS
`
`(% WEIGHT CHANGE vs. INITIAL)
`
`-10
`
`-5
`
`0
`
`°/o
`
`5
`
`10
`
`I5
`
`20
`
`Nylon {
`VITON®A{
`ECO{
`HYPALON®48{
`
`Neoprene W
`
`Graphite—
`Filled TFE
`
`I we I
`I: I III.
`I
`5959494“
`I
`
`I
`
`I
`
`|
`
`I
`
`I
`
`E1 1
`
`E
`3333
`
`I
`
`I
`
`I
`
`I
`
`I
`
`F
`
`I
`
`I
`
`I
`I
`I
`
`|
`
`I
`
`I
`
`14-Day Exposure }
`
`14-Day Drying
`
`CFC-12/M.O. ‘V///. ‘
`elder‘,
`'
`Temp.
`Final
`Temp.
`Final
`
`HFC-134a/PAG
`
`176°
`
`268°
`
`Page 9 of 14
`
`

`
`l)n\\llllI:I(I(‘(l I'rnm .\ \l£ lnlurnuliunail h_\ Rium-at llumillnn. \\ ('(lll('\(|:I_\. _l:mu:ir_\ lll. ZIIN.
`
`FIGURE 8
`
`DESICCANT COMPATIBILITY
`
`HFC-134a, PAG OIL
`
`4A-XH-5
`
`4A-XH-6
`
`XH-7
`
`4A-XH-5
`
`(CFC-12/3GS)
`
`Residual Water, Wt. °/o (init)
`
`<1.0
`
`Conditioned Samples:
`Water Capacity, °/o of Fresh
`
`Crush Strength, % of Fresh
`
`Attrition, Multiple of Fresh
`
`Fluoride, Wt. %
`
`Chloride, Wt. %
`
`91
`
`77
`
`1.0
`
`1.7
`
`—
`
`<1.0
`
`95
`
`82
`
`1.32
`
`0.02
`
`—
`
`<1.0
`
`98
`
`89
`
`1.0
`
`0.04
`
`—
`
`<1.0
`
`97
`
`92
`
`1.0
`
`——
`
`0.05
`
`Exposure Conditions: 180°F, 14 Days
`Molecular Sieve Sample Conditioning: 550°F in Nitrogen purge for at least 2 hours
`
`source of chloride and fluoride is HCI and HF
`
`from decomposition of the refrigerants. Excessive
`amounts of these acids are produced by reaction
`on the adsorption surfaces of incompatible
`molecular sieves. The molecular sieve removes
`
`acids from the system by reaction with the zeolite
`crystal. This reduces the water capacity of the
`desiccant as shown in the data.
`
`The 4A-XH-5 took up 1.7% of its weight in fluo-
`ride and suffered some damage in the compati-
`bility test with HFC-134a and PAG. Its water
`capacity was reduced to about 91% of fresh
`water capacity. The crush strength was reduced
`by 23%; the attrition tendency was unaffected.
`Attrition is the relative mass of fine particles
`produced in a shake test. The damage was done
`by HF produced as a result of decomposition of
`the refrigerant in contact with the molecular sieve
`surfaces. Type 4A-XH-5 is not compatible with
`HFC—134a and should not be used with this
`
`refrigerant.
`
`Both 4A-XH-6 and developmental molecular
`sieve (XH-7) took up negligible amounts of
`fluoride in the test (0.02—0.04 wt. °/o). There was
`minimal loss in water capacity with these molecu-
`lar sieves. There were minor losses in physical
`properties that are seen as a result of any expo-
`sure to refrigerant and oil. Types 4A-XH-6 and
`XH-7 are appropriate for evaluation in refrigera-
`tion systems. However, type XH-7 is recom-
`mended for applications where superior physical
`properties are required; i.e., automotive air
`conditioning.
`
`HFC-134a Compatibility with
`Metals/Lubricants
`
`CFC-12 and existing mineral oils have proven to
`be very stable mixtures when in contact with
`typical metals found in automotive air condition-
`ing systems—copper, steel and aluminum.
`Certain types of polyalkylene glycols (PAGS) are
`now being considered as lubricants for HFC-
`134a in automotive systems. Several tests have
`been run to determine the stability of HFC-134a
`and PAG oils when exposed to copper, steel and
`aluminum. Figure 9 summarizes tests that were
`run in sealed Pyrex‘ tubes with refrigerant/oil
`mixtures containing about 85 volume % refriger-
`ant in contact with steel 1010, copper and alumi-
`num 1100 tied together at one end with copper
`wire and copper wire spacers between the met-
`als. The tests were run at 268°F for 11.8 days
`to simulate a 10-year life of a car.
`
`Figure 10 shows the results of a similar test
`which was conducted at higher temperature but
`for a shorter duration. Nylon was included in this
`test to gain data on automotive system exposure.
`Both with and without nylon, the HFC-184a/PAG
`combination showed less halide (CI or F‘) forma-
`tion and equal or less visual change to the liquid
`and coupons than did the CFC-12/paraffin combi-
`nation. As the CFC-12/paraffin combination has
`been proven in actual service, both tests indicate
`that HFC—134a/PAG solutions have acceptable
`chemical stability. In several other tests, results
`have confirmed that the HFC-134a molecule is at
`
`least as stable chemically as CFC-12.
`
`Page 10 of 14
`
`

`
`l)0\\llIlI:I(I(‘(I lrum \ \l-L IllI(‘l'Il:llilIll;lI h_\ Ilium-;r llumiltnn. \\ 1-(lm-~(|:r\. _l:mu:rr_\ III. Zilli-
`
`FIGURE 9
`
`HFC-134a/MATERIALS COMPATIBILITY
`
`85/15 VOL. % REFRIGERANT/OIL SOLUTION
`STEEL 1010/COPPER/A1-1100
`
`PPM
`
`Effect Rating
`
`Refrigerant
`
`HFC-134a
`
`CFC-12
`CFC-12
`
`Oil (500 SUS)
`
`Halides
`
`LIQ
`
`Fe
`
`Cu
`
`PAG
`
`Naphthenic
`Paraftinic
`
`<0.2
`
`423
`(a)
`
`0
`
`4
`O
`
`0
`
`3
`3
`
`0
`
`2
`0
`
`Al
`
`0
`
`2
`0
`
`Rating Range:
`
`0 = No Effect
`5 = Severe Effect
`
`Test conditions: 268°F for 11.8 Days
`
`(a) Tube broke
`
`FIGURE 10
`
`HFC-134a/MATERIALS COMPATIBILITY
`
`90/10 VOL. % REFRIGERANT/OIL SOLUTION
`STEEL 1010/COPPER/A1-1100
`
`PPM
`
`Effect Rating
`
`Refrigerant
`
`Oil (500 SUS)
`
`Nylon
`
`Halides
`
`LIQ
`
`Fe
`
`Cu
`
`Al
`
`Nylon
`
`HFC-134a
`HFC-134a
`CFC-12
`CFC-12
`
`PAG
`PAG
`Paraffinic
`Paraffinic
`
`N
`Y
`N
`Y
`
`<0.2
`<0.2
`611
`2287
`
`O
`O
`1
`5
`
`2
`0
`2
`4
`
`0
`0
`0
`3
`
`0
`0
`0
`4
`
`—
`0
`—
`5
`
`Rating Range:
`
`0 = No Effect
`5 = Severe Effect
`
`Test conditions: 350°F for 3.0 Days
`
`Although these tests indicate that HFC-134a/
`PAG is stable in the presence of typical air
`conditioning system metals, there has been
`considerable discussion in the industry about
`copper plating or corrosion. Additional tests are
`being run to determine if this is a valid concern.
`
`Lubricants
`
`The previous section described how testing has
`confirmed that the HFC-134a/PAG combination
`
`has acceptable chemical stability for use in
`automotive systems. There are other criteria to
`consider when selecting lubricants. These in-
`clude solubility in the refrigerant and the lubrica-
`tion characteristics of the refrigerant/oil mixture.
`
`Page 11 of 14
`
`Solubility
`
`Existing automotive air conditioning mineral oils
`(500 SUS viscosity) are completely soluble in
`CFC-12 over a wide temperature range. This
`ensures that the oil moves freely around the
`system and returns to the compressor at a rate
`sufficient to provide acceptable lubrication. Many
`of the PAGs that are being evaluated for use in
`automotive systems are not completely soluble
`with HFC-134a. One example is shown in Figure
`11 where the PAG shows completely different
`solubility characteristics in HFC-134a compared
`to CFC-12/mineral oils. Whereas the CFC-12/
`
`mineral oil separates into two phases below the
`solubility curve, the HFC-134a/PAG does so
`
`

`
`])n\\llln;I(l(‘(l lrum \ \l-L lnlurnuliniml h_\
`
`lfizinczi llumiltnn. \\ (‘(lll(‘\(|:[\. _l:inu:ir_\ III. 101(-
`
`adequate lubrication for the compressor. CFC-12/
`mineral oils have demonstrated good perfor-
`mance in this area. Recently, a number of labora-
`tory tests have been run on certain PAGs to
`compare their lubricity with mineral oils. Tests
`have been run with oil only and also with oil/
`refrigerant mixtures. Figure 12 summarizes Falex
`Load to Failure testing on a candidate PAG
`versus existing mineral oils (no refrigerant). The
`lubricants were tested with a Falex Pin and Vee
`
`Block friction and wear testing machine made by
`the FaviIle—LeVally Corporation. AlSl-1137 steel
`vee blocks were used along with #8 steel pins.
`Each lubricant sample was run at 250 lbs of
`pressure for 5 minutes. The pressure exerted on
`the rotating pin by the vee blocks was then in-
`creased to 300 pounds and held at this pressure
`for 3 minutes. The pressure was increased by
`100 lb increments (3 minutes at each pressure)
`until failure. The higher the pressure needed to
`achieve failure, the better the lubricant.
`
`Since the oil/refrigerant mixture (rather than the
`oil alone) will provide compressor lubrication,
`comparative bench tests have been run to deter-
`mine the influence of refrigerant on PAGs and
`existing mineral oils. Tests similar to those de-
`scribed above were conducted to determine the
`
`load carrying ability of the oils under an atmo-
`sphere of refrigerant gas in the Falex Pin and
`V—B|ock test (Figure 13). The data show the load-
`carrying capability of the PAG is superior to the
`existing mineral oils. Further, the load-carrying
`ability of the PAG is significantly enhanced by the
`presence of an extreme pressure additive.
`
`FIGURE 13
`
`LOAD-CARRYING ABILITY OF OILS UNDER
`AN ATMOSPHERE OF EITHER HFC-134a OR
`CFC-12 IN THE FALEX PIN/V-BLOCK TEST
`
`Failure
`Load
`
`Torque
`at Fall
`
`Refrigerant
`
`Oil (500 SUS)
`
`(lbs)*
`
`(in-lbs)‘
`
`HFC-134a
`HFC-134a
`
`CFC-12
`CFC-12
`
`PAG
`PAG + EP
`Additive
`
`Naphthenic
`Paraffinic
`
`1750
`
`4150
`
`1250
`1500
`
`28
`
`46
`
`24
`33
`
`‘The larger the numbers, the better
`
`FIGURE 11
`
`SOLUBILITY OF CANDIDATE PAG
`WITH HFC-1348
`
`100
`
`(D0
`
`2, 40
`
`Two Phases
`Above Line
`
`92G
`
`0)O 6
`
`!
`75 20
`D.
`EG)
`*'
`
`o
`
`-20
`
`-40
`
`Two Phases Below Line
`
`0
`
`20
`
`4o
`
`60
`
`80
`
`100
`
`Weight °/o Refrigerant in Refrig./Oil Mixture
`
`PAG (500 SUS)
`
`— — Mineral Oi|r’CFC-12
`
`FIGURE 12
`
`FALEX TEST RESULTS
`
`(No Refrigerant)
`
`Oil (500 SUS)
`
`PAG
`
`(No additives)
`Naphthenic
`Paraffinic
`
`Viscosity
`cst @ 40°C
`
`Failure
`Load (lbs)
`
`100
`
`115
`1 13
`
`1100
`
`700
`700
`
`above the curve. This means that in parts of the
`system that operate above the minimum solubility
`temperature and at certain oil/refrigerant ratios, two
`phases will occur and the oil rich portion might
`collect at these locations. This could cause flow
`
`restrictions, reduced heat transfer and possibly
`insufficient oil return to the compressor. Individ-
`ual system testing needs to be done to determine
`the impact that partial solubility might have on
`system performance and compressor durability.
`
`Lubricity
`
`Another key criteria for selecting a lubricant is the
`ability of the refrigerant/oil mixture to provide
`
`Page 12 of 14
`
`

`
`])n\\llll|iI(l('(l [rum .\ \l£ lnlurnuliunul h_\
`
`liizim-at llumiltnn. \\ 1-(lm-~(|:r\. _l:inu:ir_\ III. Zilli-
`
`Lubrication of moving parts under heavy-load
`conditions can be enhanced by the addition of an
`appropriate extreme pressure (EP) additive to an
`oil. Selection of the oil additive is based on its
`
`miscibility in the refrigerant/oil combination, and
`its performance in screening tests such as the
`four—ball wear test and the Falex V—notch failure-
`load test. Results of the four-ball wear tests are
`
`described in Figure 14. The use of the EP addi-
`tive with the PAG appears to have increased the
`wear scar (.357 vs .303). But, based on the
`accuracy of the tests, these two numbers are
`considered to indicate similar wear.
`
`FIGURE 14
`
`LUBRICITY OF OILS UNDER AN ATMOSPHERE
`OF EITHER CFC-12 OR HFC-1348 IN A
`FOUR-BALL WEAR TEST
`
`Refrigerant
`
`Oil (500 SUS)
`
`S‘
`
`F*
`
`HFC—134a
`HFC—134a
`
`CFC-12
`CFC-12
`
`Table key:
`
`PAG
`PAG + EP
`Additive
`
`Naphthenic
`Paraffinic
`
`0.303
`
`0.023
`
`0.357
`
`0.373
`0.368
`
`0.030
`
`0.072
`0.071
`
`S = ball scar wear in mm
`F = coefficient of friction
`‘The smaller the number. the better
`
`Based on this testing and knowledge gained
`through contacts with the automotive industry, it
`is believed that certain types of PAGs are cur-
`rently the best available lubricant for use with
`HFC—134a. However, there are some issues
`
`that still require resolution:
`
`What is the long-term impact of incomplete
`solubility on compressor durability?
`
`What is the role of the PAG in copper plating?
`
`Will the hygroscopic nature of the PAGs
`create unacceptable moisture levels in the
`system?
`
`Certain PAGs exhibit less than desired lu-
`
`bricity for steel-on—aluminum surfaces. Can
`this deficiency be corrected by additive
`technology or are other lubricant structures
`required?
`
`Other non—PAG materials are being evaluated as
`lubricants for use with HFC-134a. These materi-
`
`als are in the early development stage and are
`not yet available for widespread testing. As more
`knowledge is gained. they will be made available
`to the industry in limited quantities for specific
`application testing.
`
`Conclusions
`
`- HFC-134a is the refrigerant of choice to
`replace CFC-12 in automotive air
`conditioning.
`
`- Although the properties of HFC-134a are
`similar to CFC-12, it is not a “drop-in” re-
`placement for most automotive applications.
`Depending on the system design, modifica-
`tions will be required to ensure acceptable
`cooling performance and long-term system
`durability.
`
`- Due to its low ozone depletion and global
`warming potentials, the use of HFC-134a
`will provide a significant positive environ-
`mental impact.
`
`- Heat transfer coefficients for HFC-134a are
`
`significantly better than those of CFC-12.
`
`- Based on tests performed, the permeation
`rates for Nylon-lined and HYPALON3 48
`hoses with HFC—134a are very similar to
`those of CFC-12/Nylon and should be
`acceptable for use.
`
`- Compatibility testing of several commonly
`used plastic and elastomeric materials has
`demonstrated that suitable materials are
`
`available, or can be developed for seals,
`gaskets and O-r