`
`SASSREEBURGEORD
`
`
`
`Status Report on the Development of HFC-245fa as a
`Blowing Agent.
`
`M. C. BOGDAN, D. J. WILLIAMS, P. B. LOGSDON, R. C. PARKER
`AlliedSignalInc.
`
`ABSTRACT
`December 31,1995 marked the end to the manufacture
`of CFCs in the U.S. for use as blowing agents. The foam
`industry has successfully converted to HCFC-141b and
`other blowing agents. This, however, is not the last tran-
`sition that the industry needs to make.
`.
`HCFCs have specific phaseout dates, according to the
`Montreal Protocol. which differ across the. globe. Cur-
`rently, the manufacture of HCFC-141b for use as a blow-
`ing agent in the U.S., will cease on January 1, 2003. The
`identification of the next generation ofblowing agents has
`begun as a response to this market need.
`AlliedSignal is committed to supplying a blowing agent
`to meet customer needs in the foam industry. In 1994, at
`the Polyurethane World Congress, AlliedSignal «an-
`nounced HFC-245fa (1,1,1,3,3-. pentafluoropropane) as
`their “third generation” blowing agent. candidate to
`replace HCFC-141b. HFC-245fa was selected on the basis
`of the optimum combination of possessed attributes; it is
`a liquid, nonflammable blowing agent. with acceptable
`environmental characteristics and good toxicological
`properties, Testing performed to date on foams prepared
`with HFC-245fa have demonstrated improved physical
`properties and k-factor. aging versus those made with
`HCFC-141b. This paper discusses the recent work con-
`ducted by AlliedSignal in continuing the development. of
`HFC-245fa ¢
`:
`Before 1996, AlliedSignal’s major research efforts have
`been in defining the properties of HFC-245fa and proving
`it as a feasible blowing agent. We are sensitive to the com-
`plexities ofdeveloping new foam technologies and want to
`minimize any transition difficulties through customer
`partnering, Because of this our focus has now shifted to
`developing a database of information to assist our cus-
`tomers in selecting and converting to HFC-245fa. .
`This paper contains a discussion of the latest toxicity
`and environmental data. It will focus on answering some
`commonly asked customer questions in the areasof poly-
`ol solubility, premix vapor pressures, andcell gas content.
`Research work is currently underway to optimize the
`use of HFC-246fa in various industry applications which
`focuses on improving k-factor and physical properties.
`This work, along with a comparison of Total Equivalent
`Warming Impact (TEWDof various blowing agents being
`considered, is also included.
`
`INTRODUCTION
`
`re production ofHCFC-141b is currently scheduled to
`cease as of January 1, 2003. In order to meet future
`energy and environmental requirements, alternative
`blowing agents to HCFCs, currently used in insulating
`foam applications, will be required.
`Although there are several potential options for replace-
`ment blowing agents, many applications will continue to
`require a liquid , non-flammable blowing agent designed
`to produce foams with low thermal conductivity using tra-
`ditional processing techniques. The most promising class
`of materials to fill this need are the liquid HFCs. Allied-
`Signal
`is committed to supplying an alternative HFC
`blowing agent to meet these needs.
`In 1994 at the Polyurethane World Congress, AdHed-
`Signal announced that HFC-245fa(1,1,1,3,3- pentafluoro-
`propane) is our primary blowing agent candidate to
`replace HCFC-141b. We consider HFC-245fa to be. the
`optimum replacement candidate for HCFC-141b since it
`has similar physical properties,
`is non-flammable ,
`is
`acceptable environmentally and toxicologically and can
`meet the performance requirements of the foam insula-
`tion market.
`/
`:
`With the transition from HCFC-141b on the horizon,
`there is much work to begin for an effective transition.
`Each foam market segment has its own set of testing and
`trials which must be completed before products contain-
`ing a new blowing agent can be commercially sold. ‘The
`extent and time requirements vary dramatically from seg-
`ment to segment. In addition, there ara Several phases of
`toxicity testing which must be completed and government
`approvals which must be obtained prior to HPC-245fa
`becoming a commercial product.
`AlliedSignal has presented a continuing series of
`papers since 1992, updating the industry on our liquid
`HFC blowing agent development program. This paper
`provides the most recent information on the physical
`properties and toxicology of HFC-245fa. With the stronger
`pushto utilize energy efficient appliances, reduce energy
`utilization in current structures and as well as pressure to
`create new structures which are more “environmentally
`friendly”, we have included a discussion and application of
`otal Equivalent Warming Impact or TEWIon refrigera-
`tor trials performed to date. Finally, we have included
`information relating to the manufacture of foam. with
`
`394
`
`Page 1 of 10
`
`POLYURETHANES EXPo '96
`
`ARKEMAEXHIBIT1159
`
`ARKEMAv. HONEYWELL
`PGR2016-00012
`
`
`
`

`

`Table 1. Blawing agent physical propertiesne
`Table 2, Blowing agent flammability propérties une
`To HEC-BaSia_CEC-11 HOFC.141b Gyclonentone “TPanion”sneatsCECTtHCPO-1410Cyclopantanep-Pentone’
`
`HFC-2451a CFC-11 HCFC-141b Cyclopentana _n-Pentane
`
`FlashPaint, “F
`Molecular Weight
`134.0
`137.4
`416.9
`70.0
`72.90
`59.5
`74,8
`89.7
`120.7
`96.3
`Boiling point. °F
`Tag Closed Cup
`None
`None
`None
`35"
`60°
`{ASTM DS6-67}
`.
`Liquid density, g/cc @68°F
`4.92
`4.49
`t.24
`0.75
`0.63
`Selatlash Closed Tester
`None
`None
`None
`Vapor Pressure, psia S68°F
`17.8
`12.8
`10.0
`49
`a3
`{ASTM 03928-87)
`Vapor Tharmal Conduativity,
`Tag Open Cup
`None
`None
`None
`{ASTM 0310-85)
`
`BTU inthe 87°F © 117°F 0.057=0.084.084
`
`
`
`VaporFlame Limits, Vol %3 None=7.6-17.7"None 1.4.9.4 1.3-B.0°
`
`(ASTM E621-85)
`BTUinh ft °F @ 77°F
`Q.061
`0,072
`0.063
`, 2.104
`
`
`FlameLimits, volume % None”=None’ 7.6-17.771.4-9.4> 13-48."
`Minimum ignition Energy, mj
`NIA
`N/A
`48,000
`G.2-03
`
`Flash Point, °F
`None
`Nene
`Nene
`35
`~40
`Heat of Combustion, kcal/mat
`244
`220
`7286
`839
`* ASTM B64). ambiont condainns, match ignition, ery ax
`*Tast method not spectied
`® Tossmetndd ned npecibed
`° Ambion! conditions, match ignition
`
`
`
`rey
`
`Table 3, Blowing agent environmental properties
`
` clopentane|n-Pentans
`
`
`
`
`
`[oT]
`pai{ai[oeyo to
`Saone Besighon Poles
`
`
`Aunaephene Lioling yoars 84|“80|aa114|FewDays [Fo Ome
`
`Global Waring Faioniia ot
`100 yr time horizon (CO, = 1)
`
`
`[4000["ea0|1300 “3
`
`
`
`
`VOG Status [No|No}Ro|No) Yes
`CFGand HCFC Atemasvaa” 1994
`* Aadate excom HC24bia data torn “Energy and Global Warming Impactsot NolinKingandNexl Gengration
`bot Rhaly fobe conckdornd a VOC
`
`Preesura,pola Tampermure, °F
`
`AHFC2agls ——HCFClath —at—CFE1) — B— Cyctopentane:
`
`
`Figure 1. Vapor pressures of some foam blowing agents
`
`HFC-245fa: Solubility, vapor pressure, a discussion of the
`effect of surfactants on foam properties, cell gas content,
`foam aging and k-factor vs, mean temperature. The paper
`concludes with a discussion ofAlliedSignal's plans for con-
`tinuing development and commercialization of HFC-
`245fa.
`
`PHYSICAL PROPERTIES
`The physical properties of HFC-245fa are shown in
`Table 1. CFC-13, HCFC-141b, n-pentane and cyclopen-
`tane are included for comparative purposes. The molecu-
`lar weight of HFC-245fa is approximately the same as
`that of CFC-11, therefore use levels in foams are about
`equal to those used with CFC-11 to achieve a given densi-
`ty. The boiling point of HFC-245fa is lower than those of
`the other blowing agents listed. This lower boiling point,
`and therefore higher vapor pressure at any given temper-
`ature, can provide significant improvement in certain
`foam properties, particularly at lower temperatures. The
`vapor pressure of HFC-245fa and other foam blowing
`agents as a function of temperature are shown in Figure
`1. The higher vapor pressure of HFC-245fa impacts the
`processing and storage of premixes. The lower boiling
`point and higher vapor pressure of HFC-245fa have con-
`tributed to stronger foams with better k-factors over a
`range of temperatures. These issues are discussed in the
`section in the réport entitled k-Factor vs. Mean Tempera-
`ture and Time.
`
`FLAMMABILITY CHARACTERISTICS
`The ability of HFC-245fa to be used in existing foam
`processing equipment with minimum modifications is
`
`Page 2 of 10
`
`“USTaa
`
`important. Currently, most rigid foam processing equip-
`mentis designed to handle either non-flammable or mod-
`erately flammable liquid blowing agents. HFC-245fa is
`considered “non-flammable” and has no flamelimits. The
`flammability characteristics of HFC-245fa are given in
`Table 2. The flammability properties of CFC-11, HCFC-
`141b, cyclopentane, and n-pentane are included for com-
`parison. Based on this test data, we believe HFC-245fa
`can be used safely on all existing foam processing equip-
`ment. In fact, this material has successfully been used in
`low pressure, high pressure and spray equipment. It has
`also been successfully processed into refrigerator cabi-
`nets. The details will be discussed later in this paper.
`
`ENVIRONMENTAL PROPERTIES
`HFC-245fa is a non ozone depleting chemical. The envi-
`ronmental properties of HFC-245fa are given in Table 3.
`Also included in the table are the data for HCFC-141b,
`CFC-11, cyclopentane and n-pentane. The atmospheric
`lifetimes of these blowing agent molecules is a key factor
`in the determination of these environmental properties.
`The Global Warming Potential (GWP) of HFC-245fais rel-
`atively low compared to CFC products and other commer-
`cially available products today , such as HFC-134a, which
`leads us to believe that the GWP of HFC-246fa will be
`environmentally acceptable. A recent reduction of the
`GWP of HFC-245fa has occurred due to a change in the
`atmosphericlifetime of methyl chloroform. Because ofits
`atmospheric lifetime an application to exempt HFC-245fa
`from being considered a VOC has beenfiled and weantic-
`ipate this exemption will be granted shortly.
`The assessment of GWP should not only be limited to
`the direct contribution of the blowing agent such as HFC-
`245fa, but also the assessmentof the indirect contribution
`of the carbon dioxide emissions resulting from the energy
`required to operate a system over its normal life. This
`analysis is referred to as TEW]or “total equivalent warm-
`ing impact”,
`In spring of 1995, The American Home Appliance Man-
`ufacturers (AHAM) conducted a trial with HFC-246fa.
`The formulations used to produce these refrigerator cabi-
`nets were unoptimized. Refrigerators manufactured with
`HCFC-i4lb were used as a control. The cabinets were
`assembled and a DOK Cabinet Energy test and a Reverse
`
`Bogdan, Williams, Logsdon, Parker | 395
`
`

`

`
`
`HCFC-141b
`
`270.2
`
`271.7
`0.5
`
`5
`
`Tabla 6. Summary of toxicity testing conducted on HFC-245fa to
`
`Table 4.'Results of AHAM refrigeratortrial
`date
`AHAM REFRIGERATOR
`
`Acuic Demol
`LD3q> 2000
`
`
`mghe,
`TRIAL
`feporied
`reponed
`
`
`
`
` britation
`Isvitation noted
`Noinvitation
`
`noted
`
`jolted
`
`Reverse Heat Leakage,
`OOE Cabinet Energy,
`Acute
`LCy>
`
`
`8TU/hr
`KwH/day
`Inhalation
`103,300 ppm
`
`Screen- Mice
`HcFC-141b HFG-245!a|HCFC-141b HFGC-245ia
`lowing agent
`
`Cardiac
`Negative at
`10,000 ppm
`Sensitization
`10,000 and
`Threshold
`
`
`Hyman
`
`
`Lympocyte
`
`
`Mouse
`
`Micronucleus
`
`ppm
`Not Active
`
`
`
`_TEWI analysis AHAM refrigerator trial
`Table 5.
`LCy >
`LC. -
`Acute (4) Hr
`
`Inhalation:
`200,000 ppm.|62,000 ppm
`Rais
`
`
`
`XXxX- Test results net reported
`Note: No exposure level was reported for cyclopentane only a classification as a weak sensitizer
`
`verage value
`% increase vs
`
`
`
`
`e Assumptions:
`® 100yr time horizion
`
`
`+ 20 yr. appliance lifetime
`
`
`© 100% of HIC-245fareleased
`after20 yr.
`
`
`« Contibution from refrigerant
`(184a) is constant
`
`
`capacity is not adequate enough to meet the needs of the
`population. Additional improvements in the performance
`of HFC-245fa foam can be expected as formulation opti-
`mization is done over the next few years. Hence, the ener-
`gy consumptionofrefrigerators prepared with HFC-245fa
`should be even lower. This points out the importance of
`not only considering the direct chemical effect of a sub-
`stance but also considering the total energy requirements
`of the system in which it is used.
`Yn addition to the environmental aspects, the discussion
`of the long term. breakdown products of CFC substitutes,
`into trifluoroacetic acid (TFA) hasbeen. occurring.
`Research work has recently been completed which identi-
`fies the long term breakdown products of HFC-245fa as
`HF and CO. Therefore, the use of HFC-245fa does not
`lead to the formation of TFA in the environment.
`
`TOXICOLOGICAL PROPERTIES
`It is critical that the toxicity of any alternative foam
`blowing agent candidate be fully defined. This data is
`required not only to set industrial exposure guidelines
`during storage, handling, and use of the blowing agent
`during the foam manufacturing process, butis also impor-
`tant in risk assessment analyses of the final foam product
`in its intended use and is critical to considerations in
`product stewardship and responsible care.
`/
`The HCFC and HFC molecules commercialized over the
`past several years,
`including HCFC-141b, have been
`extensively tested for toxicity characteristics. HFC-245fa
`is currently undergoinga similar toxicity testing protocol.
`A summary of the toxicity testing conducted on HFC-
`245fa to date is included in Table 6. Data on HCFC-141b
`and hydrocarbonsis also included for comparison.
`The test results support that HFC-245fa has a very low
`order of acute toxicity equivalent to or better. than those
`reported for HCFC-141b. We are continuing our toxicity
`testing and have summarized in Table 7 our anticipated
`work. We anticipate the results of the 28 day inhalation
`and 13 week inhalation study with HFC-245fa to be com-
`pleted in 1996.
`
`Heat Leakage test were performed on each, The results
`are contained in Table 4. The refrigerators produced with
`HFC-245fa had equivalent energy performance to the
`ones produced with HCFC-141b even though there was a
`6% higher k-factor for foams produced with HFC-245fa.
`This is why actual analysis of the foam in its application
`at the temperatures it will be exposed to is important.
`There is an extensive discussion of k-factor vs. mean tem~
`perature included in a latter portion of this paper.
`If this energy data for the unoptimizedrefrigerators is
`combined with the GWP for HFC-245fa, a TEWIanalysis
`of the refrigerators energy requirements can be calenlat-
`ed. The results are also displayed in Table 5. A compari-
`son of the energy requirement for a refrigerator made
`with HFC-245fa, HCFC-141b and cyclopentane ‘are
`included.
`:
`It is important to understand the assumptions which
`are madein this calculation. These include: the refriger-
`ators are operational in North America, the 100 yr. GWP
`is used for each chemical to determine its direct contribu-
`tion to global warming in terms of kg CO, generated, the
`life of the appliance is assumed to be 20 years, at the end
`of the 20 year life all the blowing agent will be released to
`the environment, and for the sake of consistency the fiuo-
`rocarbon used as the compressor fluid in these examples
`is HFO-134a.
`.
`:
`The result is that the total requirements , both direct
`and indirect, to operate a refrigerator produced with the
`unoptimized HFC-245fa formulations has less TEWI in
`its lifetime than one produced with cyclopentane. Also, it
`requires approximately 8.3% less TEWI due to direct
`energy requirements to operate than one produced with
`cyclopentane. This is a critical concern in regions where a
`consistent. supply of electricity or energy production
`
`396 / Bogdan, Williams, Logsdon, Parker
`
`Page 3 of 10
`
`

`

`Table 10. Compatibility of viton gasket with blowing agents and
`dlowing agent premix formulations
`
`Viton Gasket Compatibllity
`
`HFC-245fa
`
`HCFG-141b
`
`|}
`
`CFC-11SBA fe
`
`OBA & Premix
`
`HNoat Blowing agent
`
`9
`
`10
`5
`% Weigh! Gain
`
`15
`
`Table 7. Additional toxicological studies
`Planned Studies
`Optfonyt Stadies
`
`
`28 Dayinhalation
`©
`Reproductive toxicity
`13 Week Inhalation
`*
`Sub-chronic toxicity
`Repeated dose surdies
`Rat & Rabbit Teratology
`+
`° Environmental
`Glebsl compliance
`& Octanol/water partion coefficient
`> Algae
`* Daphnja
`@
`Fish
`
`
`
`
`
`
`Table 8. HFC-245fa compaitbility with plastics and elastomers
`
`
`Compalibic
`Not Compatible
`
`+ Polyethylene
`* Nitrile ( snarginal)
`+ Polypropylene
`+ HNBR
`
`
`« PTFE
`@ Viton
`
`
`* EPDM
`+ Epichlorohydrin
`> Neoprene
`:
`* BunN
`
`
`
`Material is immersed inHEC-245fa for 2 weeksal room temperature
`
`Pass/Fail Criteria: Swelling < 20%or shrinkage < 5%
`
`*’
`
`
`
`Table 9. Formulations for base premix
`
`Thanol R470;
`
`
`Fergostab BB4667
`
`
`
`
`Dabco R8020 PT
`
`
`Eastman Chemical
`®Huntsman Chemical
`© air Products & Chemicals
`
`premix
`Table 11. Resin formulations "BASF
`
`THERMAL AND HYDROLYTIC STABILITY
`In our previous paper™, we reported that HFC-245fa
`has been evaluated for storage stability at temperatures
`up to 200°C with no evidence of blowing agent decompo-
`sition. This testing was done on neat HFC-245fa, HFC-
`245fa with 300 ppm water added, and both neat and “wet”
`HIFC-245fa in the presence of 302 stainless steel, 3003
`aluminum,or both. Additional testing has been done with
`cald rolled steel, brass and lead with similar results.
`Limited testing has also been completed on several gasket
`Hosin2
`fe
`materials and plastics. These results are summarized in
`
`Table 8.
`Rast
`
`Tt is important te note that these tests represent the
`Piyracol 975
`
`suitability for contact of these materials with neat HFC-
`Picracel 624
`
`
`245fa and do not necessarily represent their suitability for
`
`
`Voranal WO [coumrerersaierurdpennrongreneeoe Nemas SoehaaneeEPREST
`
`
`contact with “B-components” containing HFC-245fa as is
`
`ThanA-A70K (eee es er
`
`
`the case in foam processing equipment. A test was con-
`Torol235
`ducted immersing Viton gasket material at room temper-
`Torete 2541
`ature for 2 weeks to neat samples of CFC-11SBA, HCFC-
`Toile 254c
`141b and HFC-245fa andalso to neat “B-components” and
`fe
`Torate 203
`“B-components” containing each of the blowing agents.
`Murenot 4052
`The formulation for the “B-components”usedis contained
`in Table 9. The results of the testing are contained in
`0
`10
`20
`30
`49
`50
`60
`Table 10. Although only weight gain is reported in this
`paper other dimensional measurements were also taken.
`The datafor length, width and hardness follow the same
`trend as weight. Therefore, before using any “B-compo-
`nents” in equipment independent of blowing agent it is
`highly recommended thata brief stability test be conduct-
`ed on the “wetted” gaskets and seals in the unit,
`
`
`
`Goldschmidt Chemical
`©) Ain Products & Chemicals
`“Eastman Chemica?
`Dayer
`
`Table 32.
`Solublilty of HFC-245ta In Polyols and Rosina
`
`HFC-245{fa Dissolved phpp
`
`70
`
`polyol blends tested , listed in Table 11, represent a typi-
`cal spray , appliance, PIR and panel formulation. A known
`amount of polyol or resin and an excess amount of HFC-
`245fa was added to a calibrated tube. The tubes were
`sealed and allowed to equilibrate for 24 hours. If separa-
`tion occurred the height of the blowing agent was mea-
`sured and the amount dissolved in the polyol was calcu-
`lated. The data available todate is summarized in Table
`12. Additional testing is currently in progress and will be
`presented as it becomes available.
`:
`
`POLYOL SOLUBILITY
`
`Solubility of a blowing agentis a critical variable in the
`foaming reaction. It is determines both the density and
`insulation properties of the foam. The solubility of HFC-
`245fa was tested in various polyols and polyol blends. The
`
`Bogdan, Williams, Logsdon, Parker ¢ 397
`
`
`
`Page 4 of 10
`
`

`

`—_ Qo
`
`0
`
`5
`
`10
`Wt. % HFC-245fa In P824
`
`15
`
`20
`
` Component
`Catalysis
`Polycat 5
`
`PABCO TMR-30"
`Palycat 46!
`
`‘Curithane 52°
`DABCO TMR'
`
`DABCO ‘TMR-2'
`
`DABCO TMR-3'
`Compasibilizers
`
`
`Lupranate M-20S
`MDI
`
`
`Figure 2..Vapor pressure HFC-245!a in P824
`Lupranste M-70L?
`Mondor MR‘
`
`
`Papi 27°
`Rubinate M°
`
`Surfactants
`
`
`
`Table 13. Raw materials and blenc’s used for blowing agent sta-
`bility testing
`WT RATIO TESTED
`MATERIAL
`
`ave(COMPONENT: HFC-245fa
`
`70:30
`Polyols
`
`
`1:10
`Catalysis
`
`Surfactoms
`
`
`MDI
`
`
`
`
`
`Pressure(psia)
`
`CFC- 11 Vapor Pressure at 130 F
`
`nNa
`
`nNo
`
`= an
`
`The chemical decomposition of a blowing agent such as
`HIFC-245fa produces a mineral acid (HF) whichis reactive
`with catalysts in a premix system and thus can influence
`foam reactivity. This decomposition reaction can occur
`during extended storage of a premix or during the foam
`reaction. Since the premix is a mixture ofmaterials , there
`is obvious synergy between the components which influ-
`ence the chemical reactions that occur during the foaming
`process and the storage of the premix. This test. ,however
`does not take this into account . It was designed to detect
`any reaction between the HFC-245fa and any single raw
`material only.
`The affected substances are catalysts. During the intro-
`duction of HCFC-141b, it was discovered that use of
`potassium carboxylate catalysts such as DABCO K-15*
`resulted in increased levels of decomposition products in
`the cell gas of foams produced with HCFC-141b. This
`reaction was primarily found to occur in polyisocyanurate
`foams.®) DABCO K-15* was found to be chemically reac-
`tive with HFC-245fa and isocyanates under the test con-
`ditions.
`This test determines the chemical sensitivity between
`HFC-245fa and raw materials which is an important con-
`sideration. However, the critical concern is what.occurs in
`the final foam product. This is discussed in detail] along
`with the results of toxicity testing on HFC-1234ze in a
`later portion of this paper.
`
`PREMIX AND POLYOL VAPOR PRESSURE
`
`From both an environmental and a processing perspec-
`tive the vapor pressure of the premixis critical. It is a key
`part of determining how to effectively minimize fugitive
`emissions as well as defining the type of container neces-
`sary for shipment and storage of a system.
`Two types of vapor pressures measurements have been
`taken vapor pressure vs, blowing agent concentration and
`vapor pressure vs.
`temperature. The measurement of
`vapor pressure vs. blowing agent concentration gives in-
`sight into processing parameters and how to minimize
`fugitive emissions. Figure 2 contains a graph of vapor
`pressure vs. wt. % HFC-245fa in Pluracol 824. The vapor
`pressure of HFC-245fa blend is lower than the vapor pres-
`sure of neat CFC-11 at concentration of less than 20 % at
`70. °F. Therefore, if the blowing agent is added to a cooled
`premix (i.e. below the boiling point of HFC-245fa) fugitive
`* Air Products
`
`
`
` Tegostab B-8404"
`
`Tegostab B-8458"
`
`
`Tegostab B-8467"
`‘Tegostwh B-2465’
`Tegostah B-S4Pt™
`
`Varanal
`360
`Palyotls
`Voranol ¥370"
`
`Voranol V490°
`
`
`‘ThanotR-575*
`‘Thanol R-370°
`‘Thancl R420"
`Thonot R-470x4
`Terme 2541"
`Thanal R-650Xx!
`
`
`Stepanol 2352"
`Thanol SF-265*
`7. Goldschmidt Chemical
`8, Eastman Chemical
`9. Hoechst-Celanese
`v.40, Sicpan Co.
`
`
`
`
`
`
`
`
`
`Nows:
`
`1, Air Products & Chemicals
`2. Huntsman Chemical
`3. BASF
`4. Bayer
`5, Dow Chemical
`6.
`IC)
`
`Solubility of the blowing agentin the “B-components”is
`not only a functionof its solubility in the polyol butalso is
`influenced by the surfactant used in the formulation This
`is further discussed in the Surfactant portion of. this
`report.
`
`COMPATIBILITY WITH FOAM RAW
`MATERIALS
`In our previous papers, we have reported some limit-
`ed data-on compatibility of various raw materials with
`HFC-245fa, Additional raw materials have been tested
`and the results shown in Table 14. A series of sealed tube
`studies were done to determinethe stability of HFC-245fa
`in the presence of certain polyurethane raw materials.
`Each raw material was mixed with HFC-245fa in a weight
`ratio similar to that which would be typically be found in
`a blended polyol component and placed in sealed glass
`tubes. The ratios used are listed in Table 13, The tubes
`were then placed in a 300 °F oven for 24 hours. The test
`was run under these conditions to simulate the exotherm
`temperature during the manufacture of a polyurethane
`foam. The.tubes were removed from the oven and the con-
`tents analyzed for decomposition products by gas chro-
`matography.
`
`‘398 / Bogdan, Williams, Logsdon, Parker
`
`Page 5 of 10
`
`

`

`Premix and blowing agent concentration
`Table 18.
`HFC-245fu Concentration
`
`Premix Concentration
`66% Polyether
`34 %
`65% Polyester
`35%
`
`28%
`72% Sucrose
`
`emissions should be minimal. This was demonstrated in a
`recent AHAMrefrigerator trial where HFC-245fa was
`added subsurface to a precooled premix in an open pro-
`cessing tank , Typically premix blending is done in closed
`containers. Therefore, the loss of less than 4% during the
`mixing of the blowing agent into the premix and the
`transfer of the premix into the tanks of the processing
`equipmentis considered excellent.
`Vapor pressure measurements are also useful in deter-
`mining the appropriate shipping containers . Vapor pres-
`sure is a very critical concern in the systems business
`because currently a significant portion of the systems sold
`are shipped in drumsor totebins. These containers tradi-
`tionally have low pressure ratings. The vapor pressure of
`three commercial spray foam systems containing HFC-
`245fa were measured. These were sucrose, polyether and
`polyester polyol spray formulations. The premix and blow-
`ing agent concentrations are listed in Table 15. The con-
`tainers were filled to 93% of capacity to simulate drum
`storage conditions. The results are contained in Table 16.
`There is a marked difference in vapor pressureof the sys-
`tems tested. Jt is in part due to the different quantity of
`HFC-2465fa dissolved in each.
`At the low temperature of 50°F all three systems have
`similar vapor pressures. This supports the findings that
`fugitive emissions should not pose a problem at process-
`ing temperatures below 60°F. However, at elevated tem-
`peratures the vapor pressure is directly proportional to
`the quantity of HFC-245fa used in the system. The for-
`mulations exceed the vapor pressure rating of most totes
`(1 atm) at 68°F. Drum pressure ratings vary dependent
`upon the drum gauge chosen. This is an area where more
`data needs to be generated.
`
`STABILITY DURING THE FOAMING
`REACTION
`
`The stability of HFC-245fa during the foaming reaction
`is an area under study. It was discovered that undercer-
`tain conditions during the foaming reaction it was possi-
`ble for HCFC-141b to decompose into HCFC-1131a and
`HCFC-151a. The decomposition took place more readily in
`PIR than PUR foams and washighly correlated to the lev-
`els of potassium octoate present in the system.®)
`A well defined research program has begun to identify
`what degomposition products, if any, were formed during
`foaming with HFC-245fa ,toxicity testing would then be
`conducted on these products and conditions which pro-
`mote decomposition reactions would be identified. The ini-
`tial foams produced with HFC-245fa were analyzed for
`cell gas content™. Only one decomposition product was
`identified, HFC-1234ze ( 1,3,3,3 tetrafluoropropene). It is
`possible for this structure to exist in both a cis and a trans
`orientation. The decomposition reaction is shown in Table
`17.
`Data on foams prepared from the Class I* and Class I]*
`panel formulations and the spray formulation reported in
`our previous papers”
`indicate that very little or ne de-
`composition of the blowing agent occurs during the foam-
`
`Table 16. Vapor pressure measurements vs temperature in spray
`foam formulations
`
`
`
`30
`
`anh o
`
`
`
`Pressure(psia)
`
`50
`
`59
`
`68
`
`77
`
`a6
`
`95
`
`Temperature ° F
`-~ POLYETHER -= POLYESTER --SUCROSE
`
`Tabie 17, Reaction mechanism for the formation of HFC-1234ze
`
`
`CF,-CH,-CF,
`HFC -245fa
`
`CF,-CH =CF+ HF
`HFC-i234ze
`
`Table 18. Variables for decomposition study
`Variable
`Level 1 Level 2
`
`Trimer Catalyst
`DABCO™ K-15"
`DABCO™ TMR-27
`Reactivity
`Low- 50 see
`Hi- 32 sec
`Index
`256
`350
`Amine Cataly5
`DABCO™ TMR-30
`Polyeat™ gt?
`Potyol
`Stepanpo! PS-2352 7
`Terate 25417
`Blowing Agent Package
`35 pbw HFC-245fa/
`25 phw HFC-245fal
`0 pbw water
`1 pbw water
`"Air Products & Chemicals
`1Stepan Chemical
`Sl Hoerchst -Cetanese
`
`ing reaction. As with HCFC-141b, it was found that the
`decomposition reaction oceurs primarily in PIR foams. It
`was also discovered that the reaction mechanism favored
`the formation of the trans isomer over the cis isomer in a
`4:1 ratio, In response to this information, HFC-1234ze
`trans and HFC-1234ze cis isomers were purified and sub-
`mitted for limited toxicity testing. Toxicity testing on
`HFC-1234ze trans isomer has been completed. The lHimit-
`ed amountof data availabie on HFC-1234ze trans implies
`that it is not highly toxic or mutagenic, Toxicity testing is
`currently scheduled for the cis isomer.
`Anew program was initiated to determine what condi-
`tions favored the decomposition reaction in polyisocyanu-
`rate foams. The program was modeled after the work com-
`pleted with HCFC-141b®, A series of 16 experiments
`were run to define the importanceof 6 key variables. The
`variables chosen were: trimer catalyst type { potassium
`carboxylate, DABCO™ K-15(T1) vs. a quaternary ammo-
`nium carboxylate DABCO™ TMR-2 (T2)], reactivity {fast
`(R2) vs. slow (R1) gel time] , foam index [250 (11) vs. 350
`(i2)], amine catalyst type [a volatile monoamine, N,N-
`dimethyleyclohexylamine, Polycat™ 8 (A2), vs. a less
`volatile triamine, DABCO ™TMR-30 (A1)}, polyol type
`{Stepanpol™ PS-2352 (P1) vs. Terate™2541 (P2)], and
`blowing agent package [HFC-245fa only (B1) vs. H¥C-
`245fa plus water (B2)]. Two levels or types of each vari-
`able were set . They are shown in Table 18. This arrived
`
`*This numerical flame spread 18 not intended to reflect hazards presented
`by this or any other under actual! Gre conditions.
`
`Bogdan, Williams, Logsdon, Parker | 399
`
`Page 6 of 10
`
`

`

`Table 29.
`Formulation used for surfactant evaluation
` B- component
`~~Somponent_
`Yoranot 370
`70
`i
`Terot 375°
`30
`Polyeat 5
`0.7
`Polycat 4)?
`0.5
`Water
`ia
`Surfactant
`25
`
`HFC-245fa a
`24
`A-compongent
`Mondur MR“?
`129
`Index 125
`Temperature 60 ° F
`Dow Chemical
`OQXID Incorporated
`&pix Products & Chemicals
`* Bayer
`
`Stepan refers to Stepanol PS-2352
`Terate refers to Terate 2541
`
`Main Effects for HFL-1234ze trans
`
`wn +
`
`ae we Raa)
`
`Jable 21.
`
`Surfactant vs. closed cell content
`
`100
`
`95
`
`30
`
`85
`
`Table 19..16 Experiments for study on formation of HFC-1234ze
`Content 30
`%Closed
`2 HE
`
`fians
`
`LOW
`
`polyol
`
`wucr
`
`index
`
`amine
`
`nine
`
`sire
`
`Be404
`
`68409
`
`Ba418
`Surfactant
`
`B8433
`
`BA462
`
`Figure 4.. Analysis of experimental data
`
`at the 16 experiments listed in Table 19. On each foam
`reactivity, cell gas content, peak exotherm and density
`were determined. Analysis of the data using the Taguchi
`method yield s the results in Figure 4. Each point repre-
`sents an average for all the experimental results for that
`variable.
`Based upon experimental error analysis, the two. vari-
`ables of the six tested with the strongest effect to produce
`the HFC-1234ze trans isomer is reaction time and trimer
`concentration level. In experiments where the reaction
`time was 30 seconds and DABCO K-15 was used, the lev-
`els of HFC-1234ze trans produced were on average two
`times higher than in reactions where the reaction time
`was 50 seconds and DABCO TMR-2 was used.It is impor-
`tant to realize that in any reaction the catalyst levels are
`what determines the reactivity of the system. Therefore,
`these should not be considered independenteffects as also
`the case with the amine catalyst and index results. This
`study is continuing to determine if there is any effect on
`the HFC-1234ze trans levels in the foam as it ages. It is
`important to note here that under these conditions HFC-
`1284ze trans can be generated but the limited toxicity
`data indicates that HFC-1234ze trans does not appear. to
`be highly toxic or mutagenic.
`
`a critical part of optimization of the k-factor of a foam. In
`the transition from CFC-11 to HCFC-141b,it was discov-
`ered that the best foam performance in terms of both k-
`factor and physical properties was achieved with the use
`of different surfactants. Some very preliminary work has
`been completed using Goldschmidt surfactants studying
`the effect different surfactants have on the physical prop-
`erties of foams produced with HFC-245fa. The initial
`work was done with the appliance formulation listed in
`Table 20. The foam was produced in L-molds and samples
`were taken to determine the effect each surfactant had on
`closed cell content, k-factor, flow and physical properties.
`The tests suggest the use of Goldschmidt B8404 and
`B8462 produced foams with higher closed cell contents
`(Table 21) and lower k-factors than those produced with
`other surfactants (Table 22). They also produced foams
`with better flowability (Table 23) and physical properties
`than other Goldschmidt surfactants. As a result of this
`work, the use. of Goldschmidt B8404 and B84