Purdue University
`Purdue e-Pubs
`International Refrigeration and Air Conditioning
`Conference
`
`1996
`
`The Toxicity of Refrigerants
`
`J. M. Calm
`USA
`
`School of Mechanical Engineering
`
`Follow this and additional works at: http://docs.lib.purdue.edu/iracc
`
`Calm J. M. "The Tox c ty of Refr gerants" (1996). International Refrigeration and Air Conditioning Conference. Paper 317.
`http://docs.l b.purdue.edu/ racc/317
`
`Th s docu e as bee ade ava ab e
`add o a
`o
`a o .
`Co p e e p oceed gs ay be acqu ed
`He
`ck/Eve s/o de
`.
`
`oug Pu due e-Pubs, a se v ce o e Pu due U ve s y L b a es. P ease co ac epubs@pu due.edu o
`
` p
`
` a d o CD-ROM d ec y o e Ray W. He
`
`ck Labo a o es a ttps://e g ee
`
`g.pu due.edu/
`
`Page 1 of 7
`
`Arkema Exhibit 1146
`
`

`
`
`
`
`THE TOXICITY OF REFRIGERANTS
`
`
`
`
`
`
`
`James M. Calm, P.E.
`
`Engineering Consultant
`10887 Woodleaf Lane
`
`
`
`
`
`
`
`Great Falls, VA 22066-3003 USA
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`This paper presents toxicity data and exposure limits for refrigerants. The data address both acute (short-term,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`single exposure) and chronic (long-term, repeated exposure) effects, with emphasis on the former. The refrigerants
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`covered include those in common use forthe last decade, those used as components in alternatives, and selected
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`candidates for future replacements. The paper also reviews the toxicity indicators used in both safety standards
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`and building, mechanical, and fire codes.
`It then outlines current classification methods for refrigerant safety and
`
`
`
`
`
`
`
`
`
`
`
`
`relates them to standard and code usage.
`
`INTRODUCTION
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Most of the dominant refrigerants for the past fifty years have been or are being replaced, to protect the strato-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`spheric ozone layer or as a precaution to address global warming. Much to the credit of the air-conditioning and
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`refrigeration industry, both chemical and equipment manufacturers have resisted compromise to either safety or
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`performance in developing replacements. None of the alternative refrigerants that have been commercialized are
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`highly toxic or even toxic, as classified bgl federal regulations.1 Scrutiny of the new refrigerants %hows them to be
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`as safe or safer than those they replace.
`Still, safety concerns have surfaced as significant factors in regulations
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`for the new refrigerants. These concerns do not arise from increased hazard levels, but from lack of familiarity and
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`necessary information. The rapid phaseout schedule for chlorofluorocarbon refrigerants required introduction of
`
`
`
`
`
`
`
`
`
`
`
`
`new chemicals before complete data were available.
`
`
`
`
`
`
`
`
`
`Most of the early refrigerants ~ before the 19305 — were flammable, toxic, or both. The advent of fluorochemi-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`cals ushered in a new era of safety, as illustrated by the dramatic demonstration by Thomas Midgley in April 1930.3
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`in announcing the development of fluorochemicals to the American Chemical Society, he inhaled Fi—12 and blew
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`out a candle with it. Although this dramatic performance suggested that the new refrigerant was neither toxic nor
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`flammable, it would clearly violate current safety practices.
`
`
`
`
`
`
`
`
`
`
`
`As subsequent testing established the low toxicity of the new refrigerants, recognition evolved that the primary
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`safety risks were the pressure hazards inherent to any compressed gas, asphyxiation from possible displacement
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`of air, and frostbite with skin contact at low temperatures. These concerns were, however, common to the volatile
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`compounds used before fluorochemicals. As the level of safety improved, so did expectations. Rules evolved to
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`also address acute exposure hazards under emergency conditions, for example potential decomposition in tires
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`into carbonyl halides as well as hydrochloric and hydrofluoric acids. Likewise, safety provisions also addressed the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`potential for cardiac sensitization and the effects of chronic exposures for both technicians and building occupants.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The resulting regulations restricted the use of refrigerants, set quantity limits in occupied areas, imposed isolation
`
`
`
`
`
`
`
`
`
`
`
`
`requirements for refrigerant-containing components and machinery rooms, and prescribed a range of detection,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`ventilation, pressure relief, emergency discharge, and other safety provisions.
`
`
`
`
`
`
`
`
`
`
`
`
`
`More recent focus on the effects of refrigerants on the environment spawned two significant safety measures,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`namely system tightening and modification of service practices to reduce venting. While their motivation was envi-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`ronmental protection, to curtail avoidable emissions, the result also lowers the likelihood and concentrations of
`
`
`
`
`
`
`
`
`
`refrigerant exposures.
`
`
`
`
`
`
`
`
`
`
`
`CODE ACCEPTANCE OF ALTERNATIVE REFRIGERANTS
`
`
`
`
`
`
`
`
`international treaties, most noticeably the Montreal Protocol and Framework Convention on Climate Change,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`have focused on the global issues of environmental protection, information sharing, and assistance to developing
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Countries. These treaties, and revisions to them, have fostered scientific assessments and set phaseout schedules
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`for substances of concern. While federal laws govern the production and trade of alternative chemicals, most ordi-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`nances for application of refrigerants are adopted and enforced at the local level. They are included in building,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`mechanical, and fire safety codes, which govern building construction, system installation, equipment operation
`
`
`
`
`
`
`
`
`
`
`
`and maintenance, system modification including refrigerant conversion, and ultimate demolition. Although most
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`are based on national or regional model codes, the introduction of alternative refrigerants occurred so rapidly that
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`the cautious process of regulatory revision has not caught up yet. One cause of delay has been the time needed to
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`complete toxicity tests, publish the findings, modify impacted standards, develop and adopt code revisions, and
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`prepare design professionals, contractors, technicians, and building and fire prevention officials.
`
`
`
`
`
`
`Page 2 of 7
`
`157
`
`
`
`Page 2 of 7
`
`

`
`
`Toxicity Testing
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Facing unprecedented testing and phaseout requirements, the chemical industry formed an international con-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`sortium to accelerate the development of toxicology data for substitute fluorocarbons, both for refrigerant and other
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`uses. The cooperative effort, named the Programme for Alternative Fluorocarbon Toxicity Testing (PA!-Tl"), was
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`sponsored by the major producers of chlorofluorocarbons (CFCS). The PAI-'l' research entailed more than 200 in-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`dlV|dUa| toxicology tests, by more than a dozen laboratories in Europe, Japan, and the United States. The first tests
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`were launched in 1987, to address R-123 and R-134a (PAl-‘I’ l). Subsequent programs were initiated for R—141b
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`(PAl-‘T ll), R-124 and Fl-125 (PAl-'l' lll), Fl-225ca and R-225cb (PAFT IV), R—32 (PAI-‘I’ V), and — still undenNay— the
`
`
`
`
`
`
`
`
`mechanistic, metabolic, and pharmacokinetic aspects of the toxicology of fluorocarbons.4v5 Extensive additional
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`data were developed, or contributed from prior tests, by individual chemical manufacturers.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The tasks of assembling and interpreting the resultant data were expanded by the need, in some cases, to col-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`lect comparative information for the refrigerants being replaced. Whereas their introduction largely preceded the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`more rigorous, current testing and classification requirements of the codes, the amount of information needed was
`
`significant.
`
`
`
`
`Safety Standards
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Most code provisions for refrigerant and refrigeration-system safety can be traced to either ASHRAE Standard
`
`
`
`
`
`
`
`
`
`
`
`15, Safety Code for Mechanical Flefrigeration,5 or to general code provisions developed for occupancies where
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`more hazardous materials are used. Standard 15 prescribes safeguards for design, construction, installation, and
`
`
`
`
`
`
`
`
`
`
`
`
`
`operation of refrigerating systems.7=3 Many of the specific requirements are based on safety classifications from
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Standard 34, Designation and Safety Classification of Fiefrigerants.9=10 This standard is the definitive source for
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`assignment of refrigerant number designations.
`It also provides a safety classification system and assigned classi-
`
`
`fications.
`-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`An effort is underway, by the committees responsible for the two standards, to move determination of data
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`needed for Standard 15 into Standard 34, and to rewrite the application requirements in Standard 15 parametrically,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`based on the referenced data.
`In doing so, the committees are refining the methods to determine refrigerant quan-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`tity limits for occupied areas, both to increase consistency and to take advantage of the new data and understand-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`ing. They also are addressing the toxicity, flammability, and fractionation concerns arising from use of zeotropic
`
`
`
`
`
`and azeotropic refrigerant blends.
`
`
`
`
`
`
`Parameters
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Building, fire, and mechanical codes vary throughout the United States. They are based on state or local
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`amendments to regional model codes or,
`in a few locations,
`locally-developed codes. Nevertheless, the data
`
`
`
`
`
`
`
`
`
`needed for compliance are fairly consistent. They include:
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`LC50: The "lethal concentration for 50% of tested animals," sometimes referred to as the median lethal con-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`centration, is a primary measure of acute toxicity by inhalation of gases.
`It most commonly is measured with
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`rats for exposures of four hours. A number of federal regulations (e.g., reference 1) and most building, fire,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`and mechanical codes deem substances to be toxic for one-hour LC 0 concentrations of 200 - 2 000 ppm
`and highly toxic for less than 200 ppm.* Typical LC50 concentrations For one hour are double those for four
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`hours. 1- 2 Multipliers of 1.6-4 have been suggested,13=14 and one study found a range of 1.5-5.7 for 20
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`tested chemicals.15 None of the refrigerants identified in table 1, or blends of them, qualify as toxic or highly
`
`
`
`
`
`
`
`
`
`toxic based on the LC50 data and the stated criteria.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`IDLH: The concentration deemed to be "immediately dangerous to life and health," set by the National Insti-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`tute for Occupational Safety and Health (NIOSH). This measure was initially developed as a criterion for res-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`pirator selection in the 1970s as part of the Standards Completion Program (SCP). The SCP definition for
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`IDLH was "the maximum concentration from which,
`in the event of a respirator failure, one could escape
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`within 30 minutes without a respirator and without experiencing any escape-impairing (e.g., severe eye irrita-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`tion) or irreversible health effects." The 1994 revision defines an IDLH condition as one "that poses a threat of
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`exposure to airborne contaminants when that exposure is likely to cause death or immediate or delayed
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`permanent adverse health effects or prevent escape from such an environment." The revised and added
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`* The definitions for toxic and highly toxic also include LD5o (median lethal dosage) criteria for mortality by inges-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`tion and contact. LD5o values generally are not determined for or applicable to gases and volatile substances, such
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`as refrigerants, since the standard test methods are not suited for them. Similarly, the likelihood of ingestion of or
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`prolonged contact with the quantities involved is remote. Some LD5o data, mostly based on solutions of refriger-
`
`
`
`
`
`
`
`
`
`
`
`
`
`ants in liquids, are provided in reference 20.
`
`
`
`
`Page 3 of 7
`
`158
`
`
`
`Page 3 of 7
`
`

`
`
`
`
`
`
`
`
`
`
`
`
`lDLHs in 1994 are based on additional toxicity criteria and data. Whereas the lDLHs derived for the SCP
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`were set at 100% of the lower-flammabilitylimit (LFL),
`if there were no known health hazards below those
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`concentrations, the 1994 lDLHs are reduced to 10% of the LFL. Most fire codes use the IDLH, based on the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`SCP definition, as a criterion for ventilation rates and emergency discharge treatment. Standard 15 also uses
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`the SCP IDLH as one of several criteria to determine refrigerant quantity limits for occupied areas. Use of the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`IDLH concentrations for these purposes has been challenged as inconsistent with their definition.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`PEL: The "permissible exposure limit" is the concentration level established by the Occupational Safety and
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Health Administration (OSHA). Without qualification, the PEL implies a time weighted average (TWA) for an
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`8-hour work shift in a 40-hour work week. Consistent data include similarly-defined occupational exposure
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`limits set by manufacturers (such as the Allowable Exposure Limit, AEL), the Threshold Limit Value (TLV)
`
`
`
`
`
`
`
`
`
`
`
`
`
`adopted by the American Conference of Governmental Industrial Hygienists (ACGlH),15 and the Workplace
`
`
`
`
`
`
`
`
`
`
`
`
`Environmental Exposure Level
`(\NEEL) guides developed by the American Industrial Hygiene Association
`
`
`
`
`
`
`
`(AlHA).17 Where designated as a PEL-C (e.g., for R-11), the PEL is a ceiling concentration that shall not be
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`exceeded. Most codes use either PEL or TLV-TWA values as the maximum activation levels for leak-detector
`
`
`
`
`
`
`
`
`
`
`
`
`
`alarms; Standard 15 and the new International Mechanical Code (IMC) use the TLV-TWA. While the PEL and
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`TLV-TWA are similarly defined, and the original PELs were based on TLVs, the PEL values have not been
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`revised since 1971. More protective limits published in 1989 were vacated by a court order in 1992. ACGIH
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`publishes annual TLV updates. Still, neither PELs nor TLVs have been set for most alternative refrigerants, for
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`which the primary recourse is use of WEELs or other consistent measures.
`
`
`
`
`
`
`UL group: The Underwriters Laboratories classification reflects the comparative life hazard of refrigerants in
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`the absence of flames or surfaces at high temperatures. Group 1
`is the most toxic (e.g., R-764) and group 6
`
`
`
`
`
`
`
`
`
`
`
`
`
`the least.18 This measure is used to classify refrigerants in older codes, still in effect in some jurisdictions.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Standard 34 safety group: This classification consists of a letter (A or B), which indicates the toxicity class,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`followed by a number (1, 2, or 3), which indicates the flammability class. Toxicity classes A and B signify
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`refrigerants with lower toxicity and higher toxicity, respectively, based on prescribed measures of chronic
`
`
`
`
`
`
`
`
`
`
`(long—term, repeated exposures) toxicity. Flammability class 1 indicates refrigerants that do not show flame
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`propagation in air when tested by prescribed methods at specified conditions. Classes 2 and 3 signify refrig-
`
`
`
`
`
`
`
`
`
`
`
`
`
`erants with lower flammability and higher flammability, respectively; the distinction depends on both the LFL
`
`
`
`
`
`
`
`
`
`
`
`
`
`and heat of combustion (HOC).9 Some of the mechanical codes written before 1993 used an older safety
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`classification system from earlier editions of Standard 34. They included groups 1 (no flame propagation and
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`low degree of toxicity), 2 (TLV-TWA less than 400 ppm), 3a (flammable with low LFL or high HOC), 3b
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`(flammable with high LFL and low HOC), 4a (mixtures of groups 1 and 3a that are nonflammable as formu-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`lated, but could become flammable upon fractionation), and 4b (mixtures of groups 1 and 3b that are non-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`flammable as formulated, but could become flammable upon fractionation). Excluding the group 2 refriger-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`ants, the ranked order from the lowest to highest flammability hazard was 1, 4b, 4a, 3b, and 3a. One motive
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`for the current classification system, introduced in 1992, was to provide a more rational system. Based on
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`current understanding and usage, however, the author questions whether two toxicity classes provide suffi-
`
`
`
`
`
`
`
`
`
`
`
`
`
`cient distinction and whether the classification criterion should be a measure of acute, rather than chronic,
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`toxicity or a combination of acute and chronic toxicity.
`
`
`
`
`
`
`
`quantity limits for occupied areas: The primary criterion to determine whether refrigeration systems, or
`
`
`
`
`
`
`
`
`
`
`
`
`refrigerant-containing components, are allowed in occupied areas of buildings are quantity limits set by the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`codes or Standard 15. Nearly all limits set in the codes were transcribed from Standard 15, though a few
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`intended and unintended revisions were made in the process.
`
`
`
`
`
`
`
`
`
`
`As discussed above, efforts are underway to develop consensus quantity limits for new refrigerants, including
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`blends. The following additional data are likely to be needed:
`
`
`
`
`
`
`
`
`
`
`cardiac sensitization: An acute effect in which the heart is rendered more sensitive to the body’s own cate-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`cholamine compounds or administered drugs, such as epinephrine, possibly resulting in irregular heart beat
`
`
`
`
`
`
`
`
`
`
`
`
`(cardiac arrythmia), which could be fatal.” LOEL is the "lowest—observed effect |evel," the lowest concentra-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`tion at which sensitization occurs in tests, normally to beagle dogs treated with epinephrine to simulate the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`effects of stress. NOEL is the "no-observed effect level," the highest exposure concentration at which no
`
`
`
`
`
`
`
`sensitization is observed.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`anesthetic EC50: The concentration of a substance that caused the temporary loss of ability to perceive
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`pain and other sensory stimulation to 50% of test animals, normally measured for 10 minute exposures.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`EC50 refers to the "effective concentration for 50% of specimens."
`
`
`
`
`
`
`
`
`
`
`
`l'-'lD5o: The concentration that resulted in 50% decrease in respiratory rate, normally measured in mice. A
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`maximal effect generally occurs in less than 30 minutes; the response to R-717 (ammonia), as an example, is
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`reported to take approximately two minutes.
`
`Page 4 of 7
`
`159
`
`
`
`Page 4 of 7
`
`

`
`
`
`
`
`
`
`
`
`a__*,
`
`
`
`
`
`
`
`Table 1: Refrigerant Toxicity Data, Exposure Limits, and Classifications 3 (data and limits in ppm v/v)
`cardiac
`anes-
`
`
`
`
`'
`'
`IDLH 9
`thetic
`sensitization d
`refrig *3
`UL‘
`
`
`
`safety
`
`
`
`
`
`
`
`
`
`
`
`
`1.050 P LOEL NOEL E050 6 R0501 SCP 1994erant- PEL h
`
`
`
`
`group
`group *3
`
`
`5 0001'
`50 000
`
`
`
`
`50 000 '
`—-
`——
`
`
`11
`
`12
`
`22
`
`23
`
`32
`
`
`26 200
`
`
`760 000
`
`
`220 000
`
`
`>663 000
`
`
`>760 000 P
`
`
`52 000 P
`
`
`600 000
`
`>800 000 P
`
`
`>800 000 P
`
`
`
`32 000
`
`>230 000 P
`
`
`>300 000 P
`
`
`567 000 P
`
`61 647
`
`
`
`128 000 °
`
`
`143a >540 000 P
`
`
`
`383 000 P
`152a
`
`
`>110 000
`213
`
`
`>300 000 F
`290
`
`
`
`>300 000
`C313
`
`
`
`
`
`113
`
`114
`
`115
`
`116
`
`123
`
`
`124
`125
`
`
`134a
`141b
`
`1421:»
`
`
`
`
`600
`
`600a
`717
`
`744
`
`1270
`
`
`230 000
`
`
`570 000 F
`
`
`2 000 U
`
`
`
`W
`650 000 '<
`
`
`
`
`
`
`
`
`
`5 000
`
`
`50 000
`
`
`50 000
`
`
`>800 000
`
`
`250 000
`
`
`
`1 100
`
`25 000
`
`
`25 000
`
`
`800 000
`
`
`200 000
`
`
`
`35 000
`
`254 000
`
`
`140 000
`
`
`186 000 P
`
`
`86 000 P
`
`
`
`
`
`
`5 000
`
`25 000
`
`
`150 000
`
`
`P1
`
`
`20 000
`
`
`25 000
`
`100 000
`
`
`
`75 000
`
`5 000
`
`
`50 000
`
`
`
`300 000
`
`
`150 000
`
`
`400 000
`
`
`100 000
`
`
`100 000
`
`
`
`5 000
`
`
`50 000
`
`
`—-
`4
`——
`
`--
`--
`--
`200 000
`
`
`
`
`10 300
`
`10 000
`
`
`75 000
`
`
`
`50 000
`--
`25 000
`
`
`
`
`23 000
`250 000
`
`
`
`
`23 000 P
`--
`
`
`40 000
`
`
`
`140 000
`
`
`>10 000 P
`
`
`
`
`
`205 000 P
`25 000
`
`
`250 000
`
`
`
`
`
`
`250 000 >540 000 P
`
`
`
`
`50 000
`200 000
`
`
`
`
`
`300 000 >113 000 P
`
`
`
`
`50 000
`280 000
`
`
`
`
`-- >600 000 P
`
`
`
`
`
`
`
`
`
`
`
`
`4 500
`50 000
`
`-—
`--
`4 000'
`
`
`
`
`--
`--
`
`
`50 000 '
`-—
`--
`
`--
`_-
`-_
`20 000 S
`
`
`—~
`
`--
`25 000
`
`
`-—
`-—
`—-
`
`130 000
`
`
`200 000 T
`
`
`—-
`--
`--
`
`-—
`--
`500
`50 000
`
`--
`
`
`
`
`303
`—
`
`
`
`A1
`
`A1
`
`A1
`
`A1
`
`A2
`
`A1
`
`A1
`
`A1
`
`A1
`
`B1
`
`A1
`
`A1
`
`A1
`
`
`A2
`
`
`A2
`
`A2
`
`A1
`
`A3
`
`A1
`
`
`A3
`
`A3
`
`B2
`
`A1
`
`A3
`
`
`
`2 000 C1 0001‘
`
`
`
`15 000
`1 000
`
`
`
`
`1 000 ‘P
`
`
`1 ooo I
`
`
`1 000 p
`
`
`1 000
`
`
`1 000
`
`
`1 000 m
`
`
`1 000 '
`
`
`10-30 '
`
`1 000 P
`
`
`
`1 000 P
`
`
`1 000 P
`
`
`
`500 P
`
`1 000 P
`
`
`
`
`4-5
`
`
`
`5 I
`
`
`6 6
`
`
`
`
`
`5 (b)
`
`6 l
`
`
`5(b)
`
`5(b)
`2
`
`
`5(a)
`
`2 000
`
`15 000
`
`
`
`
`
`
`
`1 000 P
`
`
`1 000 P
`
`
`1 000 P
`
`
`1 000
`
`
`--
`
`
`
`800 "1
`
`600 '
`
`50 V
`
`5 000
`
`
`1 000 '
`
`
`
`——
`2 100
`-
`
`
`
`--
`300
`
`40 000
`
`
`—
`
`
`5(a) I
`6
`
`
`5(a)
`6 '-
`
`
`
`
`
`
`
`
`
`
`
`
`
`"""D'(Q"‘l.'DQ.0D'FD
`.c1-1:10:13-zr
`5<::r+rn~
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Please see the source publications (identified in reference 20) to verify these data and examine their limitations.
`
`
`from ANSI /ASHRAE Standard 34-1992 and addenda thereto 9:10
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`4-hr LC50 rat; federal and fire code toxicity classifications are based on 1-hr LC50 rat
`
`
`
`
`
`
`dog with epinephrine challenge
`
`
`
`
`10-min ECSO mouse or rat
`
`
`
`30-min RD50 mouse
`
`
`
`
`
`
`
`NIOSH IDLH values from the Standards Completion Program (SCP) and 1994 revision 13
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`time-weighted average (TWA) for 8 hr/day and 40 hr/wk
`
`
`
`
`
`
`
`
`
`
`
`
`comparative life hazard where group 1 is the most toxic and group 6 the least
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`‘The SCP IDLH and OSHA PEL are 10 000 and 1 000 ppm, respectively; ARI recommends 5 000 and C1 000 ppm
`
`
`
`
`
`
`
`
`
`based on the cardiac sensitization potential.19
`
`2-hr ALC rat
`
`
`
`
`
`
`
`industry or manufacturer recommendation
`
`
`
`
`ACGIH Threshold Limit Value - Time-Weighted Average (TLV-TWA) 15
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`anesthetic effects observed in rats at this concentration during ALC, LC5o, or other studies
`
`
`
`
`
`
`
`4-hr Approximate Lethal Concentration (ALC) rat
`
`
`
`AIHA Workplace Environmental Exposure Limit (VVEEL) 17
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`response observed at 200 000 ppm in anesthetized dogs using tracheal cannulae (intubation); no effect found at
`
`
`
`
`
`
`
`
`
`
`600 000 ppm, in a separate study, by simple inhalation
`
`
`15-min LC5o rat
`
`
`
`
`
`
`
`
`based on the lower flammability limit (LFL)
`
`
`17-min EC50 mouse
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`wide dispersion found in the literature: 6 586 - 19 671 for 1 hr and 2 000 - 4 067 for 4 hr
`
`
`
`
`
`
`
`
`
`ACGIH Threshold Limit Value - Time-Weighted Average (TLV~TWA) = 25 ppm
`
`
`
`
`
`
`
`5-min LCLO human = 90 000 ppm
`
`
`
`
`
`
`
`
`
`
`
`
`
`Page 5 of 7
`
`160
`
`
`
`Page 5 of 7
`
`

`
`
`
`
`
`
`
`
`
`
`
`Working drafts of the proposed method to determine quantity limits, identified _as.Recommended Quantity Limits
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`(RQLs), use the preceding acute—toxicity data to determine an intermediate limit, identified as the Acute Toxicity
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Exposure Limit (ATEL). The RQL is then set at the lowest of the ATEL, the oxygen deprivation level (ODL, the cal-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`culated concentration that will reduce the oxygen concentration in normal air to below 19‘/2% by volume) of 69 100
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`ppm, and 25% of the LFL. One proposal to establish RQLs for blends uses the same calculation method as for sin-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`gle compounds, after determining a mole-weighted average for each parameter based on the values for the blend
`
`
`
`
`
`
`
`
`
`components.
`
`
`
`
`Findings
`
`
`
`
`
`
`
`Table 1 summarizes data for common, single-compound refrigerants. The quantity limits for occupied areas are
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`not included, since the calculation method still is being revised. Those limits and corresponding data and limits for
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`blends will be presented in a subsequent paper.
`
`
`
`
`
`
`
`
`
`In general, those shown are the
`Multiple values were located for approximately half of the data in the table.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`most conservative found in the published literature, except that the highest published concentrations are included
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`for no—effect levels and where a study found a lower bound to, but did not actually establish, an end point. Space
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`limitations in this paper prevent inclusion of the more than 200 pertinent references from which the data in table 1
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`were obtained. Specific or multiple corroborating sources and additional data are identified in reference 20.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Comparison of the acute—toxicity data for R-123 to those for R-11 show it to be as safe, or safer, with respect to
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`acute toxicity. The same conclusion results from comparison of the data for R—134a to those for R—12. An earlier
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`paper showed that chronic exposures can be maintained well below recommended limits.2
`
`
`
`
`
`
`
`
`
`
`
`
`
`Another point that is evident with the assembled data is that the hydrocarbons proposed as replacements for
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`fluorochemicals are generally more toxic. The LC5 values for R-32, R-125, R-134a, R-290 (propane), R-600
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`(butane), R-600a (isobutane), and R-1270 (propene) al indicate very low acute toxicity. The cardiac sensitization
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`and anesthetic effect indicators, however, suggest that R-600 and R-600a pose higher risks than R-134a and that
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`the inhalation lethality of R-1270, while very low,
`is higher than for either R-32 or R-125. Although not addressed
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`herein, these hydrocarbons also introduce much higher explosivity, flammability, and heat release concerns.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The compiled data result from a fairly extensive data search. Two caveats accompany the data presented.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`First, the table constitutes a work in progress, to provide data for interim use. Some of the values may be super-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`seded as further information is located or new data become available. Second, users must satisfy themselves with
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`the suitability and appropriateness of the data for specific uses. The data or resultant determinations also must be
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`approved by the code official having jurisdiction where required. While care has been taken in assembly of the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`data, the effort cannot be viewed as exhaustive and no attempt was made to verify the data. They are intended for
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`use by knowledgeable professionals, and offered without warranty of any kind.
`
`
`
`
`
`
`
`
`Additional Data Sources
`
`
`
`
`
`
`
`
`Other sources for the data include the PAFT summaries,5 published scientific literature, manufacturers, and
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`material safety data sheets. A number of databases are available to assist in finding the data, among them the
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Chemical Abstract Service (CAS), Hazardous Substances Data Base (HSDB), and Registry of Toxic Effects of
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Chemical Substances (RTECS). Additional sources, including a number of compilations, are identified in reference
`
`
`
`
`
`
`
`
`
`
`20.
`
`
`
`CONCLUSIONS
`
`
`
`
`
`
`
`The toxicity data presented herein provide an interim means to address code requirements for use of alternative
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`refrigerants, subject to the caveats indicated. These data also may be useful to evaluate proposed changes to
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`safety standards for refrigerants and refrigeration. While the data show the alternative refrigerants to be of compa-
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`rable or lower toxicity than those they replace, and especially so for acute effects, safe use depends on adherence
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`to proper application, handling, and service procedures.
`
`
`
`
`
`ACKNOWLEDGMENTS
`
`
`
`
`
`
`
`
`
`This paper was prepared as an account of work supported in part by the U.S. Department of Energy under grant
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`number DE-FG02-91CE23810, Materials Compatibility and Lubricant Research (MCLR) on CFC-Refrigerant Sub-
`
`
`
`
`
`
`
`
`
`
`
`
`stitutes, managed by the Air-Conditioning and Refrigeration Technology Institute (ARTI). Additional program fund-
`
`
`
`
`
`
`
`
`
`
`
`
`
`ing was provided by air-conditioning and refrigeration manufacturers through the Air-Conditioning and Refrigeration
`
`
`
`
`
`
`
`
`
`
`
`
`Institute (ARI). The ARTI Project Manager guiding the toxicity data project is Mr. Glenn C. Hourahan. Support by
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Page 6 of 7
`
`161
`
`
`
`Page 6 of 7
`
`

`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`the cited parties does not constitute an endorsement, warranty, or assumption of liability for the data and views
`
`
`
`expressed herein.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The author appreciates the assistance of Mr. Hourahan and other ARTI staff in this work. Numerous individuals
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`contributed to the underlying assembly and interpretation of safety data; notable among them are William J. Brook
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`and Clem L. Warrick (DuPont Fluoroproducts), Susan G. Cairelli and Elaine Mann (NIOSH), Michael Collins and
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Paul H. Dugard (lCl