Manufacturers have commercialized more than 50 new refrigerants (including blends) in the last decade, and they are examining additional candidates. Users should expect a number of new introductions as the phaseout approaches for R-22, which now is the most widely used refrigerant. A similar flurry of service fluids occurred with the phaseout of R-12 and R-502; R-12 was the most widely used refrigerant until a few years ago.

This article provides two tables that summarize selected physical, safety, and environmental data for the old and current refrigerants and for leading replacement candidates. The data in the two tables are the same, but are presented in different manners.

Table 1 is sorted by standard refrigerant designations. Table 2 contains the same information sorted by the normal boiling points of the refrigerants. Table 1 lends itself to finding information on a specific refrigerant. The sort order for Table 2 rearranges the refrigerants in coarse proximity of candidacy for similar applications, to facilitate comparisons.

The data in these tables were taken from the ARTI Refrigerant Database, an information system on alternative refrigerants, associated lubricants, and their use in air conditioning and refrigeration.1 The database consolidates and facilitates access to property, compatibility, safety, environmental, application, and other data.2 It also provides an extensive bibliographic reference system.

Refrigerant Data Tables

The parameter descriptions that follow are in the same sequence as presented in the tables, going from the left to the right columns.

Identifiers. The number shown is the standard designation based on those assigned by or recommended for addition to ANSI/ASHRAE Standard 34-1997, Designation and Safety Classification of Refrigerants, and pending addenda thereto.3 These familiar designations are used almost universally, usually preceded by "R-", "R", the word "Refrigerant", composition-designating prefixes (e.g. "CFC-", "HCFC-", "HFC-", or "HC-"), or manufacturer trade names.

The chemical formula indicates the molecular makeup of the single-compound refrigerants, namely those consisting of only one chemical substance. The blend composition is shown for refrigerant blends, namely those consisting of two or more chemicals that are mixed to obtain desired characteristics. The composition consists of two parts. The first identifies the components, in order of increasing normal boiling points and separated by slashes. The second part, which is enclosed in parentheses, indicates the mass fractions (as percentages) of those components in the same order. The tables also indicate the common names by which some refrigerants are frequently identified.

Physical properties. The molecular mass is a calculated value based on the atomic weights recognized by International Union of Pure and Applied Chemists (IUPAC).4 It indicates the mass in grams of a mole of the refrigerant or, for blends, the mass-weighted average of a mole of the mixture.

The normal boiling point (NBP) is the temperature at which liquid refrigerant boils at standard atmospheric pressure, namely 101.325 kPa (14.6959 psia). The NBP and most dimensional units in the tables are shown in both metric (SI) and inch-pound (IP) units of measure. The bubble point temperature - at which a bubble first appears, hence the temperature at which boiling begins - is shown as the NBP for blends. Unlike single-compound refrigerants that boil at a single temperature for a given pressure, the dissimilar volatilities of components cause mixture boiling to span a range between the bubble point and dewpoint temperatures. The dewpoint is so named because it is the condition at which condensation begins when the blend is cooled.

The critical temperature (Tc) is the temperature at the critical point of the refrigerant, namely where the properties of the liquid and vapor phases are identical. Unless actually determined, the Tc values shown for blends are the mass weighted averages of the component Tc's, sometimes referred to as the pseudo-critical temperature.

The critical pressure (Pc) is the pressure at the critical point.

The NBP and critical properties suggest the application range for which an individual refrigerant might be suitable. Those with extremely low NBP lend themselves to ultra-low temperature refrigeration, for example in cryogenic applications. Those with high NBPs generally are limited to high-temperature applications, such as chillers and industrial heat pumps. Both capacity and efficiency decline when condensing temperatures approach the Tc in a typical vapor-compression (reverse Rankine) cycle, the one most commonly used. The Pc will exceed the operating pressure except in transcritical cycles, which are uncommon except for R-744 (carbon dioxide). It is useful to compare relative operating pressures since practical cycles usually are designed to condense at 70% to 90% of the Tc (on an absolute basis) and, therefore, at corresponding fractions of the Pc.5

Safety data. The first value is the occupational exposure limit, namely the Threshold limit value - Time Weighted Average (TLV-TWA) or a consistent measure. It is an indication of chronic (long-term, repeat exposure) toxicity of the refrigerant. Two consistent toxicity indices are the Workplace Environmental Exposure level (WEEL) guide and the permissible exposure limit (PEL). Some countries and manufacturers refer to them as the acceptable exposure limit (AEL), industrial exposure limit (IEL), or occupational exposure limit (OEL). These measures indicate adopted limits for workplace exposures for trained personnel for typical workdays and work weeks.

The lower flammability limit (LFL) is the lowest concentration at which the refrigerant burns in air under prescribed test conditions. It is an indication of flammability.

The heat of combustion (HOC) is an indicator of how much energy the refrigerant releases when it burns in air, assuming complete reaction to the most stable products in their vapor state. Negative values indicate endothermic reactions (those that require heat to proceed) while positive values indicate exothermic reactions (those that liberate heat).

The ASHRAE Standard 34 safety group is an assigned classification that is based on the TLV-TWA (or consistent measure), LFL, and HOC. It comprises a letter (A or B) that indicates relative toxicity followed by a number (1, 2, or 3) that indicates relative flammability. These classifications are widely used in mechanical and fire construction codes, to determine requirements to promote safe use. Most of these code provisions are based on ASHRAE Standard 15, "Safety Code for Mechanical Refrigeration." Some of the classifications shown are followed by the lower case letter "r," which signifies that SSPC 34 has recommended revision or addition of the classification shown, but final approval and/or publication is still pending. Similarly, a "d" indicates pending deletion.

Blends were assigned dual classifications such as A1/A2 in the past, to indicate the safety groups both as formulated and for the worst case of fractionation. That practice changed to assignment of a single safety group reflecting the worst case of fractionation for specified leak and refill scenarios.

Environmental data. The atmospheric lifetime (tatm) is an indication of the average persistence of refrigerant released into the atmosphere until it decomposes, reacts with other chemicals, or is otherwise removed. While tatm factors into additional environmental parameters, it also is significant in its own right. It suggests atmospheric perseverance and therefore the potential for accumulation of released refrigerants (and other chemicals). Long atmospheric lifetime implies the potential for slow recovery from environmental problems, both those already known and additional concerns that may be identified in the future.

The values shown for the refrigerant lives are composite atmospheric lifetimes. The lifetimes also can be shown separately for the tropospheric (lower atmosphere, where we live), stratospheric (the next layer, where global depletion of ozone is a concern), and higher layers since the atmospheric chemistry changes between layers.

The ozone depletion potential (ODP) is a normalized indicator, based on a value of 1.000 for R-11, of the ability of refrigerants (and other chemicals) to destroy stratospheric ozone molecules. The data shown are the values adopted by international scientific assessment. The ODPs shown for blends are mass-weighted averages.

The ODPs in the tables are modeled ODP values, the most indicative of environmental impacts. There are several other ODP indices, including semi-empirical, time-dependent, and regulatory variations.

Semi-empirical ODPs are calculated values that incorporate adjustments for observed atmospheric measurements. This approach is conceptually more accurate, but it is difficult to measure precisely the data needed for representative adjustments.

Time-dependent ODPs use chemicals other than R-11 as the reference. Normalizing values to short-lived compounds emphasizes near-term impacts, but discounts long-term effects. Time-dependent ODPs are not often cited, particularly since the release of ozone-depleting substances already has peaked and recovery of the stratospheric ozone layer is underway.

Regulatory ODPs generally are old data used to set phaseout steps, determine compliance with the Montreal Protocol, and allocate production quotas in national regulations. Because of the political and competitive complexities in changing consumption targets and production allocations, these values commonly are left unchanged even when newer scientific findings improve the quantification precision. The ODP values listed in the annexes to the Montreal Protocol, for example, have not been updated since 1987 for chlorofluorocarbons (CFCs) and 1992 for hydrochlorofluorocarbons (HCFCs). A note in the protocol indicates that the values "are estimates based on existing knowledge and will be reviewed and revised periodically," but that has not happened yet.6

The global warming potential (GWP) is a similar indicator of the potency to warm the planet by action as a greenhouse gas. The values shown are relative to carbon dioxide (CO2) for an integration period of 100 years. The GWPs shown for blends are mass-weighted averages.

GWP values can be calculated for any desired integration period, commonly referred to as the integration time horizon (ITH). Short ITH periods emphasize immediate effects but overlook later impacts, while long ITH periods incorporate the later effects. The most common GWP values, including those cited herein, are for an ITH of 100 years.

A variant of GWPs includes an offset for cooling (more correctly negative radiative forcing) resulting from ozone destruction related to the released refrigerant (or other chemical), since ozone is itself a potent greenhouse gas. Once again, the idea behind these net GWPs is conceptually more accurate, but they are not frequently cited since the offset is difficult to quantify with current understanding.

Both the ODP and GWP are calculated from the tatm, measured chemical properties, and other atmospheric data. The tatm, ODP, and GWP all should be as low as possible in an ideal refrigerant, but those goals must be assessed along with criteria for performance, safety, and both chemical and thermal stability in use.5

New Environmental Data

The values for tatm, ODP, and GWP change as understanding of atmospheric science expands and the chemical kinetics involved become better understood. They also change when newer measurements are made for both specific and reference chemicals and as modeling of atmospheric chemistry improves. These factors have driven periodic reviews and consensus assessments by the scientific community.

The tatm, ODP, and GWP values shown in Tables 1 and 2 reflect data from the latest international scientific assessments of ozone depletion and climate change, published in February 1999 and June 2001 respectively.7,8 The tables include additional data from selected scientific publications for refrigerants not addressed in these assessments. The data indicated for blends are calculated values based on data for the components and the nominal compositions.

One reason why readers may see diverging values for environmental data, beyond differences associated with parameter choices and whether the data are current, has to do with accuracy. Some manufacturers and authors round the data, and errors propagate when rounded values are used for blend calculations. Also, some sources mislabel halocarbon or absolute GWP (HGWP and AGWP, respectively) values as GWPs; References 7 and 8 provide further information on these indices.


The database from which the summary data in Tables 1 and 2 were extracted was a part of the HVACR Research for the 21st Century (21-CR) initiative, a research program of the Air-Conditioning and Refrigeration Technology Institute (ARTI). This article updates an earlier version by the same authors inHeating/Piping/Air Conditioning Engineering, 71(8):27-33, August 1999. The new version addresses additional refrigerants and revised data that have emerged since then.ES

EDITOR'S NOTE: The images associated with this article can not be effectively transfered to this website and as a result are not included here. Please reference the print version of the November 2001 issue of Engineered Systems to view the images.

Additional References

7. WMO, Scientific Assessment of Ozone Depletion: 1998, report 44, WMO Global Ozone Research and Monitoring Project, Geneva, Switzerland; UNEP, Nairobi, Kenya; NOAA, Washington, DC; NASA, Washington, DC; and the European Commission, Directorate General XII - Science, Research and Development, Brussels, Belgium; February 1999.
8. Intergovernmental Panel on Climate Change (IPCC, of the WMO and UNEP), Climate Change 2001: The Scientific Basis - Contribution of Working Group I to the IPCC Third Assessment Report, Cambridge University Press, Cambridge, UK, 2001.