A refrigerant is a substance or mixture used as the working fluid in a system that removes heat from a lower-temperature region and delivers it to a higher-temperature region. Refrigerants circulate through refrigerators, air conditioners, industrial cooling equipment, and heat pumps. Most commercial systems exploit evaporation and condensation, although some operate with a gas throughout the cycle or reject heat above the fluid’s critical temperature. The same substance may serve as a refrigerant in one installation and as a secondary heat-transfer fluid or process material in another.
Refrigerant selection connects thermodynamics, chemical engineering, equipment design, safety, and environmental protection. A useful fluid must provide suitable operating pressures and heat-transfer performance while remaining compatible with compressors, lubricants, seals, and construction materials. Toxicity, flammability, atmospheric lifetime, ozone depletion, and contributions to climate change also influence its acceptability. Changes in these priorities have repeatedly transformed the industry, from early use of ammonia and carbon dioxide, through widespread adoption of fluorinated compounds, to the development of lower-climate-impact fluids and redesigned refrigeration systems.
Thermodynamic function and refrigeration cycles
Refrigeration does not create “cold” as a separate physical quantity. It transfers thermal energy away from a space or material. Because spontaneous heat transfer proceeds from higher to lower temperature, maintaining a cold region while rejecting heat to warmer surroundings requires an additional input of work or another energy source. This requirement follows from the second law of thermodynamics. A refrigerant enables the transfer by passing through a sequence of states in which its pressure, temperature, and often phase change.
In a conventional vapor-compression cycle, low-pressure refrigerant absorbs heat in an evaporator and boils. The resulting vapor enters a compressor, which raises its pressure and temperature. In the condenser, the hot fluid rejects heat to air, water, or another heat sink and usually becomes liquid. An expansion valve or capillary tube then reduces its pressure before it returns to the evaporator. Expansion through this restriction is approximately a constant-enthalpy process, rather than a process in which useful mechanical work is recovered. Part of the liquid normally flashes into vapor as its pressure falls.
The heat absorbed during boiling is associated with latent heat. A refrigerant can therefore transport substantial energy without requiring a large temperature increase during evaporation. For an idealized steady evaporator, its cooling capacity is
where is refrigerant mass flow rate and is specific enthalpy. The refrigeration coefficient of performance is
A COP above one does not violate energy conservation: the work input moves heat rather than being converted directly into all the delivered cooling. For an ideal reversible refrigerator operating between reservoirs at absolute temperatures and , the Carnot limit is . Real equipment performs below this limit because of compression losses, pressure drops, and finite temperature differences.
Not all refrigerants operate in this simple four-component cycle. In absorption refrigeration, a refrigerant is absorbed into another substance, pumped as part of a liquid solution, and subsequently separated using heat. Ammonia–water systems use ammonia as refrigerant and water as absorbent; water–lithium bromide systems use water as refrigerant. Gas refrigeration cycles can instead transfer heat through compression and expansion without condensation. A reversible heat pump uses much the same working-fluid circuit as an air conditioner, but changes the useful destination of heat: cooling performance concerns the evaporator load, whereas heating performance concerns heat delivered at the condenser or gas cooler.
Properties governing refrigerant selection
A refrigerant’s usefulness depends on its entire operating envelope, not simply its normal boiling point. The normal boiling point is the temperature at which a liquid boils at one standard atmosphere; actual evaporators and condensers operate at pressures chosen for their required temperatures. R-134a, for example, has a normal boiling point near −26 °C, but can evaporate at other temperatures when its pressure changes. The saturation pressure–temperature relationship determines compressor pressure ratio, vessel requirements, and whether parts of the system operate above or below atmospheric pressure. Subatmospheric operation can allow air and moisture to enter through leaks.
The critical point sets another important boundary. Above the critical temperature, a fluid cannot be liquefied by pressure alone. Carbon dioxide has a critical temperature of approximately 31.0 °C and a critical pressure of 7.38 MPa. In many warm-weather carbon dioxide systems, high-side heat rejection therefore occurs without condensation in a gas cooler. Such operation is called transcritical because the cycle includes both subcritical and supercritical states. By contrast, fluids with substantially higher critical temperatures may condense under the same ambient conditions. Neither arrangement is inherently best in every application: heat-sink temperature and cycle design strongly affect performance.
Latent heat alone does not determine compressor size or energy consumption. Suction-vapor density influences the cooling capacity delivered per unit of compressor displacement. A fluid with low vapor density may require a large compressor even if it absorbs considerable heat per kilogram. Liquid and vapor viscosities, thermal conductivity, and surface tension affect heat-transfer coefficients, pressure losses, and the distribution of two-phase flow. Compressor discharge temperature also matters because excessive temperatures can accelerate oil degradation and stress components. Accurate calculations consequently require an appropriate equation of state and transport-property data across the relevant range.
Chemical and material properties impose additional constraints. A refrigerant must remain sufficiently stable in contact with metals, oils, motor insulation, and elastomeric seals. Moisture can promote corrosion, react with some lubricants, or freeze at a restriction. Oil miscibility affects lubricant return to the compressor: strong miscibility can aid circulation but can also alter viscosity and heat-transfer behavior, while poor miscibility may require deliberate oil separation and return arrangements. Changing refrigerant chemistry can therefore require a different lubricant and different seals.
Selection is consequently a system-level trade-off. An efficient fluid in one compressor or climate may perform poorly in another. Charge quantity, heat-exchanger geometry, controls, part-load operation, service conditions, and permitted safety measures all influence the outcome. A replacement fluid’s pressure similarity to an older refrigerant is useful information, but does not establish that the replacement is technically interchangeable.
Chemical families and representative refrigerants
Refrigerants include inorganic compounds, hydrocarbons, and several families of fluorinated organic compounds. The distinction between “natural” and “synthetic” refrigerants is common in industry, but it is not a formal division based on safety or industrial production. Natural refrigerants are substances that occur in natural processes, such as ammonia, carbon dioxide, water, and hydrocarbons. Refrigerant-grade supplies of these substances are nevertheless manufactured or purified to meet demanding specifications. Their occurrence in nature does not by itself establish low toxicity, low flammability, or suitability for a particular application.
Ammonia, designated R-717, offers high latent heat and favorable heat-transfer properties and is widely used in industrial refrigeration. Its toxicity, some flammability, and incompatibility with copper-containing components shape system design. Carbon dioxide, R-744, is nonflammable and provides high volumetric refrigeration capacity, but typically requires much higher operating pressures than many conventional fluorocarbon systems. Water, R-718, can be effective where freezing is not a limitation, including absorption chillers; low-temperature evaporation requires a deep vacuum and produces a large vapor volume. Air, R-729, is used in gas-cycle refrigeration, notably in aircraft environmental-control systems.
Hydrocarbon refrigerants include propane, R-290, and isobutane, R-600a. Their favorable thermodynamic properties and very low direct climate impacts have encouraged use in domestic refrigerators, commercial cabinets, and some heat pumps. Their high flammability limits permissible charge and influences equipment construction and installation. Ethane and ethylene are used in specialized low-temperature stages. Although hydrocarbons are chemically familiar fuels, refrigerant-grade material is subject to purity and moisture specifications distinct from those of ordinary fuel supplies.
Chlorofluorocarbons, or CFCs, contain carbon, chlorine, and fluorine; R-11 and R-12 are prominent historical examples. Hydrochlorofluorocarbons, or HCFCs, additionally contain hydrogen, making them generally less persistent in the lower atmosphere than fully halogenated CFCs, although they still deplete stratospheric ozone. R-22 became especially important in air conditioning. Hydrofluorocarbons, or HFCs, contain hydrogen, fluorine, and carbon but no chlorine. R-134a and R-32 are examples. HFCs do not cause the chlorine-driven ozone depletion associated with CFCs and HCFCs, but many are potent greenhouse gases.
Hydrofluoroolefins, or HFOs, are unsaturated fluorinated compounds containing a carbon–carbon double bond. That bond commonly makes them more reactive in the lower atmosphere, shortening their atmospheric lifetime. R-1234yf and R-1234ze(E) are important examples. Some newer refrigerants are unsaturated compounds that also contain chlorine and are classified as hydrochlorofluoroolefins. Chemical-family labels alone do not specify environmental performance: atmospheric lifetime, breakdown products, flammability, and thermodynamic behavior depend on the individual molecule.
Numbering, mixtures, and temperature glide
Standardized refrigerant designations allow engineers to identify fluids without relying on proprietary trade names. ASHRAE Standard 34 and ISO 817 provide widely used frameworks for designations and safety classification. For many saturated acyclic halocarbon refrigerants, a three-digit number encodes molecular composition: the first digit is the number of carbon atoms minus one, the second is the number of hydrogen atoms plus one, and the third is the number of fluorine atoms. A leading zero is omitted, and remaining valences are filled by chlorine. Thus R-22 corresponds to chlorodifluoromethane, . Lowercase suffixes distinguish certain molecular isomers.
Other families require additional conventions. Unsaturated compounds use an expanded numbering scheme, while inorganic refrigerants generally occupy the 700 series, where the number commonly reflects 700 plus relative molecular mass. Ammonia is consequently R-717 and carbon dioxide R-744. Specific hydrocarbons and other organic fluids also have established designations outside the simple halocarbon formula. The letter R identifies a refrigerant designation, not a performance rating. Neither a larger number nor a later assignment implies greater efficiency, lower toxicity, or a lower climate impact.
Mixtures allow properties to be adjusted when a single compound cannot meet all requirements. The 400 series generally identifies zeotropic blends, while the 500 series identifies azeotropic blends. In an azeotrope, liquid and vapor have the same composition at the azeotropic condition, so boiling can resemble that of a pure fluid. A zeotropic mixture has different equilibrium liquid and vapor compositions. At a given pressure, evaporation therefore proceeds over a temperature interval rather than at one fixed saturation temperature. This interval is called temperature glide. R-407C, a mixture of R-32, R-125, and R-134a, has appreciable glide; R-410A, a mixture of R-32 and R-125, has much smaller glide under typical conditions.
For zeotropic blends, the bubble point marks the onset of boiling of a liquid and the dew point marks the onset of condensation of a vapor. These distinct temperatures matter when calculating superheat, subcooling, and heat-exchanger temperature differences. Glide can help match a changing water or air temperature in some counterflow heat exchangers, but can also complicate distribution and control. Comparing two refrigerants using an unspecified “evaporating temperature” can therefore be misleading if one is a pure fluid and the other has substantial glide.
Composition may also change through fractionation. Vapor leaking from a two-phase mixture need not have the same composition as the original charge, and withdrawal of vapor from a storage cylinder can alter the composition delivered. Charging and servicing procedures account for these effects. Commercial blends are defined by specified component mass fractions, and uppercase suffixes distinguish different formulations using the same component set. R-407A and R-407C, for example, are not interchangeable names for one composition.
Historical development
Early artificial refrigeration grew from investigations of evaporation, gas expansion, and heat. In 1755, William Cullen demonstrated cooling through the evaporation of ether under reduced pressure. Such experiments established that evaporation could lower temperature, but did not yet constitute a practical continuously operating refrigeration industry. Jacob Perkins received an 1834 patent for an apparatus implementing a vapor-compression cycle with a volatile fluid. During the nineteenth century, inventors including James Harrison and Carl von Linde helped turn mechanical refrigeration into commercially useful equipment for brewing, food storage, and other industries.
Early working fluids included ether, ammonia, carbon dioxide, sulfur dioxide, and methyl chloride. Their disadvantages varied: some were highly flammable, some toxic or corrosive, and some demanded difficult operating pressures or large machinery. Ammonia became especially important in industrial plants, where trained operators and machinery rooms could accommodate its hazards. Carbon dioxide also found uses, including marine refrigeration. The expansion of domestic refrigeration created different requirements because equipment would operate in occupied homes, often without specialist supervision and with relatively little maintenance.
In the late 1920s, research associated with General Motors and Frigidaire sought refrigerants that combined useful thermodynamic behavior with low flammability and low acute toxicity. Thomas Midgley Jr., Albert Henne, and Robert McNary were central figures in the development of chlorofluorocarbon refrigerants. R-12 entered commercial use in the early 1930s. CFCs and later HCFCs helped support mass adoption of domestic refrigerators, air conditioning, commercial refrigeration, and centrifugal chillers. Their apparent chemical stability was an operational advantage, but the same persistence later proved important to their environmental effects.
In 1974, Mario Molina and F. Sherwood Rowland described how long-lived chlorofluorocarbons could reach the stratosphere and release chlorine that destroys ozone. The discovery of severe Antarctic ozone depletion, reported in 1985 by Joseph Farman, Brian Gardiner, and Jonathan Shanklin, intensified international attention. The 1987 Montreal Protocol established a framework for controlling ozone-depleting substances. CFC replacements included HCFCs as transitional fluids and HFCs that lacked chlorine. R-134a replaced R-12 in many applications, while blends such as R-404A and R-410A became prominent in refrigeration and air conditioning.
Concern about greenhouse warming subsequently shifted attention from ozone depletion alone to the climate effects of refrigerants and equipment energy use. Research and commercialization expanded around carbon dioxide, ammonia, hydrocarbons, HFOs, and lower-global-warming-potential blends. In 2016, the Kigali Amendment added controls on HFCs to the Montreal Protocol framework. The historical pattern is therefore not a simple succession from inferior to universally superior fluids: each transition changed the balance among operating efficiency, safety, atmospheric effects, cost, and the practical constraints of existing machinery.
Safety classification and equipment hazards
Refrigerant safety concerns include toxicity, oxygen displacement, flammability, high pressure, cold contact, and hazardous decomposition products. These hazards are partly independent. A nonflammable fluid can still create an asphyxiation risk, while a low-toxicity fluid can be dangerous if released rapidly into an enclosed space. A liquid release can cause severe local cooling and cold injury. Stored pressure introduces mechanical hazards in vessels, piping, hoses, and trapped liquid sections, where thermal expansion can produce very high pressures if adequate pressure relief is absent.
ASHRAE Standard 34 and ISO 817 classify refrigerants using toxicity and flammability categories. The toxicity letters A and B distinguish lower and higher toxicity according to the standards’ criteria. They do not mean “harmless” and “poisonous” in an absolute sense. Flammability categories include 1, 2L, 2, and 3. Class 1 denotes no flame propagation under the specified test conditions; 2L identifies lower-flammability refrigerants with low burning velocity; class 3 includes highly flammable fluids. R-134a and carbon dioxide are A1, R-32 and R-1234yf are A2L, propane and isobutane are A3, and ammonia is B2L.
These classifications help establish requirements, but do not replace a full assessment of the installation. The quantity released, room volume, ventilation, ignition sources, occupancy, and location of components all affect consequences. A small sealed refrigerator and a large distributed supermarket system can present very different risks even when charged with the same chemical. Equipment and installation standards translate these distinctions into provisions concerning charge limits, leak detection, electrical components, ventilation, pressure relief, and placement. Exact provisions vary by equipment type, jurisdiction, and edition of the applicable standard.
Ammonia illustrates why fluid properties must be considered together. Its strong odor often provides warning at low concentrations, but odor is not a substitute for exposure controls. It is toxic at sufficiently high concentrations, can burn under particular conditions, and requires compatible materials. Carbon dioxide is odorless and nonflammable, yet elevated concentrations have physiological effects beyond simple oxygen displacement. Its high operating pressures and the possibility of solid carbon dioxide formation during rapid depressurization introduce additional engineering issues.
Fluorinated refrigerants can form irritating or toxic decomposition products when exposed to flames, very hot surfaces, or unsuitable process conditions. Hydrofluoric acid is an important possible product; chlorine-containing compounds can produce additional hazardous substances. Flammability testing also has defined boundaries, so an A1 classification does not establish chemical inertness under every conceivable condition. Safety engineering consequently considers the refrigerant in normal operation, foreseeable leaks, maintenance, fire exposure, and decommissioning rather than treating a designation as a complete hazard description.
Atmospheric effects and international controls
Refrigerants affect the atmosphere principally through ozone depletion, greenhouse warming, and, for some compounds, persistent breakdown products. Ozone depletion occurs when sufficiently long-lived chlorine- or bromine-containing substances reach the stratosphere and release reactive halogen species. Chlorine can participate in catalytic reaction cycles that destroy many ozone molecules before it is removed from active circulation. Ozone-depletion potential, or ODP, expresses the integrated effect of a substance relative to CFC-11. Most HFCs and HFOs have zero ODP because they contain neither chlorine nor bromine, but zero ODP does not imply zero environmental impact.
Global warming potential, or GWP, compares the time-integrated radiative effect of an emission with that of an equal mass of carbon dioxide over a specified period. A 100-year horizon is widely used in refrigerant regulation and accounting. Values depend on scientific assessment methods and updates; for example, the IPCC Fourth Assessment Report gave R-134a a 100-year GWP of 1,430 and R-32 a value of 675. A numerical GWP should therefore be accompanied by its time horizon and assessment basis. Carbon dioxide has a GWP of one by definition, while many conventional HFCs have values in the hundreds or thousands.
Direct emissions arise during manufacturing and filling, equipment operation, servicing, accidents, and disposal. Indirect emissions arise from the energy used to operate equipment, depending on the carbon intensity of the energy supply. A lower-GWP refrigerant can substantially reduce the consequences of leakage, but a system that consumes much more electricity may offset part of that benefit. Total equivalent warming impact and life-cycle climate performance are accounting approaches that combine several of these contributions. Their results depend on assumptions about leakage, service life, efficiency, refrigerant recovery, and future electricity generation.
The Montreal Protocol, adopted in 1987, controls ozone-depleting substances through schedules that differ among substance groups and country categories. Its Kigali Amendment, adopted in 2016 and entering into force in 2019, establishes schedules to reduce HFC production and consumption, generally expressed in carbon-dioxide-equivalent terms. These measures are phase-downs of groups of substances, not a universal prohibition on every fluorinated refrigerant. National and regional rules add application-specific limits, service restrictions, reporting duties, and recovery requirements. Existing equipment may therefore remain subject to different provisions from newly manufactured equipment.
The atmospheric chemistry of newer fluids is also under study. Some HFOs and related compounds can degrade to trifluoroacetic acid, a highly persistent substance in aquatic environments. Formation yield and transport differ among precursor compounds, and environmental assessment depends on concentrations, exposure pathways, and effects. This issue is distinct from both ozone depletion and direct greenhouse warming: a refrigerant can have a short atmospheric lifetime and low GWP while still producing a persistent degradation product.
Applications and system-specific choices
Domestic refrigerators generally have small sealed charges, modest cooling duties, and long operating periods. Isobutane is widely used because it can provide efficient performance with a small refrigerant inventory. Its flammability requires equipment designed for that fluid rather than casual substitution into older machinery. Domestic and commercial heat pumps face different demands: they may need to operate over a broad outdoor-temperature range and deliver water at temperatures substantially above indoor air temperature. R-32, propane, and various fluorinated blends are among the fluids used, with selection shaped by efficiency, pressure, discharge temperature, installation location, and safety requirements.
Supermarket refrigeration combines medium-temperature food display with lower-temperature frozen storage, often across many cabinets and long pipe networks. Historically, large centralized direct-expansion systems used substantial charges of high-GWP refrigerants such as R-404A. Alternatives include carbon dioxide booster systems, cascade arrangements, and secondary-loop systems. In a cascade plant, separate refrigerant circuits exchange heat through an intermediate heat exchanger, allowing each fluid to operate in a suitable temperature range. Carbon dioxide systems may employ parallel compression, ejectors, or other cycle modifications to improve performance, especially under demanding ambient conditions.
Industrial cold stores, food processing, ice production, and large freezing installations often use ammonia because its thermodynamic and heat-transfer properties suit high-capacity plants. Some systems combine ammonia in a central machinery area with carbon dioxide or a secondary liquid distributed to occupied or processing spaces. This arrangement can reduce the ammonia inventory or keep it out of particular areas. Water–lithium bromide absorption chillers serve applications where heat is available from steam, hot water, or combined heat-and-power equipment. Because water is the refrigerant, these chillers are generally used for chilled-water duties above the freezing point rather than deep-freezing applications.
Mobile air conditioning imposes constraints from compact heat exchangers, vibration, accident exposure, and variable compressor operation. R-134a became widely used after the replacement of R-12; R-1234yf has subsequently been adopted extensively in newer vehicles, while carbon dioxide is used in some systems. Electric vehicles also use refrigerant circuits to integrate cabin conditioning with battery and powertrain thermal management. The best arrangement depends on vehicle architecture and operating conditions rather than refrigerant identity alone.
Specialized low-temperature systems use multistage, cascade, mixed-refrigerant, or gas-expansion processes. Ethane and other low-boiling fluids can serve in lower cascade stages, while helium is essential in many very-low-temperature applications. At these extremes, ordinary room-temperature assumptions about phase behavior and compressor arrangements no longer suffice. A distinction is also necessary between the primary refrigerant and a secondary coolant: glycol solution circulated through a building, for example, may carry heat to a central chiller without itself undergoing the refrigeration cycle.
Refrigerant management and continuing development
A refrigerant remains an engineered material throughout its service life. Refrigerant-grade specifications address chemical identity, composition, moisture, acidity, noncondensable gases, and other impurities. Contamination can change operating pressures, impair heat transfer, promote corrosion, and damage lubricants or insulation. Air in a condenser raises the effective condensing pressure and occupies heat-transfer surface, while moisture may freeze at a metering device. Reliable identification matters especially in recovered material, where mixed refrigerants can no longer be assumed to have the properties of the original labeled product.
Recovery, recycling, reclamation, and destruction are distinct activities. Recovery removes refrigerant from equipment and stores it externally. Recycling reduces contaminants using processes such as filtration and oil separation, commonly for reuse under applicable rules. Reclamation processes used refrigerant to a specified purity standard, usually with analytical verification; AHRI Standard 700 is a widely used reference for refrigerant specifications. Destruction instead breaks down the chemical in an authorized treatment process. Recovery does not automatically establish that a fluid is suitable for return to service.
Changing a system to a different refrigerant can require much more than replacing its charge. Compressor operating limits, lubricant chemistry, expansion-device capacity, control settings, elastomer compatibility, and pressure ratings may all change. A replacement may also alter cooling capacity or efficiency even when nominal pressures are similar. Converting equipment from a nonflammable to a flammable refrigerant changes the safety basis of the installation and may fall outside its certification. The expression “drop-in replacement” therefore has limited technical meaning unless the permissible application and required modifications are specified.
Current development increasingly treats the refrigerant and hardware as a coupled design problem. Smaller-diameter heat-exchanger passages and compact heat exchangers can reduce refrigerant charge, but may introduce distribution and pressure-drop challenges. Improved compressors, variable-speed drives, ejectors, economizers, and optimized controls can compensate for disadvantages of a particular fluid or exploit its strengths. High-temperature heat pumps require working fluids that combine suitable critical temperatures with acceptable thermal stability, pressure, and safety characteristics. Blend development uses composition as an additional design variable, although glide and fractionation remain practical constraints.
Research also depends on dependable property measurements and environmental characterization. Experimental work establishes vapor–liquid equilibria, transport properties, lubricant interactions, and decomposition behavior, while models predict performance over wider operating ranges. Atmospheric studies examine reaction rates and products rather than relying solely on molecular-family labels. Alongside new chemistry, reducing leaks, improving recovery, and maintaining efficient operation remain central engineering tasks because even a well-chosen refrigerant performs poorly when charge loss, contamination, or unsuitable control prevents the system from operating as designed.