CRP Project

Solstice™ yf Refrigerant Is Safe

In 2007-2009, the Society of Automotive Engineers (SAE) conducted a comprehensive cooperative research program (CRP) to assess HFO-1234yf refrigerant.

Read Results of CRP


SAE International carried out three Cooperative Research Programs (CRP): Phase I, Phase II and Phase III regarding HFO-1234yf. Altogether eighteen international, independent research institutes were commissioned to conduct tests on various safety, performance, efficiency and suitability related aspects, including:

  • Gradient
  • Creative Thermal Solutions
  • ILK
  • Ineris
  • Hughes Associates, Inc.
  • Exponent Scientific Engineering & Consulting
  • TNO Quality of Life
  • Hamner Institute
  • WIL Research Laboratories

The Program was also sponsored by international vehicle makers and tier one/two suppliers, including Audi, BMW, Chrysler, Daimler, Fiat, Ford/Volvo, General Motors/Opel, Hyundai, Porsche, PSA, Renault, Shanghai Automotive, Tata, Jaguar Land Rover, Toyota, Volkswagen, DuPont, Honeywell, Conti Tech, Dayco, Delphi, Denso, Doowan, Dow, Freudenberg, Goodyear, Hutchinson, Maflow, Egelhof, Parker Hannifin, Sanden, Trelleborg, Valeo, and Visteon.

SAE is a global association of more than 133,000 engineers and related technical experts in the aerospace, automotive and commercial-vehicle industries. SAE is internationally recognized for quality evaluations of electronics, reliability and safety regulations and procedures in aviation ( SAE is recognized globally as the preeminent name in automobile technology and testing evaluation.

Risk Assessment

The CRP was completed in three phases. The different phases concentrated on safety and risk assessment, toxicity, flammability, air conditioning system efficiency and performance and material compatability.

Phase 1 was proposed in October 2007 and completed in February 2008. The preliminary results were presented at the VDA (Verband der Automobilindustrie) Winter meeting.

Phase 2 was completed in April 2009. That same month, the results were presented to the U.S. Environmental Protection Agency (EPA).

Phase 3 was completed in November 2009 and presented to the EPA that same month. Phase 3 differed from the other two phases by making available more detailed experimental data and vehicle measurements. Fault tree scenarios were greatly expanded.


Fault tree analysis is a top-down, deductive failure analysis in which an undesired state of a system is analyzed using boolean logic to combine a series of lower-level events. This analysis method is mainly used in the field of safety engineering and reliability engineering to determine the probability of a safety accident or a particular system level (functional) failure. It is used by NASA for all its space programs.

In the CRP for HFO-1234yf, proven fault tree analysis methods were completed for key exposure scenarios, including vehicle and repair exposures.

Also, qualitative evaluation of other scenarios with limited impact were conducted in particular parking garages and tunnels.

Fault tree inputs were based on measured data, public databases, and consensus by the vehicle makers, heating ventilation and air conditioning and safety engineers. The results were compared to other non-AC related automotive risks.

Refrigerant Concentration

The CRP also looked at the impact of the refrigerant concentration after an accidental release of the coolant.

Concentrations were measured based on a normal refrigerant charge in cars from four different vehicle manufacturers and four different types of vehicles.


  • The health-based limit (HBL) was never exceeded.
  • Lower Flammability Limit (LFL) exceeded only in a few instances in severe collision scenarios, and generally away from possible ignition sources – typically in recirculation mode, low blower, near floor.
  • Refrigerant concentration alone is not enough to produce a risk of refrigerant ignition or exposure to hydrofluoric acid (HF). All measured concentrations and the scenarios leading to these were considered in the Fault Tree Analysis.


Under the EU CLP Regulation (Classification, Labeling and Packaging), a substance must be tested for flammability using a standard OECD test. Based on the test results, the Lower and Upper Flammability Limits have been determined and HFO-1234yf is classified as flammable category 1.

Moreover, extensive additional testing has been conducted by Chillworth Laboratories addressing several critical parameters such as minimum ignition energy, energy of combustion and flame propagation. Based on these parameters, HFO-1234yf has proven to be a very mildly flammable gas compared to propane or gasoline. Under the CRP computer modeling, laboratory testing and in-vehicle testing has been conducted. Several automotive OEMs have conducted a real crash test with no refrigerant ignition.

Ineris performed contract work to determine potential electrical ignition sources in the passenger compartment (Monforte and Caretto, 2009). Based on this work, the SAE CRP concluded that the most credible ignition sources are high powered battery shorts (greater than 50 amps), matches, and fires started by other components in the vehicle that are unrelated to the refrigerant.

Ineris and Hughes Associates performed ignition testing on hot cylindrical bodies with flow and geometry variables found in the engine compartment. Ineris found that PAG lubricant itself could be ignited by the hot body at temperatures of 400°C. Ignition testing with HFO-1234yf alone did not ignite until the hot body was raised above 1000°C.

Investigations with HFO-1234yf and PAG lubricant at concentrations from 1-7 mass percentage lubricant lowered the geometry dependent ignition temperature of HFO-1234yf to 750 ± 50°C (see page 51, SAE final report). This temperature was confirmed by testing at Hughes Associates, who found a single ignition at 700°C during engine compartment testing.

Honeywell´s safety data sheet declares the auto ignition temperature of HFO-1234yf to be at 405°C. The auto ignition temperature is the lowest temperature at which a gas spontaneously ignites in a homogeneous mixture with air. The auto ignition temperature is measured in a laboratory setup, where the mixture is gradually heated up. Such laboratory conditions cannot be duplicated in real-life situations.

Ineris confirmed the auto ignition of pure HFO-1234yf in the laboratory at 405°C. Upon continued investigation, Ineris found four typical geometries in the automobile, the ignition temperature of pure HFO-1234yf on a hot cylindrical body to be on the order of 1050°C due to flow and geometry variables (see page 26). Investigations with PAG lubricant at concentrations from 1-7 mass percent lowered the geometry dependent ignition temperature to 750 ± 50°C. This temperature was confirmed by testing at Hughes Associates, who found a single ignition at 700°C during an engine compartment flammability test.

In summary:

  • HFO-1234yf is substantially different from flammable refrigerants such as propane
  • Much more energy is required to ignite HFO-1234yf
  • Much lower burning velocity/much lower flame stability
  • The energy released when the refrigerant ignites is much less than other flammable substances used in vehicles
  • Although very difficult to ignite, the possibility of refrigerant ignition does exist; therefore the analysis indicated was fully addressed in the Phase 3 fault tree analysis and indicated that HFO-1234yf was safe to use

Hydrofluoric Acid

All fluorinated refrigerants – including R-134a used in cars – today can decompose to produce Hydrofluoric Acid (HF) when exposed to sufficient energy sources. This process is called thermal decomposition. The amount of HF generated depends on the amount of refrigerant in contact with an energy source and the time of contact. However, in the decades of safe usage of previous fluorinated refrigerants, there is not a single reported injury based on HF.

Tests of HF generation were conducted with an actual vehicle (Hughes Associates) and with “bench-top” studies (INERIS, Honeywell, DuPont). Measured HF concentrations were evaluated using National Research Council Acute Exposure HF concentration due to HFO-1234yf thermal decomposition below the lower flame limit after refrigerant release and exposure to flame source. The result: the highest HF breath level concentration achieved was well below AEGL-2.

In the test, the ignition of the lighter had to be augmented with a strong spark to get a butane flame and there was no ignition of refrigerant. In the engine compartment, HF concentrations were measured after HFO-1234yf was released onto a hot surface. The result: highest HF breath level concentration was well below AEGL-2 in the passenger cabin.
This was based on: extreme engine operation (700°C), hood seal was removed allowing direct path to cabin air intake, secondary failure and blower on outside air, high speed (full suction from engine).