HFO-1234yf SNAP Workshop Meeting

November 20, 2009

Washington, DC

Meeting Summary

Prepared by ICF International

December 8, 2009TABLE OF CONTENTS

  TOC \o "1-3" \h \z \u    HYPERLINK \l "_Toc246911295"  Background	 
PAGEREF _Toc246911295 \h  4  

  HYPERLINK \l "_Toc246911296"  1.	HFO-1234yf SAE Standards	  PAGEREF
_Toc246911296 \h  4  

  HYPERLINK \l "_Toc246911297"  1.1	Systems Design	  PAGEREF
_Toc246911297 \h  4  

  HYPERLINK \l "_Toc246911298"  1.2	Refrigerant	  PAGEREF _Toc246911298
\h  5  

  HYPERLINK \l "_Toc246911299"  1.3	Service Equipment	  PAGEREF
_Toc246911299 \h  5  

  HYPERLINK \l "_Toc246911300"  1.4	Technician Requirements:	  PAGEREF
_Toc246911300 \h  5  

  HYPERLINK \l "_Toc246911301"  1.5	SAE Certification	  PAGEREF
_Toc246911301 \h  5  

  HYPERLINK \l "_Toc246911302"  2.	CRP Risk Assessment of HFO-1234yf	 
PAGEREF _Toc246911302 \h  6  

  HYPERLINK \l "_Toc246911303"  2.1	Corrosive Leaks	  PAGEREF
_Toc246911303 \h  6  

  HYPERLINK \l "_Toc246911304"  2.3	Flammability	  PAGEREF _Toc246911304
\h  6  

  HYPERLINK \l "_Toc246911305"  2.4	HF Generation	  PAGEREF
_Toc246911305 \h  7  

  HYPERLINK \l "_Toc246911306"  2.5	Input Service Scenarios	  PAGEREF
_Toc246911306 \h  7  

  HYPERLINK \l "_Toc246911307"  3.	Atmospheric Impacts	  PAGEREF
_Toc246911307 \h  8  

  HYPERLINK \l "_Toc246911308"  4.	Vehicle Makers’ Suggested Rule
Modifications	  PAGEREF _Toc246911308 \h  9  

 

ATTACHMENT 1- SAE/OEM HFO-1234yf SNAP Workshop
Presentation………………………………...……..11	

Background

The Environmental Protection Agency (EPA) has issued a proposed rule for
the chemical HFO-1234yf in motor vehicle air conditioning under EPA’s
SNAP Program. Representatives of the automotive industry, including the
SAE International, have requested to meet with the EPA to update the SAE
standards relevant to HFO-1234yf, present SAE's research on the health
and safety risks of HFO-1234yf, and to offer comments on the proposed
use conditions in EPA's proposed rule. The following is a summary of the
meeting that took place between automotive industry representatives and
EPA on November 20th, 2009. 

HFO-1234yf SAE Standards 

Ward Atkinson, SAE Climate Chairman

There are currently thirteen HFO-1234yf -related documents to be
published in the first quarter of 2010 by the SAE International
Technical Standards Board. The documents are related to automotive
system design, refrigerant purity, automotive servicing equipment,
automotive technician requirements and SAE certification. Among these,
SAE would like the EPA to incorporate relevant standards into the final
HFO-1234yf rulemaking. Industry will also refer to HFO-1234yf as
R-1234yf when it is utilized as a production refrigerant. 

Automotive Systems Design

The five SAE standards related to automotive systems design using
HFO-1234yf are:

J639 Safety Standards for Motor Vehicle Refrigerant Vapor Compression
Systems,

J2772 Measurement of Passenger Compartment Refrigerant Concentrations
Under System Refrigerant Leakage Conditions, 

J2773 R-744 and HFO-1234yf Refrigerant Standard for Safety and Risk
Analysis for use in Mobile Air Conditioning Systems, 

J2842 HFO-1234yf and R-744 Design Criteria and Certification for OEM
Mobile Air Conditioning Evaporator and Service Replacements

J2064 R-134a and HFO-1234yf Refrigerant Automotive Air-Conditioning Hose
and Assemblies. 

Together, these standards establish safety requirements, a method of
testing and verification for use in risk assessment, and a procedure to
evaluate the refrigerant concentration for mobile air conditioning (MAC)
systems. The standards also minimize exposure to personnel during normal
use or servicing of a MAC system utilizing toxic or flammable
refrigerants and establish hose design requirements to minimize
permeation of the refrigerant and to maintain functionality in the -30°
to 125°C range.  

Further, service equipment requirements and design requirements such as
in J639 reference the standard J1739, “Potential Failure Mode and
Effects Analysis in Design (Design FMEA), Potential Failure Mode and
Effects Analysis in Manufacturing and Assembly Processes (Process FMEA),
and Potential Failure Mode and Effects Analysis for Machinery (Machinery
FMEA).”

Refrigerant Purity

The J2844 standard - “Refrigerant Purity and Container Requirements
for New HFO-1234yf Refrigerant Used in Mobile Air-Conditioning
Systems” - establishes container specifications and minimum purity
levels required for use of HFO-1234yf in MAC systems. 

Service Equipment

The five SAE standards related to service equipment for HFO-1234yf are:

J2843 HFO-1234yf Recovery/Recycling/Recharging Equipment for Flammable
Refrigerants for Mobile Air-Conditioning Systems, 

J2851 HFO-1234yf Refrigerant Recovery Equipment for Mobile Automotive
Air-Conditioning Systems, 

J2888 HFO-1234yf Service Hose, Fittings and Couplers for Mobile
Refrigerant Systems Service Equipment, 

J2912 HFO-1234yf Refrigerant Identification Equipment for Use with
Mobile Air Conditioning Systems, and 

J2913: HFO-1234yf Refrigerant Electronic Leak Detectors, Minimum
Performance Criteria. 

Collectively, these standards establish minimum equipment (including
those used for refrigerant identification purposes) requirements for the
recovery/recycling/reclaiming and recharging of HFO-1234yf and detail
specific service equipment to be used in the service and maintenance of
HFO-1234yf systems.  The standards also establish minimum performance
criteria for electronic probe- type leak detectors intended for use in
automotive air conditioning systems utilizing HFO-1234yf as well as
performance and operating feature requirements for the
recovery/recycling/reclaiming and recharging of HFO-1234yf.

Technician Requirements

The J2845 standard - “Technician Training for Safe Service and
Containment of Refrigerants Used in Mobile A/C Systems (R-744 and
HFO-1234yf)” - requires that technicians be trained to recognize which
refrigerant is being handled and how to handle both refrigerants safely.


SAE Certification

The SAE J2911 standard - “Procedure For Certification That
Requirements For Mobile Air Conditioning System Components, Service
Equipment and Service Technicians Meet SAE J Standards” - establishes
a procedure for certifying that the requirements in SAE J Standards for
MAC system components, service equipment and service technicians are
met.

Closing Remarks: 

More standards have been established for HFO-1234yf than for R-134a. 
SAE has developed standards for HFO-1234yf similar to those for R-134a,
and has added a few additional standards which enable the industry to
address design and service requirements of HFO-1234yf (summary slide #24
in presentation).  All HFO-1234yf standards are to be published in the
first quarter of 2010.

CRP Risk Assessment of HFO-1234yf

William Hill, MAC Expert Consultant and Dr. Tom Lewandowski, Gradient
LLC

The Cooperative Research Program (CRP) is sponsored by vehicle makers
(e.g., BMW, Daimler, Chrysler, Ford, General Motors, Toyota, Hyundai,
Jaguar-Land Rover, etc.) and Tier One/Two Suppliers (e.g., DuPont,
Honeywell, Delphi, Goodyear, etc.). This study evaluated HFO-1234yf
material compatibility, air-conditioning system efficiency and
performance, and safety and risk assessments. The CRP Risk Assessments
found no significant risks associated with the use of HFO-1234yf in
mobile air-conditioning systems. 

CRP’s methodology utilized Fault Tree Analysis (FTA) methods for key
exposure scenarios including vehicle exposures and repair exposures, and
qualitatively evaluated other limited impact scenarios. These results
were compared to non- air conditioning (AC) related automotive risks. In
this last phase of the CRP risk assessment, new experimental data
replaced modeling assumptions, fault trees were expanded, and results
were presented for the global vehicle fleet. This provided more reliable
numbers relating to refrigerant concentrations in vehicles and potential
HF generation and a more comprehensive conceptualization of the chain of
events that could lead to exposure (e.g., by including a break-down of
factors that are speed-dependent in the FTA).  For example, four vehicle
makers obtained refrigerant concentration measurements in the passenger
cabin after simulated refrigerant release.  The refrigerant
concentrations did not exceed the health-based exposure limit of 115,000
ppm, but did exceed the lower flammability limit (LFL) in a few severe
collision instances.  It was noted that for either the refrigerant to
ignite or for HF exposure to rise above a health-based limit, a series
of events or conditions had to occur in addition to an accidental
release of the refrigerant.

Corrosion Leaks

During evaluation of risks associated with evaporator leaks due to
corrosion (for example, due to water, road salt, aging, etc.), it was
found that the specified type of leak grew slowly, and the slow release
of refrigerant as the leak grew larger created cabin concentrations of
HFO-1234yf below 2%. This concentration is significantly below the LFL
of 6.2%, and thus poses no significant risk.  The previous assessment
had indicated that large corrosion leaks could produce HFO-1234yf
concentrations in the cabin above the LFL but further consideration
shows that the charge will be lost long before leak holes reach this
size (i.e., within several hours).  Corrosion leaks of the evaporator
were therefore not addressed by the FTA, which focuses on
collision-related leaks as well as leaks related to parts fatigue (e.g.,
abraded hoses). 

Chemical Toxicity

Chemical toxicity was also found to pose an insignificant exposure risk.
Toxicity testing was performed by TNO, WIL, and Hamner Institutes.  This
analysis was conducted in the first two phases of the CRP and was not
repeated in this most recent, third phase, as exposures were not of
concern from the standpoint of refrigerant toxicity.

Flammability

The risk posed by flammability was also concluded to be low, as
evaluated by considering the possibility and effect of a flame
occurring. HFO-1234yf has a low burning velocity (1.5
centimeters/second) as compared to 152a and propane (both greater than
20 centimeters/second), meaning that it would have a limited effect if
ignited. HFO-1234yf is also difficult to ignite due to its high minimum
ignition energy (( 5000 milliJoules) as compared to methane and gasoline
(both less than 1 milliJoule). Furthermore, possible ignition sources
for HFO-1234yf are limited. The only sources considered to be credible
were a match, a failure or short of a positive temperature coefficient
(PTC) heater, and hot surfaces in the engine compartment under extreme
vehicle use. Although the flammability risk posed was concluded to be
very low, the risk was still addressed in the FTA.

HF Generation

HF generation testing was conducted in both an actual vehicle and in
bench-top studies. In a passenger cabin scenario with refrigerant
release and exposure to flame source, HF concentration due to HFO-1234yf
thermal decomposition below the LFL was significantly below the AEGL-2
limit (95 ppm over 10 minutes) inside the passenger cabin.  Moreover,
there was no direct ignition of the refrigerant HFO-1234yf; instead, a
strong spark was used to produce a butane flame, which was then used to
light the HFO-1234yf.  In an engine compartment scenario, where
HFO-1234yf was released onto the hot surface, the highest HF breath
level concentration measured in the vehicle cabin (with the cabin air
intake fan running at maximum) was also well below the AEGL-2 limit.
This is a conservative estimate given that the data were based on
extreme engine operation (700ºC) with the hood seals removed allowing
for a direct path to the cabin air intake and full suction from the
engine.  The presenter also noted that individuals can perceive
irritation from HF at concentrations of 3 to 5 ppm, and that those
exposed would quickly move to avoid further exposure.

EPA staff asked if the analysis considered exposure to first responders.
 Dr. Lewandowski explained that the fault tree analysis assumed that
professional first responders would be using appropriate protection. 
However, the FTA did include scenarios for exposure to a “good
Samaritan,” such as a member of the general public or a volunteer
first responder, who might not have appropriate protective equipment. In
these scenarios, the fault tree analysis determined limited exposure
risks.  

Input Service Scenarios 

New CFD modeling applied to service-related refrigerant releases
suggests minimal ignition hazard. Service pits were not considered in
the FTA because they are assumed to be covered by existing OSHA
regulations on confined spaces that require ventilation for service
pits.  SAE assumes that EPA will require that only certified
professional technicians conduct AC service, which SAE standards will
also require.  Upon further clarification from the presenters, it was
determined that a “certified professional technician” is not written
intending to exclude certified non-professional technicians.  However,
it would require technicians, whether professional or not, to receive
certification before conducting AC service.

One automotive representative asked about the possible timing for a
follow-up rule for service technicians under §609 of the Clean Air Act.
 EPA staff stated that they could not answer that question.

Closing Remarks: 

The FTA considered many scenarios for exposure to HFO-1234yf.  In the
analysis, it was found that:

Risk Comparison to Other Vehicle Events

Risk Scenario	Risk per vehicle per operating hour	Citation

Risk of being in a police reported vehicle collision	= 5 x 10-5	NHTSA,
2007

Risk of vehicle collision due to vehicle brake failure	= 3 x 10-7	New
York State DMV

Risk of highway vehicle fire (any cause)	= 1 x 10-7	Ahrens, 2008; BTS,
2004

Risk of an airbag-related fatality associated with a vehicle collision
=2 x 10-10	NHTSA, 2001

Risk of vehicle occupant/former occupant experiencing HF exposure above
health based limits associated with an HFO-1234yf ignition event	= 5 x
10-12	CRP, 2009

Risk of vehicle occupant being exposed to an open flame due to
HFO-1234yf ignition 	= 9 x 10-14	CRP, 2009

Risk Comparison to Other Workplace and Service Events

Risk Scenario	Risk per 

working hour	Citation

Non-fatal recordable injury at work (all occupations)	=2 x 10-5	NSC,
2004

Risk among automotive repair tecnicians of being struck by object
resulting in lost work days	=2 x 10-6	BLS, 2007

Risk of service technician exposure to an open flame due to HFO-1234yf
ignition 	= 1 x 10-20	CRP, 2009

Risk of recordable incident involving repair technicians being struck
with high pressure HFO-1234yf equipment hose 	= 3 x 10-8	CRP, 2009



Based on these results, the presenter stated that use of HFO-1234yf in
MAC systems poses an extremely low level of risk for vehicle operators
and service workers.  In response to questions, Dr. Lewandowski
clarified that Phase III of the CRP study explicitly compares HFO-1234yf
to R-744 (CO2), but does not directly compare HFO-1234yf’s risks to
those for R-134a

Atmospheric Impacts 

Dr. Tim Wallington, Ford Motor Company

The Ford Motor Company conducted a review of findings by scientific
studies and articles that evaluated atmospheric impacts caused by
HFO-1234yf. Section 3 summarizes the relevant findings and discrepancies
between the Company’s review and the Significant New Alternatives
Program’s (SNAP) evaluation.  

Vehicle makers support the following SNAP findings:

 “HFO-1234yf has an ozone-depletion potential (ODP) of nearly zero,”


“The global warming potential (GWP) of HFO-1234yf is 4…” and

“TFA is produced from atmospheric oxidation of HFO-1234yf.”

TFA Accumulation

 

Vehicle makers also agree with the SNAP finding that the “…projected
maximum TFA concentration in rainwater should not result in a
significant risk of aquatic toxicity.” However, they disagree with a
statement in a report by ICF International (2009) that TFA is diluted in
fresh water bodies with the exception of “vernal pools and similar
seasonal water bodies that have no significant outflow capacity.”
Vehicle makers believe there is limited TFA accumulation in seasonal
water bodies based on Boutonnet et al. (1999).  Vehicle makers also
stated that no impacts were observed on vernal pool ecosystems in an
experimental study conducted by Benesch et al. (2002).

HFO-1234yf Emission Rate Estimates

Regarding the appropriateness of analyzing environmental impacts of
HFO-1234yf based on the current emission rate for HFC-134a, vehicle
makers recommend that EPA instead use 50 g per vehicle per total
lifecycle emission rate of HFO-1234yf in future environmental analyses.
This value was based on a 2009 study by Papasavva et al. 

VOC Designation

In assessing the classification of HFO-1234yf as a volatile organic
compound (VOC), the vehicle makers suggest that ethane is more similar
to HFO-1234yf than ethylene based on their photochemical ozone creation
potential (POCP) values and maximum incremental reactivity. Vehicle
makers state that ethane is not classified as a VOC and HFO-1234yf has a
reactivity comparable to or less than that of ethane and therefore will
not contribute to a >1-4% increase in ground level ozone.  Vehicle
makers cited Wallington et al. (2009), Derwent et al. (1998) and Carter
(2009) in this assertion.

EPA staff asked if all documents cited in the presentation were
available.  Dr. Wallington clarified that all the documents are in the
public domain.  He said that Wallington et al., 2009 has recently been
submitted for publication and is available upon request.  EPA staff
indicated that they would like to see this paper if it is accepted for
publication by a peer-reviewed journal. 

Closing Remarks: 

The vehicle makers’ opinion is that there are sufficient environmental
data now available for HFO-1234yf SNAP approval and the chemical’s
exemption from VOC restrictions.

Vehicle Makers’ Suggested Rule Modifications 

Jim Baker, MAC JAB Consulting

Jim Baker stated that strict interpretation of the proposed rule
effectively prevents the use of HFO-1234yf. As proposed, the rule
mitigates risk to zero and vehicle makers believe the proposed under
hood concentration limits are unachievable.  

Summarizing results from the CRP, Jim Baker stated that when the vehicle
is off, HFO-1234yf concentrations do not reach levels of concern and
when the vehicle is on, the non-collision under hood risk is the same as
the risk associated with R-134a because of HF generation.  The exposure
frequency of the HFO-1234yf direct expansion system is also twenty times
lower than that of the R-152a secondary loop, which has been approved by
SNAP. However, HFO-1234yf would not be suitable for a secondary loop
because the loss of efficiency would be prohibitive for this
refrigerant. Exposure frequency of HFO-1234yf is less even if exposure
concentrations are not limited and HF generation is considered. Thus,
exposure to the refrigerant poses a near zero risk. Based on the CRP
risk assessment, no concentration limits are required. 

The presenter stated that the proposed rule, as written, creates a
mitigation dilemma. For example, ventilating the cabin or installing
refrigerant discharge devices is required to comply with interior
concentrations limits.  Cabin ventilation could draw in additional
hazardous fumes from non-cabin fires, etc., during and after collision
events in every collision with ventilation—resulting in higher risk
compared to if they do not add ventilation. Currently, not all
collisions release refrigerant. However, with the installation of
directed refrigerant discharge, HFO-1234yf would be released by every
collision—again, resulting in higher risk. 

Passenger Compartment

Proposed Rule: Engineering strategies and/or devices must be
incorporated into the system such that leaks into the free space of the
passenger compartment do not result in HFO–1234yf concentrations of
6.2% v/v or above in any part of the free space inside the passenger
compartment for more than 15 seconds.

Comment: Unnecessary based on near zero risks identified in the SAE
Cooperative Research Program’s FTA risk assessment.

Proposal: Vehicle makers shall be responsible for conducting risk
assessments, and implementing appropriate risk mitigation as necessary,
to reduce risk of injury and/or property damage to an acceptable level.
Such risk assessments shall be based upon SAE J1739 (FMEA), or its
equivalent, and kept on file. Risk assessments shall cover all aspects
of vehicle use, including normal use, aging, service, and vehicle
collision.

One automotive representative clarified that “risk to an acceptable
level” can be informally interpreted as “okay to put onto the
market.” Mr. Baker suggested that EPA could consider that an
acceptable level of risk would be equivalent to the level of risk for
HFC-152a.  An acceptable risk can exist when exceeding the LFL for
HFO-1234yf when also considering additional factors.  Mr. Baker noted
that HFO-1234yf has unique flammability characteristics, since it
requires a two-stage ignition.  He stated that it takes approximately
5000 times more energy to ignite HFO-1234yf than for hydrocarbons or
gasoline. The cost to recharge after every collision if using a
refrigerant discharge device, e.g., squib valve, would also need to be
considered, even though this cost is not the main barrier. 

Although further alternatives to the proposal were not presented,
automotive representatives suggested using wording that was similar to
wording used in the third use condition for hybrids—specific, but also
flexible. This would also be relevant to the following proposed use
conditions.

Engine Compartment and Vehicle Electric Power Source

Proposed Rule: Engineering strategies and/or devices must be
incorporated into the system such that leaks into the engine compartment
or vehicle electric power source storage areas do not result in
HFO–1234yf concentrations of 6.2% v/v or above for any period of time.

Comment: Unnecessary based on near zero risks identified in the SAE
Cooperative Research Program’s FTA risk assessment. This requirement
is not possible to meet and therefore prevents the use of HFO-1234yf.

Proposal: Vehicle makers shall be responsible for conducting risk
assessments, and implementing appropriate risk mitigation as necessary,
to reduce risk of injury and/or property damage to an acceptable level.
Such risk assessments shall be based upon SAE J1739 (FMEA), or its
equivalent, and kept on file. Risk assessments shall cover all aspects
of vehicle use, including normal use, aging, service, and vehicle
collision.

In response to a question from EPA staff, Mr. Baker also stated that
that engineering controls such as flame retardant on the interior of the
vehicle are not in consideration at the moment, since risks of fire
caused from the refrigerant are nominal.

Exhaust Manifold Surfaces and Hybrid/Electric Vehicle Electric Power
Sources

Proposed Rule: HFO–1234yf MVAC systems must incorporate protective
devices, isolation and/or ventilation techniques in areas where
processes, procedures or upset conditions such as leaks have the
potential to generate HFO–1234yf concentrations at or above 6.2% v/v
in proximity to exhaust manifold surfaces and hybrid/electric vehicle
electric power sources.

Comment: Unnecessary based on near zero risks identified in the SAE
Cooperative Research Program’s FTA risk assessment. This requirement
is not possible to meet and therefore prevents the use of HFO-1234yf.

Proposal: Vehicle makers shall be responsible for conducting risk
assessments, and implementing appropriate risk mitigation as necessary,
to reduce risk of injury and/or property damage to an acceptable level.
Such risk assessments shall be based upon SAE J1739 (FMEA), or its
equivalent, and kept on file. Risk assessments shall cover all aspects
of vehicle use, including normal use, aging, service, and vehicle
collision.

Closing Remarks: 

In a response to a statement by one automotive representative about the
risks of gasoline, EPA staff clarified that the SNAP Program’s
statutory authority refers to the overall risks to human health and the
environment when comparing an alternative to ozone depleting substances
to the other available alternatives.

Vehicle makers agree with the proposed requirements for unique fittings,
labeling, high-pressure compressor cutoff switch, failure mode and
effect analysis, as well as the proposed requirement that the
refrigerant be used in new MAC systems only. Vehicle makers would like
to be able to choose HFO-1234yf as an option as soon as possible and
believe that the removal of HFO-1234yf concentration limits from the
rule is necessary for implementation of HFO-1234yf in vehicles.
Additionally, the automotive representatives would like the SAE
standards and CRP risk assessment to be considered in the final rule
making. However, both are not yet published and/or contain CBI
information.  EPA staff responded that they would check with EPA’s
Office of General Counsel for advice.

The acute exposure guideline (AEGL)-2 is the airborne concentration of a
substance above which it is predicted that the general population,
including susceptible individuals, could experience irreversible or
other serious, long-lasting adverse health effects or an impaired
ability to escape. (EPA, 2008)

HFO-1234yf Meeting Summary

Final

December 8, 2009

Page   PAGE  10 

