The move to a low-global-warming (LGW) refrigerant—replacing R-134a with R-1234yf—has come to a fork in the road that leads to an unexpected new choice. Although they aren't making a commitment, a number of automakers are looking hard at the possible advantages of taking the path which until 2011 they didn't even know existed. At the same time, some of their competitors have backtracked to another path they previously had chosen to bypass.
This is happening at a time when it appears a second R-1234yf supplier is emerging, to compete with the DuPont-Honeywell joint venture. Some reduction in price could be on the horizon.
The forthcoming competition was promised by Arkema, one of the world's largest producers of refrigerants. It recently announced it has begun construction of R-1234yf plants, first in Asia, then in Europe. Honeywell holds patents on R-1234yf, which are being challenged in courts in the U.S. and Europe. An Arkema executive at the recent SAE Thermal Management Systems Symposium in Troy, MI, told SAE Magazines that the company received strong indications from U.S. and European regulators that it soon would receive fair and reasonable access to those patents.
Arkema says its projected production would help satisfy what it anticipates will be worldwide demand for LGW refrigerant. The European Union (EU) has a 2017 date for phase-out of R-134a on all new vehicles, also the year the U.S. EPA expects Corporate Average Fuel Economy (CAFE) credits to spur changeover in the U.S.
Low-cost flame arrestors possible
With a recent SAE Cooperative Research Program (CRP) affirming the safety of the mildly flammable R-1234yf, interest in mitigation strategies would seem likely to fade. However, DuPont, while emphatically stating there is no need for additional mitigation to use R-1234yf safely, showed how “belt and suspenders” mitigation could be used at minimum cost by any OEMs with specific vehicle designs they feel could use it.
The approach, explained by Mary E. Koban, DuPont's technology lead for R-1234yf, consists of judicious placement of flame arrestors for exhaust manifolds and turbo housings. They could be as sophisticated as ceramic coatings or as common as steel shields and perhaps even fracture-resistant stainless steel condenser lines. There are even simpler arrestors, she added, such as kitchen staples such as steel wool and stainless wire mesh pads.
In a DuPont lab, the company tested both pad materials for arrestor effectiveness on an insulated BAM ignition chamber, with a sprayed-in mixture of R-1234yf and 1% PAG oil (polyalkylene glycol), a test designed to force ignition. The flame duration from the forced ignition was under one second with or without an arrestor placed at the chamber entrance. But with an arrestor, the flame did not propagate outside the pocket in the chamber. The arrestor, she said, can be positioned so an air gap is maintained for insulation of an exhaust manifold.
R-445A blend development continues
R-1234yf safety data notwithstanding, work continues on the newest LGW possibility, R-445A, the three-part refrigerant blend proposed in 2011 by Mexichem. R-445A is being evaluated by another SAE CRP, which has 10 automakers and four compressor suppliers.
The blend—85% R-1234ze (in the same hydrofluoro-olefin family as R-1234yf), 6% R-744 (CO2 as a refrigerant), and 9% R-134a—has a global warming number under the 150 limit set by the EU. Although the primary ingredient has similarly mild flammability, in combination with its other ingredients it is lower overall than R-1234yf. Both R-1234yf and R-445A carry the ultralow ASHRAE (American Society of Heating, Refrigeration and Air-Conditioning Engineers) designation of A2L.
All three R-445A components are in production, and estimated cost of materials may save an automaker about $25 for a 1.0-lb (0.45-kg) per car charge compared with R-1234yf, based on recently quoted prices for the three ingredients and reported prices for R-1234yf. System components are available. The refrigerant has similar performance to R-134a at high loads, somewhat less at low loads, so further system performance development obviously is needed. However, according to a CRP statement, three OEMs tested R-445A in systems with an internal heat exchanger and reported results comparable to R-134a. Further, tests also indicate R-445A may perform better than the other refrigerants in a heat pump (reverse) cycle.
An obvious issue is the impact of R-744, a higher-pressure gas, in a blend with lower-pressure ingredients. A leakage-caused drop from 6% to 2% would create a measurable performance loss. There reportedly are R-744 leakage concerns that remain to be solved, particularly at internal system temperatures above 75°C (167°F), as a seven-year life for an A/C refrigerant charge is expected.
And how should the refrigerant charge be installed, both on the assembly line and in a service garage, when one ingredient is so very different?
The charging equipment being proposed for OE and service-shop use is fully automated. Although each machine would work differently, both would use refrigerant identifiers to confirm accurate ratios. The system is based on test data that shows the two lower-pressure ingredients (R-1234ze and R-134a) form a very stable binary blend.
For factory fill, the binary blend and R-744 are injected separately into a holding tank with capacity for a full shift, the mixture agitated and ratios checked with the identifier, adjusted if necessary, then dispensed into a vehicle A/C system. The process is repeated for each car.
The R-445A system could replace an R-1234yf charging station on an assembly line during a weekend shutdown, so vehicle production would not be interrupted, explained Mike Hope, Group Leader, Climate Control Systems at Jaguar Land Rover.
For service, R-445A remaining in a vehicle is recovered into an internal cylinder, composition is checked with the identifier, and, if necessary, R-744 is injected from a separate container to reconstitute the ternary blend to specification. A pump in the internal vessel mixes the reconstituted blend, which is checked with the identifier once more and, if correct, charged into the vehicle.
Leak detection is a problem, because in a static vehicle system the composition is different throughout. So there is no practical, assured way for a conventional electronic leak detector to be calibrated. An approach under consideration, already used for single compound refrigerants in some European factories and service shops, would be to remove all refrigerant from the system and fill with a leak tracer gas, such as 5% hydrogen/95% nitrogen. An SAE standard (J2970) provides an equivalency measurement of leaks detected with tracer gas vs. R-134a or R-1234yf, but to date no tracer gas detector has certified to it.
Backtracking to R-744 'pathway'
Daimler, which led objections to R-1234yf, is a member of the R-445A CRP but also is working with other members of the VDA (German automakers association). They are in a new effort to develop a competitive R-744 system, a program halted in 2010. Although estimated cost for an R-744 system has dropped in the past few years, it reportedly still will be more expensive.
R-744 is a high-pressure system with a very low (31°C/88°F) condensation temperature, which reduces efficiency in high ambient temperatures. Because no other automakers have joined this effort, VDA members are developing VDA standards, and details on progress have not been presented.
As a result, the SAE Interior Climate Control Standards Committee standards development for R-744 systems and equipment, halted in 2010, will not resume at this time.