MONTREAL PROTOCOL ON SUBSTANCES THAT DEPLETE THE OZONE LAYER-2

Notes:
Fluids given with a green background are fluids which were not mentioned in the XXVI/9 Task Force report.
* Indicates refrigerants pending official ASHRAE 34 approval, submitted June 2015.
** Indicates refrigerants pending official ASHRAE 34 approval, submitted January 2016.
DR-7 has changed nominal composition slightly from originally R-1234yf/32 (64/36) to R-1234yf/32 (65/35).

As in the previous Decision XXVI/9 Task Force report, the data sources for Tables 2-1 through 2-4 are as follows:

• “GWP (RTOC)” values are taken from the 2014 RTOC report (UNEP, 2014) where available (they are based on (WMO, 2014)); where not available the value is calculated based on values for pure fluids from the 2014 RTOC report (UNEP, 2014).
• “GWP (IPCC5)” values are taken from the IPCC AR5 report (IPCC, 2014) for pure fluids; for mixtures values are calculated based values for pure fluids from the IPCC AR5 report (IPCC, 2014).
• For Tables 2-1 and 2-2, refrigerant designations, safety classes and compositions are taken from the AHRI AREP program where available, and where not available from ASHRAE 34 public review (ASHRAE, 2015).
• All other data in Tables 2-1 through 2-4 are taken from the 2014 RTOC report
  (UNEP, 2014).

2.3 Refrigerant classification and standards

Refrigerants are classified by the refrigerant standard ISO 817 and ASHRAE 34 into 8 classes depending on toxicity and flammability, for instance: A1, A2L, A3 or B2L. The first a letter
A or B indicates the toxicity of the fluid:

– A, lower chronic toxicity, have an occupational exposure limit of 400 ppm or greater

– B, higher chronic toxicity, have an occupational exposure limit of less than 400 ppm

The suffix 1, 2L, 2 or 3 indicates the flammability:

– 1, no flame propagation, measured at 60 °C

– 2L, lower flammability, burning velocity not higher than 10cm/s, energy of combustion below 19 MJ/kg and not flammable below 3.5 % volume concentration.

– 2, flammable, energy of combustion below 19 MJ/kg and not flammable below 3.5 % volume concentration.

– 3, higher flammability.

These safety classes are used by the system safety standards, such as ISO 5149, IEC 60335-2- 24, IEC 60335-2-40, IEC 60335-2-89, EN378 and ASHRAE 15.

With the introduction and potentially wide use of refrigerants that are flammable, have higher toxicity and/or operate at notably higher pressures than the conventional ODS refrigerants or alternative non-flammable HFC refrigerants, consideration of safety matters has become more important. Accordingly, more attention is presently being paid to the requirements of safety standards and regulations that directly relate to refrigerants that exhibit these characteristics.

For instance, the safety standards sets upper limits on how much refrigerant charge is allowed in a refrigerant circuit, primarily depending on the safety class, location of the equipment, and on the type of people who have access to the equipment; the amount of charge is related to the cooling or heating capacity of the equipment. Using a wall mounted split A/C unit in a 30m2 room as an example, the safety standard, in this case IEC 6335-2-40, allows 413 g of HC-290 per refrigeration circuit, while for HFC-32 it allows 5.6 kg charge, due to the different flammability characteristics of the substances. Clearly a more than 10 times higher charge allows higher cooling capacity with HFC-32, and requires higher level of optimising for the low charge when using HC-290.

2.4 Likelihood of new molecules and new radically different blends

There are alternative refrigerants available today with negligible ODP and (lower or) low GWP, but for some applications it can be challenging to reach the same lifetime cost level of the systems while keeping the same performance. The search for new alternative fluids may yield more economical system designs, but as will be explained below, the prospects of discovering new, radically different fluids are minimal.

The alternative refrigerants must have suitable thermodynamic properties, which determine the efficiency and capacity of the system. In addition, they need to satisfy several other criteria, such as zero ODP, low GWP, low toxicity, stability in the system, materials compatibility, acceptable cost, and, if possible, non-flammability, low flammability or lowrisk due to flammability. These requirements are difficult to balance.

The list of proposed R-410A and HCFC-22 replacement candidates includes singlecomponent refrigerants (HFC-32, HC-290, HC-1270, HFC-161, R-717, R-744). The list also includes blends, which, in addition to the listed single-component candidates, comprise the unsaturated HFC’s such as HFC-1234yf and HFC-1234ze(E), along with traditional HFC refrigerants to achieve the desired attributes of the blend, e.g., low GWP, low flammability, or lubricant compatibility. Through ongoing evaluation studies, the performance potentials of these alternatives are being established.

Significant efforts have been done in the past to find new fluids. A recent study (McLinden, 2015) started with a database of over 150 million chemicals, screening the more than 56,000 small molecules and finding none of them ideal. It can be concluded from the study that the prospects of discovering new refrigerants that would offer better performance than the fluids currently known are minimal.

2.5 Road to availability of alternative refrigerants

As discussed in the Task Force Decision XXVI/9 Report (UNEP, 2015), developing a new fluid is a process where uncertainties are addressed, both regarding what is technically feasible and regarding what can be accepted by the market. It is a process structured in discrete steps, where some are visible to the industry. The commercialisation of a newmolecule is complicated and can take significant time, while for mixtures consisting of existing molecules the commercialisation is much faster. Once the fluid is launched in the market, the availability is largely controlled by where there is a market need.

The technical uncertainty includes how to produce the fluid, and whether the preferred properties can be attained. The market uncertainty includes uncertainty about what properties the customer prefers, and what fluids the competitors will market.

The development process requires a series of investments, such as researching the toxicity of candidate fluids, or doing field tests at potential customers with a candidate fluid. The investment pattern is similar from fluid to fluid, and companies therefore manage the process with a state-gate process (Cooper, 1988). The state-gate process is a process, where a “gate” is placed just in-front of each major investment, and a “gate” is simply a decision point where management evaluates whether or not to accept the next investment or stop the development project. While the exact gates are not visible from outside the company, some of the steps will be visible in the market. Examples of such steps could be:

– Research, possibly in collaboration with a few selected system builders;

– Fluid released for small scale testing in industry test programs (with a research acronym);

– R-number applied for through ASHRAE 34 (or ISO 817) and is normally accepted;

– Testing in the market to see whether the market is interested in larger capacities;

– Broad market launch (large scale production set-up);

– Market adoption, where the market actually starts using the refrigerant in larger quantities.

Within the context of development of low GWP refrigerants, one of the most important incentives is the occurrence of relevant legislation that hinders competing fluids or opens pathways for new fluids and creates some measure of certainty for investments into the market.

The investment sizes and time needed for each step for new molecules (pure refrigerants) are much larger and longer than for refrigerant mixtures. Especially the research and toxicity evaluation are expensive in the early phases, and the production set in the later stages, are expensive for new molecules. While for new mixtures, the major uncertainty is related to the market, and the large investments are primarily on research, especially market research, to find a composition which matches the needs of the customers as well as possible, and on the market launch with investments in marketing.

This means that the commercialisation of a new fluid can take 10 years, while for mixtures the commercialisation takes closer to 5 years. A issue for the new mixtures is that many contain one of the two new molecules, HFC-1234yf or HFC-1234ze(E), which may have had only limited production until recently.

Once the fluid is launched in the market, companies will invest in sales where they see the greatest market potential. There is a limit to how many different refrigerants customers, service companies, and sales channels in a given market will accept, and market shares obtained in the early phase tends to be relatively easy to sustain, why companies can be very picky about where they launch a product. Although current availability to the markets and market launch plans for specific fluids are proprietary information, there is however the general rule of thumb that new fluids will be available where a sufficiently large share of users request it.

Two examples of the step from commercial productions to market launch in specific markets are as follows:

• Commercial production of HFC-1234ze(E) started at the end 2014 in manufacturing plants in the US. It is now already commercially used in chillers by companies in the US, EU and Japan; besides this, it is also applied in one-component foam applications. Within two years after the start of commercial production, it is currently commercially available in the US, Europe and most of Asia.
• Commercial production of HCFC-1233zd(E) started by mid-year of 2014 in plants located in the USA. It is used in low pressure centrifugal chillers, which have been
  released in Europe, the Middle East and other 50 Hz markets; besides that, it is also used in foam applications as a replacement for HFC-245fa. Within two years after the start of commercial production, it is currently commercially available in the US, Europe and most of Asia.

2.6 Energy efficiency in relation to refrigerants

Assessing the energy efficiency associated with a refrigerant is a complicated process, and the results depend on the approach taken. Energy efficiency of refrigeration systems is in addition to the refrigerant choice related to system configuration, component efficiencies, operating conditions, operating profile, system capacity, and system hardware, among others, which makes a consistent comparison difficult in many instances.

One approach is to start with a target refrigerant and use a system architecture suitable for this specific refrigerant, while comparing it with a reference system for the refrigerant to be replaced.

Another approach is to screen for alternative refrigerants suitable for a given system architecture. The common methods for determining the efficiency in this case can be placed into one of three categories:

– theoretical and semi-theoretical cycle simulations

– detailed equipment simulation models, and

– laboratory tests of the equipment.

In a refrigerant selection process, great reliance is placed on cycle simulations for selecting best candidate fluids for further examination either by equipment simulation models or tests of actual equipment.

Most often, cycle simulations employ a refrigerant’s thermodynamic properties along with fixed values for temperatures inside the system and fixed compressor isentropic efficiency.

These models are popular among refrigeration practitioners because they are simple in principle and easy to use. However, the shortcomings to be kept in mind includes not taking into account the heat transfer properties and pressure drops in a system. Detailed simulation models do not have this shortcoming (Domanski, 2006).

Laboratory tests provide the ‘most trusted’ information about performance of a refrigerant in a given system. It must be recognized that tests of a new refrigerant in a system optimized for a different refrigerant do not demonstrate the performance potential of the refrigerant tested (Abdelaziz, 2015). In addition to system ‘soft-optimization’, which includes adjustment of the refrigerant charge and expansion device, ‘hard optimization’ is necessary, which includes, among others, optimization of the compressor (including the size), refrigerant circuitry in the evaporator and condenser, and the overall system balance.

Hard optimization is a rather involved process. Usually, it is most effectively implemented by concurrent detailed simulations and extensive testing. It can be particularly complicated with blends of significant temperature glide, which offer special challenges in heat exchanger design. Hence, overall system design and successful optimization play a significant role in achieving the refrigerant performance potential in a commercialized product. In practice the hard optimization is also limited by the cost of the system, as the success in the market depends on a cost/performance trade-off. In addition, it is also constrained by commercial availability (e.g., manufacturing ability) for certain components, such as availability of preferred compressor displacement, heat exchanger dimensioning and capability to produce
preferred circuitry.

To illustrate the difficulties of assessing the energy efficiency associated with a refrigerant, consider the tests under high ambient temperature conditions described in Chapter 3 of this report:

• Testing temperatures differs from test program to test program.
• Obviously no single temperature can accurately match a real geographical location, so the results do not relate directly to the actual energy consumption in a real situation.
• The units (including technologies) used for testing varied within the same test programs.
• In some tests only the refrigerant is changed, in others the oil or even the compressor.
• Differences in test protocols further contributed to differences in results, for example,adjusting the expansion device, adjusting the charge, or adjusting compressor displacement to match compressor and heat exchanger capacity.
• The cost/performance ratio is an important factor (see above) but it is difficult to analyse in the test programs as other long term parameters need to be considered.

2.7 Climate impact related to refrigerants2

There are a number of difficulties in assessing the climate impact including the difficulties of obtaining reliable and accurate data on system leakage rates and determining the carbon 2 This section includes substantial contributions from J. Steven Brown, Ph.D., P.E., of The Catholic University of America, Washington, D.C., USA emissions generated, now and in the future, and in producing the energy necessary to power the RAC&HP system.

Climate impact related to refrigerants consists of direct and indirect contributions. The direct contribution is a function of a refrigerant’s GWP, charge amount and leakage rates (annual, catastrophic, and during servicing and decommissioning) from the air-conditioning and refrigeration (RAC&HP) equipment. The indirect contribution accounts for the kg CO2- equivalent emissions generated during the production of the energy consumed by the RAC&HP equipment, its operating characteristics and the emissions factor of the local electricity production. The relative importance of the direct and indirect contributions will depend on the type of system. Systems that are “more leaky”, e.g., automotive vehicle air conditioning, typically have larger relative contributions from direct warming than would “tighter systems”, e.g., hermetically sealed chiller systems, although this can be offset for systems that have much shorter operating periods or where power is supplied from a source with low carbon content.

There are several metrics that measures the total emissions from a system. Most common are Total Equivalent Warming Impact (TEWI) and Life Cycle Climate Performance (LCCP) which attempts to quantify the total global warming impact by evaluating the RAC&HP system during its lifetime from “cradle to grave” (IIR, 2016). Sometimes, a TEWI calculation may be simplified by neglecting broader effects including manufacture of the refrigerant and equipment, and disposal of the refrigerant and equipment after decommissioning. More indepth analyses not usually performed also look at the emissions associated with the production and disposal of the equipment, e.g., including the mining and recycling of the metal used to manufacture compressors, heat exchangers, and other components.

To summarize, the most important factors determining the climate impact are:

• The GWP of the refrigerant multiplied with the amount leaking from the system, this is the direct contribution.
• Energy consumption of the system multiplied with the amount of CO2 generated per unit of energy, this is part of the indirect contribution.

The uncertainty on energy consumption and leakage makes determining the total climate impact difficult.

2.8 The GWP classification issue

To minimize direct climate impact a lower GWP refrigerant can be used. The RTOC 2014 Assessment Report included a taxonomy of GWP values, including what constitutes high, medium, and low GWP (again given in Table 2-5 below). This taxonomy is based on fixed GWP values.

Table 2-5 defines “low” as smaller than 300 and “high” as more than 1000. There are sources that define low as lower than 25, as lower than 100, or as lower than 150 (which results from the 2006 EU MAC directive). It will be clear that “high”, “medium” and “low” are qualifiers, related to a scale, and that a number definition of these levels would be a non-technical choice. This also because it is somehow related to what is acceptable in specific applications.

Table 2-5: Classification of 100 year GWP levels

For instance, there is a relationship between the pressure, the GWP and the flammability of a refrigerant, as illustrated in Table 2-6, which is also from the RTOC 2014 Assessment Report (UNEP, 2014). The trend can be described as “the higher the pressure, the higher the minimum GWP which is needed for fluids to be non-lammable”. The exception to this relationship is R-717 and R-744, which do not fit this pattern. Therefore, what could be an“acceptable” GWP for a high pressure fluid replacing R-410A may not be “acceptable” for a low pressure fluid replacing HFC-134a. But this does not relate to a definition of a GWP classification in an absolute sense.

3 Suitability of Alternatives under High Ambient
Temperature (HAT) Conditions

This chapter updates information on the results of research projects for testing alternative refrigerants at HAT conditions and on further considerations with respect to designing products using alternatives in new and retrofit applications. For a comprehensive understanding of high ambient and the implications on refrigerant selection, please refer to chapter 7 of (UNEP, 2015), which remains relevant.

3.1 HAT considerations

HAT conditions are an important issue for the design of refrigeration and AC systems. While 35 °C has been designated as comparison condition for performance of standard ambient, there is no definition currently for what constitutes a High Ambient Temperature and consequently a high ambient temperature country (or region). Ambient temperatures are also used in cooling load calculation and building envelope design.

A high ambient temperature can be defined as the incidence over a number of hours per year of a certain temperature. If this temperature is set above the standard ambient of 35 °C, the question becomes at what incidence this occurrence will be considered to constitute a high ambient condition.

Chapter 7 of the XXVI/9 report (UNEP, 2015) lists several methods to define HAT conditions using weather profiles, cooling degree days, bin weather data, or the occurrence of a certain temperatures above the globally accepted standard temperature. One of the most used is to define values of ambient dry bulb, dew point, wet bulb temperature, and wind speed corresponding to the various annual percentiles of that are exceeded on average by 0.4%, 1%, 2%, and 5% of the total number of hours in a year equivalent to 8,760 hours (ASHRAE, 2013). These values correspond to 35, 88, 175, and 438 hours per year respectively, for the period of record. By defining the value of the high ambient temperature and using information about the percentiles, the appropriate design conditions for refrigeration and air conditioning equipment can be adopted. There are important references that present the information about temperatures profiles and percentiles of incidence in the several regions of the planet.

As the ambient temperature increases, system load increases and capacity decreases. With increasing ambient temperatures, the condensing pressure and compressor discharge temperatures also increase, thus leading to possible reliability issues. ISO and EN (European Standards) prescribe pressures corresponding to certain design temperatures for the safe operation of a system. This information is required by design engineers to specify material and pipe wall thickness requirements in a system. Table 3-1 is taken from EN378-2:2008 and the equivalent ISO 5149 based on IEC 60721-2-1. The table does not specify what is classified as high ambient temperature.

Table 3-1: Specified Design Temperatures

While normally systems are designed for 35 °C (T1 in ISO 5151:2010) with appropriate performance (cf. for example, under standards requirements) up to 43 °C in some countries, the high ambient temperature condition requires a design at 46 °C (T3 in ISO 5151:2010) with appropriate operation up to 52 °C.

3.2 Testing at HAT conditions

Most of the research and development has traditionally been made at the “standard ambient” of 35°C dry bulb temperature; even lower temperatures are used for some tests (e.g., under AHRI Standard 210/240). The performance of units at different ambient temperatures would then be simulated or extrapolated. The status of the following projects testing refrigerants used in specific equipment operating under high ambient temperature conditions are discussed below:

• “Promoting low GWP Refrigerants for Air-Conditioning Sectors in High-Ambient Temperature Countries” (PRAHA) and “Egyptian Project for Refrigerant Alternatives” (EGYPRA);
• the Oak Ridge National Laboratory (ORNL) “High-Ambient-Temperature Evaluation Program for Low-Global Warming Potential (Low-GWP) Refrigerants”, Phase I (and ongoing Phase II); and
• the AHRI Low GWP Alternative Refrigerants Evaluation Program (AREP) Phase I (and ongoing Phase II).

3.2.1 PRAHA and EGYPRA projects

To shed light into what can be considered as sustainable technologies for high ambient temperature conditions. UNEP and UNIDO launched a project to study and compare refrigerants working in machines specifically built for those refrigerants and operating at high ambient temperatures. PRAHA was launched in 2013 and completed at the end 2015. The project is implemented at the regional level in consultation with National Ozone Units of Bahrain, Iraq, Kuwait, Qatar, Oman, Saudi Arabia, and the UAE to ensure incorporating the project outputs within the HCFC Phase-out Management Plans (HPMPs) particularly for the preparation of post 2015 policies and action-plans.

Building up on PRAHA and the linkage to country phase-out plans, Egypt adopted a similar initiative as part of the HPMP to test refrigerant alternatives for air-conditioning units built in Egypt. The initiative EGYPRA tests more blends in different applications. The initiative was launched back in June 2014 and is expected to have the results by the end of 2016.

Both projects built custom made units and testing was done at independent labs at 35, 46, and 50 °C ambient temperatures for PRAHA and an additional 27 °C for EGYPRA, with an “endurance” test at 55°C ambient to ensure continuous operation for two hours when units are run at that temperature. The proposed refrigerants are shown in table 32 below:

Table 3-2: Alternative refrigerants used in PRAHA and EGYPRA projects

The main finding of the PRAHA project is that some of the alternative refrigerants with higher relative volumetric capacity than HCFC-22 show better COP than was theoretically expected and that the R-410A units and their equivalents are more technologically advanced than the HCFC-22 unit because the development of HCFC-22 units has been stopped for some time. The test results from the rooftop packaged units with larger capacity than the other categories were better than those for smaller units possibly indicating that the capacity of the air conditioner affects the outcome hakroun, 2016). Results indicate a way forward in the search for efficient low-GWP alternatives for high ambient temperatures especially when coupled with a full system redesign.

The summary of results will be available when the final report will be published in April 2016. The outcome of the PRAHA project, other than the testing results, are:

• There is a need to do risk assessment studies at HAT conditions in countries that experience HAT conditions;
•  There is a need for a full product re-design taking into consideration the technical issues of heat exchanger optimisation, expansion device selection, charge optimisation, excessive pressure, temperature glide, flammability, oil, and energy efficiency issues;
• The economic impact is still to be considered when the availability and cost of components have been determined by market factors. Todays’ price on the market of
  components is not representative for the cost in the longer term, so alternative methods have to be used to analyse future cost;
• There is need for field testing of the units once a design and alternative refrigerants have been selected by the concerned OEMs

The work by PRAHA and EGYPRA will facilitate the technology transfer and the exchange of experience with low-GWP alternatives for air-conditioning applications operating in highambient temperature countries. The other indirect objective is to encourage the development of local/regional codes and standards that ease the introduction of alternatives needing special safety or handling considerations, and to ensure that national and regional energy efficiency programs are linked to the adoption of low-GWP long term alternatives (PRAHA, 2013).

3.2.2 ORNL project

The ORNL project aims to develop an understanding of the performance of low-GWP alternative refrigerants to HCFC and HFC refrigerants under HAT conditions in mini-split air conditioners under Phase I and in roof-top units under the ongoing Phase II.

Phase I: ORNL in cooperation with a panel of international experts designed a test matrix of 84 tests. ORNL and the panel selected the refrigerants based on their GWP, commercial availability and physical properties while considering whether information about the characteristics of the refrigerants is readily available. ORNL conducted tests using two “softoptimized” ductless mini-split air conditioners have a cooling capacity of 5.25 kWh (1.5 TR).

One unit is designed to operate with HCFC-22 refrigerant (2.78 coefficient of performance [COP], equivalent to a 9.5 energy efficiency ratio [EER]). The other is designed to use R- 410A refrigerant (3.37 COP, equivalent to an 11.5 EER).

Table 3-3: Alternative refrigerants used in ORNL Project

The ORNL/TM-1015/536 report has the following conclusion appearing as part of its Executive Summary (reproduced here without changes):

The test results from this evaluation program demonstrate that there are several viable alternatives to both R-22 and R-410A at high ambient temperatures. In some cases, there was a significant improvement in the performance of the alternatives over that of the baseline, in terms of both COP and cooling capacity. In other cases, the performance of the alternatives fell within 10% of the baseline, which suggests that parity with baseline performance would likely be possible through additional engineering design.

The R-22 alternative refrigerants showed promising results at high ambient temperatures: although both of the A1 alternative refrigerants lagged in performance, some of the A2L refrigerants showed capacity within 5% and efficiency within approximately 10% of the baseline system. The A3 refrigerant (R-290) exhibited higher efficiency consistently; however, it did not match the cooling capacity of the baseline system. The most promising A2L refrigerants exhibited slightly higher compressor discharge temperatures, while the A3 refrigerant exhibited lower compressor discharge temperatures.

The R-410A alternative refrigerants are all in the A2L safety category. Most of them showed significant potential as replacements. R-32 was the only refrigerant that
showed consistently better capacity and efficiency; however, it resulted in compressor discharge temperatures that were 12–21°C higher than those observed for the baseline refrigerant. These higher temperatures may negatively impact compressor reliability. DR-55 and HPR-2A had higher COPs than the baseline and matched the capacity of the baseline at both the hot and extreme test conditions. R- 447A and ARM-71a had lower cooling capacity than the baseline at all ambient conditions. The system efficiency of R-447A showed improvement over the baseline at high ambient temperatures; for ARM-71a, the efficiency was similar to the baseline at all test conditions.

The efficiency and capacity of the alternative refrigerants could be expected to improve through design modifications that manufacturers would conduct before introducing a new product to market. However, given that the scope of this study covered only soft-optimized testing, no detailed assessment can be made of the extent of potential improvements through design changes. Within the bounds of what is possible in optimizing the units for soft-optimized tests, the ORNL test plan included only minor optimizations, including refrigerant charge, capillary tube length, and lubricant change. Therefore, these are conservative results that probably could be improved through further optimization. Additional optimization, including heat transfer circuiting and proper compressor sizing and selection, would likely yield better performance results for all of the alternative refrigerants.

Losses in cooling capacity are typically easier to recover through engineering optimization than are losses in COP. The primary practical limit to improvements in
capacity is the physical size of the unit; but that is not expected to be a significant concern in this case, based on the magnitude of the capacity losses exhibited in this
evaluation program. Thus, the COP losses and the increases in compressor discharge temperature are particularly important results of this testing program, in that these
variables will be the primary focus of future optimization efforts.

This performance evaluation shows that viable replacements exist for both R-22 and R-410A at high ambient temperatures. Multiple alternatives for R-22 performed well.
Many R-410A alternatives matched or exceeded the performance of R-410A. These low-GWP alternative refrigerants may be considered as prime candidate refrigerants
for high ambient temperature applications. Before commercialization, engineering optimization carried out by manufacturers can address performance loss, the
increase in compressor discharge temperature that many alternatives exhibited (particularly the R-410A alternatives), and any safety concerns associated with flammable alternatives. (Abdelaziz, 2015)

Phase II: ORNL started the second phase of the program testing “Low GWP Refrigerants in High Ambient Temperature Countries” in February 2016, covering roof-top air conditioners in this phase. Results will be published in the second half of 2016.

3.2.3 AREP project

AREP, a project launched by the Air-Conditioning, Heating, and Refrigeration Institute (AHRI) is a cooperative research program to identify suitable alternatives to high GWP
refrigerants without prioritizing them.

In the first phase of the project, 21 companies evaluated 38 refrigerant candidates for replacing HCFC-22 and three HFCs, R-410A, HFC-134a, and R-404A (Amrane, 2013), in
applications varying from air conditioners and heat pumps (both package, split and variable refrigerant flow), chillers (screw and centrifugal), refrigeration (commercial and ice
machines), refrigerated transport, and bus air-conditioning. Phase II tested 17 refrigerants plus doing more tests at high ambient temperature conditions.

Table 3-4: AREP Phase II low-GWP High Ambient Testing

* L-41-2 at wet suction, no HAT
**HFC-32 with same charge and with optimized charge
*** HFC-32 with standard POE oil and with prototype POE oil

AREP concluded its first phase in 2013 and the second phase began in early 2014.

For high ambient testing, seven entities tested three residential split units, four rooftops, one chiller, and several compressors. The results were compared to baseline of R-410A units. Refrigerant candidates are shown in table 3-5 below (Schultz, 2016a). Refrigerants were tested at 115 °F (46.1 °C) and 125 °F (52.6 °C). The tests were either a drop-in or a soft optimization with a change in refrigerant charge and/or expansion device.

The conclusion from AREP-II (Schultz, 2016a) is that general trends in HAT performance are similar for all alternative refrigerants. Systems with alternatives generally provided similar to higher capacities than R-410A systems at HAT conditions, i.e., showing a smaller decrease in capacity as ambient temperatures increase.

AREP-II was conducted by several entities with different test protocols which contributed to differences in results. The different tests varied from drop-in to soft-optimized tests adjusting the expansion device for similar superheat, or adjusting the charge for a similar sub-cooling. The results shown below are extracted from some of the test reports of AREP-II that are available publically on-line and are as presented at a special ASHRAE session in January 2016. They do not represent a conclusion on behalf of AREP-II since there was none published.

A drop-in test for ARM-71a, R-454B (DR-5A), HPR2A, R-446A (L-41-1), and R-447A (L- 41-2) at 125 °F (51.6 °C) showed slightly less capacity for all systems compared to systems with R-410A. ARM-71a, DR-5A, and HPR2A charged systems resulted in 3-6% better efficiency. The discharge pressure at that condition was lower for all refrigerants compared to R-410A, while the discharge temperatures were equal or slightly higher than R-410A (Burns, 2016).

A test using a prototype oil for HFC-32 prevented breaks in the operation of the unit at 52 °C as it did when the original polyolester (POE) oil was used. HFC-32 showed higher
compressor (isentropic) efficiency and lower volumetric efficiency with POE oil than R-410A at rating conditions, but better efficiencies in both tests with prototype HFC-32 oil (Li, 2016).

In another test, a variable speed drive was used to equalize the capacity with the R-410A base unit capacity. DR-55 and DR-5A required no to little adjustment to the compressor displacement, while HFC-32 required a 8% reduction in compressor displacement to match heat exchanger capacity. All three refrigerants were slightly less sensitive to variations in ambient temperatures resulting in somewhat higher capacities and efficiencies at high ambient temperatures compared to R-410A (Schultz, 2016b).

AREP-II included a system drop-in test for a water chiller, a category not tested by the other projects, for ARM-71a, R-454B (DR-5A), HPR2A, R-446A (L-41-1), and R-447A (L-41-2), and HFC-32 at various temperatures between 30 °C and 46 °C. The test showed a similar degradation in relative efficiency for all refrigerants of about 30% between and 35 and 46 °C.

The degradation in relative cooling capacity on the other hand was less steep. HFC-32 showed an increase in discharge pressure at higher temperatures; however, using a receiver resulted in reducing that pressure (Zoughaib, 2016).

3.2.4 Common remarks on the three testing projects

A summary of the four projects is found in Table 3-5 below. The table outlines the types of equipment and alternative refrigerants tested, the conditions at which the tests were carried out, and the constraints on the prototype building or optimization process.

As mentioned in Chapter 2 above, assessing the R/AC equipment energy efficiency
associated with a refrigerant is a complicated process, and the results depend on the approach taken. Energy efficiency of R/AC systems is in addition to the refrigerant thermodynamic and transport properties related to system configuration, component efficiencies, operating conditions, operating profile, system capacity, and system hardware, among others, which makes a consistent comparison difficult in many instances. Laboratory tests provide the ‘most trusted’ information about performance of a refrigerant in a given system. It is recognized that tests of a new refrigerant in a system optimized for a different refrigerant do not demonstrate the performance potential of the refrigerant tested (Abdelaziz, 2015). In addition to system ‘soft-optimization’, which includes adjustment of the refrigerant charge and expansion device, ‘hard optimization’ is necessary, which is a rather involved process and includes, amongst others, optimization of the compressor (including the size), refrigerant circuitry in the evaporator and condenser, and the overall system balance. Hard optimization is usually most effectively implemented by concurrent detailed simulations and extensive testing. It can be particularly complicated with refrigerant blends characterized by a significant temperature glide, which offer special challenges in heat exchanger design. Hence, overall system designand successful optimization play a significant role in achieving the refrigerant performancepotential in a commercialized product. In practice, the hard optimization is also limited by the cost of the system, as the success in the market depends on a cost/performance trade-off. In addition, it is also constrained by commercial availability (e.g., manufacturing ability) for certain components, such as availabilit y of preferred compressor displacement, heat exchanger dimensioning and capability to produce preferred circuitry.

Table 3-5: Summary of the three testing projects

The tests under high ambient temperature conditions described above illustrate the difficulties of assessing the energy efficiency associated with a refrigerant, considering:

• Testing temperatures differs from test program to test program.
• Obviously no single temperature can accurately match a real geographical location, so the results do not relate directly to the actual energy consumption in a real situation.
• The units used for testing vary within the same test programs.
• In some tests only the refrigerant is changed, in others the oil is changed or even the compressor.
• Differences in test protocols further contributed to differences in results, for example: adjusting the expansion device for similar evaporator superheat, adjusting the charge for a similar sub-cooling, or adjusting compressor displacement to match compressor capacity to heat exchanger capacity.

The efficiency and capacity of the alternative refrigerants could be expected to improve through design modifications that manufacturers would conduct before introducing a new product to market. However, given that the scope of the research mostly covered softoptimized testing, no detailed assessment can be made of the extent of potential improvements through design changes (Abdelaziz, 2015). Soft optimization affected limited areas such as capillary tube length or expansion device changes, refrigerant charge, and the type of lubricant. While the PRAHA project included a change of compressors, suppliers had no time to properly design those compressors for the particular applications. Results could probably be improved through further optimization such as heat transfer circuiting and proper compressor sizing and selection; however, there is a particular need for a redesign of systems including new components.

Losses in cooling capacity are typically easier to recover through engineering optimization than are losses in COP. The primary practical limit to improvements in capacity and COP is the physical size of the unit. COP losses and the increases in compressor discharge temperature are particularly important in so far that these variables will be the primary focus of future optimization efforts. Before commercialization, engineering optimization carried out by manufacturers can address performance loss and the increase in compressor discharge temperature that many alternatives exhibited as well as safety concerns associated with flammable alternatives. The cost/performance ratio in the long term will be an important factor.

3.3 Further considerations

Chapter 7 of the September 2015 XXVI/9 Task Force Report (UNEP, 2015) discussed additional topics related to HAT conditions: a definition of options for HAT conditions, design considerations, research projects, energy efficiency and regulations related to energy efficiency, current/future alternative chemicals and technologies for air conditioning under HAT conditions, and considerations for refrigeration systems under HAT conditions including not-in-kind technologies (UNEP, 2015). That information is not repeated here, but in view of the initial results of the testing under HAT conditions discussed above, some of those considerations are further highlighted below.

On energy efficiency: In regions with HAT conditions, legislations which set minimum energy performance standard (MEPS) values on air conditioners are emerging quickly. Most of the countries require third party verification of declared performance. Higher minimum energy efficiencies are being announced on a regular basis, and this tendency may continue.

As examples, Bahrain recently announced MEPS values and regulation of labelling air conditioners and Saudi Arabia is moving closer to releasing their regulation for large air conditioners and chillers expected in the second quarter of 2016.

When selecting new refrigerants, it is important to consider further increases on the current minimum energy efficiency requirements. To the extent increases in MEPS are not met by current models, this offers the opportunity for manufacturers to implement new refrigerants while redesigning equipment for those new refrigerants.

On design and availability: The design for HAT conditions needs special care to avoid excessive condensing temperatures and getting close to the critical temperature for each type of refrigerant. Other issues such as safety, refrigerant charge quantity, and improving the energy efficiency for both partial and full load have to be taken into consideration In HAT conditions, the cooling load of a conditioned space can be up to three times that for moderate climates. Therefore larger capacity refrigeration systems may be needed which implies a larger refrigerant charge. Due to the requirements for charge limitation according to certain safety standards, the possible product portfolio suitable for high ambient conditions is more limited than for average climate conditions when using the same safety standards.

As concluded from the testing projects, special design of both components and products is needed for the new alternatives to meet the performance of systems in both capacity as well as efficiency requirements. While the commercialization process of refrigerants can take up to ten years, as seen from chapter 2, the commercialization of products using these alternatives will take further time.

As HCFC-22 air conditioning products get phased-out for some applications, the industry is turning to available technology using higher GWP refrigerants with higher discharge pressures like R-410A or comparable pressures like R-407C, depending on the application.

One exception is HFC-32 which has seen a limited release for room air conditioners following the change-over to HFC-32 in Japan. HC-290 products, which have potential due to the favourable performance of HC-290 compared to HCFC-22 at HAT conditions, are not yet commercially available in many countries although some of the local suppliers are busy researching and designing such products.

On retrofits: It is important to note that any change of refrigerant in an existing design requires careful considerations. Theoretical calculations can give an idea about what is generally to be expected with a change in refrigerant, but specific details on the system design are needed. Modifying the electrical connections to meet the requirements needed for flammable refrigerants is an additional cost that needs to be taken into consideration. This cost is the same for A3 or A2L refrigerants and mostly due to changing the location of the controls in order to reduce the risk of a spark.

For HAT conditions, the design and sizing of heat exchanger will impact how the system capacity and energy efficiency is influenced by a change in refrigerant. For system builders this means that each system design needs to be optimized for each type of refrigerant. This requires an investment similar to what has been spent on optimizing the system for the current refrigerant, and for highly cost optimized systems this investment might be considerable.

On safety: Standards for the new refrigerants (that are mostly flammable), like ISO 5149, EN 378, IEC 60335-2-40 for air conditioners and heat pump systems and IEC 60335-2-89 for some commercial refrigeration appliances, are available, although IEC 60335-2-89 needs to be adapted to allow larger charges of flammable refrigerants that are required for the bigger capacities of air conditioners working at HAT conditions. IEC standards are a de facto legal requirement in several countries as the Certification Body (CB) scheme is the actual requirement for import and sales of products. In some countries, the implementation of old standards in the legislation, for instance building codes or other mandatory safety regulations, blocks the uptake of especially flammable refrigerants.

Another important aspect of safety standards is that their value is tied to the degree of compliance, and this makes training of system builders and service technicians an important part of implementing safety standards. The cost of the above certification, including the thirdparty certification cost and including the training cost, should be considered.

Although risk assessment work on flammable refrigerants is an on-going research in some countries, there is a need for a comprehensive risk assessment for A2L & A3 alternatives at installation, servicing and decommissioning practices at HAT conditions and field testing of units.

4 BAU and MIT scenarios for Article 5 and non-Article 5
Parties for 1990-2050: Refrigeration and Air Conditioning

This chapter is organized as follows:

4.1 Expansion of scenarios

4.2 Method used for calculation

4.3 HFC consumption and production data

4.4 Non-Article 5 scenarios

4.5 Article 5 scenarios

4.6 Demand and benefit numbers

4.1 Expansion of scenarios

The previous Decision XXVI/9 paragraph 1 (c) asks to revise the scenarios: “Taking into account the uptake of various existing technologies, revise the scenarios for current and future demand elaborated in the October 2014 final report on additional information on alternatives to ozone-depleting substances of the Technology and Economic Assessment Panel’s task force on decision XXV/5, and improve information related to costs and benefits with regard to the criteria set out in paragraph 1 (a) of the present decision, including reference to progress identified under stage I and stage II of HCFC Phase-out Management Plans”.

The current Decision XXVII/4 requests to expand the scenarios to the period 1990-2050, twenty years after 2030 which was the last year in the scenarios used in the XXVI/9 Task Force report (UNEP, 2015).

The following scenarios have again been calculated, which apply to the R/AC sector only for this first report of the XXVII/4 Task Force submitted to OEWG-37:

a. A BAU scenario: In non-Article 5 countries this implies consideration of the F-gas regulation in the EU and regulations in the USA making certain HFCs unacceptable for certain sub-sectors by specific dates. This implies that, in the BAU calculation, certain high GWP substances in specific subsectors are replaced by low or lower GWP substances. In this way it responds to comments the XXVI/9 Task Force
already received at OEWG-36 and at MOP-27. The changes incorporated mainly apply to commercial refrigeration and, to a small degree, to stationary air conditioning. In Article 5 Parties, economic growth percentages expected for the period 2015-2050 are virtually the same as the ones in the XXVI/9 report.

b. An MIT-3 scenario: A 2020 completion of conversion in non-Article 5 Parties of all R/AC sub-sectors and the start of the manufacturing conversion of all R/AC subsectors in 2020 in Article 5 Parties, now with consequences for the period 2020-2050.

c. An MIT-4 scenario: This is the same as the MIT-3 scenario, but with the assumption of 2025 for the start of the manufacturing conversion for stationary AC in Article 5 Parties, now with consequences for the period 2020-2050.

d. An MIT-5 scenario: This is the same as the MIT-3 scenario, but with the assumption of a 2025 completion of conversion in non-Article 5 Parties of all R/AC sub-sectors and the start of the manufacturing conversion of all R/AC sub-sectors in 2025 in Article 5 Parties, now with consequences for the period 2020-2050

For Article 5 Parties, manufacturing conversion projects would need preparation to be funded; it would also take a certain period of time before conversion projects would have been approved by a funding authority, so that they can be initiated. Finally, experience with CFCs and HCFCs has shown that, the slower the conversion of manufacturing, the longer the servicing tail will be, i.e., the longer servicing of equipment will be required (see sections 4.5.3 and 4.5.5 taken from UNEP, 2015).

In this chapter the 1990-2050 scenarios will be given in the following sequence. First, the BAU scenario for non-Article 5 Parties will be dealt with, which will be analysed in tonnes and ktonnes CO2-eq. This is then followed by the MIT-3 and MIT-5 scenarios for non-Article 5 Parties; new manufacturing and servicing figures are given in ktonnes CO2-eq. (not in tonnes). As a next step, the Article 5 scenarios are given. Again, first the BAU scenario for

Article 5 Parties will be dealt with, which will be presented in tonnes and ktonnes CO2-eq. This is then followed by the MIT-3, MIT-4 and MIT-5 scenarios for Article 5 Parties; new manufacturing and servicing figures are given in tonnes and in ktonnes CO2-eq.

4.2 Method used for calculation

A “bottom-up” method has been used to predict the demand for R/AC equipment, as in the XXVI/9 Task Force report (UNEP, 2015). The RTOC 2010 Assessment Report (RTOC, 2010) describes the bottom-up method used here. A bottom up method derives the size of banks from information obtained from outside (accountancy reports on trade and exports, if possible, supplemented with a trend analysis). The banks serve to calculate emissions using agreed emission parameters. As a result, the demand (or “consumption”) can be calculated, which consists of (1) what is supplied to the existing banks (i.e., to compensate for leakage), and (2) what is added to the bank (i.e., in new equipment that has been charged) ), less (3) what is recovered and reused from the bank (i.e., material reclaimed from equipment decommissioned). In a spreadsheet analysis, this can be seen as one stream of refrigerant into a bank with equipment that has been manufactured over a number of years. In summary, the
refrigerant demand or the annual sales of new or virgin refrigerant are equal to the amount of refrigerant introduced into the refrigeration and AC sector in a country (or regions) in a given year. It includes all the chemicals used for charging or recharging equipment, whether the charging is carried out in the factory, in the field after installation, or whether it concerns recharging with the appropriate equipment during maintenance operations.

In this type of “bottom-up” approach, one therefore evaluates the consumption of a certain refrigerant based on the numbers of equipment in which the fluid is charged, e.g. refrigerators, stationary air-conditioning equipment, and so on. It requires the establishment of an inventory of the numbers of equipment charged with substances (which then forms the total inventory, or the “bank”), and the knowledge related to their average lifetime, their emission rates, recycling, disposal, and other parameters. The annual emissions are estimated as functions of all these parameters during the equipment lifetime.

Further information on sub-sectors, equations used and information on how the installed base is being considered can be found in the XXVI/9 Task Force report (UNEP, 2015).

As in the XXVI/9 report, the GWP for low GWP replacement refrigerants has been chosen as follows. In domestic refrigeration the use of isobutane is assumed with a very low GWP. In cases where the replacement refrigerant is known (ammonia, hydrocarbons) the very low GWP factors have been used. In case of commercial refrigeration, one can assume the use of carbon dioxide, pure low-GWP refrigerants or refrigerant blends in supermarkets, low GWP hydrocarbons in mass produced units, blends or carbon dioxide in condensing units (where an average GWP of 300 is used). For stationary AC as a whole, an average GWP of 300 has also been used (as an estimated average between very low GWP refrigerants and others, such as HFC-32 and various blends under investigation). The choice has been made on the basis of averaging and is not related to GWP considerations presented in Table 2-6. In MACs, replacement refrigerants are assumed to have negligible GWP.

Growth rates for equipment production have been slightly changed (in two cases) compared to the rates used in the XXVI/9 Task Force report, in order to give a more realistic approach in the period after 2030. In this report, the growth rates apply as given in Table 4-1 below.

A number of considerations substantially complicate the calculations. This includes the preference to apply certain alternatives in specific equipment (and often under certain conditions), combined with the fact that the R/AC banks -the amounts present in the equipment- need recharging (i.e., servicing) over the entire lifetime of the equipment. Details on lifetime and annual leakage are given in (Table 5-2 in) the XXVI/9 report (UNEP, 2015).

Table 4-1: Growth rates for high-GWP HFCs in the various R/AC manufacturing subsectors during the periods 2010-2020 and 2020-2050 (a negative growth rate may still imply a positive growth in the number of equipment, assumes that low-GWP alternatives are increasingly applied in a BAU scenario and also in MIT scenarios until the conversion starts in specific years)

The calculation method covers the period from 1990 until 2050, as requested in Decision XXVII/4. Table 4-1 gives the growth rates assumed for new manufacturing in the various R/AC sub-sectors. The growth rate assumed in the manufacturing sector is only one parameter in the scenario calculations. The total demand for a sub-sector is calculated using parameters such as equipment lifetime, equipment leakage, charge at new manufacturing etc.

This implies that, if the annual growth rate in a manufacturing sub-sector would be 1-5%, the annual growth rate of the total demand for that subsector can be several percent higher, e.g., varying from 3 to about 10%. These percentages can also be derived from the BAU Tables 4- 7 and 4-8 and from the more detailed tables in the Annex.

Depending on the application sector, uncertainties are different either because activity data include different uncertainties or because emission factors may vary significantly from one country to the other. The 2010 RTOC Assessment Report (RTOC, 2010) describes a simple approach that gives a quality index expressed in percentages. Further elaboration can be found in the XXVI/9 report. Uncertainties in banks are estimated at 12.5-22.5%, uncertainties in emissions 12.8-37%, specific numbers are dependent on the sub-sector. For the total R/AC sector, the uncertainty range in the demand calculated ranges from -10% to +30%.

As mentioned in the XXVI/9 report (UNEP, 2015), estimates should be cross-checked with reported HFC consumption and production data, specified per refrigerant (or blend).

4.3 HFC consumption and production data

Estimates for global 2012 and 2015 HFC production can be made by combining UNFCCC data, manufacturer’s estimates for production capacity as well as global emission data. This has been given in the XXVI/9 Task Force report (UNEP, 2015).

Data on HFC emissions are reported annually by developed countries, i.e., the Annex I Parties under the UNFCCC Kyoto Protocol; these emission data are estimated (calculated) by national agencies.

HFC consumption and production data are also reported to the UNFCCC. Even when certain consumption and production data are missing (i.e., data not reported by some countries, or reported as HFCs in general), these reports enable a first estimate for the production of most HFCs in the Annex I Parties to be made.

Estimates for HFC production in the developing countries are often made by developed country chemical manufacturers (Kuijpers, 2015). Based on global consumption calculations, estimates for HFC production were also made by McCulloch (2015). Furthermore, global emissions data for several HFCs are available from certain literature sources, e.g. from Montzka (2015). Recently, Chinese HFC (and HCFC) production data up to the year 2013 were reported by Kaixiang (2015). Further HFC production estimates from Chinese manufacturers were also obtained through May-July 2015 (Kuijpers, 2015).

This report uses almost the same estimates (as in the XXVI/9 report) for HFC production of the four main HFCs in Table 4-2 below (here has been added the estimated production of HFC-134a in one country). These four HFCs are the main ones used in the R/AC sector, except for HFC-134a which is also applied in several other sectors (foams, aerosols, MDIs). It shows a total HFC production of about 510 ktonnes for these four HFCs, forecast for the year 2015 (about 930 Mt CO2-eq., if calculated in climate terms). The global production capacity for these HFCs is estimated much higher, at a level of 750 ktonnes (Campbell, 2015).

It needs to be emphasised that the global HFC production (for the four main HFCs) determined in this way is estimated to have a ± 10% uncertainty for the separate HFC chemicals. These production data for certain years are reasonably reliable global estimates and have been used in order to check the demand data determined via the bottom-up method used, which are given in the sections below for the R/AC sector.

Note: (*) Global production is equal to non-Article 5 plus Article 5 country (China, minor other) production
Note: (**) Estimate from Rigby (2013)
Note: (^) New value for this report

4.4 Non-Article 5 scenarios up to 2050
4.4.1 BAU scenario

The figures below present the results of the Non-Article 5 scenario calculations:
• Non-Article 5 BAU scenario with subdivision for refrigerants.
• Non-Article 5 BAU scenario with subdivision for the various R/AC sub-sectors.

Figure 4-1: Non-Article 5 BAU scenario with subdivision for the various refrigerants or refrigerant blends in tonnes and ktonnes CO2-eq.

Figure 4-1 shows the current and projected future non-Article 5 refrigerant BAU demand, with a subdivision for the commonly used high-GWP refrigerants and low-GWP refrigerants.

The demand is given in tonnes and in GWP weighted terms (in ktonnes CO2-eq.). The amount of low-GWP refrigerants in the BAU scenario increases rapidly after 2020, because the BAU scenario includes the EU F-gas regulation as well as the US measures that enter into force as of 2016-2021 (e.g. low GWP in manufacturing of MACs). Over the period 2010-2050, the importance of R-410A and also R-407C for stationary AC becomes more and more dominant, with an increase of a factor 2 in tonnes and in GWP weighted tonnes between 2015 and 2050.

Figure 4-2 shows the non-Article 5 refrigerant BAU demand, with a subdivision for the different R/AC sub-sectors (n.b., all graphs start in the year 1990). The demand is given in tonnes and in ktonnes CO2-eq.). By 2030-2050, stationary AC accounts for more than 80% of the GWP adjusted tonnage (even when using low growth percentages, as given in Table 4-1). This is due to the fact that only a small amount of regulatory restrictions have been built in.

Figure 4-2: Non-Article 5 BAU scenario; subdivision for the various R/AC sub-sectors

Figure 4-3 below is in principle the same as in the XXVI/9 report (UNEP, 2015). It is surprising that, with the assumptions used, the percentage of R-404A in manufacturing decreases sharply, servicing remains (with the assumptions on the servicing percentage) and increases again after 2033 due to economic growth. The low GWP fraction remains very moderate in ktonnes CO2-eq., but that number implies a much larger percentage in tonnes.

Figure 4-3: New manufacturing and servicing parts of the non-Article 5 BAU scenario
with a subdivision for the various R/AC sub-sectors

4.4.2 MIT-3 scenario

The following figures are for the MIT-3 scenario, for non-Article 5 Parties, in the various R/AC sub-sectors. This is the scenario where all sub-sectors are assumed to have converted by the year 2020. The total demand, the new manufacturing and the servicing demand are shown in Figs 4-4, 4-5 and 4-6.

Figure 4-4: Total demand for the Non-Article 5 MIT-3 scenario with a subdivision for the various R/AC sub-sectors

In MIT-3, the conversion in all sub-sectors to replace high-GWP refrigerants with a variety of refrigerants with an average GWP of 300 is assumed to be complete by 2020. Manufacturing capacity is converted in equal portions per year during the period 2017-2020. This is a major difference with the BAU scenario in which stationary AC is not addressed in this manner.

Figure 4-4 shows the steep decrease in the years before 2020, after which the curve flattens due to continued servicing needs. Since some high-GWP equipment will have been manufactured until 2020, and has an average 12 year lifetime, supplies of high-GWP refrigerants will continue to be required in decreasing amounts until about 2032.

During 2010-2015, stationary AC and commercial refrigeration demands are assumed to increase quickly (see above). With transition in new manufacturing as of 2020, high-GWP refrigerants in these sector decrease, being replaced by low-GWP refrigerants that will account for more than 80% of total demand between 2020 and 2050.

What becomes again clear here, that is that the minimum demand is reached by 2032-2033, after which the demand increases again, in particular due to growth in stationary AC. So, there is a large mprovement in climate impact, although with a GWP of 300, the large refrigerant volumes considered still have a certain climate impact.

In Figure 4-5, the new manufacturing demand for the R/AC sub-sectors for high-GWP chemicals is given. By 2020, the demand for high-GWP refrigerants in new equipment manufacture falls to < 5% of the 2019 peak.

Figure 4-5: Non-Article 5 MIT-3 scenario for new manufacturing demand for high-GWP refrigerants in the various R/AC sub-sectors in ktonnes CO2-eq. (assuming manufacturing conversion over a period of 3 years, 2017-2020).

Figure 4-5: Non-Article 5 MIT-3 scenario for new manufacturing demand for high-GWP refrigerants in the various R/AC sub-sectors in ktonnes CO2-eq. (assuming manufacturing Figure 4-6: Non-Article 5 MIT-3 scenario with the servicing demand for the various subsectors in ktonnes CO2-eq. (assuming manufacturing conversion over the period 2017- 2020)

Figure 4-6 shows the volumes of high-GWP refrigerants that will be needed for servicing the installed equipment in the MIT-3 scenario. This varies between sectors (see table in section above) and decreases rapidly between 2020 and 2032, increases again due to economic growth after 2033.

4.4.3 MIT-5 scenario

This is the scenario where, for non-Article 5 Parties, all sub-sectors are assumed to have converted by the year 2025. The total, the new manufacturing and servicing demand are shown in Figs 4-7, 4-8 and 4-9.conversion over a period of 3 years, 2017-2020).

Figure 4-7: Non-Article 5 MIT-5 scenario by R/AC sub-sectors in ktonnes CO2-eq.(compare Figure 4-4 for MIT-3)
Figure 4-7 includes both manufacturing and servicing, and is similar to Fig 4-5 for MIT-3.
Figure 4-8 shows the same data for HFCs used in new manufacturing only.

Figure 4-8: Non-Article 5 MIT-5 scenario for new manufacturing demand for the various R/AC sub-sectors in GWP weighted terms (compare Fig. 4-5 for MIT-3)

In 2020, demand for new manufacturing is at about 180 Mt CO2-eq, and demand for servicing is also at about 180 Mt CO2-eq, but after 2025, the picture becomes different. New manufacturing demand decreases to less than 30 Mt CO2-eq, whereas this value is reached around the year 2037 in servicing. This is an issue that needs to borne in mind, i.e., that servicing will be delaying non-Article 5 reductions expressed in Mt CO2-eq. After a minimum in the demand in new manufacture and service, demand will increase again after 2035-2037 (5 years later than in MIT-3), due to economic growth assumed.

Figure 4-9: Non-Article 5 MIT-5 scenario with the servicing demand for the various subsectors in GWP weighted terms (compare Fig. 4-6 for MIT-3)

4.5 Article 5 scenarios
4.5.1 BAU scenario

Figure 4-10 below shows the Article 5 refrigerant BAU demand, with a subdivision for the different high GWP refrigerants and the low-GWP ones, both in tonnes and in GWP weighted terms (in CO2-eq.).

The low-GWP refrigerants applied here are again only visible in tonnes and cannot be really seen in the scale when adjusted for GWP, shown in GWP weighted terms. In the 2020-2030 period, the high-GWP refrigerant R-404A, which is used in commercial refrigeration, becomes increasingly important in GWP weighted terms.

The demand calculated for the year 2015 is about 300 ktonnes, a higher value than calculated for the BAU demand in non-Article 5 Parties (210-220 ktonnes, see above).

The combined demand for non-Article 5 and Article 5 Parties of 510 ktonnes is somewhat higher than the 475 ktonnes estimate for global HFC production for the R/AC sector in 2015 (about 7% higher); total HFC production also includes HFC-134a production for other sectors.

However, the above is likely to be caused by differences between production and calculations for stationary AC (see above; for the specific sub-sector the differences between amounts produced and calculated will be larger).

Figure 4-11 shows the Article 5 refrigerant BAU demand for the different sub-sectors. The demand is again given in tonnes and in GWP weighted terms (ktonnes CO2-eq.). The BAU model predicts that between 2015 and 2050, overall demand increases by a factor of 7-8, to about 4.5 Gt CO2. Stationary AC increases substantially, but the commercial refrigeration sub-sector also is important in GWP terms, due to the use of the high-GWP R-404A.

Figure 4-10: Article 5 BAU scenario with a subdivision for the various refrigerants and refrigerant blends in tonnes and ktonnes CO2-eq.

Figure 4-11: Article 5 BAU scenario with a subdivision for the various sub-sectors in tonnes and ktonnes CO2-eq.

Figure 4-12: Article 5 BAU scenario with new manufacturing and servicing demand for the various refrigerants (both in ktonnes CO2-eq.)

Figure 4-12 shows the demand for new manufacturing and for servicing. When manufacturing increases rapidly, the demand for servicing initially lags behind the volumes used for manufacturing. However, after a certain period it catches up, the servicing volumes become comparable to those used in manufacturing as follows in the BAU scenario (in ktonnes, not in ktonnes CO2-eq.):

• 2015: new manufacturing 195 kt, servicing 100 kt
•  2020: new manufacturing 300 kt, servicing 200 kt
• 2030: new manufacturing 530 kt, servicing 515 kt
• 2050: new manufacturing 915 kt, servicing 1080 kt