Atmospheric Treaties and
Replacement Refrigerants for Vapour Compression¹

Ian Maclaine-cross MAIRAH

Abstract

Keywords: refrigerants, consumption, emissions, global warming

1  Introduction

In the 1970s scientists measured atmospheric changes due to refrigerants and then predicted worse to come (Section 2.). Governments met to agree atmospheric treaties (Section 3.) and scientists invented potentials to express the relative environmental merit of refrigerants (Section 4.). Fluorocarbon refrigerant imports and exports are available from customs data (Section 5.) which allows calculation of consumption and potential emissions (Section 6.). For natural refrigerants only manufacturers data was available (Section 7.). By 2005 fuels and refrigerants with large contributions to global warming will be more expensive (Section 8.).

2  Climate Impacts

Solar radiation incident on the earth may be reflected by aerosols, clouds or the surface and adsorbed by the atmosphere or surface. The surface adsorbs about 120 W/m² on average which convects to the atmosphere, thermally radiates to the sky or evaporates moisture. Many atmospheric gases adsorb the surface thermal radiation making the earth's average temperature about 40 K higher than without adsorption. This is the greenhouse effect.

Ice cores from Greenland and Antarctic glaciers give the composition of the atmosphere since 1000 AD. The cores show stable concentrations of anthropogenic greenhouse gases followed by small increases in the 19th century and large increases during the 20th century (IPCC 2001a). Radiative forcing is the surface heating effect of an atmospheric change and is positive for greenhouse gases and negative for many aerosols. Radiative forcing from carbon dioxide increased between 1750 and 2000 by about 1.46 W/m² and from halocarbons about 0.34 W/m² (IPCC 2001a). Radiative forcing increases are magnified by the loss in reflection from snow and ice.

Ocean thermal inertia delays much climate impact of increased radiative forcing by hundreds of years however last century average surface temperatures rose by 0.6±0.2 K and sea levels between 100 and 200 mm (IPCC 2001a). Recent mathematical models of climate predict these measurements. For the 21st century models predict a rise in average temperature from 1.4 to 5.8 K and in sea level 0.09 to 0.94 m depending on assumptions about population and economic activity known as emission scenarios. Predicted rainfall changes are unfavourable to Australia. Cereal crops will become more difficult to grow while Australia's customers and competitors have improved growing conditions (IPCC 2001b).

World fossil fuel reserves and resources contain over 5 Eg carbon but the worst 21st century emission scenarios require only 2 Eg. Oil and gas have half the climate impact of coal per energy unit but only about 1 Eg are available (IPCC 2001c). Oil and gas depletion will increases emissions. Limiting anthropogenic radiative forcing from carbon dioxide in 2100 to twice 2000 levels requires a reduction in world consumption of fossil fuel to about 50% of 1990 levels well before 2100. Much of this emission reduction gives a net saving but the rest may cost up to US$100/MgC_eq (IPCC 2001c).

3  Atmospheric Treaties

The UNFCCC applies to all anthropogenic greenhouse gas emissions except those covered by the Montreal Protocol (UNEP 1998). The Kyoto Protocol assigns country quotas for total emissions as carbon dioxide equivalent of six of the UNFCCC gases. These six gases are carbon dioxide, methane, nitrous oxide, hydrofluorocarbons, perfluorocarbons and sulphur hexafluoride. The global warming potentials (GWP) used as coefficients in calculating the carbon dioxide equivalent of the last five gases are those approved by the Intergovernmental Panel on Climate Change (IPCC 1994).

Neither the Montreal or Kyoto Protocols require of Australia so called end use controls such as: licenses, certificates, training courses, codes of practice, recovery, recycling, new equipment or service procedures. Import and production controls include: bounties, excises, subsidies, quotas, tariffs and taxes. End use controls are more expensive and markedly less effective than import and production controls. Polluters like the ineffectiveness of end use controls and activists and regulators often lack the knowledge to assess them. Australia ultimately relied on import and production controls to meet its Montreal commitments.

4  Environmental Potentials

The Montreal Protocol (UNEP 1998) annexes tables of ozone depletion potentials (ODP Table 1) relative to R11. Treaty requirements depend on multiplying masses by ODP and summing to give an equivalent mass of R11.

For the UNFCCC (UNEP 1992) and Kyoto (UNEP 1997) reference is made to global warming potentials (GWP) relative to carbon dioxide approved from time to time by the IPCC (1994). WMO (1999) gives revised potentials based on more data with estimated uncertainty ±35%. Table 1 gives the potentials used here. For comparison, ammonia and hydrocarbon refrigerants have ODP 0.0 and GWPs 1 and 3 respectively independent of period (Maclaine-cross and Goedhart 1999).

5  Imports and Exports

2903491062

Chlorodifluoromethane;

2903300067

1,1,1,2-tetrafluoroethane (HFC 134a) (CAS 811-97-2);

2903491069

1,1-dichloro-1-fluoroethane (HCFC 141b) (CAS 1717-00-6);

2903499064

Halogenated derivatives of acyclic hydrocarbons containing two or more different halogens nes;

2903300068

Fluorinated, brominated or iodinated derivatives of acyclic hydrocarbons (excluding 1,1,3,3,3-Pentafluoro-2(trifluoromethyl) prop-l-ene (PFIB) & excluding 1,1,1,2-Tetrafluoroethane (HFC 134a) (CAS 811-97-2));

8703

Motor cars and other motor vehicles principally designed for the transport of persons (other than public transport type), including station wagons and racing cars.

The ACS export database uses the Australian Harmonised Export Commodity Classification (AHECC) with an eight digit code. Those used here with their descriptions are:

29033000

Fluorinated, brominated or iodinated derivatives of acyclic hydrocarbons;

29034900

Halogenated derivatives of acyclic hydrocarbons containing two or more different halogens nes;

8703

Motor cars and other motor vehicles principally designed for the transport of persons (other than public transport type), including station wagons and racing cars.

The HTISC and AHECC codes are used to purchase data from the Australian Bureau of Statistics (ABS). As imports or exports increase broad categories are refined and redefined. The blanks on the left hand side of the tables indicate that the category did not exist at that time. The codes on the right are correct for the last item in the row but for some items the codes and definitions have changed slightly. Code changes are usually introduced to the entire database at an arbitrary time so the first quantity in a new code will be for a reduced period. The data period chosen was the financial year since refrigerant delivery rates are lowest in early winter at its end and tax calculations generate accurate data.

Table 2 summarizes the free on board price of the shipment calculated from the customs value and mass in the ACS databases. The 99/00 step in the R134a import price resulted from shortages caused by Californian regulations requiring R134a as an aerosol propellant.

Table 3 gives annual imports and exports of fluorocarbon refrigerants from the ACS databases.

6  Emissions

Table 4 contains estimates of total imports and potential emissions calculated from the data in Tables 1 and 3. Many chemicals not used as refrigerants are include in the monohalocarbon and bihalocarbon exports in Table 3. These exports are believed an upper limit for the uncertainty of potential refrigerant emissions. The GWP used in calculating the UNFCCC potential emissions is exact by treaty and the uncertainty estimate in the UNFCCC GWE is ±10%. The GWP used in calculating the total potential GWE have an uncertainty of ±35% and this is the uncertainty of the GWE. Australia's 1990 anthropogenic carbon dioxide emissions were 348529.97 Gg (AGO 2001). These emissions do not change when GWP values are revised and so are most suitable for comparisons.

The Montreal Protocol (UNEP 1998) imposed a cap on Australia's ozone depleting consumption of 548 Mg between 1st January 1996 and 31st December 2003. Australia has a voluntary cap of 220 Mg at present. Tables 3 and 4 show a shift from R22 to R134a after the voluntary cap was reached in 98/99. This increased Australia's UNFCCC GWE but reduced the total refrigerant GWE between 98/99 and 99/00 for both short and medium terms.

From Table 3 one can calculate Australian R134a consumption for 99/00 as 2546 Mg. For refrigeration use transfers to bank are about 75 Mg and actual emissions about 30 Mg. Subtracting these gives 2441 Mg for R134a consumption by mobile air conditioners (MAC). ABS (2000) estimated the total number of enclosed motor vehicles registered in Australia on the 31st October 1999 as 11934797. If 60% of these had MACs charged with R134a, the unit R134a consumption was 341 g/year. R134a emissions from MAC occur as leakage and during maintenance (Maclaine-cross and Goedhart 1999), repairs, collision and disposal.

The carbon dioxide equivalent of 341 g of R134a is 545.4 kg. An average passenger vehicle in Australia during 98/99 used 1684 L (ABS 2000b). One litre of petrol burns to 2.15 kg of carbon dioxide so the fuel consumption emits 3621 kg CO₂. The R134a consumption of a representative Australian MAC is 15.1% of the exhaust global warming emissions of a passenger vehicle.

7  Manufacture

Australia manufactures large quantities of ammonia for fertiliser but little of refrigerant grade (R717). About half this is actually used as refrigerant. A typical bulk price for refrigerant ammonia is 3 $/kg. The liquid density of ammonia is typically twice that of fluorocarbons making ammonia the cheapest vapour compression refrigerant available on a liquid volume basis. Ammonia is corrosive to aluminium and copper so cannot replace hydrocarbons or fluorocarbons in most existing equipment.

Hydrocarbon refrigerants are made from a variety of feed-stocks and many compositions are available in Australia (Maclaine-cross and Goedhart 1999). They are classified here into medium R290/600a [60/40] and high R290/170 [94/6] pressure grades. A typical bulk price is 9 $/kg but the charge mass of R290/600a is usually one third that of fluorocarbons and of R290/170 40%. Hydrocarbons are effectively about half the price of fluorocarbons and directly replace them in major applications.

8  The Future

Kyoto applies to emissions but controls will be applied where reliable measurement is easy changing prices throughout the economy to discourage emissions. Greenhouse gas excises and tariffs at ports, wells, generating, metallurgical and chemical plant replacing many existing taxes or equivalent measures are highly probable with net taxation unchanged. These measures would also apply to Montreal gases since many are substitutes for Kyoto gases but with greater GWP. Kyoto imposes no obligation to control exports of coal, gas, oil or anything and explicitly recognizes that developing countries may increase their emissions and imports.

The IPCC (2001c) found the marginal cost of GWE reduction as US$100/MgC_eq, Aust $167/MgC_eq or Aust $45.5/MgCO_{2 eq}. Economic theory says that net incentives must equal this for policy success. For fuels and refrigerants this is equivalent to:-

petrol

11.2 cents/L so 10 cents/L or more reduction possible;

natural gas

2.5 $/GJ possible price increase;

coal

133 $/Mg but less on low grade or high volatile coals;

electricity

6 cents/kWh for 100% coal fired and 0 cents/kWh for renewables;

wood

100 $/Mg for native and 0 $/Mg for plantation;

LPG

7 cents/L so no price change likely;

R717

4.6 cents/kg;

R290/600a

14 cents/kg;

R22

77 $/kg not required by Kyoto but to protect Montreal cap from swapping between R134a and R22 mixtures;

R134a

59 $/kg calculated using Kyoto GWP of 1300 (IPCC 1994).

These values are roughly double some currently applying in Denmark and other European countries but their measures include a phased prohibition of fluorocarbon refrigerants which will be complete well before 2008. Denmark prohibited R22 in 2000 and already taxes carbon dioxide emissions.

The effects of implementing Kyoto in Australia will be:

• Electric utilities will favour natural gas over coal and new plant will be wind or other renewables wherever possible;
• Real energy and electricity prices will rise initially in some states and fall after 10 years as wind turbines are cheaper to purchase and operate;
• Fluorocarbon refrigerants will disappear from MACs in three years and in ten years all refrigeration and air conditioning.

9  Conclusion

Production data from manufacturers and import and export data from Australian Bureau of Statistics gives Australian consumption of replacement refrigerants for vapour compression since 1996 in Tables 3, 4 and 5. Correcting natural refrigerants masses for lower liquid density, the total market has grown to 7400 Mg in 99/00. Ammonia, fluorocarbons and hydrocarbons had 22%, 77% and 1% market shares respectively with hydrocarbons sales growing 15%/year. Typical bulk prices for refrigerants 717, 134a and 290/600a were 2, 8 and 5 $/L liquid respectively.

Australia has easily kept total HCFC consumption below 220 ozone depleting Mg which is well below the 548 Mg Montreal cap. The Kyoto potential GWE of refrigerants had grown from 0 to the equivalent of 0.95% of 1990 carbon dioxide emissions in 99/00. The potential 100 year GWE from all fluorocarbon refrigerants of 2.3% indicates the Kyoto GWEs future value. Australians concerned about predicted increased drought and flooding this century will however focus on the 20 year GWE of 6.1% of 1990 CO₂ emissions.

An economically rational implementation of the Kyoto Protocol in Australia appears highly probable. Electric energy price increases up to 4 cents/kWh in some localities could be expected to last up to 10 years. The price increase in refrigerants would be about 4.55 cents/kg GWP e.g., about 59 $/kg for R134a.

10  Acknowledgement

11  References

ABS 2000a, Motor Vehicle Census Australia, 31st October 1999, Australian Bureau of Statistics, Canberra, 18th August, 32 p.

ABS 2000b, Survey of Motor Vehicle Use Australia, Australian Bureau of Statistics, Canberra, 21st July, 43 p.

AGO 2001, National Greenhouse Gas Inventory 1999, Australian Greenhouse Office, Canberra, ISBN 1 876536 98 5, April, 207 p.

IPCC 1994, Radiative Forcing of Climate Change and an Evaluation of the IPCC IS92 Emissions Scenarios, J.T. Houghton et al. eds, Cambridge University Press, 339 p.

IPCC 2001a, Summary for Policy Makers, Climate Change 2001: The Scientific Basis, Third Assessment Report of Working Group I, Intergovernmental Panel on Climate Change, http://www.ipcc.ch/, Shanghai, 20th January, 20 p.

IPCC 2001b, Summary for Policy Makers, Climate Change 2001: Impacts, Adaption and Vulnerability, Third Assessment Report of Working Group II, Intergovernmental Panel on Climate Change, http://www.ipcc.ch/, Geneva, 16th February, 17 p.

IPCC 2001c, Summary for Policy Makers, Climate Change 2001: Mitigation, Third Assessment Report of Working Group III, Intergovernmental Panel on Climate Change, http://www.ipcc.ch/, Accra, 3rd March, 17 p.

Maclaine-cross, I. and Goedhart, J. 1999, Hydrocarbon Refrigerants in Australian Cars, Chapter 11 in Developments and Trends in ODS Phase Out, Yearbook 1997-1999, GTZ Project PROKLIMA, Eschborn Germany, 156 p.

UNEP 1992, The United Nations Framework Convention on Climate Change, United Nations Environment Program, New York.

UNEP 1997, The Kyoto Protocol to the United Nations Framework Convention on Climate Change, United Nations Environment Program, New York.

UNEP 1998, The 1987 Montreal Protocol on Substances that Deplete the Ozone Layer, with amendments up to Montreal 1997, United Nations Environment Program, New York.

WMO 1999, Scientific Assessment of Ozone Depletion: 1998, World Meteorological Organization, ISBN 92-807-1722-7, February, Geneva, p. 10.27.

Footnotes:

¹Maclaine-cross, I., 2001, AIRAH Journal, July, Vol.55, No.7, pp.20-23