United States        Office of Air and       EPA 430-R-04-006
Environmental        Radiation (6205J)       June 2004
Protection Agency
Analysis of Costs to Abate
International Ozone-Depleting
Substance Substitute Emissions

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Acknowledgements


This report was produced by the Alternatives and Emissions Reduction Branch of the U.S. EPA's Office
of Air and Radiation, Office of Atmospheric Programs.  Key staff from the Office of Air and Radiation
who worked on this report include Erin Birgfeld,  Jeanne  Briskin, Jeff Cohen, Dave Godwin, Anhar
Karimjee, Colm Kenny, Suzanne Kocchi, Bella Maranion, Scott Monroe, Margaret Sheppard and Deanne
Upson.

The report would not have been complete without the efforts and contributions of many individuals and
organizations.  Much of the  underlying assumptions and data used in  the analysis were provided by
numerous people and companies cited throughout the report.  In addition, many individuals provided a
peer review  of the particular chapter matching his or her  expertise.  These individuals include: Paul
Ashford, Ward Atkinson, Dave Bateman, Donald Bivens, Nick Campbell, Jim Crawford, Hugh Crowther,
William Dietrich, Tony Digmanese, Maureen Hardwick, Kenneth Hickman, William Hill, John Owens,
Eugene  Smithart, William Walter, Kert Werner, and Robert Wickham. This review does not constitute
endorsement by the person or his/her employer of the results or conclusions of this report.

The staff of the Energy Policy and Programs Practice at ICF Consulting deserves special  recognition for
their expertise, efforts in preparing many of the  individual analyses, and for synthesizing  this report.
These individuals  include:  Mollie Averyt,  Philip Groth, Daniel Lieberman, Marian Martin Van Pelt,
Pamela Mathis, Iliriana Mushkolaj, Lauren Pederson, and Katrin Peterson.

Questions concerning this report should be directed to:

David S. Godwin, P.E.
Environmental Engineer
Alternatives and Emissions Reduction Branch
Global Programs Division
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW
Mail Code 6205J
Washington, D.C. 20460
1-202-343-9324 (phone)
1-202-343-2363 (fax)
Godwin.Dave@epa.gov
                                                                                           -i-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table of Contents

Acknowledgements	i
Table of Contents	ii
List of Tables	iii
Executive Summary	ES-1
1   Analysis of Costs to Abate International HFC Emissions from Refrigeration and Air-
Conditioning	1
  1.1    Introduction	1
  1.2    Baseline Emission Estimates	5
    1.2.1    Emission Estimating Methodology	5
    1.2.2    Baseline Emissions	10
  1.3    Costs of HFC Emission Reduction from Refrigeration/AC	11
    1.3.1    Description and Cost Analysis of Abatement Options	11
    1.3.2    Calculation of Indirect Emissions and Costs for Options Applicable to Stationary
    Equipment	34
    1.3.3    Summary of Technical Applicability and Market Penetration of Abatement Options	36
  1.4    Results	41
  1.5    Summary	42
  1.6    References	43
2   Analysis of Costs to Abate International HFC Emissions from Aerosols	49
  2.1    Introduction	49
  2.2    Baseline Emission Estimates	50
    2.2.1    Emission Estimating Methodology	50
    2.2.2    Baseline Emissions	51
  2.3    Costs of HFC Emission Reductions for Aerosols	51
    2.3.1    Description and Cost Analysis of Abatement Options	52
    2.3.2    Summary of Technical Applicability and Market Penetration of Abatement Options	56
  2.4    Results	57
  2.5    Summary	58
  2.6    References	59
3   Analysis of Costs to Abate International HFC Emissions from Foams	61
  3.1    Introduction	61
  3.2    Baseline Emission Estimates	62
    3.2.1    Emission Estimating Methodology	62
    3.2.2    Baseline Emissions	64
  3.3    Costs of HFC Emission Reductions from Foams	64
    3.3.1    Description of Abatement Options	64
    3.3.2    Description of Abatement Options	67
    3.3.3    Summary of Technical Applicability and Market Penetration of Abatement Options	85
  3.4    Results	90
  3.5    Summary	92
  3.6    References	93
4   Analysis of Costs to Abate International HFC and PFC Emissions from Fire Extinguishing....96
  4.1    Introduction	96
  4.2    Baseline Emission Estimates	98
    4.2.1    Emission Estimating Methodology	98
    4.2.2    Baseline Emissions	99
  4.3    Costs of HFC and PFC Emission Reductions from Fire Extinguishing	100
    4.3.1    Description and Cost Analysis of Abatement Options	101
    4.3.2    Summary of Technical Applicability and Market Penetration of Abatement Options	110
  4.4    Results	112
  4.5    Summary	112
  4.6    References	113
                                                                                         -11-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
5   Analysis of Costs to Abate International HFC and PFC Emissions from Solvents	116
  5.1    Introduction	116
  5.2    Baseline Emission Estimates	117
    5.2.1    Emission Estimating Methodology	117
    5.2.2    Baseline Emissions	118
  5.3    Cost of HFC and PFC Emission Reductions for Solvents	119
    5.3.1    Description and Cost Analysis of Abatement Options	119
    5.3.2    Summary of Technical Applicability and Market Penetration of Abatement Options	125
  5.4    Results	126
  5.5    Summary	127
  5.6    References	127
Appendices


List of Tables

Table 1-1: Percent of On-Road Vehicle Fleet Assumed to Have Operational Air-Conditioning Units	8
Table 1-2: Estimated Percent of Refrigeration/Air-Conditioning HFC Emissions Attributable to MVACs....9
Table 1-3: Distribution of Refrigeration and Air-Conditioning Sector HFC Emissions by End-use, Region,
    and Year	9
Table 1-4: Baseline HFC Emission Estimates from Refrigeration and Air-Conditioning (MMTCE)	10
Table 1-5: Assumptions on Duration and Applicability of Emission Reduction Options	13
Table 1-6: Summary of Assumptions for Leak Repair for Large Equipment	15
Table 1-7: Summary of Assumptions for Recovery/Recycling from Small Equipment	17
Table 1-8: Summary of Assumptions for Distributed Systems for New Stationary Equipment	21
Table 1-9: Summary of Assumptions for HFC Secondary Loop Systems for New Stationary Equipment. 23
Table 1-10: Summary of Assumptions for Ammonia Secondary Loop Systems for New Stationary
    Equipment	25
Table 1-11: Summary of Assumptions for Enhanced HFC-134a Systems for New MVACs	26
Table 1-12: Summary of Assumptions for HFC-152a DX Systems in New MVACs	28
Table 1-13: Summary of Assumptions forCO2 Systems in New MVACs	31
Table 1-14:  Net Annual Emissions and Energy Costs of Replacement Options in the United States in a
    600,000 Square Foot Supermarket	35
Table 1-15: Summary of Technical Applicability of Abatement Options by Region, Percent3	37
Table 1-16: Incremental Maximum Market Penetration of Technology Options into New Equipment by
    Region, Expressed as a Percent of Emissions from New Refrigeration/Air-Conditioning  Equipmenta38
Table 1-17: Incremental Maximum Market Penetration of All Abatement Options by Region, Expressed  as
    a Percent of Total Refrigeration/Air-Conditioning Emissions	39
Table 1-18: Percent Reduction off Baseline Emissions of All Abatement Options by Region	40
Table 1-19: United States Emission Reductions in 2020 and Break-Even Costs for Refrigeration and Air-
    Conditioning 	41
Table 1-20: Non-U.S. Annex I Emission  Reductions in 2020 and Break-Even Costs for Refrigeration and
    Air-Conditioning	42
Table 1-21: Non Annex I Emission Reductions in 2020 and Break-Even Costs for Refrigeration and Air-
    Conditioning 	42
Table 2-1: Baseline HFC Emission Estimates from MDI Aerosols (MMTCE)	51
Table 2-2: Baseline HFC Emission Estimates from Non-MDI Aerosols (MMTCE)	51
Table 2-3: Technical Applicability and Incremental Maximum Market Penetration of Aerosol Options
    (Percent)	56
Table 2-4: Emission Reductions off the Total Applicable Aerosols Baseline (Percent)	57
Table 2-5: United States Emission Reductions in 2020 and Break-Even Costs for Aerosols	57
Table 2-6: Non-U.S. Annex I Emission Reductions in 2020 and  Break-Even Costs for Aerosols	58
Table 2-7: Non Annex I Emission Reductions in 2020 and Break-Even Costs for Aerosols	58
Table 3-1: U.S. EPA's Vintaging Model Emissions Profile for the Foams Sector	63
Table 3-2: Baseline Emissions in MMTCE	64
Table 3-3:  Base Case Assumptions fora Contractor Using HFC-134a	67
                                                                                        -in-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 3-4: Assumptions and costs used in the cost analysis to substitute HFC-134a Hydrocarbons	68
Table 3-5: Base Case Assumptions fora Contractor Using HFC-134a and HFC-152a	69
Table 3-6: Assumptions and costs used in the cost analysis to substitute HFC-134a with Hydrocarbons 70
Table 3-7: Assumptions and costs used in the cost analysis to substitute HFC-152a with Hydrocarbons 71
Table 3-8: Base Case Assumptions fora Hypothetical Contractor Using HFC-134a/CO2 (LCD)	72
Table 3-9: Assumptions and costs used in the cost analysis to substitute HFC-134a/CO2with CO2	73
Table 3-10: Assumptions and costs used in the cost analysis to substitute HFC-134a/CO2 with
    CO2/Alcohol	74
Table 3-11: Base Case Assumptions for a Hypothetical Spray Foam Contractor Using HFC-245fa/CO2
    (water)	74
Table 3-12: Assumptions and costs used in the cost analysis to substitute HFC-245fa/CO2 (water) with
    CO2 (water)	76
Table 3-13: Assumptions and costs used in the cost analysis to substitute HFC-245fa/CO2 (water) with
    Hydrocarbons	77
Table 3-14 General Assumptions Applicable Both End-of-Life Options	81
Table 3-15 Assumptions Applicable to the Manual Process with Foam Incineration	81
Table 3-16 Assumptions Applicable to the Automated Process with Foam Landfilling	81
Table 3-17: Automated Process with Foam Landfilling	84
Table 3-18: Manual Process with Foam Incineration	85
Table 3-19: Reduction Efficiency of Foam Options (Percent)	85
Table 3-20: Technical Applicability of Foam Options for the US, Europe, and Japan  (Percent)	86
Table 3-21: Technical Applicability of Foam Options for the CEITs, China, and the Rest of the Developed
    World (Percent)	87
Table 3-22: Incremental Maximum Market Penetration Expressed as Percent of New Emissions for which
    the Options are Technically Applicable	88
Table 3-23: Incremental Maximum Market Penetration Expressed as Percent of All Emissions for which
    the Options are Technically Applicable	89
Table 3-24: Emission Reductions off Total Foams Baseline for the US, Europe, and Japan (Percent) ....90
Table 3-25: Emission Reductions off Total Foams Baseline for the CEITs, China, and the Rest of the
    Developed World (Percent)	90
Table 3-26: Emission Reductions in 2020 and Costs of Abatement for Foams in the US	91
Table 3-27: Emission Reductions in 2020 and Costs of Abatement for Foams in the non-US Annex I
    Countries	91
Table 3-28: Emission Reductions in 2020 and Costs of Abatement for Foams in non-Annex I Countries 92
Table 4-1: Baseline HFC and PFC Emission Estimates from Fire Extinguishing (MMTCE)	100
Table 4-2: Assumed Breakout of Total GWP-Weighted Baseline Fire Extinguishing Emissions	100
Table 4-3: Summary of Technical Applicability of Abatement Options	110
Table 4-4: Incremental Maximum Market Penetration, Expressed as Percent of Annual Installation of New
    Class A or Class B Systems	110
Table 4-5: Incremental Maximum Market Penetration Expressed as Percent of Entire Installed Base
    (Class A or Class B)	111
Table 4-6: Emission Reductions off Total Fire Extinguishing Baseline	111
Table 4-7: United States Emission Reductions in 2020 and Break-Even Costs for Fire Extinguishing ... 112
Table 4-8: Non-U.S. Annex I  Emission Reductions in 2020 and Break-Even Costs for Fire
    Extinguishing	112
Table 4-9: Non Annex I Emission Reductions in 2020 and Break-Even Costs for Fire Extinguishing	112
Table 5-1: General Overview of Solvent Technologies Used Globally	117
Table 5-2: Baseline HFC and PFC Emission Estimates from Solvents (MMTCE)	119
Table 5-3: Retrofit Techniques for Batch Vapor Cleaning  Machine (Less than 13 Square Feet)	122
Table 5-4: Technical Applicability and Incremental Maximum Market Penetration of Solvent Options
    (Percent)	125
Table 5-5: Emission Reductions off the Total Solvent Baseline (Percent)	126
Table 5-6: United States Emission Reductions in 2020 and Break-Even Costs for Solvents	126
Table 5-7: Non-U.S. Annex I  Emission Reductions in 2020 and Break-Even Costs for Solvents	126
Table 5-8: Non Annex I Emission Reductions in 2020 and Break-Even Costs for Solvents	126
                                                                                         -IV-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Executive Summary
Background

Since the introduction of the Montreal Protocol on Substances That Deplete the Ozone Layer, over a
decade ago, international use of ozone-depleting substances (ODSs) has declined significantly. The key
component of the  Montreal Protocol is the stipulation that the production and consumption of all
identified compounds that deplete the ozone layer are to be phased out by 2030 in developed countries
and 2040 in developing countries.  The use of hydrofluorocarbons (HFCs), and  to a much lesser extent
perfluorocarbons  (PFCs),  has  allowed   the   rapid  phaseout   of   chlorofluorocarbons   (CFCs),
hydrochlorofluorocarbons (HCFCs), and halons in the U.S. and other countries in applications for which
other alternatives are not readily available.   HFCs have generally been selected for applications where
they provide superior technical (reliability) or safety (low  toxicity and flammability) performance.  In
many cases, HFCs provide equal or better energy efficiency as compared to other available alternatives,
thereby reducing long-term environmental impacts.  HFCs are expected to  replace a significant portion of
past and current demand for ODSs in  insulating  foams, refrigeration and air-conditioning, propellants
used in metered dose inhalers, specialized fire protection equipment, and in other applications.

The  end-use sectors in which transition away from ODS use  and production is occurring  include the
refrigeration and air-conditioning,  aerosols, foams,  solvents, and  fire-extinguishing industries.   An
increasing reliance on HFC use in these processes has in turn increased concern  over emissions  of these
gases.  Emissions that result from the ODS-substitute sectors contribute to a group of gases known as the
high global warming potential (GWP) gases, which includes HFCs, PFCs, and sulfur hexafluoride (SF6).
These gases are many times more effective (on  a per ton basis)  than CO2  in trapping heat in the
atmosphere.  The global warming potentials of the high GWP gases range  from 140 to over 23,900 times
the global warming capability of CO2,  and in some cases these gases  remain in the atmosphere for
hundreds or thousands of years. Currently, the most widely used high GWP gas is HFC-134a, which has
a shorter lifetime of about 15 years and a GWP of 1300.


Overview


The major focus of most previous analyses regarding GHG emission reduction opportunities has  been on
energy-related  CO2 emissions.  Along with continued efforts on methane and nitrous oxide reductions,
reduced emissions of the high GWP gases have the potential to make a significant contribution to  cost-
effective GHG reductions. This report has been developed, in part, to better characterize the role of the
high GWP gases as part of a comprehensive GHG mitigation approach internationally.  To that end, this
report assembles the costs of reducing quantities  of high GWP gas emissions from each of the major
ODS-substitute end-use sectors into an international marginal abatement cost (MAC) curve that shows the
total emission reductions achievable—at increasing monetary values of carbon—in the year 2020.

The three main objectives of this report—and of the analysis that supports the  generation of the MAC
curve—are to:  (1) present the U.S. EPA's current forecasts of global high GWP gas emissions from ODS-
substitute sectors through 2020 under a "business-as-usual" scenario that assumes no further actions, apart
from normal and expected incremental technological improvements, are taken to reduce emissions; (2)
use available cost and technical data to describe those technologies and practices that  can reduce these
emissions  from the major emissions sources analyzed (some of which are expected to be voluntarily
adopted by industry); and (3) estimate the  costs of reducing high GWP gas emissions for each major
                                                                                        -ES-1-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
source included in this analysis.  Forecasts of world emissions are estimated and summarized by region
for 2005, 2010, 2015, and 2020.  These emissions baselines, when combined with cost, efficiency, and
technical data on the specific technologies and practices available to reduce high GWP gas emissions,
provide the set of data used to calculate  regional, sectoral, and total global MAC curves.  Aggregate
results of the analysis—conducted by country—are presented (for the purposes of this report) by region in
each of the following chapters.  The three regions presented include: (1) the United States, (2) non-U.S.
Annex I countries (that is, all Annex I countries except for the U.S.), and (3) non-Annex I countries (that
is, all countries in the world except for Annex I countries).:

This report is divided into the following five chapters, each corresponding with one of the ODS-substitute
source-categories.

    •   Chapter 1. HFC Emissions from Refrigeration and Air-Conditioning
    •   Chapter 2. HFC Emissions from Aerosols
    •   Chapters. HFC Emissions from Foams
    •   Chapter 4. HFC and PFC Emissions from Fire Extinguishing
    •   Chapters. HFC and PFC/PFPE Emissions from Solvents

For each emission source, this report presents the following information:

    •   Baseline Emissions of High  GWP Gases.  A section  in each chapter discusses the forecasted
        emissions from that source through 2020 in the three  key regional groups (the U.S., Non-U.S.
        Annex I and Non-Annex I countries).  Emissions baselines are estimated based on a "business-as-
        usual" scenario that assumes no further actions, apart from normal and expected incremental
        technological improvements, are taken to reduce emissions.
    •   High  GWP Gas Emission  Reduction  Options and  Associated Costs.    Several available
        technologies or practices ("options") have the potential to reduce high GWP gas emissions within
        each sector. The U.S. EPA uses discounted cash flow analysis to estimate the present and future
        cost of achieving the reductions associated with each option discussed.  Costs  are presented in
        terms of year 2000 United States dollars per metric ton of carbon equivalent ($/TCE). The costs
        associated  with emissions  reductions  have been identified  and  categorized  as one-time
        investments and/or annual operating and maintenance (O&M) costs. In most cases, data on these
        costs were available; where sufficient data were not available,  EPA has presented the options
        qualitatively.

The framework for this analysis is global in scope, consistent with the intent to develop inputs useful for
macroeconomic studies. Given this broad view, this report does not present highly detailed analyses of the
individual sources of high  GWP gas emissions, nor can it comprehensively evaluate the comparative
advantages or technical challenges of alternative  technologies  in specific industry sectors.  It does,
however, provide a valuable resource for policy makers internationally in characterizing the role of the
high GWP gases  as part of a comprehensive greenhouse gas (GHG) mitigation approach.  While these
gases represent only a small portion of global GHG emissions, they include the most quickly growing
sources of emissions.  While the focus of most previous  work on GHG emission reduction opportunities
has been on energy-related CO2 emissions, reduced high  GWP gases can also contribute to cost-effective
GHG reductions.  This analysis estimates the costs of innovative reduction efforts that can be  undertaken
by various industries, and is meant to supplement carbon dioxide, nitrous oxide, and methane economic
1 The term "Annex I" refers to members of the United Nations Framework Convention on Climate Change
(UNFCCC) listed under Annex I, which includes developed countries and countries with economies in transition. A
comprehensive list of countries and the regions under which they are categorized for the purposes of this report is
provided in Appendix B.
                                                                                           -ES-2-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
analyses in order to provide a broader understanding of the costs of a comprehensive GHG mitigation
strategy that covers all of these gases;  it will  also help to identify opportunities to  reduce the costs
associated with such a strategy.


Baseline Emission  Estimates


Until recently, few countries have made  significant efforts to track and project the use and emissions of
HFCs and PFCs used as ODS substitutes. However, to the extent that country-specific ODS substitute
emission information is available—e.g.  from National Communications submitted under  the  United
Nations Framework Convention on  Climate  Change (UNFCCC)—each country's data was used as the
basis for projecting future emissions.

In the absence of reported data, the  following approach was used. First, a "Vintaging Model" of ODS-
containing equipment and products was used to  estimate the U.S. use and emissions of ODS substitutes.
Next, emissions from non-U.S. countries were estimated  for each  ODS consuming  end-use  in  each
country.  In developing these estimates, it was  initially assumed that the international transitions from
ODSs to HFCs and other substitutes follow the same substitution patterns as the United States.  These
U.S.-based substitution scenarios  were  then  customized to each region or  country using  adjustment
factors that take into consideration differences in historical and projected economic growth,2 the timing of
the phase-out, and the distribution of ODS and substitute use across end-uses in each region or country.
In some sectors, specific adjustments or  methodologies were used that were specific only to that sector.
The methodology used to estimate  and adjust emissions is described further in Appendix A.

Exhibit ES-1 shows the emission estimates of high GWP gases from the ODS and ODS-substitute end use
sources in the years 2005, 2010, 2015, and 2020. The exhibit shows a continuing rise in emissions from
each  of these sectors throughout  the forecasted period.  This growth  in HFC  and PFC emissions is
expected as substitution in ODS markets reaches maximum market penetration.3
 Exhibit ES-1: Baseline World ODS Substitute High GWP Gas Emissions (MMTCE)
Source of High GWP Gases Used as ODS Substitutes
Refrigeration and Air-Conditioning
Partially and Fully Fluorinated Solvents
Aerosols
Foams
Fire Extinguishing
TOTAL
2005
59.92
1.07
12.26
2.94
2.26
78.45
2010
94.50
1.14
14.44
5.30
2.26
117.66
2015
128.73
1.20
15.92
7.76
2.65
156.26
2020
158.80
1.26
17.60
16.38
3.19
197.23
 Notes:
 Forecast assumes a "business-as-usual" scenario as described above.
 The emissions forecast includes only direct emissions. Indirect emissions—those that result from the production of energy
 required in the manufacturing and operation of the emitting sources—are not included.
 Sums might not add to total due to independent rounding.	

Each chapter of this report presents similar tables, each of which present data specific to one of the three
key regional groups analyzed: the United States, Non-U.S. Annex I countries, and Non-Annex I countries.
 Economic growth is used as a proxy for the growth in the use of and emissions from products containing ODSs and
ODS substitutes. Further research is warranted to assess how closely these two variables have been related in the
past and may be related in the future.

3 Although many developed countries are well into the phaseout of ODS production, continued substitution with
high GWP gases, especially in developing countries, is expected much further than 2020, as long-lived ODS-
containing products, such as air-conditioners and foams, are retired and replaced with ODS substitutes.
                                                                                            -ES-3-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
At the present time,  industry decisions to  pursue emission reductions  depend greatly  on the cost-
effectiveness of the available reduction options. A major purpose of this report is therefore to estimate
the cost-effectiveness of various  emission reduction options and to determine the  quantity of future
emission reductions that might be achievable across a  range  of possible market  values of carbon-
equivalent emission reductions.


Economic Analysis of Options for Reducing  Emissions  of High GWP
Gases


In addition  to  baseline emission estimates, each  source-specific chapter of this report  analyzes  the
technical and economic viability of several emission mitigation options.  The cost of implementation and
the resulting emission reductions  associated  with  each option provide the bases for the analysis.  The
reduction options considered in this report were identified from various reports and literature on emission
reductions, industry publications, and industry contacts. The most promising options to reduce high GWP
gas emissions from ODS substitute sources broadly include: substituting other gases for high GWP gases
in a variety  of applications, where safety and performance requirements can be met;  implementing new
technologies that use and/or emit significantly lower amounts of the high GWP gas; and various process
and handling  options that reduce emissions during the  manufacture, use, and disposal of products
containing high GWP gases.

Data regarding options that are assumed to have  already reached full market penetration or that  are
assumed not to be feasible in particular countries or regions are not included  in the MAC analysis  for
those countries.  Where particular options are deemed applicable, several specific assumptions and data
elements contribute to the calculation of the MAC.

Discounting Costs and Benefits

This report  uses discounted cash flow  analysis to estimate the cost  of achieving reductions through
instituting each potential mitigation option available to the  several sources.  All costs are presented in real
year 2000 United States dollars.  Consistent  with the approach EPA has used in developing high GWP
gas and methane MAC curves for the United States, the practice of using discounted cash flow analysis
reflects the decision-making process that manufacturers use when considering investments in emission
reduction practices (see EPA, 1999; EPA, 2001). This decision-making process is typically a cost-benefit
analysis, comparing the positive cost (financial burden) to the negative cost (financial and/or emission
reduction benefit)  associated with each  potential mitigation option.  Data to support estimates of both
costs and  benefits of options are  available  in  most cases; where  sufficient data are not available,  the
options are summarized qualitatively, but are not included in the MAC curves presented.

The positive costs are typically categorized as either one-time (capital) investment costs—such as may be
incurred when installing new equipment or applying a retrofit  option—or as O&M costs—such as  are
generally applied annually to maintain an option through manual labor, routine repairs, electricity or other
fuel use, etc. Mitigation options may have one or both of these types of costs, and both ultimately affect
the viability of the option. The costs used in this  international analysis improve upon those used in  the
analysis that  supported  U.S. High GWP  Gas Emissions 1990-2010: Inventories, Projections,  and
Opportunities for Reductions (EPA, 2001).  International costs are extrapolated from these values using
adjustment factors that vary  by country  and region according to the most pertinent variable (specific to
each option  or technology).  Factors used to adjust these costs include relative labor rates  (for labor-
intensive options)  and relative electricity prices (for energy-intensive options).  For some specific
                                                                                          -ES-4-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
options—specifically those that are considered in the context of an international market in which prices
are not highly variable by country—costs are assumed to be the same internationally.

Negative costs, or benefits, include savings that are achieved by reducing or mitigating emissions of a
product, sale or recycling of recovered product,  or by improving process energy efficiency.  The  value
associated with a particular product may vary by country, and where possible, this variability is taken into
account in estimating the benefit.

Together,  all of the cost data described here, expressed as costs or savings per metric  ton of carbon
equivalent  emissions  reduced,  contribute to the calculation of a break-even price specific to each
mitigation option.  This figure represents the point at which an  entity (individual, corporation, industry,
etc.)—regardless of environmental, legal, or policy concerns—will  be financially indifferent in deciding
whether to  institute an emissions mitigation option.   The concept  of the  break-even price  and its
implications are discussed briefly in the Marginal Abatement Cost Curves section below, and—along
with the general equation used to estimate its value—in greater detail in Appendix D.

Marginal Abatement  Cost Curves

The  world high GWP gas MAC curve is shown in Exhibits ES-2 and ES-3 at four and twenty percent
discount rates, respectively. Exhibit ES-3 also incorporates a forty percent tax rate.  Each of these two
curves uses the appropriate schedule of emission reductions and costs  for all of the high GWP gases as
presented in Exhibits ES-4 and  ES-5. The  MAC curve illustrates emission reductions achievable as the
value of carbon increases ($/TCE).

The  MAC curve is derived by rank ordering individual reduction opportunities by $/TCE and plotting the
corresponding emission reductions cumulatively. Each point along the MAC curve shows the cost of
abating an  additional ton  of carbon-equivalent gas at the margin.  The break-even price (expressed in
dollars per metric ton of carbon equivalent emissions and calculated as described in Appendix D) of each
option determines its placement with respect to the y-axis.  Moving away from the origin, each point on
the curve represents a cumulative sum of emission  reductions with  respect to the x-axis.   Points
corresponding to  a zero or negative  $/TCE value along the y-axis demonstrate a market in which the
benefits of reducing the high GWP gas—represented by avoided costs of HFCs, for example—pay for the
emission reduction effort alone.  As discussed above, these negative costs imply that it is cost-efficient to
adopt the measure (i.e., the option will result in financial  savings).  Positive break-even values imply
financial costs that only the introduction of some  external value of emission reductions (determined by an
emissions trading market,  through taxes or  incentives, or through some other  externally imposed value)
can outweigh, thereby making an investment viable.

As seen  in  Exhibit ES-2—which  is in effect an economic supply curve measuring the quantity of
emission reductions supplied by industry to the market at increasing values of carbon—as one follows the
curve away from  the origin, its elasticity decreases drastically at  or around a certain point.  At values
above  this  point  (roughly $200/TCE), the curve is  relatively  inelastic; thus, subsequent or marginal
increases in the value of carbon will have a decreased effect  on emission reductions. Reductions past this
point are unlikely to occur without some external  benefit or incentive, neither of which is within the  scope
of this publication.  It is, however, interesting to note the quantifiable emission reductions that are cost-
effective barring any type  of external variable; these options for which break-even prices fall below zero
contribute to a total potential emission reduction  of about 18 MMTCE.  A total of 65 MMTCE reduction
in emissions is achievable at a carbon value less than about $200/TCE.
                                                                                           -ES-5-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Exhibit ES-2: 2020 Marginal Abatement Cost Curve for All OPS Substitutes at Four Percent Discount Rate (No Tax Rate)


           0%       5%       10%      15%       20%      25%      30%       35%

uT
o
o
o
G.
+J
0)
.1
HI
c
o
•s
«
O
"o

ra


$1 snn
(t-i /inn
ti 9nn
ti nnn
•RRnn
tfinn
f;4nn
i;9nn

•fin
($200)
C

I
f
I
I
I
I
^J


r •
) 10 20 30 40 50 60 7
Sum of Emissions Avoided (MMTCE)











0
Exhibit ES-3: 2020 Marginal Abatement Cost Curve for All ODS Substitutes at Twenty Percent Discount Rate (40% Tax

Rate)
            o%
5%
10%
15%
20%
25%
30%
35%
      $1,800
  -.  $1,600 -

  HI

  O

  t   $1,400 -
  o
  o
  o
      $1,200 -
  


  "5
       $800 --
       $600 --
       ($200)
                      10        20       30        40        50


                             Sum of Emissions Avoided (MMTCE)
                                                 60
                                                  70
                                                                                                -ES-6-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Exhibit ES-4: World Emission Reductions in 2020 and Break-Even Costs for ODS Substitutes (4% Discount Rate ,
SECTOR
Solvents
Aerosols (Non-MDI)
Aerosols (Non-MDI)
Foams
Refrigeration/AC
Foams
Foams
Refrigeration/AC
Foams
Solvents
Refrigeration/AC
Solvents
Aerosols (Non-MDI)
Solvents
Refrigeration/AC
Foams
Foams
Foams
Refrigeration/AC
Foams
Fire Extinguishing
Foams
Refrigeration/AC
Refrigeration/AC
Fire Extinguishing
Foams
Refrigeration/AC
Foams
Fire Extinguishing
Aerosols (MDI)
Option Name
Retrofit
HFC to HC
HFC to NIK
Spray HFC-245fa/CO2 (HFC-227ea/365mfc) to HC
Leak Repair
PU One Component HFC-152a to HC
XPS: HFC-134a/C02to CO2/Alcohol
Replace DX with Distributed System
PU One Component HFC-134a to HC
HFC to HFE
Recovery
NIK Semi-Aqueous
HFC-134ato 152a
NIKAqueous
Enhanced HFC-134a in MVACs
XPS: HFC-134a/CO2to CO2
Appliance HFC-134ato HC
Appliance End-of-Life: Automated Process
HFC-152a in MVACs
PU Continuous & Discontinuous Panel HFC to HC
FK-5-1-12
Spray HFC-245fa/CO2 (HFC-227ea/365mfc) to CO2
Ammonia Secondary Loop
HFC Secondary Loop
Inert Gases
Appliance End-of Life: Manual Process
C02 for New MVACs
Appliance HFC-245fa (HFC-227ea/365mfc) to HC
Water Mist
Dry Powder Inhalers
1 0% Tax Rate)
Reduction Cumulative
Emission off Baseline Cumulative % Reduction
Break-Even Price Reduction of of Option Reductions from 2020
($/TCE)a Option (MMTCE) (Percent) (MMTCE) Baseline
Low
$ (134.18)
$ (22.76)
$ (21.37)
$ (17.97)
$ (1 1 .74)
$ (9.40)
$ (8.50)
$ (57.61)
$ (0.07)
$
$ 0.49
$ 0.80
$ 1.91
$ 6.67
$ (275.47)
$ 1 1 .48
$ 17.35
$ 23.82
$ (73.65)
$ 43.83
$ 83.66
$ 96.45
$ 1 1 .47
$ 9.66
$ 71.10
$ 175.09
$ (23.44)
$ 201.80
$ 111.76
$ 1,691.25
High
$ (134.18)
$ (22.76)
$ (21 .37)
$ (15.13)
$ (1 1 .74)
$ (9.40)
$ (8.50)
$ (4.78)
$ (0.07)
$
$ 0.49
$ 0.80
$ 1.91
$ 6.67
$ 7.02
$ 1 1 .48
$ 17.35
$ 23.82
$ 30.81
$ 43.83
$ 85.22
$ 96.45
$ 123.84
$ 125.39
$ 136.29
$ 175.09
$ 180.72
$ 201 .80
$ 287.80
$ 1,691.25
0.013
1.078
1.078
0.457
3.593
-
-
11.108
0.136
0.426
8.510
0.098
3.991
0.196
5.029
1.700
0.039
1.336
4.921
-
0.495
0.343
5.918
8.993
0.396
0.793
4.237
0.225
0.057
3.409
0.0%
0.5%
0.5%
0.2%
1.8%
0.0%
0.0%
5.6%
0.1%
0.2%
4.3%
0.0%
2.0%
0.1%
2.5%
0.9%
0.0%
0.7%
2.5%
0.0%
0.3%
0.2%
3.0%
4.6%
0.2%
0.4%
2.1%
0.1%
0.0%
1.7%
0.013
1.091
2.169
2.626
6.219
6.219
6.219
17.327
17.463
17.889
26.399
26.497
30.487
30.683
35.712
37.412
37.451
38.787
43.708
43.708
44.203
44.546
50.463
59.456
59.852
60.645
64.882
65.107
65.163
68.572
0.0%
0.6%
1.1%
1.3%
3.2%
3.2%
3.2%
8.8%
8.9%
9.1%
13.4%
13.4%
1 5.5%
1 5.6%
18.1%
1 9.0%
19.0%
19.7%
22.2%
22.2%
22.4%
22.6%
25.6%
30.1%
30.3%
30.7%
32.9%
33.0%
33.0%
34.8%
 Costs vary by country/region based on one-time or annual adjustment factors (e.g., electricity price, fuel price, etc.); therefore, the lowest and highest costs are shown.
                                                                                                                                                          -ES-7-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Exhibit ES-5: World Emission Reductions in 2020 and Break-Even Costs for ODS Substitutes (20% Discount Rate / 40% Tax Rate)
Emission Reduction Cumulative
Reduction of off Baseline Cumulative % Reduction
Break-Even Price Option of Option Reductions from 2020
SECTOR Option Name ($/TCE)a (MMTCE) (Percent) (MMTCE) Baseline

Solvents
Aerosols (Non-MDI)
Aerosols (Non-MDI)
Refrigeration/AC
Solvents
Refrigeration/AC
Foams
Solvents
Aerosols (Non-MDI)
Foams
Foams
Foams
Solvents
Foams
Foams
Refrigeration/AC
Foams
Foams
Refrigeration/AC
Refrigeration/AC
Foams
Refrigeration/AC
Fire Extinguishing
Foams
Refrigeration/AC
Fire Extinguishing
Fire Extinguishing
Refrigeration/AC
Foams
Aerosols (MDI)

Retrofit
HFCtoHC
HFC to NIK
Leak Repair
HFCtoHFE
Recovery
Spray HFC-245fa/CO2 (HFC-227ea/365mfc) to HC
NIK Semi-Aqueous
HFC-134ato 152a
PU One Component HFC-152a to HC
XPS: HFC-134a/CO2to CO2/Alcohol
PU One Component HFC-134a to HC
NIKAqueous
Appliance End-of-Life: Automated Process
PU Continuous and Discontinuous Panel HFC to HC
Replace DX with Distributed System
XPS: HFC-134a/CO2to CO2
Spray HFC-245fa/CO2 (HFC-227ea/365mfc) to CO2
HFC Secondary Loop
Ammonia Secondary Loop
Appliance End-of-Life: Manual Process
HFC-152a in MVACs
FK-5-1-12
Appliance HFC-134ato HC
Enhanced HFC-134a in MVACs
Inert Gases
Water Mist
C02 for New MVACs
Appliance HFC-245fa (HFC-227ea/365mfc) to HC
Dry Powder Inhalers
Low
$ (132.14)
$ (21 .95)
$ (20.75)
$ (3.78)
-
$ 0.49
$ (13.98)
$ 2.14
$ 2.49
$ 2.77
$ 12.82
$ 14.08
$ 17.89
$ 37.16
$ 56.26
$ 15.99
$ 122.09
$ 122.55
$ 45.33
$ 55.34
$ 191.65
$ 55.55
$ 334.91
$ 404.45
$ (35.34)
$ 369.25
$ 467.17
$ 422.48
$ 1,115.58
$1,691.25
High
$ (132.14)
$ (21 .95)
$ (20.75)
$ (3.78)
-
$ 0.49
$ 1.53
$ 2.14
$ 2.49
$ 2.77
$ 12.82
$ 14.08
$ 17.89
$ 37.16
$ 56.26
$ 72.96
$ 122.09
$ 122.55
$ 173.22
$ 180.96
$ 191.65
$ 205.42
$ 336.75
$ 404.45
$ 468.38
$ 505.32
$ 781 .28
$ 839.80
$1,115.58
$1,691.25

0.013
1.078
1.078
3.593
0.426
8.510
0.457
0.098
3.991
-
-
0.136
0.196
1.336
-
11.108
1.700
0.343
8.993
5.918
0.793
4.921
0.495
0.039
5.029
0.396
0.057
4.237
0.225
3.409

0.0%
0.5%
0.5%
1.8%
0.2%
4.3%
0.2%
0.0%
2.0%
0.0%
0.0%
0.1%
0.1%
0.7%
0.0%
5.6%
0.9%
0.2%
4.6%
3.0%
0.4%
2.5%
0.3%
0.0%
2.5%
0.2%
0.0%
2.1%
0.1%
1.7%

0.013
1.091
2.169
5.761
6.187
1 4.697
15.155
15.253
19.243
19.243
1 9.243
19.379
19.575
20.912
20.912
32.020
33.719
34.062
43.055
48.972
49.765
54.686
55.181
55.219
60.248
60.645
60.701
64.938
65.163
68.572

0.0%
0.6%
1.1%
2.9%
3.1%
7.5%
7.7%
7.7%
9.8%
9.8%
9.8%
9.8%
9.9%
1 0.6%
10.6%
16.2%
17.1%
17.3%
21 .8%
24.8%
25.2%
27.7%
28.0%
28.0%
30.5%
30.7%
30.8%
32.9%
33.0%
34.8%
1 Costs vary by country/region based on one-time or annual adjustment factors (e.g., electricity price, fuel price, etc.); therefore, the lowest and highest costs are shown.
                                                                                                                                                          -ES-8-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
References

EPA (Environmental Protection Agency). 2001.  U.S. High GWP Gas Emissions 1990-2010: Inventories,
Projections,  and Opportunities for Reductions.  U.S.EPA #000-F-97-000. Office of Air and Radiation,
U.S. Environmental Protection Agency. Washington, DC, June 2001.

EPA  (Environmental  Protection Agency).  1999.  U.S.  Methane  Emissions 1990-2020:  Inventories,
Projections,  and Opportunities for Reductions. U.S. EPA #430-R-99-013. Office of Air and Radiation,
U.S. Environmental Protection Agency, Washington, DC, September 1999.
                                                                                        -ES-9-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
1   Analysis of Costs to Abate International HFC  Emissions
     from Refrigeration and Air-Conditioning
1.1        Introduction


A number of hydrofluorocarbons (HFCs) are used in refrigeration and air-conditioning systems and are
emitted to the atmosphere during equipment operation  and repair. Specifically, emissions occur during
product and equipment manufacturing, during equipment servicing, and from the disposal of equipment
and used  refrigerant  containers.  Emissions also  occur during  equipment operation, as  a result of
component failure,  leaks, and purges.  The use of refrigeration and air-conditioning equipment also
generates "indirect" emissions of greenhouse gases (primarily carbon dioxide) from the generation of
power required to  operate the equipment.  In many, but  not  all, refrigeration and air-conditioning
applications, these indirect emissions outweigh the  direct emissions and, hence, energy efficiency has a
major impact on the total greenhouse gas emissions  of an application.  To the extent possible, both direct
and indirect emissions are considered in  the refrigeration/air-conditioning  analysis; however, options
aimed solely at improving energy efficiency rather than  abating HFC emissions are not explored in detail.
HFCs used in this  sector have global warming potentials (GWPs) that range  from 140 to  11,700; the
majority of HFCs used today in the refrigeration and air-conditioning sector have GWPs from 1,300 (i.e.,
HFC-134a) to 3,300 (i.e., R-507A).

The refrigeration and air-conditioning sector includes eight major end-uses:

    •   Household refrigeration;
    •   Motor vehicle air-conditioning;
    •   Chillers;
    •   Retail food  refrigeration;

    •   Cold storage warehouses;
    •   Refrigerated transport;
    •   Industrial process refrigeration; and
    •   Residential  and small commercial air-conditioning/heat pumps.
Each end-use is composed of a variety of different equipment types that have historically used ozone-
depleting substances (ODSs) such as chlorofluorocarbons (CFCs) or hydrochlorofluorocarbons (HCFCs).
As the  ODS phase-out is taking effect under the Montreal  Protocol, equipment is being retrofitted or
replaced to use HFC-based substitutes or intermediate substitutes—HCFCs—that will eventually need to
be replaced by non-ozone depleting alternatives. In time, HCFCs are expected to be replaced with HFCs
or other alternative refrigerants. These end-uses are  explained in more detail below.

Household Refrigeration

The household refrigeration end-use consists of household refrigerators and freezers. HFC-134a is the
primary substitute  for CFC-12  in domestic refrigeration units  in most developing countries,  with
hydrocarbon refrigerant, especially isobutane (HC-600a), dominating much of the European  market and
continuing to grow in market share. HC-600a is also  gaining market share in Japan (Kuijpers, 2002).
                                                                                         -1-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
The charge size of a typical household refrigeration unit has decreased over the past 15 years to about
0.17 kilograms for new HFC-134a units (with HC-600a systems being about 40 percent smaller), and the
equipment has an expected lifetime of about 20 years.  This end-use is one of the largest in terms of the
number of units in use; however, because the charge sizes are small and the units are hermetically sealed
(and, therefore, rarely require recharging), emissions are relatively low.  Thus, the potential for reducing
emissions through leak repair is small.  In most Annex I countries, where regulations are in place that
require the recovery of refrigerant from appliances prior to disposal, the retirement of old refrigerators is
not expected to result in significant refrigerant  emissions.  Refrigerant emissions at disposal from
developing countries, where refrigerant  recovery is not generally required, are expected to be greater.
Emissions from the insulating foam in household refrigerators/freezers are discussed in a separate chapter
of this report.

Motor Vehicle Air-Conditioning

Motor  vehicle air-conditioners (MVACs)  refer  to the air-conditioning  systems contained in motor
vehicles (e.g., cars, trucks, and buses). Currently, the quantity of refrigerant contained in a typical car air-
conditioner is approximately 1 kilogram  (typically from 1 to 1.2 kilograms for vehicles containing CFC-
12 systems, and an  average  of 0.8 kilograms for vehicles containing HFC-134a systems) (Atkinson,
2000).  Due to concerns over the environmental impact of refrigerants, the average charge size of MVACs
-as well as associated leak rates—have been reduced over time, and this trend is expected to continue.
The expected lifetime of MVACs is about 12 years. Refrigerant use in this sector is significant because
there are more than  700 million motor vehicles  registered  globally (Ward's,  2001).  In  developed
countries, CFC-12 was the refrigerant used in MVACs until it was phased out of new cars  from 1992 to
1994, after which all air-conditioners installed in new automobiles used HFC-134a.   HFC-134a is also
used as a retrofit chemical for existing CFC-12 systems (UNEP,  1998).  CFC-12 is still used in MVACs
in developing  countries and its availability  in some developed countries (e.g., the  United States) has
resulted in its use for servicing  older MVACs that were originally manufactured as CFC-12 systems.  A
variety of HCFC/hydrocarbon refrigerant blends are approved for use in the United States by the U.S.
EPA as replacements for CFC-12 in MVACs, although these blends have not been endorsed by vehicle or
system manufacturers for such use. Globally, HCFC/hydrocarbon blends have captured only a small and
declining share of the retrofit market.  Climate change concerns associated with the use of HFC-134a
resulted in the research and development into other MVAC alternatives.  Possible alternatives to HFC-
134a systems include transcritical carbon dioxide systems, hydrocarbons, and HFC-152a systems, all of
which are under study and development (SAE, 2000).

CA7/7/ers

Chillers are used to regulate the temperature and reduce humidity in offices, hotels, shopping centers, and
other large buildings, as well as in specialty applications on  ships,  submarines,  and nuclear power plants
and other industrial applications. The four primary types of chillers are centrifugal, reciprocating, scroll,
and screw—each of which is named for the type of compressor employed.  Chillers are long-lasting
relative  to most  air-conditioning and refrigeration equipment.  Most operating chillers will remain in
service for more than 20 years, and some will last 30 years or more. A wide  variety of chillers are
offered, with cooling capacities from 7  kW to over 30,000 kW  (RTOC,  2003).  The charge size of a
chiller depends mostly on cooling capacity, and ranges from less than 25 kilograms (reciprocating) to
over 2,000 kilograms (centrifugal). HCFC-123 has been the refrigerant of choice as a retrofit option for
newer existing CFC-11 units and HFC-134a has been the refrigerant of choice as a  retrofit option for
newer existing CFC-12 units.   The replacement market for CFC-12 high-pressure chillers  and CFC-11
low-pressure chillers is dominated by both HCFC-123  chillers and HFC-134a chillers in developed and
developing countries.  Following the phase-out of the production of HCFCs  (in 2030 for developed
                                                                                             -2-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
countries and 2040 for developing countries),1 recycled, recovered, and reclaimed HCFCs will continue to
be used in most countries.  This trend is not the case, however, in the EU, where restrictions on the use of
HCFCs in new equipment exist, where the production of HCFCs is not permitted beyond 2010, and where
recycled HCFCs may not be reused beyond 2015. In the EU, HFC-134a will be an important option for
chillers, but because of its global warming impact,  it is being targeted for phase-out there.  Ammonia
chillers are, therefore, being used as an HFC alternative in some EU countries (Kuijpers, 2002).

Additionally, HFC-245fa is a potential refrigerant for use in new design low-pressure chillers to  replace
the market currently dominated by HCFC-123 (in low-pressure chillers), and HFC-134a (in high-pressure
chillers).  However, due to a variety of reasons, the commercialization of this chiller technology is not
likely to occur in the next ten years or so, if at all. High-pressure chillers that currently use HCFC-22 will
ultimately be replaced by several HFC refrigerant blends and HFC-134a chillers. Likewise, most existing
CFC-114 chillers will be replaced with HFC-236fa or HFC-134a chillers, for use primarily in specialty
applications (e.g., on ships, submarines,  and nuclear power plants).

Retail Food Refrigeration

Retail  food  refrigeration includes  refrigerated  equipment found in supermarkets, convenience  stores,
restaurants, and other food service establishments.  This equipment includes small reach-in  refrigerators
and freezers, refrigerated display cases,  walk-in coolers and freezers, and large parallel systems.  Charge
sizes range  from 6 to   1,800 kilograms,  with a lifetime  of 15 to  20 years.   Convenience stores and
restaurants  typically  use  stand-alone  refrigerators,  freezers, and  walk-in  coolers.    In  contrast,
supermarkets usually employ large  parallel systems that connect many  display  cases to  a  central
condensing unit by means of extensive piping. Because the piping required for connection of all the cases
can be miles long, these systems contain very large refrigerant charges and often experience high leak
rates.

During the earlier phases of the CFC phase-out in developed countries, the  use of HCFC-22 in retail food
refrigeration was  expanded considerably.  Retail food equipment is being retrofitted with HCFC-based
blends,  although HFC  blends are  also  used as a retrofit  refrigerant.  The HFC blend R-404A is the
preferred refrigerant being used  in new retail food equipment in developed countries, while R-507A is
also used extensively in the market (Kuijpers, 2002). In developing countries, distributed systems  as well
as centralized systems that use HFCs, hydrocarbons, ammonia, and carbon dioxide are being developed
(both with and without secondary loops) (Kuijpers, 2002).

Cold Storage Warehouses

Cold storage warehouses are used to store meat, produce, dairy products, and other perishable goods. The
expected lifetime of a cold storage warehouse is 20 to 25 years, and while  charge sizes vary widely with
system size  and design, a  rough  average is about 4,000 kilograms.  Warehouses in developed countries
have historically used CFC-12 and R-502 refrigerants, and use HCFC-22 and HFC-134a as replacements
in new equipment. Once HCFCs are phased out, R-404A and R-507A are  expected to replace HCFC-22
in new warehouses.  Retrofits are also possible; for example, existing CFC-12 cold storage warehouses
can be retrofitted with R-401A, and existing R-502 warehouses can be retrofitted with R-402A.  Not all
1 Several scientific studies have found that HCFC-123 in chillers can bring environmental benefits compared to
other refrigerants as a result of its low ODP, very low direct GWP, very short atmospheric lifetime, and the fact that
its design allows for both extremely low emissions (documented at less than 0.5 percent annually [Calm et al., 1999;
Smithart, 2003]) and the highest full-load efficiency (RTOC, 2003). Nonetheless, the emission projections
developed for this analysis assume full compliance with the current HCFC phase-out schedule.
                                                                                             -3-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
cold storage  warehouses currently  use  halocarbon refrigerants.   Many  facilities, for example, use
ammonia in secondary loop brine systems.

Refrigerated Transport

The refrigerated transport end-use includes refrigerated ship holds, truck trailers, railway freight cars, and
other shipping containers. The average charge sizes are relatively small (7 to 8 kilograms). The expected
lifetime of a  refrigerated transport system is 12 years.  Trailers, railway cars, and shipping containers
using CFC substitute refrigerants are commonly charged with HFC-134a, R-404A, and HCFC-22 (UNEP,
1999a). Ship holds, on the other hand, rely on HCFC-22 (UNEP, 1999a) and ammonia.  In addition to
HFC-134a, R-404A can also be used in new equipment.  Existing equipment can be retrofitted with
R-401A R-402A, R-404A, R-507A, and other refrigerants. In addition, refrigerated transport equipment
includes  systems that operate  based on the evaporation and expansion of liquid carbon dioxide or
nitrogen.

Industrial Process Refrigeration

Industrial  process refrigeration  includes complex,  often custom-designed, refrigeration systems used
within  the chemical industry, petrochemical industry, pharmaceutical industry,  oil and gas industry,
metallurgical  industry, sports and leisure facilities, and many other applications. Charge sizes typically
range from 650 to 9,100 kilograms  (although charge sizes for some applications can fall  outside this
range), and the average  lifetime is approximately 25 years.  Ammonia, hydrocarbons, HCFC-123, and
HFC-134a are expected to be the most widely used substitute refrigerants for new equipment in the near
future  (UNEP,  1999a).  Assuming no changes  to the requirements  of the Montreal Protocol  and no
additional restrictions on refrigerant choice, upon completion of the HCFC phase-out, HFC-134a, R-404A
and R-507A are expected to be the primary refrigerants used in this end-use.

Residential and Small Commercial Air-Conditioning/Heat Pumps

Residential and small commercial air-conditioning (window units,  unitary air-conditioners,  packaged
terminal air-conditioners) and heat pumps are another source of HFC emissions. Most of these units are
window and through-the-wall units, ducted central air-conditioners, and non-ducted split systems.  The
charge sizes of the equipment in this  sector range based on cooling capacity requirements, and are on the
order of 0.5 to 10 kilograms  for residential systems, and about  10 to 180 kilograms for  commercial
systems.  The average lifetime of equipment is 15 years. Residential and commercial air-conditioning has
been relying almost  exclusively on HCFC-22 refrigerant.  R-410A, R-407C, and HFC-134a are currently
being used to replace HCFC-22  in some new equipment for most end-uses, and this trend is expected to
continue as HCFC-22 is phased out.   In particular, R-410A is expected to dominate the U.S. residential
market in the future, while R-407C is expected to replace HCFC-22 mainly in retrofit applications and
some new residential equipment. Other countries may experience different patterns of R-410A and R-
407C use.
                                                                                            -4-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
1.2        Baseline Emission Estimates


1.2.1  Emission Estimating Methodology

Description of Methodology

A full description of the emission model used to calculate ODS substitute emissions from all sectors is
provided in Appendix  A.   Specific information on  how the model calculates  refrigeration and air-
conditioning emissions is described below.

The U.S. EPA's Vintaging Model and data from industry are used to simulate the aggregate impacts of
the ODS phase-out on the use and emissions of various fluorocarbons and their substitutes in the United
States (see Appendix A).  Emission estimates for  non-U.S. countries incorporate estimates of the
consumption of ODSs by country, as provided by the United Nations Environment Programme (UNEP,
1999b).  The estimates for the European Union (EU) were provided in aggregate and each country's gross
domestic product (GDP) was used as a proxy to divide the consumption of the individual member nations
from the EU total.  Estimates  of country-specific  ODS consumption as reported under the Montreal
Protocol were then used in conjunction with Vintaging Model output for each ODS-consuming sector. In
the absence of country level data, preliminary estimates of emissions were calculated by assuming that the
transition  from ODSs to HFCs and other substitutes  follow the  same general substitution patterns
internationally as observed in the United States.  From this preliminary assumption, emission estimates
were then tailored to individual countries or regions by applying  adjustment factors to U.S. substitution
scenarios,  based on relative differences in (1) economic growth; (2) rates of ODS  phase-out; and (3) the
distribution of ODS use across end-uses in each region or country, as explained below and in further
detail in Appendix A.

Emission Equations

For refrigeration and air-conditioning  products, emission calculations  are split into two  categories:
emissions  during equipment lifetime, which arise from annual leakage and service losses, and disposal
emissions, which occur at the time of discard. The first equation calculates the emissions from leakage
and service, and the second equation calculates the  emissions resulting from disposal of the equipment.
These service/leakage emissions and disposal emissions are summed to calculate the total emissions from
refrigeration and air-conditioning. As new technologies replace older ones, improvements in their leak,
service, and disposal emission rates are assumed to occur.

Emissions from any piece of equipment include both the amount of chemical  leaked during equipment
operation and the amount emitted during service. Emissions from leakage and servicing can be expressed
as follows:

                             ESj  = (la + /SJ * X QCj-M   fori=1^k

Where:

Es =  Emissions from Equipment Serviced. Emissions in year/ from normal leakage and servicing of
       equipment.
la  =  Annual Leak Rate. Average  annual leak rate during normal equipment operation (expressed as a
       percentage of total chemical  charge).
                                                                                           -5-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
ls   =   Service Leak Rate. Average annual leakage from equipment servicing (expressed as a percentage
        of total chemical charge).
Qc =   Quantity of Chemical in New Equipment.  Total amount of a specific chemical used to charge
        new equipment in a given year by weight.
j    =   Year of emissions.
        Counter.  From 1 to lifetime (k).
        Lifetime.  The average lifetime of the equipment.
Note: It is recognized that leakage rates are not a function of the total system, but change with system pressure and
temperature.  For instance, when equipment charges are diminished due to  refrigerant losses (leakage), system
pressures are also reduced somewhat and the leakage rate changes. This becomes appreciable once the entire liquid
refrigerant is gone.  The average leak rates used in the equation above account for this effect.  They also account for
the range of equipment types (from those that do not leak at all to those with high leaks) and service practices (i.e.,
proper refrigerant recovery and refrigerant venting).
Emissions also result at equipment disposal.   The disposal emission  equations assume that a certain
percentage of the chemical charge  will  be emitted to the atmosphere when that vintage is discarded.
Disposal emissions are thus a function of the quantity of chemical contained  in the retiring equipment
fleet and the proportion of chemical released at disposal:

                              Edj = QCj.k+i x [1 - (rm  x re)]

Where:

Ed =   Emissions from Equipment Disposed. Emissions in year/ from the disposal of equipment.
Qc =   Quantity of Chemical in New Equipment.  Total amount of a specific chemical used to charge
        new equipment in year/-£+./, by weight.
rm =   Chemical Remaining. Amount of chemical remaining in equipment at the time of disposal
        (expressed as a percentage of total chemical charge).
re  =   Chemical Recovery Rate. Amount of chemical that is recovered just prior to disposal (expressed
        as a percentage of chemical remaining at disposal (rm)).
j    =   Year of emissions.
        Counter.  From 1 to lifetime (k).
        Lifetime.  The average lifetime of the equipment.

Finally, lifetime and disposal emissions are summed to provide  an estimate of total emissions.
                              Ej = ESj + Edj
Where:
E  =   Total Emissions.  Emissions from refrigeration and air-conditioning equipment in year/.
Es  =   Emissions from Equipment Serviced.  Emissions in year/ from normal leakage and servicing
        (recharging) of equipment.
Ed  =   Emissions from Equipment Disposed. Emissions in year/from the disposal of equipment.
j    =   Year of emissions.

Regional Variations/A djustments

From the general methodology described in Appendix A,  the  following regional assumptions were
applied:

    •   Adjustment for EC-Regulation No. 2037/2000.  Countries in the European Union are assumed
        to  be  in  full  compliance with EC-Regulation No.  2037/2000, which stipulates  that no  new
                                                                                             -6-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
       refrigeration and air-conditioning equipment be manufactured with HCFCs as of January 1, 2002.
       The EC regulation also bans  the use  of HCFCs  in  all equipment after January 1, 2015.
       Compliance with these regulations will likely  lead to increased use of HFCs to replace HCFCs,
       and is assumed to correspond to increased emissions of 20 percent in 2005, 15 percent in 2010,
       and 15 percent in 2020, relative to what the EU baseline otherwise would be (see Appendix A for
       an explanation of how these baseline refrigeration and air-conditioning emissions are calculated
       before applying this adjustment). These relative emission increases were determined by running
       a Vintaging Model scenario wherein the uses  of HCFCs were  assumed to comply with  the
       regulation.
       Recovery/recycling adjustments. The emissions from developing (or "non-Annex I") countries,
       Countries with Economies in Transition (CEITs) and Turkey were increased by approximately 20
       percent over initial estimates to reflect the assumed low levels  of recovery and recycling of
       refrigerants  from  small  end-uses  (i.e.,  MVACs,  commercial/residential   air-conditioning,
       refrigerated transport,  and other  appliances), relative to the United States. This assumed increase
       in emissions from lower levels of recovery/recycling was derived based on Vintaging Model test
       runs,  wherein emissions were first  projected assuming an 80-percent baseline recovery rate (to
       reflect the assumed status quo in developed countries) and then projected again assuming a  30-
       percent baseline recovery rate (to reflect the assumed status quo in developing countries).  The
       GWP-weighted  emissions   in  the  latter  low-recovery  scenario were  determined   to  be
       approximately 20 percent higher than the former high-recovery scenario (ICF Consulting, 2002a).

       Market adjustments. The baseline assumes that hydrocarbon and ammonia refrigerants and
       other non-HFC or low-emitting options  will  penetrate international markets  more than in  the
       United States, because of more relaxed safety  standards, greater acceptance of non-HFC choices
       by industry, end-users, regulators, and insurance companies, and increased public and regulatory
       scrutiny to reduce HFC emissions.  To  reflect this, baseline emission estimates of non-U.S.
       countries were reduced by the following amounts:
           -   European Union:        30%
           -   Japan:                 30%

           -   Non-EU Europe:        25%
           -   CEITs:                 20%
           -   Australia/New Zealand:  10%
           -   All Other Countries:    20%

       These assumptions were based  solely on qualitative information on current and  future  global
       market penetration of  low-GWP refrigerants, as well as low-emission technologies and practices.
       For example, hydrocarbon  technology is  believed to now dominate the domestic refrigeration
       market  in Western Europe, particularly in Germany and  Scandinavia.  Hydrocarbon domestic
       refrigerators are produced  by major manufacturers  in Germany,  Denmark, Italy, Japan, UK,
       France, Spain,  and Sweden.  Some  of the largest manufacturers in  China, India, Indonesia,
       Australia, Korea, and Cuba are also producing domestic refrigerators that use hydrocarbons
       (Greenpeace, 2001; Japan  Times,  2002).  To reflect this and many other trends,  baseline
       emissions from non-U.S. countries are adjusted downwards as shown above.

       Redistribution of emissions by end-use, based on MVAC analysis.  Based on a variety of
       available data on international motor vehicle sales, air-conditioning usage, and MVAC emissions,
       a separate analysis was conducted to  estimate  total MVAC emissions by region. These MVAC
       emission estimates by region were then used to determine the relative  share of refrigeration/air-
                                                                                            -7-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
       conditioning emissions attributable to MVACs, and to reapportion emissions from all other end-
       uses  accordingly, relative to the  end-use  breakout calculated  for  the  United States.   The
       methodology used to perform this analysis is explained in detail below.
MVAC Analysis

Because  MVACs  are  expected to account  for  a  much higher proportion  of total refrigeration/AC
emissions in the United States compared to most other developed and, especially, developing countries
(due to the number of air-conditioned automobiles on the road), this end-use has been modeled separately
to achieve a higher degree of accuracy in emission estimates.

Specifically, for all countries for which data on historical vehicle sales were available in Ward's World
Motor Vehicle Data (2001), country-specific models  were developed to estimate emissions from MVACs.
To populate this model, the number of operational MVACs was estimated based on annual historical sales
of passenger cars and light trucks (as  provided in Ward's (2001)) and on estimates of the percentage of
the vehicle fleet equipped with air-conditioning units. Based on quantitative and qualitative data provided
in EC (2003a), Hill and Atkinson (2003), OPROZ (2001) and Barbusse et al. (1998), Table 1-1 shows the
percent of current and future vehicle fleets assumed  to be equipped with  operational air-conditioning
units:

Table 1-1: Percent of On-Road Vehicle Fleet Assumed to  Have Operational Air-Conditioning Units
Country/Region
United States & Japan
All Other Annex I countries
Latin America & Caribbean
All Other Non-Annex I countries, Russia, CEITs
2000
95.0%
58.0%
45.0%
18.0%
2005
98.0%
65.5%
50.0%
23.0%
2010
99.0%
70.0%
55.0%
28.0%
2015
99.0%
80.5%
60.0%
33.0%
2020
99.0%
95.0%
65.0%
38.0%
As shown above, MVACs in Japan are assumed to have reached the same degree of market penetration
within the vehicle  fleet as in the United  States.   In all  other  countries,  MVACs  are  assumed  to
increasingly penetrate  the  vehicle fleet over time.  In developing countries, this rate of increase  is
assumed to be one percent each year, while in all other Annex I countries, the rate of increase is assumed
to be more rapid, reaching 95 percent of the vehicle fleet in 2020 (EC, 2003a; Hill and Atkinson, 2003).

Based on historical estimates of vehicle fleet size and assumptions about the percent of vehicle fleets with
operational air-conditioners (Baker, 2002),  current  and future  MVAC fleets were calculated for each
country.  From these estimates, MVAC emissions were calculated by assuming that, as a result of low
levels of refrigerant recovery and recycling, MVAC emissions in developing countries are approximately
20 percent greater than emission rates in the United States  (where annual  average emissions for this
analysis are assumed to be  10.9 percent2 and emissions at disposal are assumed to be 42.5 percent of the
total MVAC charge).3 MVAC charge size and loss rates were decreased in later years, as it is assumed
that MVAC systems become tighter and require less refrigerant over time.  All systems are assumed to be
HFC-134a in the baseline.

Once MVAC  emissions were estimated for those countries  for which  Ward's data were available, the
proportion of MVAC emissions as a percent of the total refrigeration/AC emissions (developed using the
methodology described above) was calculated.  These percentages were then averaged by region.  The
2 This emission rate includes emissions released during routine equipment operation from leaks, as well as those
released during the servicing of equipment by both professionals and do-it-yourselfers.

3 While these emission rates differ slightly from those reported in recent studies conducted for the European
Commission (EC, 2003a; 2003b), overall MVAC emissions based on U.S. EPA and EC estimates are comparable.

-------
Analysis of Costs to Abate International OPS Substitute Emissions
average  estimated  percentage  of  refrigeration/air-conditioning  GWP-weighted  emissions  that  are
attributable to MVACs by regional grouping are presented in Table 1-2.

Table 1-2: Estimated Percent of Refrigeration/Air-Conditioning HFC Emissions Attributable to MVACs
Country/Region
United States & Japan
All Other Annex I Countries
Latin America & Caribbean
Russia, CEITs & All Other Non-Annex I Countries
2005
37.0%
48.7%
18.7%
9.5%
2010
28.5%
40.4%
16.2%
10.7%
2015
22.7%
20.5%
10.7%
10.9%
2020
19.5%
15.9%
7.8%
13.6%
Based on the above percent of sector baseline emissions assumed to come from MVACs for each region,
the U.S. baseline emissions breakout by end-use was used to proportionally redistribute the remaining
emissions of a particular country/region.  Thus, for example, because MVACs contribute only 4.0 percent
of total sector  emissions in Latin American countries in 2005, the balance of emissions in Latin America
was distributed across all other end-uses, in proportion to the U.S. end-use  breakout. Therefore,  if the
emissions in Latin America in 2005 were estimated to be 10 MMTCE, this analysis assumes that 0.4
MMTCE are from MVACs  and the remaining 9.6 MMTCE are from other sector end-uses.  The resulting
breakout of baseline GWP-weighted HFC emissions by end-use and region are summarized in Table 1-3.
These emission breakouts by end-use help determine the maximum amount of emissions that can be
avoided by any given abatement option (since each option is applicable only to specific end-uses).

Table 1-3:  Distribution of Refrigeration and Air-Conditioning Sector HFC  Emissions by End-use, Region, and
Year
End-use
United States & Japan
All Other Annex I
Latin America &
Caribbean
All Other Non-Annex I,
Russia, CEITs
2005
Chillers
Retail Food
Cold Storage
Industrial Process
Commercial A/C
Residential A/C
Refrigerated Transport
Other Appliances
MVACs
3.2%
38.5%
1.2%
4.5%
0.7%
0.6%
13.8%
0.5%
37.0%
2.6%
31.3%
1.0%
3.7%
0.6%
0.5%
11.3%
0.4%
48.7%
4.1%
49.6%
1.6%
5.8%
0.9%
0.8%
17.9%
0.6%
18.7%
4.6%
55.3%
1.7%
6.5%
1.0%
0.9%
19.9%
0.7%
9.5%
2010
Chillers
Retail Food
Cold Storage
Industrial Process
Commercial A/C
Residential A/C
Refrigerated Transport
Other Appliances
MVACs
2.3%
42.1%
1.4%
6.1%
3.7%
5.6%
9.8%
0.4%
28.5%
1.9%
35.1%
1.2%
5.1%
3.1%
4.7%
8.2%
0.4%
40.4%
2.7%
49.4%
1.7%
7.1%
4.4%
6.5%
11.5%
0.5%
16.2%
2.9%
52.6%
1.8%
7.6%
4.6%
7.0%
12.2%
0.5%
10.7%
2015
Chillers
Retail Food
Cold Storage
Industrial Process
1.8%
42.6%
1.5%
6.6%
1.9%
43.9%
1.5%
6.8%
2.1%
49.3%
1.7%
7.6%
2.1%
49.2%
1.7%
7.6%
                                                                                             -9-

-------
Analysis of Costs to Abate International OPS Substitute Emissions


End-use
Commercial A/C
Residential A/C
Refrigerated Transport
Other Appliances
MVACs

United States & Japan
6.4%
9.9%
7.4%
1.0%
22.7%

All Other Annex I
6.5%
10.2%
7.6%
1.1%
20.5%
Latin America &
Caribbean
7.3%
11.5%
8.6%
1.2%
10.7%
All Other Non-Annex I,
Russia, CEITs
7.3%
11.4%
8.5%
1.2%
10.9%
2020
Chillers
Retail Food
Cold Storage
Industrial Process
Commercial A/C
Residential A/C
Refrigerated Transport
Other Appliances
MVACs
1.6%
41.6%
1.5%
7.0%
8.3%
13.0%
6.5%
0.8%
19.5%
1.6%
43.5%
1.6%
7.4%
8.7%
13.6%
6.8%
0.9%
15.9%
1.8%
47.7%
1.8%
8.1%
9.6%
14.9%
7.5%
1.0%
7.8%
1.7%
44.7%
1.6%
7.6%
8.9%
14.0%
7.0%
0.9%
13.6%
Note: Totals may not sum to 100% due to independent rounding.

1.2.2   Baseline  Emissions

The amount of HFC emissions from MVAC units is expected to  rise,  since HFC-134a has been the
primary refrigerant  used in the growing  automobile industry, and because it is the primary refrigerant
used to replace older CFC-12 systems.   The baseline for MVACs  assumes a mix of professionally
serviced systems and those  serviced  by unskilled people without  recovery equipment (i.e.,  "do-it-
yourselfers").  Since commercial unitary  and residential air-conditioning equipment has yet to transition
fully into HFCs, the emissions of HFCs from these end-uses in 2005  are estimated to be relatively
insignificant, but will increase substantially over time.  Retail food systems are expected to transition at
least in part to HFC-134a and HFC-containing blends, and because of certain equipment characteristics,
such as their large  number of fittings, often have higher refrigerant emission rates.  Because of the
relatively high cost of refrigerant and the large charges associated with  retail food equipment, the baseline
assumes that an inherently strong economic incentive exists to closely monitor refrigerant leakage in these
types of equipment.  Cold storage systems also have large charge sizes, but their emissions relative to
other refrigeration  and air-conditioning  end-uses  are  not  expected  to increase  significantly.   HFC
emissions from chillers are  relatively low  as a result of the  continued use  of HCFC-123  in this
application,4 as well as the low leak rates of new HFC-134a units.  The baseline emission projections
assume that the recovery and recycling of refrigerants during service  and disposal in Annex I countries
will curtail emissions across all of the end-uses.

The resulting baseline estimates of HFC emissions are summarized in Table 1-4.
Table 1-4: Baseline HFC Emission Estimates from Refrigeration and Air-Conditioning (MMTCE)
 Region                  2005          2010          2015          2020
 United States
26.87
40.08
53.95
67.84
4 Note the emissions of all CFC and HCFC refrigerants, including HCFC-123, are not included in the baseline
emission estimates.
                                                                                             -10-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Region
Non-U.S. Annex 1
Non-Annex 1
Total
2005
18.11
14.94
59.92
2010
25.66
28.77
94.50
2015
35.70
39.08
128.73
2020
38.11
52.86
158.80
Note: Totals may not sum due to independent rounding.

1.3        Costs of HFC Emission Reduction from Refrigeration/AC


This section presents a cost analysis for achieving HFC emission reductions from the emission baselines
presented above.  Each abatement option is described below, but only those options not assumed to occur
in the baseline and for which adequate cost data are available are included in the cost analysis.  To the
extent possible, this  analysis considered total equivalent warming impacts (TEWI), to  account for the
climate and cost impacts associated with energy consumption (i.e., indirect emissions).  Due to  data
limitations, a full life cycle analysis was not possible. For example, the cost and emission impacts
associated with (a) the manufacture of refrigerant and all system components, (b) the energy required for
reclamation, and (c) the recycling of all system components at the end of equipment life were not assessed
in this analysis.

The remainder of this section provides a description of the economic assumptions for these abatement
options.


1.3.1  Description and Cost Analysis of Abatement Options

HFC emissions from refrigeration and air-conditioning equipment can be reduced through a variety of
practice and technology options. Many of the options considered in this report would  entail voluntary
action by the private sector and/or further government regulation.  For example, national governments can
regulate maximum allowable leak rates for refrigeration and air-conditioning equipment, or require the
recovery of refrigerant and the proper disposal of non-reclaimable refrigerant.  Many Annex I countries
have already implemented a variety of such regulatory actions to reduce  ODS emissions.  Some of the
most widely recognized  options to reduce refrigerant emissions include the following (UNEP, 1998;
UNEP, 1999a; Crawford, 1999; U.S. EPA, 2001a):

Practice Options:

    •   Leak repair;
    •   Refrigerant recovery and recycling;

    •   Proper refrigerant disposal;  and

    •   Technician certification/HFC sales restriction

Alternative Refrigerant Options:

    •   Ammonia;

    •   Hydrocarbons;
    •   Low-GWP refrigerants; and

    •   Carbon dioxide.
                                                                                         -11-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Technology Options:
    •  Distributed systems5 for stationary commercial refrigeration equipment;
    •  Secondary loop systems for stationary equipment;
                  —  HFC secondary loop systems
                  —  Ammonia secondary loop systems
    •  Enhanced HFC-134a systems in MVACs;
    •  HFC-152a refrigerant in MVACs (direct expansion or secondary loop systems);
    •  Carbon dioxide systems in MVACs;
    •  Oil-free compressors;
    •  Geothermal (in lieu of air-to-air) cooling systems;
    •  Desiccant cooling systems; and
    •  Absorption systems.
Table 1-5 summarizes the duration and applicability of the process and technology emission reduction
options across all end-use applications considered in this analysis.  The  applicability of the  alternative
refrigerant options depends on the technology used; hence, some are explored in more detail  in the
analysis of technology options.
5 The term "distributed system" as used in this report refers to commercial refrigeration equipment applicable to the
retail food and cold storage end-uses, although the term could also be used to refer to technology applicable to other
uses, such as residential and small commercial air-conditioning.
                                                                                              -12-

-------
     Analysis of Costs to Abate International OPS Substitute Emissions
     Table 1-5: Assumptions on Duration and Applicability of Emission Reduction Options
Option Description

Practice Options
Leak Repair
Refrigerant Recovery
Proper Refrigerant Disposal
Technician Certification
Duration of Emission Potential Applicability to End-use Equipment
Reduction
(years)


5
1
NA
NA
Chillers Retail Cold Refrigerated Industrial Commercial MVACs Residential Household
Food Storage Transport Process Unitary A/C A/C Refrigeration
Warehouses Refrigeration and Small
Appliances

+ + + • + ...
+ . + + + +
.
.
Alternative Refrigerants
Ammonia
Hydrocarbons
Low-GWP Refrigerants
Carbon Dioxide
Lifetime of equipment
Lifetime of equipment
Lifetime of equipment
Lifetime of equipment
Technology Options
Distributed Systems for Stationary
Commercial Refrigeration Equipment
Secondary Loop Systems for Stationary
Equipment—HFC Primary Refrigerant
Secondary Loops Systems for Stationary
Equipment—Ammonia Primary
Refrigerant
Enhanced HFC-134a Systems in MVACs
HFC-152a  in MVACs (direct expansion
or secondary loop)
C02 in MVACs
Oil-Free Compressors
Geothermal (in lieu of air-to-air) Cooling
Systems
Desiccant Cooling Systems
Absorption Systems
Lifetime of equipment

Lifetime of equipment
Lifetime of equipment
Lifetime of equipment
Lifetime of equipment

Lifetime of equipment
Lifetime of equipment
Lifetime of equipment

Lifetime of equipment
Lifetime of equipment
+ Option is technically feasible and is addressed in the cost analysis of this report.
• Option is potentially feasible but is not addressed in the cost analysis of this report.
                                                                                                                                          -13-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
The following section describes all of these options in greater detail and presents a cost analysis for those
options not assumed to occur in the baseline and for which adequate cost data are available. The resulting
emission abatement potentials and costs of each option explored in the cost analysis are  summarized in
Section 1.4.  The technology options explored in this chapter do not include retrofit costs and, therefore,
are assumed to penetrate only the markets of new (not existing) equipment.  "New" equipment is defined
as air-conditioning/refrigeration equipment manufactured in 2005 or later.

Practice Options

Leak Repair for Large Equipment

Reducing leak rates can reduce HFC emissions to a significant degree, especially in large systems such as
chillers, cold storage warehouses, and retail food  systems that could leak large amounts of refrigerant.
Although some  of the  options available  for existing  equipment  may  be  impractical  for in-place
equipment, given the difficulty and expense of retrofitting, there are still many options available that are
economically feasible.  Some of the leak repair options used in current industry  practice include (U.S.
EPA, 1997; U.S. EPA, 1998; Calm, 1999):

    •  Use of preventive maintenance, including scheduling inspection and repairs;
    •  Monitoring of leaks using stationary leak monitors or other new technologies, such as early
       warning signals,6 remote monitoring, and diagnostics;
    •  Use of new, more durable gasket materials  that provide tighter seals and absorb less refrigerant;
    •  Augmentation of threaded joints with O-ring seals;
    •  Augmentation or replacement of gaskets and O-rings with adhesive sealants;
    •  Broader use and improvement of brazing  techniques rather than threaded or snap fittings (e.g.,
       use of sufficient silver content,7 and use of dry nitrogen or other inert gas to avoid oxidation);
    •  Focus on ensuring accessibility to field joints and use of isolation valves, which allow for greater
       ease of repair;
    •  Focus on proper  securing to reduce vibration  fractures in the pipe and  connections  from  the
       compressor and other moving parts of the system;

    •  Repair or retrofit of high-emitting systems through targeted component upgrades;8 and
    •  Performance of major modifications to the  systems.9
6 Technologies in the final stages of development expect to be able to generate early warning signals at less than five
percent charge loss in commercial refrigeration and air-conditioning systems (Gaslok, 2002).
7 For solder, a 15-percent silver content is recommended (U.S. EPA, 1997).
8 This option may include replacing the purge unit or other component upgrades that typically require the removal of
refrigerant from the machine, two full days of two technicians' time, and several thousand dollars worth of materials
(U.S. EPA, 1998).
9 This option may include modifications that are not strictly leak repair, but would result in greatly reduced leak
rates. For example, combining the installation of a new purge system, the replacement of flare joints, and other
containment options, or combining the replacement of gaskets and seals, the replacement of the motor, and the
installation of new refrigerant metering.
                                                                                              -14-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
As suggested by the  above list, leak reduction  options  range from  simple repairs to major system
upgrades.  Even in countries where maximum allowable  leak  rates are  regulated by law, further leak
reduction improvements,  such as the  replacement or upgrade  of a major system component, are still
possible.  For example, preliminary data gathered from U.S. industry indicate that leak  rates for certain
types of existing equipment in the United States range from 8 to 40 percent, whereas achievable leak rates
for new or modified equipment can be between 4 to 15 percent.  Furthermore, through additional design
and practice  changes,  leakage  could be reduced to as low as 0.25  to 5 percent in the future (Amrane,
2001).  Already, some retail food equipment has reached leak rates of less than 10 percent, and even
lower for new chiller equipment (Crawford, 2002; Amrane, 2001).

Since the lower cost leak reduction options represent significant cost savings, this analysis assumes that
they occur under the baseline.  The cost analysis therefore focuses only on the more extensive and costly
options.  This option is assumed to be technically applicable to all equipment with large charge sizes (i.e.,
chillers, retail food, cold  storage, and  industrial process refrigeration).10  This analysis assumes that 50
percent of emissions occur as a result of equipment leakage during routine operation, while the other 50
percent of emissions are released  during equipment servicing and disposal. Thus, the maximum technical
applicability  of this option is assumed to be 50 percent of emissions from large equipment (see Table
1-6).   Furthermore, because equipment will still  leak a marginal amount  once repairs are made, it is
assumed that the reduction efficiency of this option is 95 percent. The project lifetime is estimated to be
five years. Regional technical applicability and reduction efficiency for 2010 and 2020  are presented in
Table 1-6. Assumptions on maximum  market penetration for each region  and year are presented in Table
1-16.

Table 1-6: Summary of Assumptions for Leak Repair for Large Equipment
Country/Region
U.S. & Japan
Other Annex I
Latin America & Caribbean
Other Non-Annex I, Russia, CEITs
Applicable End-
Uses3
Chillers
Retail food
Cold storage
Industrial process
Reduction
Efficiency3
95%
Technical Applicability11
2010 2020
26.0% 25.9%
21.7% 27.0%
30.5% 29.7%
32.5% 27.8%
a End-Uses and Reduction Efficiency apply to all regions.
b Technical applicability is shown as a percent of total refrigeration/air-conditioning sector emissions, and equals 50 percent of total
refrigeration/air-conditioning emissions from chillers, retail food, cold storage, and industrial process refrigeration.

Cost and Emission Reduction Analysis

The following bullets describe the cost and emission inputs used to derive the final  $/TCE for the leak
repair option, the results of which are presented in Section 1.4:

    •    One-Time Costs.  A one-time cost of $1,450 is assumed for performing more significant minor
        repairs on larger systems, such as maintenance of the purge system or replacement of a gasket or
        O-ring, etc.  This cost is based on an estimate provided in U.S. EPA (1998), adjusted to 2000
        dollars, which assumes that $200 in parts and 16 hours of labor is required to perform the repair.
    •    Annual Costs. No annual costs are associated with this option.
10 In this report, the term technically applicable refers to the emissions to which an option can be applied. The Leak
Repair option is assumed to be technically applicable to all emissions from leaks (but not servicing and disposal)
from the four end-uses listed in Table 1-6. See Appendix I for a glossary of terminology.
                                                                                               -15-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •   Cost Savings. An annual cost savings is associated with reduced refrigerant loss.  The cost of
        refrigerant (assumed to be R-404A) is estimated to be $18.41/kg,n and 43 kg of refrigerant per
        unit are assumed to be saved  each year, resulting in an annual cost savings of approximately
        $791.63 per unit.
    •   Emission Reductions.  Under the leak repair option described above, approximately 43 kg of R-
        404A refrigerant can be avoided each year for five years, resulting  in  an annual emission
        reduction of approximately  36 TCE per job (i.e.,  43 kg or 0.043  metric tons of refrigerant,
        multiplied by the GWP of 3260, the reduction efficiency of 0.95, and the conversion factor of
        carbon to carbon dioxide equivalents of 12/44).
Refrigerant Recovery and Recycling from Small Equipment

Recovery and recycling of HFCs help to decrease HFC emissions during equipment service and disposal.
The approach involves the use of a refrigerant recovery device that transfers  refrigerant into an external
storage  container prior to servicing of the equipment. Once the recovery process and source operations
are  complete, the refrigerant contained  in the storage container may be  recharged back  into  the
equipment, cleaned through the use of recycling devices,  sent to a reclamation facility to be purified,12 or
disposed through the  use of incineration technologies.  Refrigerant recovery may also  be an  important
way to reduce emissions  from near-empty refrigerant containers (i.e., can heels).  Refrigerant recovery is
assumed to be widely practiced in Annex I countries in the baseline, as it is typically required by law.

This analysis assesses the  recovery of refrigerant  above that which is  already practiced (e.g., due to
regulations in many developed countries or for  economic reasons) at service and disposal only for small
equipment—i.e.,  MVACs,  refrigerated transport, household  and other small appliances,  and unitary
equipment—because it is assumed that recovery from large equipment is already widely practiced in the
baseline.13  Recovery for large equipment is assumed  to be practiced in the baseline because  of the
significant  cost savings associated with recovery of large quantities of refrigerant from this equipment.
Because emission reductions and  costs vary by scenario and end-use,  emission reductions  and costs
associated with four recovery scenarios were averaged to obtain one break-even cost. The  four scenarios
studied  are recovery/recycling of (1) MVACs at service,  (2) MVACs at disposal, (3) small appliances at
service, and (4) small appliances at disposal.

This analysis assumes that 50 percent of emissions are released during equipment servicing and disposal,
while the remaining 50 percent occur as a result of leakage during normal operations.  Thus, the technical
applicability  of this  option is 50 percent of emissions  from  small  equipment (see Table  1-7).
Furthermore, because  in the United States small appliances are considered completely recovered when 90
percent  of the refrigerant is removed from units with running compressors,  or when 80 percent of the
refrigerant  is removed  from  units  with  non-operating compressors, this  analysis assumes  that  the
reduction efficiency  of this option  is  85 percent (Contracting  Business Interactive, 2003; U.S.  EPA,
1993).  The project lifetime is assumed to be one year.  Regional technical  applicability  and reduction
11 Cost of R-404A is based on the list price quoted by DuPont Customer Service (2004).
12 Recycling cleans, and reclamation purifies, recovered refrigerant; reclamation is more thorough and involves
repeated precision distillation, filtering, and contaminant removal. Recycling is used for onsite servicing of MVACs
and other equipment and reclamation requires sending the refrigerant offsite to a reclaimer.
13 While the Society of Automotive Engineers (SAE) has issued industry standards on equipment and technician
procedures that apply to MVACs and provide for on-site recovery/recycling of HFC-134a from MVAC systems for
reuse in the serviced system, recovery from these and other small systems is still not believed to be widely practiced
in most developing countries, as a result of a lack of infrastructure (i.e., recovery/recycling equipment) (World
Bank, 2002).
                                                                                              -16-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
efficiency for 2010 and 2020 are presented in Table 1-7.  Recovery of small appliances and MVACs is
assumed to be  practiced at 80 percent in the baseline in developed countries and at 30 percent in the
baseline in developing countries.  Assumptions on maximum market penetration for each region and year
are presented in Table 1-16.

Table 1-7: Summary of Assumptions for Recovery/Recycling from Small Equipment
Country/Region
U.S. & Japan
Other Annex I
Latin America & Caribbean
Other Non-Annex I, Russia, CEITs
Applicable End-Uses3
MVACs
Refrigerated transport
Household/ other small appliances
Commercial unitary AC
Residential AC
Reduction
Efficiency3
85%
Technical
Applicability'^
2010 2020
24.0% 24.1%
28.3% 23.0%
19.5% 20.3%
17.5% 22.2%
a End-Uses and Reduction Efficiency apply to all regions.
b Technical applicability is shown as a percent of total refrigeration/air-conditioning sector emissions, and equals 50 percent of total
refrigeration/air-conditioning emissions from MVACs, refrigerated transport, household/other small appliances, and commercial unitary and
residential air-conditioning.

Cost and Emission Reduction Analysis

The following  bullets describe the  cost and emission inputs used to derive the final $/TCE for the
refrigerant recovery option, the results of which are presented in Section 1.4:

    •    One-Time Costs.  The one-time cost associated with this option is the cost of recovery and/or
        recovery/recycling equipment, which ranges based on the equipment type. While all costs were
        expressed  in terms of cost  per job  (see below), the cost of a high-pressure  recovery  unit is
        assumed to be approximately $815,  based on the average cost  of four recovery scenarios: (1)
        recovery from MVACs at service; (2) recovery  from MVACs  at disposal; (3) recovery from
        stationary  equipment at service; and (4) recovery from stationary equipment at disposal (U.S.
        EPA, 1998;2001b).
    •    Annual Costs.  All costs associated  with this option, including capital costs, were expressed in
        terms of cost per job. The cost per job was calculated by multiplying the average additional labor
        required by the technician to recover the  refrigerant charge (from five to 10 minutes, depending
        on recovery scenario) by the average labor rate ($50/hr) and the average operating costs (which
        incorporates  both the annualized costs of equipment  and energy  use).   Based on this
        methodology, the average job is estimated to cost approximately $9.50 (U.S. EPA, 1998; 2001b).
    •    Cost Savings. As a result of the average cost of recovered refrigerant (R-134a, R-404A, and R-
        407C)—calculated to  be $11.32/kg (the  average  for  the   four scenarios described  above,
        considering the likely refrigerants involved)—and an average  recoverable charge of 0.83 kg (the
        average of the four recovery  scenarios described above,  considering the 85-percent reduction
        efficiency), this option is associated with a cost savings of approximately $9.40 per job.
    •    Emission Reductions. Under the refrigerant recovery option described above, the emission of
        0.83  kg of  refrigerant can  be avoided  from small  equipment, resulting in the  reduction of
        approximately 0.38 TCE per recovery job (assuming an average GWP of 1,664).14
14 This GWP value is based on an average GWP for MVACs and stationary equipment, where the average GWP for
MVACs is assumed to be 1,300 (R-134a), and the average GWP for stationary equipment is assumed to be 2,028.5
(R-134a, R-404A, and R-407C). Therefore, the calculation is: [1300+ (1300 + 3260 + 1525.5)/3]/2 = 1664.25
                                                                                              -17-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Proper Refrigerant Disposal

One potential source of emissions from the refrigeration and air-conditioning sector is the accidental or
deliberate venting of refrigerant. The venting of refrigerant can be reduced by increasing the reclamation
of used refrigerant (discussed in more detail below) and properly disposing of refrigerant that cannot be
reclaimed (such as highly contaminated refrigerant or mixed refrigerant). Disposal costs vary by country
and region, as do transportation costs, storage costs, and access to refrigerant disposal facilities (e.g., high
temperature incinerators that handle refrigerants).  Global average ODS destruction costs are estimated to
vary between $1.80 and $2.70 per pound (approximately $4 to $6 per kilogram) (ICF Consulting, 2002b).
This option was not explored in the cost analysis  as a result of the uncertainty associated with access to
disposal facilities and cost disparities within regions.

Technician Certification/HFC Sales Restriction

By ensuring that refrigeration/air-conditioning technicians receive training in proper refrigerant handling,
including recovery/recycling  practices,  and/or by restricting the  sale of HFC refrigerants to certified
technicians  only, refrigerant emissions can be reduced. In some countries, such as the  United States,
technicians  must be certified in accordance with  national regulations to purchase  CFC and HCFC
refrigerants and service refrigeration  and air-conditioning  equipment.   Restricting  the use of HFC
refrigerants to certified  technicians would similarly reduce emissions.   The  costs of promulgating
regulations  in individual countries  are uncertain  and highly variable, and hence, this option was not
explored further.

Alternative Refrigerant Options

Ammonia

Ammonia, primarily used in water-cooled chillers, has excellent thermodynamic properties and can be
used in many types of systems.  In  addition, it has the advantage of having a strong odor, which makes
refrigerant leaks easier to detect, and is lighter than air, facilitating dispersion in the event of a release
(UNEP,  1999a).   However,  it  must  be used  carefully,  because it  is toxic and slightly flammable.
Ammonia is an explosion hazard at 16 to 25 percent in air, which creates a problem in confined spaces.
Chillers that use ammonia as a refrigerant are commercially available in Europe and elsewhere, and they
have efficiencies that are comparable to those of HFC-134a chillers in some instances.  Building and fire
codes, however, restrict  the use of ammonia in the urban areas  of the United  States and many other
countries. These safety concerns and institutional  barriers effectively limit the potential for expanded use
of ammonia chillers (Sand et al., 1997).

While the use of ammonia within public spaces  such as supermarkets is  limited in some countries by
building  codes and ordinances,  it is a potential alternative for  supermarkets if safety concerns can be
adequately addressed through engineering design such as secondary loops and isolation. Indeed, modern
ammonia systems manufactured in the United States are fully contained closed-loop systems with fully
integrated controls that regulate pressures throughout the system.  Also, all systems are required to have
an emergency diffusion system and a series of safety relief valves to protect the system and its pressure
vessels from over-pressurization and possible failure (ASHRAE,  2002). Systems with ammonia are being
built and used in Europe (Sand et al., 1997). However, the further use of ammonia as  a  supermarket
primary refrigerant may be unlikely  in the near future in the United Kingdom and other countries because
of the capital costs and issues  of  compliance with standards  and safety  regulations  (Cooper, 1997).
Ammonia would also be an option in some industrial process refrigeration and cold storage applications,
contingent upon addressing all of the relevant  concerns regarding flammability and  toxicity.   For
                                                                                             -18-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
example, ammonia is used in about 80 percent of current installations of large-size refrigeration plants, as
well as many indirect commercial refrigeration systems (RTOC, 2003).

The chemical  properties of ammonia make it  incompatible  with  current designs of residential light
commercial  unitary air-conditioning systems, which use copper for the  refrigerant tubing, in the heat
exchangers and in other components. Ammonia in the presence of water cannot be used with copper or
zinc (UNEP, 1999a); however, ammonia can be used in aluminum and steel systems.  Compatible
components would have to be developed to use ammonia. As a result of these technical and cost barriers,
as well as ammonia's flammability and toxicity, ammonia is considered an unlikely candidate for use in
commercial and residential unitary equipment (Sand et al., 1997).

Many of the existing uses of ammonia are included in the baseline analysis.  One additional option—
using  ammonia secondary loop systems in retail food and cold storage end-uses—is analyzed in more
detail below, in the section on Technology Options.

Hydrocarbons

Hydrocarbons  have  thermodynamic properties  comparable  to fluorocarbons that  make them good
refrigerants; however,  their  high flammability causes  safety   concerns.   Considering  technical
requirements apart from safety, there is potential for use of hydrocarbons in retail food refrigeration,
refrigerated  transport, household  refrigeration,  residential air-conditioning, MVACs, and commercial
unitary systems.  Currently used refrigerants include HC-600a, HC-290,  and HC-1270 (UNEP,  1999a).
In addition to good thermodynamic properties, hydrocarbons also have other advantages such as energy
efficiencies comparable to fluorocarbons, zero ODP, and very low direct GWP.

The primary disadvantage of hydrocarbons is flammability, resulting in  significant safety and liability
issues. This causes increased costs for safety precautions in factories and can necessitate design changes
in every  application, such as relocation of electrical  components to reduce the likelihood of accidents
from potential leaks  (Kruse,  1996; Paul, 1996).  This also entails  additional hardware costs for many
applications (ADL, 1999; Crawford, 2000). Hydrocarbon refrigerant use is generally restricted by U.S.
safety codes, and with the exception of industrial refrigeration, the U.S. EPA has not listed hydrocarbons
as acceptable substitutes to  ODS refrigerants (per Section 612 of the Clean Air  Act Amendments of
1990). Even if systems that are designed to use hydrocarbon  refrigerants were listed, liability concerns
would remain.  Systems using flammable refrigerants will require additional engineering and testing,
development of standards and service procedures, and training of manufacturing and service technicians
before commercialization.

Hydrocarbon domestic refrigerators have been available  in Western Europe since  the early  1990s, and
have now fully penetrated some of the new domestic refrigeration markets,  such as that in Germany.
Hydrocarbon domestic refrigerators are also available in Argentina, Australia, Brazil, China, Cuba, India,
Indonesia, Japan and elsewhere.   Similarly, hydrocarbon refrigerants are available  in other products,
although  little information is readily available regarding their market success to date (Hydro Cool Online,
2002; Calor Gas Refrigeration website, 2004; CARE Web site, 2004).

In addition, hydrocarbons have been used in MVACs for the last several years.  Some have  estimated that,
in certain parts of Australia,  280,000  vehicles contain  hydrocarbon refrigerants  (Greenchill website,
2000), although independent  data have not been  supplied  to confirm that  estimate.   The use  of
hydrocarbon refrigerants in direct expansion systems not designed for a flammable refrigerant can pose
safety concerns and is not considered acceptable by much of the global MVAC industry.  The Society of
Automotive  Engineers' (SAE) Alternate Refrigerant Cooperative Research Program has demonstrated a
                                                                                            -19-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
secondary loop system using hydrocarbon refrigerant that minimizes the possible release of flammable
refrigerant into the passenger compartment (Hill and Atkinson, 2003).

Proponents of hydrocarbon systems claim that these systems bring numerous benefits, including increased
energy efficiency, lower refrigerant cost, lower capital cost, and less  noise (HyChill website, 2004;
Greenchill website, 2000), although little independent research exists to confirm the veracity  of these
claims.   In many parts  of the  world, however, safety  issues,  public perception, and manufacturer
acceptance impedes further penetration of this option.

This analysis does not consider the use of hydrocarbons in household refrigeration because it is assumed
that  this option is  reaching maximum market penetration in the  baseline.   In those  regions  where
hydrocarbons have  not successfully penetrated markets (e.g., North America),  it is assumed  that the
perceived risk and lack of acceptance of hydrocarbon refrigerants, which has prevented adoption to date,
will continue to serve as a barrier in the foreseeable future.  The use of hydrocarbons in other refrigeration
end-uses was not considered further due to uncertainty concerning price and likely market penetration.

Low-GWP Refrigerants

The  use of low-GWP refrigerants (e.g., HFC-152a  with a GWP of 140) in place  of higher GWP
refrigerants  (e.g., HFC-134a with a GWP of 1,300)  is  another option for  reducing greenhouse  gas
emissions to the atmosphere.   The use of HFC-152a  in  MVACs is explored in this  cost analysis, as
described in detail on page 28.

Several other low-GWP refrigerants exist.  For example, carbon dioxide, discussed further below, has a
GWP of 1.  In addition, HCFC-123 and HCFC-124, which are not considered alternatives to HFCs, have
low direct GWPs, complicated by factors including their  contribution to stratospheric  ozone depletion.
While some studies (Calm et  al, 1999;  Wuebbles and Calm, 1997; U.S. EPA, 2002; RTOC, 2003)
suggest the extended use of HCFC-123 in large tonnage chillers may offer a means to reduce direct GWP-
weighted refrigerant emissions, and in some instances may reduce overall greenhouse gas emissions, this
option is not examined in this analysis, as full compliance with the current HCFC phase-out schedule is
assumed.

Carbon Dioxide

Another option is to use carbon dioxide as a refrigerant.  Prototype carbon dioxide systems have been
developed for numerous  types of systems,  including MVACs, industrial processing, refrigerated
transport, and retail food systems. Carbon dioxide has  zero ODP and a GWP of 1, and is claimed by its
proponents to be advantageous for use as a refrigerant.  However, carbon dioxide is associated with
potential safety risks  and other technical/economic  disadvantages.   Above  certain concentrations,
exposure to carbon dioxide may result in adverse health consequences.  At very high concentrations, even
for short periods of time, carbon dioxide affects the central nervous system and  is an asphyxiant.  To
protect against adverse health  effects from workplace exposure, the Occupational Safety  and Health
Administration's (OSHA) recommended  8-hour time-weighted average exposure limit is  5,000 ppm
(ACGIH, 1999). Also, carbon dioxide systems operate at high pressure, which presents a potential hazard
and  may increase the cost of designing and purchasing equipment.  In addition, potential loss  of
operational  efficiency and associated increases  in  energy use  and  indirect  emissions, refrigerant
containment issues, long-term reliability, and compressor  performance are  other potential problems
(Environment Canada, 1998).
                                                                                           -20-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
For this analysis, carbon dioxide systems are only evaluated as options for MVACs. Carbon dioxide is
being investigated for use in  other end-uses but, due to the early stage of development and a lack of
available information, those end-uses are not explored in this analysis. The MVAC option is described in
detail in the section on Technology Options.
Technology Options

Distributed Systems for Stationary Commercial Refrigeration Equipment

A distributed system consists of multiple compressors that are distributed throughout the store near the
display cases they serve and are connected by a water loop to a single cooling unit that is located on the
roof or elsewhere outside of the store. Refrigerant charges for distributed systems can be smaller than the
refrigerant charge used in a comparable traditional centralized direct expansion (DX) system.  Significant
reductions in total global warming impact from current levels may be possible with distributed systems
that use HFC refrigerants (Sand et al., 1997).  Reduced refrigerant charge sizes, in addition to increased
energy efficiency associated with such systems, could effectively decrease global warming impacts, even
with the use of fluorocarbon refrigerants.

Using  HFC  distributed systems in lieu of HFC centralized DX systems in retail  food settings offers the
potential to reduce  HFC emissions.   Distributed systems have smaller refrigeration  units distributed
among the refrigerated and frozen food display  cases, with  each  unit sending heat to a  central water
cooling system.  A distributed system would significantly reduce the refrigerant inventory and minimize
the length of refrigerant tubing and the number of fittings that are installed  in direct expansion systems,
thereby reducing leaks of HFCs (ADL, 2002).

This technology option is assumed to be applicable to the retail food and  cold storage end-uses.  The
project lifetime is assumed to be 20 years, and the emission reduction efficiency is  calculated to  be
approximately 93  percent (i.e.,  the reduced leakage from use of a  distributed  system divided by the
leakage of the baseline system, or (0.15  -  0.04 x  0.25)70.15).  This calculation is based on an assumed
annual leak rate of 4 percent (compared to 15 percent for centralized direct expansion [DX]  systems), and
a charge size that is 75 percent less than centralized  DX systems (ADL, 2002).   Regional technical
applicability and  reduction efficiency for 2010 and  2020 are presented  in Table 1-8.  Assumptions  on
maximum market penetration for each region and  year are  presented in Table  1-16  and Table  1-17.
Because the cost analysis for this option does not address the costs to retrofit existing DX systems, this
option is assumed to penetrate only new  (post-2004) retail food and cold storage installations (i.e., those
installed in 2005 or beyond).

Table 1-8:  Summary of Assumptions for Distributed Systems for New Stationary Equipment
Country/Region
U.S. & Japan
Other Annex I
Latin America & Caribbean
Other Non-Annex I, Russia, CEITs
Applicable End-Use
Sector(s)a
Retail food
Cold storage
Reduction
Efficiency3
93%
Technical Applicability11
2010 2020
43.6% 43.1%
36.3% 45.1%
51.0% 49.4%
54.4% 46.3%
a End-Uses and Reduction Efficiency apply to all regions.
b Technical applicability is shown as a percent of total refrigeration/air-conditioning sector emissions, and equals the percent of total
refrigeration/air-conditioning emissions that are assumed to come from retail food and cold storage end-uses.
                                                                                              -21-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Cost and Emission Reduction Analysis

The following bullets describe the cost and emission inputs used to derive the final $/TCE for distributed
systems, the results of which are presented in Section 1.4.  For more information on the calculation of
costs and emission savings of this alternative refrigeration technology, see Section 1.3.2 on "Calculation
of Indirect Emissions and Costs for Options Applicable to Stationary Equipment."

    •    One-Time Costs. HFC distributed systems are assumed to cost approximately $300 per ton of
        cooling capacity  to install, or $100 more per ton of cooling capacity than conventional HFC
        centralized DX systems (U.S. EPA, 200la).
    •    Annual Costs. There are no annual costs associated with this option.
    •    Cost Savings.  Based on average electricity prices in the  United  States from 1994-1999 (EIA,
        2000), this option is associated with an annual cost savings of $5.72 per ton of cooling capacity in
        the United States, as a result of an approximately 8 percent increase in  energy efficiency relative
        to centralized DX systems (ADL, 2002).   In all  other countries,  this annual cost savings was
        adjusted by average electricity prices (average of 1994-1999) based on EIA (2000).  In addition,
        this system will prevent the annual emission of nearly 0.32 kg of refrigerant per ton of cooling
        capacity, as a result of reduced leakage.  Assuming an average cost of (R-404A) refrigerant of
        $18.4I/kg  (DuPont  Customer Service,  2004), this  translates  into  annual cost savings  of
        approximately $5.85 per ton of cooling capacity.  Combined, the annual cost savings associated
        with this option is $11.56 per ton of cooling  capacity.

    •    Emission Reductions.  Because the distributed system described above use less energy than a
        typical DX system,  less  carbon dioxide is produced  in generating electricity to run  it.  This
        indirect energy benefit is calculated to be approximately 0.02 TCE per ton of cooling capacity
        each year, using average  power plant emission rates in the United  States.  In all other countries,
        the indirect emission benefit was  calculated  by multiplying the 0.02 TCE emission reduction
        calculated for the United States by a ratio of U.S. to regional or national average CO2  emission
        rates for electricity production, based on Sand et al. (1997). The reduction of direct emissions of
        approximately 0.32 kg of refrigerant equates to  approximately 0.28  TCE per ton  of cooling
        capacity per year.  Therefore, in the United  States, total annual emission reduction associated
        with this option is estimated to be 0.31 TCE per ton of cooling capacity. Furthermore, as a result
        of avoided losses at disposal (assumed to be 56 percent of original charge),  a further 0.89 kg of
        refrigerant emissions could be avoided per ton of cooling capacity, equal to a one-time  emission
        reduction of 0.79 TCE per ton of cooling capacity.
Secondary Loop Systems for Stationary Equipment

Secondary loop systems pump cold fluid to remove heat from equipment (e.g., refrigerated food display
cases) or areas to be cooled. The fluid, often a brine solution, passes through a heat exchanger to be
cooled by a refrigerant isolated from the equipment or areas cooled. These systems require a significantly
lower refrigerant charge, have lower leak rates, and can allow the use of flammable or toxic refrigerants.

Secondary loops may be used in commercial and industrial refrigeration applications,  for example, to cool
supermarket display cases without circulating toxic or flammable refrigerants throughout the store or to
reduce the needed charge of HFC refrigerants. The primary disadvantages of the secondary loop system
can be  a loss of energy  efficiency  and higher  capital  costs.   Potential benefits of secondary cooling
systems, however, include decreased charge sizes, decreased leak rates, faster defrost, lower maintenance
needs, and longer shelf lives,  which  can  result in significant cost-savings over time (Bennett, 2000;
Baxter,  2003; Faramarzi and Walker, 2003). Indeed, the reduction in size and leak rate of the refrigerant
charge could result in a reduced global warming impact, even with the use of fluorocarbon refrigerants.
                                                                                            -22-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
The use of zero GWP refrigerants could result in even lower global warming impacts (Sand et al, 1997).
Furthermore, secondary loop systems have improved temperature control compared to conventional direct
expansion systems, which can represent an important advantage in countries like the United States, where
recent regulations  on temperature control for refrigerated products such as meat, poultry,  and fish have
become more stringent.  Moreover, recent technological improvements to secondary cooling systems,
such as high-efficiency evaporative condensers and display  cases with high temperature brines,  have
increased system efficiency (Baxter, 2003; Faramarzi and Walker, 2003).  Two types of secondary loop
systems, for use in retail refrigeration and cold storage warehouses, are analyzed in more detail below.

Secondary loops could mitigate some but not all of the risks of using flammable refrigerants in residential
and commercial unitary end-uses. In addition, secondary loops also have potential applications in motor
vehicle air-conditioning, discussed further under the option of "HFC-152a  Refrigerant in MVACs." Due
to the lack of technical and cost information on secondary loop  systems in these other applications, they
are not included as options in this analysis.

HFC Secondary Loop Systems for Stationary Commercial Refrigeration Equipment

Designing new retail food and cold storage systems to operate using secondary loops with HFCs can also
reduce HFC  emissions.  As discussed above, secondary loop systems  circulate a secondary coolant or
brine from the central refrigeration system to the display cases (UNEP,  1999a; ADL,  1999).  These
systems have lower leak rates and operate at reduced charges. Additionally, pipes used in  these  systems
are  now pre-manufactured and  can be made of pre-insulated plastic instead of copper.  This  reduces
material costs and, by eliminating the need for brazing, allows for faster installation. In the United States,
installation costs have  been reduced by more than 25 percent in recent years. With continued research
and development, it is expected that  this technology will soon be as cost-effective to purchase, install, and
operate as centralized DX systems (Bennett, 2000). This technology option is assumed to be applicable to
the  retail food and cold storage end-use sectors.  This option is assumed  to reduce emissions  from
appropriate end-uses by up to  98.5 percent, based on an assumed charge of only 11 percent of the base
case (centralized DX  systems)  and an annual leak  rate of 2 percent  compared to 15 percent for an
"optimal installation" of a centralized DX system (ADL, 2002).  The project lifetime is assumed to be 20
years. The regional technical applicabilities and reduction efficiencies for 2010 and 2020 are presented in
Table 1-9. Assumptions on maximum market penetration for each region and year are presented in Table
1-16 and Table  1-17.  Because the cost analysis for this option does not address  the costs to retrofit
existing  DX  systems,  this  option is assumed to penetrate only  new (post-2004)  retail food and cold
storage installations (i.e., those installed in 2005 or beyond).

Table 1-9: Summary of Assumptions for HFC Secondary Loop Systems for New Stationary Equipment
Country/Region
U.S. & Japan
Other Annex I
Latin America & Caribbean
Other Non-Annex I, Russia, CEITs
Applicable End-Use
Sector(s)a
Retail food,
Cold storage
Reduction
Efficiency3
98.5%
Technical Applicability11
2010 2020
43.6% 43.1%
36.3% 45.1%
51.0% 49.4%
54.4% 46.3%
a End-Uses and Reduction Efficiency apply to all regions.
b Technical Applicability is shown as a percent of total refrigeration/air-conditioning sector emissions, and equals the percent of total
refrigeration/air-conditioning emissions that are assumed to come from equipment t in the retail food and cold storage end-uses.

    Cost and Emission Reduction Analysis

The following bullets  describe the cost and emission inputs used to derive the final $/TCE for HFC
secondary loop systems, the results of which are presented in Section  1.4.  For more information on the
                                                                                             -23-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
calculation of costs and emission savings of this alternative refrigeration technology, see Section 1.3.2 on
"Calculation of Indirect Emissions and Costs for Options Applicable to Stationary Equipment."

    •   One-Time Costs.  This option is assumed to cost approximately $240 per ton of cooling capacity
       to install, or $40 (20 percent) more than conventional centralized HFC direct expansion systems
       (Bennet, 2000; Smithart, 2000; ADL, 1999).   Improved  secondary loop systems now being
       manufactured may be associated with even lower cost premiums—ranging from 0 to 10 percent
       more expensive than conventional centralized DX systems (Kazachki, 2004). Such systems may
       be considered in future analyses.
    •   Annual Costs. Based on average electricity prices in the United States from 1994-1999 (EIA,
       2000), this option  is associated with an annual cost of $11.43 per-ton of cooling  capacity, as a
       result of an approximately 15 percent increase in energy consumption relative to centralized DX
       systems (ADL, 2002).  In all other countries, this annual cost was adjusted by average electricity
       prices  (1994-1999) based on EIA (2000).  Annual energy costs may actually be much lower if
       energy saving features are applied.  Indeed, new secondary loop systems with improved design
       features (e.g., use of evaporative condensing) may result in annual energy requirements that are
       roughly the same as those for conventional DX systems, or even lower (Kazachki, 2003; Baxter,
       2003). Such systems may be considered in future analyses.
    •   Cost Savings.  On a per-ton of cooling capacity basis, this system will prevent nearly 0.34 kg of
       refrigerant emissions each year in the United States, as a result of reduced leakage.  Assuming an
       average  cost of (R-404A) refrigerant of $18.4I/kg  (DuPont Customer Service, 2004),  this
       translates into an annual cost savings of $6.17 per ton of cooling  capacity.
    •   Emission Reductions. Under the HFC secondary loop system described above, the annual direct
       emission of nearly 0.34 kg (0.30 TCE) of refrigerant can be avoided annually per ton of cooling
       capacity.  However,  indirect emission (energy) penalties lower total  emission benefits by more
       than 15 percent (approximately 0.05 TCE per ton of cooling capacity) in the United States. In all
       other countries, the  indirect  emission penalty was  calculated by multiplying the 0.05  TCE
       emission penalty assumed for the United States by a ratio of U.S. to regional or national average
       CO2 emission rates for electricity production, based on Sand et al. (1997).  Additional emission
       benefits are also associated  with this option as a result of avoided  (R-404A) refrigerant losses at
       equipment disposal.  Assuming an emission rate of 56 percent  at disposal, a further 1.11 kg of
       refrigerant emissions could be avoided per ton of cooling capacity, equal to a one-time emission
       reduction of 0.99 TCE per ton of cooling capacity.

Ammonia Secondary Loop Systems for Stationary Commercial Refrigeration Equipment

The use of ammonia is very common in certain countries while strongly restricted in others. For example,
for many decades ammonia  has been used in almost all dairies, breweries,  slaughterhouses, and large
freezing plants nearly all over Europe, while its use is heavily  regulated in North America (ACHR News,
2000).   Ammonia refrigeration has  historically been  used  in large,  low-temperature  industrial
refrigeration, as well  as medium and  large chillers,  generally  in  food processing (Crawford, 1999).
However, the use of ammonia refrigerant is beginning to expand into retail food and smaller-sized chillers
in some countries, particularly those in the European Union.

Because of ammonia's toxicity and flammability, major design modifications would be required for the
majority of traditional HFC  systems.   Furthermore, since  different  countries have different  sets of
building codes,  fire  codes,  and other safety standards  relating to the  use of ammonia in building
equipment, some countries (e.g., the  United States) would need to  revise  those codes to allow  for the
expanded use of ammonia in new equipment types.
                                                                                           -24-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Ammonia can be used as the primary refrigerant in secondary loop systems in place of HFCs. Because
ammonia secondary loop systems avoid running the primary refrigerant through miles of piping to and
from  food  storage cases,  they have lower leak rates  than conventional centralized  DX systems, and
operate at reduced charges. In these types of systems, ammonia is kept out of public contact (e.g., outside
of buildings), and non-toxic fluids  are used  as  secondary coolants.  Incremental one-time costs for
ammonia systems are assumed to include expenditures for equipment needed to ensure safety.   The
annual operating costs also include net energy requirements, but, because of a lack of information, do not
cover costs associated with training of technicians and development and updating of safety protocols to
handle more hazardous refrigerants such as ammonia. This technology option is assumed to be applicable
to the retail  food and cold storage  end-uses.  The project lifetime is assumed  to be 20 years.   The
reduction efficiency of this option is 100 percent, as the ammonia completely replaces the HFC. Because
the cost analysis for this option does not address the costs to retrofit existing DX  systems, this option is
assumed to be technically applicable in only new (post-2004) retail food and cold storage installations.

Table 1-10 presents regional technical applicabilities and the reduction efficiency for 2010 and 2020.

Table 1-10: Summary of Assumptions for Ammonia Secondary Loop Systems for New Stationary Equipment
Country/Region
U.S. & Japan
Other Annex I
Latin America & Caribbean
Other Non-Annex I, Russia, CEITs
Applicable End-Use
Sector(s)a
Retail food
Cold storage
Reduction
Efficiency3
100%
Technical Applicability'^
2010 2020
43.6% 43.1%
36.3% 45.1%
51.0% 49.4%
54.4% 46.3%
a End-Uses and Reduction Efficiency apply to all regions.
b Technical Applicability is shown as a percent of total refrigeration/air-conditioning sector emissions, and equals the percent of total
refrigeration/air-conditioning emissions that are assumed to come from equipment in the retail food and cold storage end-uses.

Ammonia systems are assumed to penetrate a greater percentage of non-U.S. markets, as a result of more
relaxed safety standards  and  greater acceptance  by industry,  end-users,  regulators,  and insurance
companies in those countries. Assumptions on maximum market penetration for each region and year are
presented in Table 1-16 and Table 1-17.

    Cost and Emission Reduction Analysis

The following bullets describe the  cost and emission inputs used to  derive the final $/TCE for ammonia
secondary loop systems, the results of which are presented in Section 1.4.  For more information on the
calculation of costs and emission savings of this alternative refrigeration technology,  see Section 1.3.2 on
"Calculation of Indirect Emissions and Costs for Options Applicable to Stationary Equipment."

    •    One-Time Costs.  Ammonia secondary loop systems are assumed to cost approximately $250 per
       ton of cooling capacity to  install,  or $50 (25 percent) more than conventional centralized HFC
        direct expansion systems (Anderson, 2001).

    •   Annual Costs.  Based  on  average electricity prices  in the United States from 1994-1999 (EIA,
        2000), this option is associated with an annual cost of $11.43 per ton of cooling capacity,  as a
        result of an approximately  15 percent increase in energy consumption relative to centralized DX
        systems (ADL, 2002). In all other countries, this annual cost was adjusted by average electricity
       prices (1994-1999) based on EIA (2000). Note that next generation ammonia secondary loop
        systems may  adopt  energy saving features that  yield lower annual energy costs,  as discussed
        above for HFC secondary loop systems.
                                                                                             -25-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •   Cost Savings.  On a per-ton of cooling  capacity basis, this system will prevent 0.34  kg of
        refrigerant emissions each year in the United States, as a result of reduced leakage.  Assuming an
        average  cost of (R-404A) refrigerant of $18.4I/kg (DuPont  Customer  Service,  2004),  this
        translates into an annual cost savings of $6.26 per ton of cooling capacity.
    •   Emission Reductions.  Under the ammonia secondary loop system described above, the annual
        emission of 0.34 kg (0.30 TCE) of refrigerant can be avoided annually in the United States per
        ton  of cooling capacity.  However, indirect emission (energy) penalties  lower total emission
        benefits  by more than 15 percent (approximately 0.05 TCE per ton of cooling capacity)  in the
        United States.  In all other countries, the indirect emission penalty was calculated by multiplying
        the 0.05 TCE emission penalty assumed for the United States by  a ratio of U.S. to regional or
        national  average CO2 emission rates  for  electricity production, based on Sand et al. (1997).
        Additional emission benefits are also associated with this option as a result of avoided (R-404A)
        refrigerant losses at equipment disposal.  Assuming an emission rate of 56 percent at disposal, a
        further 1.27 kg of refrigerant emissions could be avoided per ton of cooling capacity, equal to a
        one-time emission reduction of 1.13 TCE per ton of cooling capacity.

Enhanced HFC-134a Systems in MVACs

Various options  exist to reduce emissions of HFC-134a in MVACs by reducing charge  size, leak rates,
and/or system efficiency (i.e., reducing system power consumption).  Specifically, reducing the volume of
the system components, such as the condenser and refrigerant lines, can reduce charge size.  Similarly,
leak rates can be lowered and system efficiency improved by using better system  components, such as
improved system sealing, lower permeation hoses, improved fittings, and higher evaporator temperatures
(Lundberg, 2002; Xu and Amin, 2000).   Additional savings of indirect emissions can  be obtained by
improving system efficiency, for example through the use  of oil separators  and  externally controlled
swashplate compressors.

Based on the latest science and industry estimates available at publication, enhanced HFC-134a systems
can reduce baseline direct emissions by 50 percent (SAE, 2003a). This technology is expected to become
commercial between 2004 and 2006 (SAE, 2003a). This analysis assumes a project lifetime  (i.e., MVAC
lifetime) of 12 years.  Regional technical applicabilities and the reduction efficiency are  presented in
Table 1-11.

Table 1-11: Summary  of Assumptions for Enhanced HFC-134a Systems for New MVACs
Country/Region
U.S. & Japan
Other Annex I
Latin America & Caribbean
Other Non-Annex I, Russia, CEITs
Applicable End-
Use Sector(s)
MVACs
Reduction
Efficiency3
50%
Technical Applicability11
2010 2020
28.5% 19.5%
40.4% 15.9%
16.2% 7.8%
10.7% 13.6%
a Reduction Efficiency applies to all regions, and represents the reduction in direct emissions (compared to conventional HFC-134a systems)
as a result of reduced leakage.
b Technical Applicability is shown as a percent of total refrigeration/air-conditioning sector emissions, and equals the percent of total
refrigeration/air-conditioning sector emissions that are assumed to come from MVACs.

Acceptance of this  substitute would likely vary by region, based on consumer and industry attitudes,
economic variables, and availability of competing options. Enhanced HFC-134a systems are expected to
become commercially available several years before other alternatives (e.g., carbon dioxide, HFC-152a).
Therefore, this analysis assumes that initially enhanced HFC-134a systems will most deeply penetrate the
markets of Europe, Australia and Japan, where regulations, policy,  and/or voluntary initiatives are moving
                                                                                             -26-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
fastest to reduce HFC use and emissions. Beyond 2010,  however, Europe15 and Japan are expected to
move away from HFC-134a use in MVACs,  so  this option is assumed to gain the greatest market
penetration in other developed countries, such as the United States and Canada, where industry is resistant
to switching from using HFC-134a.  In developing countries, capital cost is expected to prevent this
option from penetrating the market  at all. The cost  analysis for this option does not include any costs
associated with  retrofitting existing  HFC-134a systems. Therefore, this  option is  assumed to penetrate
only new MVACs produced after 2004.  Assumptions on  maximum market penetration for each region
and year are presented in Table 1-16 and Table 1-17.

Cost and Emission Reduction Analysis

The following bullets describe the cost and emission inputs used to derive the final $/TCE for enhanced
HFC-134a systems for MVACs, the results of which are presented in Section 1.4:

    •    One-Time  Costs.  While  enhanced HFC-134a  MVACs are yet to be  fully developed and
        commercialized, the additional capital cost of this  option is assumed to  be 40 Euros, or $45
        USD,16 per system, based on the latest available industry estimates (SAE, 2003a).
    •    Annual Costs. No annual costs are associated with this option.
    •    Cost Savings.  Enhanced HFC-134a systems will potentially reduce energy consumption by as
        much as 25 to 30 percent (SAE, 2003a). For calculation purposes, 27.5 percent is assumed.  In
        the United States, this gain  in energy efficiency is estimated to translate into a savings of 10.2
        gallons  of gasoline per vehicle per year  (Rugh  and Hovland,  2003).  Assuming an average
        gasoline price of $1.51/gallon (EIA, 2001),  this results in an annual cost savings of approximately
        $15.46 per year in the United States.17 For all other countries, this annual cost savings is adjusted
        by the estimated amount of gasoline saved per vehicle per year18 and by average regional costs of
        unleaded gasoline  in 2000.19  Additional savings  due to less maintenance are not analyzed or
        included here.
        In addition, small cost savings are also associated  with saved HFC-134a refrigerant, assumed to
        cost $7.94/kg (DuPont Customer Service, 2004). On an annual basis, these savings are estimated
15 The proposed EC Regulation on fluorinated gases, presented by the Commission on August 11, 2003, prohibits
the use of fluorinated gases with a GWP higher than 150 in new MVACs placed on the market beginning in 2009,
with the exception of manufacturers that successfully apply for a quota allocation (EC, 2003b). It is currently
undergoing review and revision.
16 This cost conversion is based on an exchange rate of $112.5/€100 (Universal Currency Converter Web site, 2003).
17 Average gasoline price is based on the reported average retail price of regular unleaded gasoline in 2000 in the
United States (EIA, 2001).
18  The estimated quantity of gasoline saved per vehicle per year varies by the percent of fuel consumed by MVACs
(as a percent of total fuel consumption), which in turn varies by MVAC usage. Based on available  data, the
estimated annual savings of gasoline per vehicle per year associated with a 27.5-percent increase in MVAC
efficiency is 2.3 gallons in Europe and 2.0 gallons in Japan (Rugh and Hovland, 2003).  For the purpose of this
report, the MVAC efficiency value for Europe was used as a proxy for the remaining countries.
19 EIA (2001) provides average 2000 prices of regular unleaded gasoline for select countries, which are used as
proxies for adjusting the annual cost savings for non-U.S. countries by region. Specifically, the average price in
Germany ($3.45/gallon) is used as a proxy for all European countries; the average price in Australia and Canada
($1.90/gallon) is used as a proxy for all other developed countries except Japan; and the average price in Mexico and
Taiwan ($2.09/gallon) is used as a proxy for all developing countries. The average gasoline price for Japan is
$3.74/gallon.
                                                                                               -27-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
       to total  approximately $0.35  per MVAC—assuming  that  conventional  HFC-134a  MVACs
       contain an average charge of 0.8 kg, that they emit 10.9 percent of this charge each year, and that
       50 percent of these emissions could be avoided through this option.
    •  Emission Reductions.  Under the  enhanced HFC-134a system described above, the annual
       emission  of approximately 0.04 kg  (i.e., 0.8 kg charge  multiplied by the  emission rate of
       10.9%/year and the reduction efficiency of 50%) of HFC-134a refrigerant could be avoided from
       reduced leakage, resulting in the annual reduction of 0.015 TCE per MVAC. Furthermore, based
       on U.S.  emission  factors for motor  gasoline  (U.S. EPA,  2003),  indirect  emission benefits
       associated with a 27.5-percent system efficiency improvement could lead to the annual reduction
       of an additional 0.025 TCE per MVAC in the United States. Overall, this option could, therefore,
       reduce 0.040 TCE per MVAC each year in the United States.   For all non-U.S. countries, the
       annual indirect emission benefit estimated for the United States (0.025 TCE) was  adjusted by the
       estimated amount of gasoline saved per vehicle per year (Rugh  and Hovland, 2003) and by the
       global average  emission factor for motor gasoline (IPCC, 1996).

HFC-152a Refrigerant in MVACs

Replacing HFC-134a refrigerant in MVACs with HFC-152a represents a significant opportunity to reduce
HFC emissions, since the GWP of HFC-152a is 140, 89 percent less than that of HFC-134a, whose GWP
is 1300.  HFC-152a is  a flammable  refrigerant but is less flammable than hydrocarbons.  HFC-152a can
be used in both direct expansion (DX) and secondary loop MVAC  systems.  Because there is still great
uncertainty associated with the future costs of HFC-152a secondary loop systems for MVACs, this cost
analysis only considers the DX option. Likewise, because there is still great uncertainty  associated with
future costs of "improved" HFC-152a MVACs—which will use improved system components to further
reduce refrigerant leak rates,  as well as externally-controlled variable  displacement compressors and
refrigerant system controls to further increase system efficiency—only the conventional DX systems are
considered in this costs analysis.

In addition to direct emission reductions associated with a lower  GWP, HFC-152a DX systems in
MVACs  also reduce indirect  emissions by improving  system efficiency by about  10  percent (SAE,
2003a). This analysis  assumes a project lifetime (i.e., MVAC lifetime) of 12 years. Regional technical
applicabilities and the reduction efficiency are presented in Table 1-12.

Table 1-12: Summary of Assumptions for HFC-152a DX Systems in New MVACs
Country/Region
U.S. & Japan
Other Annex I
Latin America & Caribbean
Other Non-Annex I, Russia, CEITs
Applicable End-
Use Sector(s)
MVACs
Reduction
Efficiency3
89%
Technical Applicability11
2010 2020
28.5% 19.5%
40.4% 15.9%
16.2% 7.8%
10.7% 13.6%
a Reduction Efficiency applies to all regions, and represents the reduction in direct emissions (compared to conventional HFC-134a systems)
as a result of lower GWP.
b Technical Applicability is shown as a percent of total refrigeration/air-conditioning sector emissions, and equals the percent of total
refrigeration/air-conditioning sector emissions that are assumed to come from MVACs.

The use of HFC-152a DX systems in MVACs would not require any significant changes to existing HFC-
134a system components apart from a safety mitigation system (e.g., a refrigerant detector and a valve to
isolate the  remaining charge from the passenger compartment), thereby rendering this option  easy to
introduce into the market.  Furthermore, compared to baseline HFC-134a systems, HFC-152a systems are
                                                                                             -28-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
expected to be more efficient and may operate at reduced refrigerant charges and  leakage  rates.20
However, because HFC-152a is a slightly flammable gas, safety systems are needed.  Thus, personnel
training would be needed to enable the safe and effective recovery and recycling of refrigerant at service
and disposal, and additional safety systems to  minimize the potential for large  leaks into the passenger
compartment may be required.  New fire-safe service equipment for refrigerant recovery/charging and
leak detection may also be required.

While the MVAC industry has demonstrated the use of HFC-152a in prototype DX (and secondary loop)
MVAC systems, the technology is still in the research and development phase.  HFC-152a systems are
expected to become commercially available between 2006 and 2008 (SAE, 2003a). Once available, it is
assumed that initially HFC-152a systems will  have the greatest market  share in Europe, Australia and
Japan,  where policies and regulations are being developed that will aim aggressively at reducing HFC-
134a emissions from MVACs. Once the technical and economic barriers associated with carbon dioxide
systems  are  resolved and such systems become more  widely used,  it is expected  that HFC-152a
popularity will then decline in the European market and later in other nations like Japan and Australia.  In
comparison, because other developed countries  (e.g., the United States  and Canada) are less open to using
non-HFC alternatives, this option is expected to gain market share slowly in those regions.  In addition,
because HFC-152a has the lowest capital cost of all MVAC options considered in this analysis, it is
expected to be the  primary alternative for MVAC markets in developing countries in later years.
Retrofitting  HFC-134a systems to HFC-152a systems is not considered technically or economically
feasible because it is assumed that additional safety systems to reduce  potential passenger exposure must
be incorporated into the system. Thus, costs associated with retrofit were not assessed, and this option is
assumed to penetrate only new (post-2004) MVACs.  Assumptions on maximum market penetration for
each region and year are presented in Table 1-16 and Table 1-17.

Cost and Emission Reduction Analysis

The following bullets describe the cost and emission inputs used to derive the final $/TCE for HFC-152a
systems for MVACs, the results of which are presented in Section  1.4:

    •    One-Time Costs. While research and development is still ongoing on HFC-152a systems, based
        on the latest available industry estimates, the capital cost of this option is assumed to be 20 to 25
        Euros per system more than a standard HFC-134a system (SAE, 2003a).   For calculation
        purposes, 22.50 Euros, or $25.31 USD,21 is used.
    •    Annual Costs. No annual costs are associated with this option.
    •    Cost Savings.  Based on industry consensus, HFC-152a systems are estimated to reduce energy
        consumption by 10 percent (SAE, 2003a), although these gains may not be realized in all weather
        conditions (Hill and Atkinson, 2003).   This gain in energy efficiency is estimated to result in a
        savings of approximately 3.9 gallons of gasoline per vehicle per  year in  the United States, which
        translates into an  annual cost savings of $5.92, based on  average U.S. prices of regular unleaded
        gasoline in 2000  (Rugh and Hovland,  2003; EIA, 2001). For all non-U.S. countries, this cost
        savings is adjusted by the estimated amount of gasoline  saved per vehicle per year22 and by the
        average regional costs of unleaded gasoline in 2000.23
20 Because these systems are still under development, this cost analysis does not consider the possible reduction in
charge and leakage rates, although efficiency improvement predictions based on SAE (2003 a) are included.
21 This cost conversion is based on an exchange rate of $112.5/€100 (Universal Currency Converter Web site,
2003).
22 It assumed that a 10 percent increase in MVAC efficiency results in the annual savings of 0.8 gallons of gasoline
                                                                                            -29-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
       Because this analysis assumes that HFC-152a would be the same price as HFC-134a, and because
       HFC-152a systems are assumed to leak at the same rate as conventional HFC-134a, (although the
       associated emissions  are less damaging to the environment due to the lower GWP), no  cost
       savings are associated with saved refrigerant.
    •  Emission Reductions.  Under  the HFC-152a system described above, the  annual emission of
       approximately 0.09 kg of HFC-134a refrigerant could be avoided in place of annual emissions of
       0.09 kg of HFC-152a—which results in the net annual reduction of 0.028 TCE per MVAC. In
       addition, based on U.S. emission factors for motor gasoline (U.S. EPA, 2003), indirect emission
       benefits associated with a 10-percent system efficiency improvement could lead to the reduction
       of an annual additional 0.009 TCE per MVAC in the United States.  Overall, this option could,
       therefore, reduce approximately 0.037 TCE per MVAC each year in the United States.  For all
       non-U.S.  countries, the annual  indirect emission benefit estimated for the United States  (0.009
       TCE) was adjusted by the estimated amount of gasoline saved per vehicle per year  (Rugh and
       Hovland,  2003) and by the global  average emission factor for motor gasoline (IPCC,  1996).
       Emission  benefits are also achieved by this option at MVAC disposal, as the emission of HFC-
       134a could be substituted by emission of lower-GWP HFC-152a.  Assuming that on average 42.5
       percent of the original MVAC  charge is lost at disposal, the one-time loss of 0.34 kg of HFC-
       134a refrigerant could be replaced by HFC-152a, resulting in the net reduction of approximately
       0.11 TCE per MVAC.
Carbon Dioxide in MVACs

Systems using carbon dioxide as the refrigerant in MVACs represent a potential opportunity for emission
reduction.  This technology uses a transcritical vapor cycle which differs from conventional MVAC
systems and requires  innovative design and engineering.  The arrangement  of components in carbon
dioxide systems is generally consistent with conventional systems; however, a suction line heat exchanger
is added and a low side accumulator is used (in place of a high side receiver, as used in most conventional
HFC-134a systems).  In addition, the individual system components are designed to reflect the extremely
high pressure levels of supercritical carbon dioxide (about 2,000 psig).

Because  carbon dioxide has  a GWP of 1, it would virtually eliminate the  climate impacts of direct
refrigerant emissions from MVACs.  Carbon dioxide systems perform most efficiently in  areas,  like
northern  Europe,  that require air conditioners for cooling and other purposes, but generally have mild
ambient temperatures.24  In addition, heat pump technology  for vehicles is under development (VDA,
2003), which may allow CO2  systems to be used for supplemental heating of the passenger compartment
(SAE, 2003a).  This may be  an important function in cars with very efficient engines where minimal
waste heat is available to warm the passenger compartment.

While CO2 has the advantage of being  non-flammable, its system operating pressure is five to 10 times
that of HFC-134a; therefore, appropriate safety features and new system/component designs are required
before this option can be brought to market.  In addition, an internal heat exchanger, which would further
cool the  high-temperature CO2 from  the  gas  cooler and  heat the low-temperature  CO2 from the
accumulator, would be needed to increase cooling  capacity  and energy efficiency to acceptable  levels.

in Europe and 0.7 gallons of gasoline in Japan (Rugh and Hovland, 2003). For the purpose of this report, the MVAC
efficiency value for Europe was used as a proxy for the remaining countries.
23 Annual  cost savings are adjusted by average gasoline prices as explained in footnote 19.
24 Compared to other refrigerant technologies, prototype carbon dioxide MVAC systems are not as efficient in
warmer climate conditions.  The MVAC industry is actively pursuing research and development activities to
improve system efficiency in warmer weather conditions (SAE, 2003b).
                                                                                           -30-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Also, in the event of a large leak, passengers could be exposed to potentially dangerous levels of CO2;
therefore, it is  assumed that safety  systems  designed to minimize passenger  exposure would  be
incorporated into the system design.

Several engineering constraints must still be overcome, including those associated with flexible lines,
increased system weight, and system leakage/leak detection methods.  In addition, because these systems
will be designed and built differently  than current MVACs,  and because the high  pressure presents
additional risks, technicians will need to be trained on how to service and maintain these new systems
safely and  correctly.  New service equipment for refrigerant charging and leak detection may also be
required.  Moreover, because of the high pressure of these systems, MVAC servicing and maintenance
would need to be performed by skilled technicians, in order to prevent safety hazards and maintain  system
performance.

The efficiency gains associated  with CO2  systems (between 20 and 25  percent [SAE,  2003a]) are
considered in this cost analysis, using 22.5 percent for calculation purposes. While efforts are ongoing to
develop "improved" CO2 systems for MVACs—which experts predict would exceed this 20 to 25 percent
energy efficiency  gain—much  uncertainty remains regarding  the  investment costs  required  to
manufacture these systems.  Therefore, these "improved" CO2 systems are not considered further in this
analysis. This analysis assumes a project lifetime (i.e., MVAC lifetime) of 12 years. Regional  technical
applicabilities and the reduction efficiency for the CO2 option considered are presented in Table 1-13.

Table 1-13:  Summary of Assumptions for C02 Systems in New MVACs
Country/Region
U.S. & Japan
Other Annex I
Latin America & Caribbean
Other Non-Annex I, Russia, CEITs
Applicable End-
Use Sector(s)
MVACs
Reduction
Efficiency3
100%
Technical Applicability11
2010 2020
28.5% 19.5%
40.4% 15.9%
16.2% 7.8%
10.7% 13.6%
a Reduction Efficiency applies to all regions, and represents the reduction in direct emissions (compared to conventional HFC-134a systems).
b Technical Applicability is shown as a percent of total refrigeration/air-conditioning sector emissions and equals the percent of total
refrigeration/air-conditioning sector emissions that are assumed to come from MVACs.

Carbon dioxide systems may be available on the market in the next four to six years (SAE, 2003a).
Because  European and  Japanese  manufacturers are most aggressively  pursuing  CO2, this option is
expected to eventually (by 2020) become the dominant market player in these world markets. In other
developed countries such as the U.S. and Canada, industry is not aggressively developing this technology
and it is assumed that this option will not be widely adopted in these markets in the near future.  Finally,
because of the high capital costs associated with this option (see details below), it is also not expected to
be adopted in developing countries until later years, assuming a projected global market shift to non-GWP
alternatives.  The project  lifetime is assumed to be 12 years, and assumptions  on maximum market
penetration for each region and year are presented in Table 1-16 and Table 1-17. Retrofitting HFC-134a
systems to carbon dioxide is not considered technically or economically feasible because of the  high
operating pressures and because it is assumed that additional safety systems to  reduce potential passenger
exposure must be incorporated into the systems.  Thus, costs to retrofit were not assessed, and this option
is assumed to penetrate only new (post-2004) MVACs.

Cost and Emission Reduction Analysis

The following  bullets describe the cost and emission  inputs used to derive the  final $/TCE for CO2
systems for MVACs, the results of which are presented in Section 1.4:
                                                                                              -31-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •   One-Time Costs.  Based on the latest available industry estimates, the capital cost of this option
       is assumed to be 80 to 120 Euros on a per system basis (SAE, 2003a). For calculation purposes,
       100 Euros, or $112.50 USD,25 was used.
    •   Annual Costs. No annual costs are associated with this option.
    •   Cost  Savings.  Based on industry consensus, it is  estimated that enhanced CO2 systems may
       reduce energy consumption, compared to a baseline HFC-134a system,  by 20 to  25 percent (an
       average of 22.5 percent is  used  for  calculation  purposes), although these gains may not be
       realized at all ambient temperatures (SAE, 2003a).  This gain in energy efficiency results  in the
       savings of approximately 8.4 gallons of gasoline per vehicle per year in the United States, which
       translates  roughly  into an  annual cost  savings of $12.67 (Rugh and Hovland, 2003; EIA, 2001).
       For all non-U.S. countries, this cost savings was adjusted by the estimated amount of gasoline
       saved per vehicle per year26  and by average cost of unleaded gasoline in 2000 (Rugh and
       Hovland,  2003; EIA, 2001).27
       In addition, small cost savings are associated with saved (HFC-134a) refrigerant, assumed to cost
       $7.94/kg (DuPont  Customer Service, 2004). On an  annual basis, these  savings are estimated to
       total  $0.69 per MVAC—assuming that  conventional  HFC-134a MVACs  contain an average
       charge of 0.8 kg, that they emit 10.9  percent of this charge each year,  and that  100 percent of
       these emissions could be avoided through this option.  The additional cost of CO2 refrigerant is
       assumed to be small and is not included in the analysis.
    •   Emission  Reductions.   Under the  CO2  system  described above, the  annual emission of
       approximately 0.09 kg of HFC-134a refrigerant could be avoided, resulting in the reduction of
       0.031 TCE per MVAC.  In addition,  based on U.S. emission factors for motor  gasoline (U.S.
       EPA, 2003), indirect emission  benefits associated with  a 22.5-percent  system efficiency
       improvement could lead to the reduction of an additional  0.020 TCE per MVAC in the United
       States.  Overall, this option could, therefore, reduce  approximately 0.051 TCE per MVAC each
       year  in the United States.  For  all non-U.S. countries, the annual indirect emission benefit
       estimated for the  United States (0.020 TCE) was  adjusted by the estimated amount of gasoline
       saved per vehicle per year (Rugh and Hovland, 2003) and by the global average emission  factor
       for motor gasoline (IPCC, 1996).  Furthermore, a one-time disposal loss  of 0.34 kg of HFC-134a
       refrigerant can also be avoided (assuming a disposal loss rate of 42.5 percent), which would result
       in a further one-time reduction of 0.12  TCE per MVAC.

Oil-Free Compressors

Oil-free compressors are available for chillers, industrial process applications, and other applications
where compressors are used.  The elimination of oil  in refrigeration/air-conditioning compressors has
been achieved through various innovative designs,  including the incorporation of magnetic or hybrid
ceramic bearings  (SKF, 2003; Smithart, 2003).  In some systems, oil may decrease  heat transfer and
reduce operating efficiency; therefore, removing oil may increase the ability to sustain system efficiency
over the life  of the equipment—thereby lowering indirect emissions of CO2  associated with producing
electricity. Eliminating the use of oil in compressors can reduce  the number of equipment components
(e.g., oil  separators and sealing, fittings and connections), allowing equipment  to be made  tighter, and
resulting  in lower leak rates.  In addition, oil-free compressors remove the need for oil changes and the
  This cost conversion is based on an exchange rate of $112.5/€100 (Universal Currency Converter Web site, 2003).
26 A 22.5-percent increase in MVAC efficiency is assumed to result in the annual savings of 1.9 gallons of gasoline
in Europe and 1.7 gallons of gasoline in Japan (Rugh and Hovland, 2003). For the purpose of this report, the
MVAC efficiency value for Europe was used as a proxy for the remaining countries.
27 Annual cost savings are adjusted by average gasoline prices as explained in footnote 19.
                                                                                            -32-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
associated refrigerant emissions that may be experienced either through the service practices used or from
refrigerant dissolved in the oil. This potential emission reduction may be offset however by an increased
frequency of compressor/bearing inspection or replacement (Digmanese, 2004), although an increasing
history of operation may prove that unnecessary.  This option was  not included in the cost analysis,
however, as a result of limited data available  on this new technology and  its  applicability in  future
markets.

Geothermal Cooling Systems

In some locations, geothermal cooling systems for residential and commercial spaces are popular and
economically sound as an alternative to conventional air-conditioning systems. Geothermal technology
transfers heat between the system and the earth and can provide both space heating and cooling.  Though
installation costs are typically 30 to 50 percent higher than conventional systems, annual costs are reduced
by 20 to 40 percent, due in large part to increased energy efficiency. Economic paybacks can accrue in as
little as three to five years.  Geothermal systems may save homeowners 20 to 50 percent in cooling costs
(Geoexchange, 2000; Rawlings, 2000). This technology has also been used in ice skating rinks. Because
of a lack of cost and market penetration data, this technology is not considered further in this analysis.

Desiccant Cooling Systems

Desiccant cooling is produced by removing moisture from an air stream using a desiccant, and then
separately cooling the dry air. The desiccant is thermally regenerated, typically by burning natural  gas or
alternatively, by capturing excess heat. Desiccant cooling  may replace the latent cooling done by some
end-uses, such as unitary systems.  Integrated desiccant cooling systems that combine a desiccant system
with a vapor compression or other cooling system have been successfully installed in some commercial
buildings (Fisher et al, 1994).

However, current designs are used primarily in niche markets that require precisely-controlled and/or low
humidity, such as hospital operating rooms and certain industrial processes. For desiccant-based systems
to be considered widely feasible options in the commercial air-conditioning  market, improvements in
efficiency, cost, size, reliability, and life expectancy must be made (Sand et al.,  1997).

Desiccant  systems  have also been  tried  in MVAC  systems, but found not to be  technically or
economically feasible.   These systems require an intermittent  source of heat;  however, because new
automobiles produce very  little  waste heat, not  enough  heat  is  produced for a desiccant system to
function.  Desiccant systems may therefore  only be feasible where there is a  large heat source, as in a
large truck or bus (Environment Canada, 1998).  Furthermore, in order for desiccant air-conditioners to
become viable options for MVACs,  the varying heat source must be controlled  during normal driving
conditions, when vehicle speed is continually changing. Current prototypes are large and heavy, and the
systems have not been shown to be cost-effective or durable enough to justify the initial investment (U.S.
EPA, 200la).

Because of the technical barriers and insufficient cost information associated with the feasibility of this
option in potentially applicable end-uses, this option was not explored further in this analysis.

Absorption Systems

Absorption systems refrigerate/cool through the use of two fluids and  some quantity of heat input, rather
than electrical input as in the more familiar vapor compression  cycle. Specifically, absorption systems
use a secondary fluid or absorbent to circulate the refrigerant (Rafferty, 2003).   These systems can be
                                                                                            -33-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
used in residential refrigeration and chiller applications and, potentially, heat pumps in residential and
light commercial applications, as described below.

Refrigeration Systems:  In the late  1990s, more than one million of an estimated 62 million refrigerators
sold annually were thermally-activated ammonia/water absorption  systems (Sand et al.,  1997).   The
refrigerants  used for  absorption refrigeration have  negligible GWPs.   Absorption  refrigeration is
commonly used in hotel rooms and for recreational vehicles because it operates quietly and has the ability
to use bottled gas as an energy source. Absorption refrigerators  are limited in size  because of design
constraints.  The thermal coefficient of performance (COP) of these  refrigerators can be increased by as
much as 50 percent (from a COP of 0.2 to 0.3) through design improvements without degrading cooling
capacity (Sand et al., 1997).  However, the low efficiency of absorption equipment means that the indirect
emissions must  be  carefully analyzed.  Inherent  design limitations make it unlikely that absorption
refrigeration  will become a significant replacement  for vapor compression refrigerators.  However,
absorption refrigeration has  great capacity and operating attributes  that permit it to  fill niche markets
(Sandetal,  1997).

Chillers: Gas-fired (as  opposed to electrically-powered) absorption water chillers are  sold in the United
States and are common in Japan. These systems are used mostly where there is a relatively short cooling
season, where electricity costs (especially demand charges) are high, and/or where fairly high-grade waste
heat is available. Although absorption chillers are far less efficient than competitive systems if waste heat
is not available, the technology is feasible and, under some economic circumstances, compares favorably
with vapor compression chillers using fluorocarbon refrigerants.  Market success will be determined by
factors such as the relative costs of natural gas and  electricity, peak load charges, and purchase costs. In
addition, absorption chillers currently have higher capital costs than vapor compression equipment, such
that significant  operating cost savings would  be necessary  to make  their  purchase  economically
competitive.

Heat Pumps: Research and development efforts are attempting to create absorption heat pumps that would
be used for heating and cooling in residential and light commercial  applications.  Several years ago, in
Europe and the United States,  generator absorber heat exchange (GAX) ammonia-water absorption heat
pumps were being developed, while field test units had been built in Japan. Absorption heat pumps could
be used to reduce global warming impacts in areas where heating load dominates, although they would
have the opposite effect in areas where cooling dominates (Sand et al., 1997).

Because these options are either still under development or are  currently optimal primarily in niche
markets, sufficient information was not available to include their costs  and reduction potential in this
analysis.


1.3.2   Calculation  of Indirect Emissions and Costs for  Options Applicable to
        Stationary Equipment

For three of the technology  options assessed for stationary  equipment28—distributed  systems,  HFC
secondary loop systems, and  ammonia secondary loop systems—a TEWI analysis  was conducted to
account for  "indirect"  emissions and costs associated with changes in energy consumption relative to
centralized direct  expansion  systems, the  conventional HFC technology assumed to be  replaced.
28 Three other technology options analyzed quantitatively in this report apply only to MVACs. For those options,
the net costs and emissions associated with changes in energy efficiency (i.e., reduced fuel consumption) are
considered in the analysis, as explained under each of the option descriptions.
                                                                                            -34-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Calculations of indirect  emissions and costs are based  on data from ADL (2002)  and EIA (2000),  as
presented in Table 1-14.

Based on Table  1-14, net energy consumption and costs in the United States were factored into the annual
costs/cost  savings of each of the three technology options.   For all non-U.S.  countries, net  indirect
emissions were adjusted based on country-specific  information on average CO2 emission rates (kg/kWh)
associated with national  electricity generation (Sand et al., 1997). Likewise,  for all non-U.S. countries,
annual costs or cost savings associated with changes in energy consumption were also adjusted based on
country-specific electricity prices for industry (average  1994-1999) (EIA, 2000).

Table 1-14:  Net Annual Emissions and Energy Costs of Replacement Options in the United States in a
600,000 Square Foot Supermarket3^


Charge Size (kg)c
HFC Leak Rate (% of charge/yr)c
Direct Emissions (kg/yr)
Change in Direct Emissions (kg/yr)
Change in Direct Emissions (kg/yr)
per ton of cooling capacityd
Change in Direct Emissions (TCE/yr)e
Energy Consumption (kWh/yr)c
Indirect Emissions (TCE/yr)s
Change in Indirect Emissions (TCE/yr)
Change in Indirect Emissions (TCE/yr)
per ton of cooling capacityd
Change in Net Emissions (TCE/yr)
Net Electricity Cost ($/yr)h
Annual Change in Energy
Consumption (kWh/yr) per ton of
cooling capacity
Estimated Net Electricity Cost
($/ton of cooling capacity^
Centralized
Direct Expansion
System (Base)
1,633
15%
245
N/A
N/A
N/A
1,200,000
198
N/A
N/A
N/A
N/A
N/A
N/A

Distributed
System
408
4%
16
(229)
(0.32)
(204)
1,100,000
182
(17)
(0.02)
(221)
($4,000)
(139)
($5.72)
Ammonia
Secondary Loop
System
180
0%
0
(245)
(0.34)
(218)
1,400,000'
231
33
0.05
(185)
$8,000
278
$11.43

HFC Secondary
Loop System
180
2%
4
(241)
(0.34)
(214)
1,400,000'
231
33
0.05
(181)
$8,000
278
$11.43
Note: Totals may not sum due to independent rounding.
N/A = not applicable.
a600,000-sq. ft. is the typical size of a supermarket in the United States (ADL, 2002).
b To apply this U.S. analysis to the rest of the world, adjustments by relative energy emission factors (kg CCWkWh) and electricity costs for
individual countries, based on Sand et al. (1997) and EIA (2000), were made.
" Based on ADL (2000).
d Assumes that conventional direct expansion systems require 5 pounds (or 2.27 kg) of refrigerant per ton of cooling capacity (Smithart, 2000).
For a 600,000 sq. ft. store using 3,600 Ib. refrigerant (ADL, 2002), this translates to 720 tons of cooling capacity.
e Assumes the refrigerant is R-404A (ADL, 2002).
' Recent technological advances on secondary loop refrigeration systems for supermarkets suggest that, with the use of improved technological
features and design/manufacturing/contractor experience, these systems can lead to significant reductions in energy consumption (Walker,
2000; Baxter, 2003; Kazachki, 2003);  however, these reductions are not assumed in this analysis.
a Assumes a national average emissions factor of 0.606 kg CCWkWh (EIA,  2004).
h Assumes average energy costs for U.S. between 1994-1999 (of approximately $0.04/kWh) based on EIA (2000).
' Calculated by dividing the change in energy consumption (kWh/yr) by the  charge size of direct expansion systems(kg) and multiplying by the
quantity of refrigerant needed on a per ton of cooling capacity basis for conventional direct expansion systems (see footnote e). E.g., for
secondary loop systems: (1,400,000 -1,200,000) kWh/yr •*• 1,633 kg charge x 2.27 kg/ton of cooling capacity = 278 kWh/ton of cooling
capacity.
                                                                                                            -35-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
i Calculated by multiplying annual change in energy consumption per ton of cooling capacity by average energy costs (assumed to be
approximately $0.04/kWh, as explained in footnote h).

Thus, for HFC  and ammonia  secondary loop systems, which  are associated with decreased energy
efficiency relative  to  centralized HFC  direct expansion  systems and, therefore,  increased indirect
emissions, the net increase in indirect emissions was  subtracted from the estimated savings in direct
emissions, and annual energy costs were accounted for in the analysis.  Conversely, for distributed
systems, which  are  associated  with increased energy  efficiency compared to centralized HFC direct
expansion systems,  the net savings in indirect emissions were added to the savings of direct emissions,
and annual  cost savings associated with lower electricity costs were  accounted for in  the analysis.
Therefore, overall,  the calculation of indirect emissions and costs  resulted in slightly lower emission
benefits and higher  annual costs for HFC and ammonia secondary loop systems, and in a slightly greater
emission benefit and lower annual costs for direct expansion systems.


1.3.3   Summary of Technical Applicability and Market Penetration of Abatement
        Options

Table 1-15 summarizes the percent of total  refrigeration/air-conditioning sector emissions that may be
technically abated by  each of the  options  explored in this analysis, based on the percent of sector
emissions from each end-use (which varies  by region), as provided in Table  1-3.  Market penetration
values for each abatement option were developed for each region when possible, to best reflect qualitative
information  available  on  region-specific   realities  and  possible  future  action.   The  commercial
refrigeration and MVAC technology options explored in this chapter are assumed to penetrate only new
(not existing) equipment, where "new" equipment is defined as equipment manufactured in 2005 or later.
Table 1-16 presents the assumed maximum market penetration for the technology options into equipment
manufactured in 2005, 2010, 2015 and 2020.  Table 1-17 presents the final maximum penetration into the
installed base of equipment, taking into account the percent of each market that is new (i.e., manufactured
in 2005 or  beyond) in  all preceding years.  Values  from Table 1-17 are multiplied  by technical
applicabilities  (Table 1-15) and the reduction efficiency to generate the percent reduction off baseline
emissions for each  option,  as presented in Table 1-18.  The text box provided in Section 1.4 provides
further explanation on how the results (percent reduction off baseline emissions) are calculated.
                                                                                             -36-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 1-15: Summary of Technical Applicability of Abatement Options by Region, Percent3
Abatement Option

Refrigerant Recovery
from Small
Equipment
Leak Repair for Large
Equipment
Ammonia Secondary
Loop
Distributed System
HFC Secondary Loop
System
Enhanced HFC-134a in
MVACs
HFC-152ainMVACs
C02 in MVACs

LO
§
fM

26.3
23.7
39.7
39.7
39.7

37.0
37.0
37.0
U.S.&
o
o
fM

24.0
26.0
43.6
43.6
43.6

28.5
28.5
28.5
Japan
LO
O
fM

23.7
26.3
44.1
44.1
44.1

22.7
22.7
22.7

o
S
fM

24.1
25.9
43.1
43.1
43.1

19.5
19.5
19.5
Europe, Australia, New
Zealand, & All Other
Developed Countries
LO O LO O
§T— T— fM
O O O
fM fM fM fM

30.7 28.3 23.0 23.0
19.3 21.7 27.0 27.0
32.3 36.3 45.4 45.1
32.3 36.3 45.4 45.1
32.3 36.3 45.4 45.1

48.7 40.4 20.5 15.9
48.7 40.4 20.5 15.9
48.7 40.4 20.5 15.9
Latin America
LO
§
fM

19.4
30.6
51.2
51.2
51.2

18.7
18.7
18.7
o
o
fM

19.5
30.5
51.0
51.0
51.0

16.2
16.2
16.2
& Caribbean
LO
O
fM

19.6
30.4
51.0
51.0
51.0

10.7
10.7
10.7
o
S
fM

20.3
29.7
49.4
49.4
49.4

7.8
7.8
7.8
Russia, CEITs, & All Other
Developing Countries
LO O LO
§T— T—
O O
tM tM tM

16.0 17.5 19.7
34.0 32.5 30.3
57.0 54.4 50.9
57.0 54.4 50.9
57.0 54.4 50.9

9.5 10.7 10.9
9.5 10.7 10.9
9.5 10.7 10.9
o
S
tM

22.2
27.8
46.3
46.3
46.3

13.6
13.6
13.6
a Expressed as a percent of total refrigeration and air-conditioning emissions.
                                                                                                                                    -37-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 1-16: Incremental Maximum Market Penetration of Technology Options into New Equipment by Region, Expressed as a Percent of Emissions
from New Refrigeration/Air-Conditioning Equipment3
Abatement Option

Ammonia
Secondary Loop
Distributed System
HFC Secondary
Loop System
Enhanced HFC-
134ainMVACs
HFC-152ain
MVACs
C02 in MVACs
U.S.
LO O LO O
O T— T— fM
O O O O
fM fM fM fM

3 5 13 20
8 15 23 30
8 15 23 30

10 65 75 65

0 5 15 25
0 5 10 10
Europe
LO O LO O
O T— T— fM
O O O O
fM fM fM fM

5 10 13 15
10 20 30 40
5 10 18 25

40 20 10 0

0 25 15 0
0 30 75 100
Japan, Australia & New
Zealand
LO O LO O
O T— T— fM
O O O O
fM fM fM fM

5 10 13 15
10 20 30 40
5 10 18 25

40 25 10 0

0 25 40 25
0 25 50 75
All Other Developed
Countries
LO O LO O
O T— T— fM
O O O O
fM fM fM fM

5 10 13 15
10 20 30 40
5 10 18 25

10 65 75 65

0 5 15 25
0 5 10 10
Latin America &
Caribbean, Russia,
CEITs, & All Other
Developing Countries
LO O LO O
O T— T— fM
O O O O
fM fM fM fM

5 10 10 10
8 15 20 25
8 10 13 10

0000

0 20 35 50
0 0 5 10
a Expressed as a percentage of new equipment for the given year. The baseline market penetration of all technology options is assumed to be zero so that only incremental market penetration is
analyzed.
                                                                                                                                    -38-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 1-17: Incremental Maximum Market Penetration of All Abatement Options by Region, Expressed as a Percent of Total Refrigeration/Air-
Conditioning Emissions3
Abatement Option







Refrigerant Recovery
from Small
Equipment
Leak Repair for
Large Equipment
Ammonia Secondary
Loop
Distributed System
HFC Secondary
Loop System
Enhanced HFC-134a
in MVACs
HFC-152ainMVACs
C02 in MVACs
U.S.



LO O LO O
O T— T— fM
O O O O
fM fM fM fM

5.0 10.0 10.0 15.0

3.0 3.0 5.0 5.0

0.2 1.4 4.3 9.4
0.5 4.1 9.9 17.9

0.5 4.1 9.9 17.9

0.8 18.8 48.3 68.3
0.0 1.3 5.8 14.1
0.0 1.3 4.6 8.3
Europe



LO O LO O
O T— T— fM
O O O O
fM fM fM fM

5.0 10.0 10.0 15.0

3.0 3.0 5.0 5.0

0.3 2.6 6.1 8.8
0.6 5.3 12.9 20.5

0.3 2.6 7.0 12.0

3.3 15.0 20.8 11.2
0.0 6.3 14.2 14.2
0.0 7.5 31.3 65.8
Japan, Australia & New
Zealand


LO O LO O
O T— T— fM
O O O O
fM fM fM fM

5.0 10.0 10.0 15.0

3.0 3.0 5.0 5.0

0.3 2.6 6.1 8.8
0.6 5.3 12.9 20.5

0.3 2.6 7.0 12.0

3.3 16.3 22.9 12.8
0.0 6.3 20.4 30.8
0.0 6.3 22.9 47.5
All Other Developed
Countries


LO O LO O
O T— T— fM
O O O O
fM fM fM fM

5.0 10.0 10.0 15.0

3.0 3.0 5.0 5.0

0.3 2.6 6.1 8.8
0.6 5.3 12.9 20.5

0.3 2.6 7.0 12.0

0.8 18.8 48.3 68.3
0.0 1.3 5.8 14.1
0.0 1.3 4.6 8.3
Latin America &
Caribbean, Russia, CEITs,
& All Other Developing
Countries
LO O LO O
O T— T— fM
O O O O
fM fM fM fM

20.0 30.0 40.0 50.0

5.0 10.0 12.0 15.0

0.3 2.6 5.6 7.0
0.5 4.1 9.4 13.7

0.5 3.2 6.6 7.9

0.0 0.0 0.0 0.0
0.0 5.0 17.1 33.4
0.0 0.0 1.3 4.6
a Expressed as a percentage of entire installed base. The baseline market penetration is assumed to be zero, unless otherwise noted.
b Shown percentage values are incremental relative to the baseline market penetration, which is assumed to be 80 percent in developed countries and 30 percent in developing countries.
                                                                                                                                            -39-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 1-18: Percent Reduction off Baseline Emissions of All Abatement Options fr
Abatement
Option




Refrigerant
Recovery from
Small
Equipment
Leak Repair for
Large
Equipment
Ammonia
Secondary
Loop
Distributed
System
HFC Secondary
Loop System
Enhanced HFC-
134a in
MVACs
HFC-152ain
MVACs
C02 in MVACs
U.S.


LO O LO O
§T— T— fM
O O O
fM fM fM fM


1.1 2.0 2.0 3.1


0.7 0.7 1.2 1.2


0.1 0.6 1.9 4.0

0.2 1.6 4.1 7.2

0.2 1.7 4.3 7.6


0.2 2.7 5.5 6.7

0.0 0.3 1.2 2.4
0.0 0.4 1.0 1.6
Europe


LO O LO O
§T— T— fM
O O O
tM tM tM tM


1.3 2.4 2.0 2.9


0.5 0.6 1.3 1.3


0.1 1.0 2.8 4.0

0.2 1.8 5.5 8.6

0.1 0.9 3.1 5.3


0.8 3.0 2.1 0.9

0.0 2.2 2.6 2.0
0.0 3.0 6.4 10.4
Japan


LO O LO O
§T— T— fM
O O O
tM tM tM tM


1.1 2.0 2.0 3.1


0.7 0.7 1.2 1.2


0.1 1.2 2.7 3.8

0.2 2.1 5.3 8.2

0.1 1.1 3.0 5.1


0.6 2.3 2.6 1.2

0.0 1.6 4.1 5.3
0.0 1.8 5.2 9.3
/ Region
Australia & New
Zealand

LO O LO O
§T— T— fM
O O O
tM tM tM tM


1.3 2.4 2.0 2.9


0.5 0.6 1.3 1.3


0.1 1.0 2.8 4.0

0.2 1.8 5.5 8.6

0.1 0.9 3.1 5.3


0.8 3.3 2.3 1.0

0.0 2.2 3.7 4.4
0.0 2.5 4.7 7.5
All Other Developed
Countries

LO O LO O
§T— T— fM
O O O
tM tM tM tM


1.3 2.4 2.0 2.9


0.5 0.6 1.3 1.3


0.1 1.0 2.8 4.0

0.2 1.8 5.5 8.6

0.1 0.9 3.1 5.3


0.2 3.8 5.0 5.4

0.0 0.4 1.1 2.0
0.0 0.5 0.9 1.3
Latin America &
Caribbean

LO O LO O
§T— T— tM
O O O
tM tM tM tM


3.3 5.0 6.7 8.6


1.5 2.9 3.5 4.2


0.2 1.4 2.8 3.5

0.2 1.9 4.4 6.3

0.2 1.6 3.3 3.9


0.0 0.0 0.0 0.0

0.0 0.7 1.6 2.3
0.0 0.0 0.1 0.4
Russia, CEITs, & All
Other Developing
Countries
LO O LO O
§T— T— tM
O O O
tM tM tM tM


2.7 4.5 6.7 9.4


1.6 3.1 3.5 4.0


0.2 1.4 2.8 3.2

0.2 2.1 4.4 5.9

0.3 1.7 3.3 3.6


0.0 0.0 0.0 0.0

0.0 0.5 1.7 4.0
0.0 0.0 0.1 0.6
                                                                                                                            -40-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
1.4
Results
Emission reduction potential for abatement options varies by region based on assumed end-use breakouts
(provided in Table  1-3) and on qualitative information regarding current and future likelihood of market
penetration by region.  The percent reduction from the baseline associated with each abatement option is
calculated by multiplying the technical applicability (from Table 1-15) by both the incremental maximum
market  penetration (from Table 1-16) and the  reduction  efficiency.   For more information  on how
emission reductions are  calculated for each option, please  see the Text Box below, which presents an
illustrative example of the emission reduction methodology.

Table 1-19 through Table 1-21 provide a summary of the potential emission reduction opportunities and
associated annualized costs for the United States, Non-U.S.  Annex I and Non-Annex I countries in 2020.
The  annualized costs to reduce one ton of carbon  equivalent (TCE) are presented for two different
discount rate scenarios: 4 percent and 20 percent.  The  tax rate  associated with each cost scenario is 0
percent and 40 percent,  respectively.  The  results are ordered by increasing costs per TCE, using the
highest cost in the region under the 4 percent discount rate.  The emissions reduced by the option and a
cumulative total of emissions reduced,  in MMTCE and percent of the regional refrigeration  and air-
conditioning baseline, are presented.

Table 1-19: United States Emission Reductions in 2020 and Break-Even Costs for Refrigeration and Air-
Conditioning	
Reduction Option
                              Cost (2000$/TCE)
                              Discount/Tax Rate
                             4%/0%    20%/40%
                                       Emission
                                     Reduction of     Percent     Cumulative    Cumulative %
                                        Option    Reduction from  Reductions   Reduction from
                                       (MMTCE)    2020 Baseline    (MMTCE)     2020 Baseline
Enhanced HFC-134a in MVACs    ($275.47)    ($35.34)       4.51
HFC-152ainMVACs            ($73.65)     $55.55        1.66
C02 in MVACs                 ($23.44)     $422.48       1.09
Distributed Systems             ($12.70)     $62.60        4.87
Leak Repair for Large Equipment   ($11.74)     ($3.78)        0.83
Refrigerant Recovery from Small
  Equipment                   $0.49      $0.49        2.09
HFC Secondary Loop            $28.87     $68.66        5.16
Ammonia Secondary Loop         $30.19     $78.79        2.74
                                                      6.7%
                                                      2.4%
                                                      1.6%
                                                      7.2%
                                                      1.2%

                                                      3.1%
                                                      7.6%
                                                      4.0%
4.51
6.17
7.26
12.13
12.96

15.05
20.20
22.94
6.7%
9.1%
10.7%
17.9%
19.1%

22.2%
29.8%
33.8%
                                                                                               -41-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 1-20: Non-U.S. Annex I Emission Reductions in 2020 and Break-Even Costs for Refrigeration and Air-
Conditioning	
                               Break-Even Cost (2000$/TCE)a
                                    Discount/Tax Rate
                                4%/0%
Reduction Option
                             Low
                          High
Low
High
                              Cumulative
Emission  Percent                %
Reduction Reduction Cumulative  Reduction
of Option from 2020 Reductions  from 2020
(MMTCE)  Baseline   (MMTCE)    Baseline
Leak Repair for Large Equipment
Distributed Systems
Refrigerant Recovery from Small
Equipment
Enhanced HFC-134a in MVACs
HFC-152a in MVACs
Ammonia Secondary Loop
HFC Secondary Loop
C02 in MVACs
($11.74)
($57.61)

$0.49
($161.67)
($4.59)
$18.59
$16.97
$113.91
($11.74)
($8.08)

$0.49
$7.02
$30.81
$123.84
$125.39
$180.72
($3.78)
$15.99

$0.49
$299.69
$164.88
$62.61
$53.67
$763.86
($3.78)
$69.97

$0.49
$468.38
$205.42
$180.96
$173.22
$839.80
0.
3.

1.
0.
1.
1.
1.
2.
,64
,07

52
,51
,31
,44
,90
84
1.7%
8.0%

4.0%
1.3%
3.4%
3.8%
5.0%
7.5%
0.64
3.71

5.24
5.74
7.05
8.50
10.40
13.24
1.7%
9.7%

13.7%
15.1%
18.5%
22.3%
27.3%
34.7%
a Costs vary by country/region based on one-time or annual adjustment factors (e.g., electricity price, fuel price, etc.); therefore, the lowest and
highest costs for the region are shown.

Table 1-21: Non Annex I Emission Reductions in 2020 and Break-Even Costs for Refrigeration and Air-
Conditioning
Break-Even Cost (2000$/TCE)a
Discount/Tax Rate
4%/n% ?n%/4n%

Reduction Option
Enhanced HFC-134a in MVACs
Leak Repair for Large Equipment
Distributed Systems
Refrigerant Recovery from Small
Equipment
HFC-152a in MVACs
HFC Secondary Loop
Ammonia Secondary Loop
C02 in MVACs

Low
($13.12)
($11.74)
($25.88)

$0.49
$26.58
$10.53
$12.47
$172.75

High
($13.12)
($11.74)
($4.60)

$0.49
$26.58
$59.19
$59.61
$172.75

Low
$448.24
($3.78)
$48.87

$0.49
$200.58
$49.87
$60.78
$830.74

High
$448.24
($3.78)
$70.09

$0.49
$200.58
$104.49
$114.41
$830.74
Emission
Reduction
of Option
(MMTCE)
-
2.12
3.17

4.91
1.96
1.93
1.74
0.30
Percent
Reduction
from 2020
Baseline
0.0%
4.0%
6.0%

9.3%
3.7%
3.7%
3.3%
0.6%
Cumulative
Reductions
(MMTCE)
-
2.12
5.29

10.20
12.16
14.09
15.83
16.13
Cumulative
%
Reduction
from 2020
Baseline
0.0%
4.0%
10.0%

19.3%
23.0%
26.7%
29.9%
30.5%
a Costs vary by country/region based on one-time or annual adjustment factors (e.g., electricity price, fuel price, etc.); therefore, the lowest and
highest costs for the region are shown.
1.5
Summary
Baseline HFC emissions from refrigeration/AC are expected to grow significantly between the years 2005
and 2020 as HFCs become increasingly used throughout the world to replace gases phased out under the
Montreal Protocol.   The  highest percentage  emissions  growth  is expected to occur  in  developing
countries.

This analysis considers the  costs and emission reduction potential of eight practice and  technology
emissions mitigation options: (1) Leak Repair for Large Equipment;  (2) Refrigerant Recovery/Recycling
from  Small Equipment; (3)  Distributed System;  (4) HFC Secondary  Loop; (5) Ammonia Secondary
Loop;  (6) Enhanced HFC-134a Systems in MVACs; (7) HFC-152a Systems in MVACs; and (8) CO2
Systems in MVACs.  The  costs and emission reduction benefits of each option were compared in each
region.   Overall, enhanced HFC-134a systems in MVACs represent the most cost-effective option for
                                                                                              -42-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
reducing emissions in all regions at a 4 percent discount rate.  Distributed systems represent the most
promising option for reducing the greatest amount of emissions worldwide, at a negative or low cost. In
developing countries,  the low-cost option  of refrigerant recovery is also promising  for significantly
reducing emissions.


1.6        References

ACGIH (American Conference of Governmental Industrial Hygienists, Inc.). 1999. Guide to
Occupational Exposure Values. Cincinnati, OH, 1999.

ACHRNews. 2000. An Argument for NH3's Superiority over Other Refrigerants. Air Conditioning
Heating Refrigeration News. Business News Publishing Company. July 27, 2000.

ADL. 2002. Global Comparative Analysis of HFC and Alternative Technologies for Refrigeration, Air
Conditioning, Foam, Solvent, Aerosol Propellant, and Fire Protection Applications. Final Report to the
Alliance for Responsible  Atmospheric Policy. Reference Number 75966. Arthur D. Little, Inc. March 21,
2002.

ADL. 1999. Global Comparative Analysis of HFC and Alternative Technologies for Refrigeration, Air
Conditioning, Foam, Solvent, Aerosol Propellant, and Fire Protection Applications. Final Report to the
Alliance for Responsible  Atmospheric Policy. Reference Number 49648. Arthur D. Little, Inc. August 23,
1999.

Amrane, Karim. 2001. Personal communication between Karin Amrane, the Director of Public Policy at
the Air-Conditioning and Refrigeration Institute (ARI), and ICF Consulting. September 2001.

Anderson, M. Kent. 2001. Personal communication between Kent Anderson, the President of the
International Institute of Ammonia Refrigeration (IIAR), and ICF Consulting. December 5, 2001.

ASHRAE. 2002. Ammonia as a Refrigerant: Position Document. American Society of Heating,
Refrigerating and Air-Conditioning Engineers, Inc. Approved by ASHRAE Board of Directors 17
January 2002. Available online at
.

Atkinson, W. 2000. Review comments on draft report, U.S. High GWP Gas Emissions 1990-2010:
Inventories, Projections, and Opportunities for Reductions [Refrigeration and Air-Conditioning
Chapter]. Sun Test Engineering. March 2000.

Baker, J.A. 2002. Mobile Air Conditioning Sector Update. Presented at the  19th Meeting of the Ozone
Operations Resource Group (OORG), The World Bank, in Washington, DC. March 28, 2002.

Barbusse, S., D. Clodic, and J.P. Roumegoux. 1998. Mobile Air Conditioning; Measurement and
Simulation of Energy and Fuel Consumptions. Presented at the Earth Technologies Forum. The Alliance
for Responsible Atmospheric Policy. October 1998.

Baxter, Van D. 2003. IEA Annex 26: Advanced Supermarket Refrigeration/Heat Recovery Systems. Final
Report Volume 1—Executive Summary. Based on information developed in Canada, Denmark, Sweden,
United Kingdom, United States (Operating Agent). Oak Ridge National Laboratory. April 2003.
                                                                                          -43-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Bennett, C. 2000. Personal communication between C. Bennett, Senior Vice President of Althoff
Industries, Inc., and ICF Consulting. December 14, 2000.

Calm, J. 1999. Emissions and Environmental Impacts from Air-Conditioning and Refrigeration Systems.
Joint IPCC/TEAP Expert Meeting on Options for the Limitation of Emissions of HFCs and PFCs. May
1999.

Calm, J.M., D.J. Wuebbles and A.K. Jain. 1999. Impacts on Global Ozone and Climate from Use and
Emission of 2,2-Dichloro-1,1,1 -trifluoroethane (HCFC-123). 42:439-474. Journal of Climate Change.
June 1999.

Calor Gas Refrigeration. 2004. Care Refrigerants Technical Information. Available online at
. Accessed on June 7, 2004.

CARE Web site. 2004. CAREing for our world. Available online at . Accessed on March 1, 2004.

Contracting Business Interactive. 2003. Refrigerant Recovery in Residential Systems. Available online at
. Accessed on July 7, 2003.

Cooper, P.J. 1997. Experience with Secondary Loop Refrigeration Systems in European Supermarkets.
Proceedings of the International Conference on Ozone Protection Technologies, pg. 511. The Alliance for
Responsible Atmospheric Policy. November 1997.

Crawford, J. 2002. Refrigerant Options for Air Conditioning. Presented at the Earth Technologies Forum.
The Alliance for Responsible Atmospheric Policy. March 26, 2002.

Crawford, J. 2000. Review comments on the draft report, U.S. High GWP Gas Emissions 1990-2010:
Inventories, Projections, and Opportunities for Reductions [Refrigeration and Air-Conditioning
Chapter]. The Trane Company. March 2000.

Crawford, J. 1999. Limiting the HFC emissions of Chillers. Joint IPCC/TEAP Expert Meeting on Options
for the Limitation of Emissions of HFCs and PFCs  held in the Netherlands. May 1999.

Digmanese, T. 2004. Peer review comments on U.S. EPA Draft Report, DRAFT Analysis of International
Costs of Abating HFC Emissions from Refrigeration and Air-Conditioning. York International
Corporation. March 19, 2004.

DuPont Customer Service. 2004. List Prices of Refrigerants (based on single 25-pound cylinders).
Customer Service (Tel: 800-441-9409), June 9, 2004.

EC (European Commission). 2003a. How to Considerably Reduce Greenhouse Gas Emissions Due to
Mobile Air  Conditioners Consultation paper from the European Commission Directorate-General
Environment. February 4, 2003.

EC (European Commission). 2003b. Proposal for a Regulation of the European Parliament and of the
Council on  certain fluorinated greenhouse gases. European Commission, Commission of the European
Parties. August 11, 2003. Available online at .
                                                                                          -44-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
EIA. 2004. Form EIA-1605, Long Form for Voluntary Reporting of Greenhouse Gases, Instruction, Data
through 2003 (Appendix C. Adjusted Electricity Emission Factors by State and Region), pp. 50. Energy
Information Administration. 2004. Available online at
. Accessed on
June 7, 2004.

EIA. 2001. Retail Motor Gasoline Prices in Selected Countries, 1990-2000 (U.S. Dollars per Gallon).
Energy Information Administration. 2001. Available online at
. Accessed in September/October 2003.

EIA. 2000. Annual Energy Outlook 2000 (Electricity Prices for Industry). Energy Information
Administration. 2000. Available online at .
Accessed on April 2, 2002.

Environment Canada. 1998. Powering GHG Reductions Through Technology Advancement, pp.185-188.
Environment Canada, Clean Technology Advancement Division. 1998

Faramarzi, R. and D. Walker. 2003. Field Evaluation of Secondary Loop Refrigeration for Supermarkets.
Presented  at the 2003 ASHRAE Winter Meeting in Chicago, IL on 26 January 2003. The American
Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.

Fisher, S.K., J.J. Tomlinson, and P.J. Hughes. 1994. Energy and Global  Warming Impacts ofNot-in-Kind
and Next Generation CFC and HCFC Alternatives. Prepared for the Alternative Fluorocarbons
Environmental Acceptability Study and U.S. Department of Energy. Oak Ridge National Laboratory.

Gaslok. 2002. Gaslok Flyer. Submitted electronically to ICF Consulting  by David Peall. Gaslok. May 13,
2002. The Gaslok Web site is available at .

Geoexchange. 2000. Information on geothermal heat pumps. Geoexchange. 2000. Available online at
.

Greenchill Web site. 2000. Fire and Ice. Sydney Morning Herald. March 18, 2000. Available online at
.

Greenpeace. 2001. Major Japanese Refrigerator Manufacturers to Produce Hydrocarbon Fridges for
Japanese Market in 2002. A Greenpeace position paper prepared for the 35th Meeting of the Multilateral
Fund for the Implementation of the Montreal Protocol in Montreal, Canada. December 5-7 2002.

Hill, William and Ward Atkinson. 2003. Peer review comments on  the U.S. EPA Draft Report, DRAFT
Analysis of International Costs of Abating HFC Emissions from Refrigeration and Air-Conditioning.
General Motors Corporation and Sun Test Engineering. October 28, 2003.

HyChill Web site. 2004. The Case for Hydrocarbons. Available online at .

Hydro Cool Online. 2002. Cool Technologies: Working Without  HFC's. Updated June 2002. Available
online at .

ICF Consulting. 2002a. Analysis on Combined Global Emission Estimates  Scenarios. Deliverable
submitted  by ICF Consulting to the U.S. EPA that included a revised analysis of the estimated level of
                                                                                           -45-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
recycling in other countries. Delivered to Casey Delhotal, Dave Godwin, and Debbie Ottinger of the U.S.
EPA Office of Atmospheric Programs. August 23, 2002.

ICF Consulting. 2002b. ODS Destruction Report. Revised draft report submitted to Julius Banks of the
U.S. EPA Global Programs Division. April 5, 2002.

IPCC. 1996. IPCC Guidelines for National Greenhouse Gas Inventories. Intergovernmental Panel on
Climate Change, United Nations Environment Programme, Organization for Economic Co-Operation and
Development, International Energy Agency.  1996.

Japan Times. 2002. Hydrocarbon Fridges Hit Environment-Savvy Japan. Japan Times. March 19, 2002.

Kazachki, Georgi. 2004. Personal communication from Georgi Kazachki, Director of Research and
Development at the Hill Phoenix to ICF Consulting. February 13, 2004.

Kazachki, Georgi. 2003. A Path Towards Perfection in Supermarket Refrigeration: a Manufacturer's
Perspective on Secondary Collant Systems. Presented by Dr. Georgi Kazachki, Director of Research and
Development at the Hill Phoenix, at the 21st International Institute of Refrigeration (IIR) Congress of
Refrigeration in Washington, DC. International Institute of Refrigeration. August 23, 2003.

Kruse, H. 1996. The State of the Art of Hydrocarbon Technology in Household Refrigeration.
Proceedings of the International Conference on Ozone Protection Technologies, pp. 179-188. The
Alliance for Responsible Atmospheric Policy. October 1996.

Kuijpers, L. 2002. Refrigeration Sector Update. Presented at the 19th Meeting of the Ozone Operations
Resource Group (OORG), The World Bank.  March 28, 2002.

Lundberg, E. 2002. An Enhanced R-134a Climate System. Presented at the 2002 SAE Automotive
Alternative Refrigerant Systems Symposium in Scottsdale, AZ. Society of Automotive Engineers. July 9-
11,2002.

OPROZ. 2001. Report on the Supply and Consumption ofCFCs and Alternatives in Argentina. Oficina
Programa Ozono. February 2001.

Paul, J. 1996. A Fresh Look at Hydrocarbon  Refrigeration: Experience and Outlook.  Proceedings of the
International Conference on Ozone Protection Technologies, pp. 252-259. The Alliance for Responsible
Atmospheric Policy. October 1996.

RTOC. 2003. 2002 Report of the Refrigeration, Air Conditioning and Heat Pumps Technical Options
Committee: 2002 Assessment. Section 8.4.2.7. Refrigeration Technical Options Committee. January 2003.

Rafferty, Kevin D. 2003. Absorption Refrigeration. Geo-Heat Center, Bulletin Vol. 19, No. 1. Available
online, at .

Rawlings, P. 2000. Personal communication  between P. Rawlings of the Geothermal Heat Pump
Consortium and ICF Consulting. December 8, 2000.

Rugh, John and Valerie Hovland. 2003. National and World Fuel Savings and CO2 Emission Reductions
by Increasing Vehicle Air Conditioning COP. Presented by John Rugh and Valerie Hovland of the


                                                                                         -46-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
National Renewable Energy Laboratory at the SAE 2003 Automotive Alternate Refrigerant Systems
Symposium in Phoenix, AZ. Society of Automotive Engineers. July 17, 2003.

SAE. 2003a. Alternative Refrigerants Assessment Workshop. Presented at the 2003 Conference on
Mobile Air Conditioning Technologies in Phoenix, AZ. Society of Automotive Engineers. July 14, 2003.

SAE. 2003b. SAE Alternate Refrigerant Cooperative Research Project: Project Overview. Slide
presentation given by Ward Atkinson at the 2003 Conference on Mobile Air Conditioning Technologies
in Phoenix, AZ. Society of Automotive Engineers. July 15, 2003.

SAE. 2000. 2000 Conference on Mobile Air Conditioning Technologies in Phoenix, AZ. Society of
Automotive Engineers. July 2000.

Sand, J.R., S.K. Fischer, and V.D. Baxter. 1997. Energy and Global Warming Impacts of HFC
Refrigerants  and Emerging Technologies. Prepared for the Alternative Fluorocarbons Environmental
Acceptability Study and U.S. Department of Energy. Oak Ridge National Laboratory. 1997.

SKF. 2003. Hybrid  bearings in oil-free air conditioning and refrigeration compressors. Evolution. SKF's
business and technology magazine. Available online at
. Accessed on October 17, 2003

Smithart, G. 2003. Peer review comments on U.S. EPA Draft Report, DRAFT Analysis of International
Costs of Abating HFC Emissions from Refrigeration and Air-Conditioning. Turbocor Inc. October  17,
2003.

Smithart, G. 2000. Personal communication between G. Smithart, Director of Environmental Affairs at
The Trane Company, and ICF Consulting. July 1, 2000.

UNEP. 1999a. Report of the TEAP HFC andPFC Task Force. United Nations Environment Programme.
October 1999.

UNEP. 1999b. Production and Consumption of Ozone Depleting Substances 1986-1998. United Nations
Environment Programme. October 1999.

UNEP. 1998.1998 Report of the Technology and Economic Assessment Panel (Pursuant to Article 6 of
the Montreal Protocol). United Nations Environment Programme. 1998.

Universal Currency Converter Web site. 2003. Available online at . Accessed in
July  2003

U.S.  EPA. 2003. Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990- 2001. EPA #430-R-03-
004.  U.S. Environmental Protection Agency, Office of Atmospheric Programs. April 2003.

U.S.  EPA. 2002. Building Owners Save Money, Save the Earth: Replace Your CFC Air Conditioning
Chiller. U.S. EPA #430-F-02-026. U.S. Environmental Protection Agency, Global Programs Division and
Climate Protection Partnerships Division. December 2002.
                                                                                         -47-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
U.S. EPA. 2001a. U.S. High GWP Gas Emissions 1990-2010: Inventories, Projections, and Opportunities
for Reductions. U.S.EPA #000-F-97-000. U.S. Environmental Protection Agency, Office of Air and
Radiation. June 2001.

U.S. EPA. 2001b. Draft Regulatory Impact Analysis: The Substitutes Recycling Rule. Prepared by ICF
Incorporated for the U.S. Environmental Protection Agency. September 2001.

U.S. EPA. 1998. Draft Regulatory Impact Analysis:  The Substitutes Recycling Rule. Prepared by ICF
Incorporated for U.S. Environmental Protection Agency. May 1998.

U.S. EPA. 1997. Options for Reducing Refrigerant Emissions from Supermarket Systems. EPA-600/R-97-
039. Prepared by Eugene F. Troy of ICF Consulting for the U.S. Environmental Protection Agency. April
1997.

U.S. EPA. 1993. Protection of Stratospheric Ozone; Refrigerant Recycling, Final Rule. Federal Register
citation 58 FR 28660. U.S. Environmental Protection Agency. May 14, 1993. Available online at
.

VDA 2003. Various presentations at the Alternative Refrigerant Winter Meeting: Automotive Air
Conditioning and Heat Pump Systems in Saalfelden, Austria. February 13-14, 2003. Verband der
Automobilindustrie, Frankfurt, Germany. Available online at . Accessed on 24 December 2003.

Walker, D. 2000. "Low-charge refrigeration for supermarkets." IEA Heat Pump Centre Newsletter. Vol.
18, No. 1/2000, pp. 13-16.

Ward's. 2001. Ward's World Motor Vehicle Data, ISBN Number 0-910589-79-8. Southfield, MI, 2001.

World Bank. 2002. CFCMarkets in Latin America. Latin America and Caribbean Region Sustainable
Development Working Paper No. 14. Prepared by ICF Consulting for the World Bank. December 2002.

Wuebbles, Donald J. and James M. Calm. 1997. An Environmental Rationale for Retention of
Endangered Chemicals. Science. 278:1090-1091. November 1997.

Xu, J. and J. Amin. 2000. Development of Improved R134a Refrigerant System. Presented at the 2000
SAE Automotive Alternative Refrigerant Systems Symposium in Scottsdale, AZ. Society of Automotive
Engineers. July 11-13, 2000.
                                                                                          -48-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
2   Analysis of Costs to Abate  International HFC Emissions
     from Aerosols
2.1        Introduction


Aerosol propellants are used in metered dose inhalers (MDIs), as well as a variety of consumer products.
Historically, the majority of aerosol applications have used chlorofluorocarbons (CFCs) as propellants;
however, efforts have been taken to transition away from CFC propellants. As a result of initiatives under
the Montreal Protocol, many pharmaceutical companies that produce MDIs have committed to develop
alternatives to CFC-based MDIs.  Furthermore, many consumer products, such as spray deodorants and
hair sprays and specialty aerosol uses, such as freeze spray and dust removal products, have successfully
been reformulated with hydrocarbon propellants or replaced with Not-in-Kind (NIK) substitutes such as
pump sprays or solid and roll-on deodorants. Such  transitions occurred in the United States as far back as
1977, when the country placed a ban on CFC propellants in non-MDI aerosols for non-essential uses.

Various hydrofluorocarbons  (HFCs)  have also been introduced as alternative propellants in aerosol
applications.   These HFCs include  HFC-134a,   HFC-152a,  and HFC-227ea,  and  are  associated,
respectively, with 100-year global warming potentials  (GWPs) of 1,300, 140, and  2,900 times that of
carbon dioxide.   Aerosol HFCs are  emitted  from pharmaceutical products (primarily MDIs)29  and
consumer products (primarily specialty aerosols).

The pharmaceutical aerosol  industry is actively working to develop HFC-propellant MDIs, a type of
inhaled therapy  used to treat asthma and  chronic obstructive pulmonary disease (COPD).  The earliest
non-CFC substitute products used HFC-134a, but eventually the industry expects products to utilize HFC-
227ea as well.  In addition to MDIs that  use propellants, dry powder inhalers (DPIs) can be used as a
substitute  for some  MDIs.   Because MDIs are medical  devices, substitute propellants must meet far
stricter performance and toxicology specifications  than would be required in most other products.  For
example, prior to entering the United States market, the  Food and Drug Administration must approve
reformulated MDIs with an alternative propellant.

Chemical  manufacturers are also  marketing HFCs, especially HFC-152a and HFC-134a, as aerosol
propellants in consumer products, primarily for use in specialty applications.  This use is particularly true
for applications  where flammability or volatile organic  compound (VOC) emissions  and their impact on
urban air quality are a concern.  If HFC use is accelerated, increased public concern  will likely facilitate
the aerosol industry's responsible use of these chemicals (UNEP, 1999).
29 Note that this analysis does not include non-MDI aerosols produced by the pharmaceutical industry such as
bandage sprays.
                                                                                         -49-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
2.2        Baseline Emission  Estimates


2.2.1   Emission Estimating Methodology

Description of Methodology

A full description of the emission model used to calculate ODS substitute emissions from all sectors is
provided in Appendix A.  Specific  information on how the model  calculates  aerosol emissions is
described below.

EPA uses a detailed Vintaging Model of ODS-containing equipment and products to estimate the use and
emissions of various ODS substitutes in the United States,  including HFCs.  Emission baselines from
non-U.S. countries were derived using country-specific ODS consumption estimates as reported under the
Montreal Protocol in conjunction with Vintaging Model output for each ODS-consuming end-use sector.
For sectors where detailed information was available, these data were incorporated into country-specific
versions of the Vintaging Model to customize emission estimates.  In the  absence of country level data,
these preliminary estimates were calculated by assuming that the transition  from ODSs to HFCs and other
substitutes follows the same  general substitution patterns internationally as observed in the United States.
From this preliminary assumption, emission estimates were then tailored to individual  countries  or
regions  by applying adjustment factors to U.S. substitution scenarios, based on relative differences in (1)
economic growth; (2) rates of ODS phaseout; and (3) the distribution of ODS use across end uses in each
region or country.

Emission Equations

All HFCs used in aerosols are assumed to be emitted in the year of manufacture.  Since there is currently
no aerosol recycling, all of the annual production of aerosol propellants is assumed to be released to the
atmosphere. The following equation describes the emissions from the aerosols sector:

                                           Ej = Qq

Where:

Ej =   Emissions. Total emissions of a specific chemical in a year j from use in aerosol products, by
        weight.
Qcj =   Quantity of Chemical.  Total quantity of a specific chemical  contained in aerosol products sold in
        the year j, by weight
j   =   Year of emissions.


For aerosols, two separate baseline emissions were created; one baseline tracks HFC emissions from the
MDI industry, while the other estimates HFC emissions  from consumer and specialty products (i.e., non-
MDI aerosols).

Regional Adjustments

The adjustment factor assumptions used in the global aerosol emissions estimating methodology, which
are described in more  detail in Appendix A, include both economic and timing adjustment factors.  The
timing factors reflect that some nations are not moving at the same  pace out of the use of CFCs and into
                                                                                          -50-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
the use of HFCs as other nations are. For all ODS end-uses, by 2005, it is assumed that Non-Annex I
(i.e., developing) countries are 75 percent through the CFC transition, and by 2010, the CFC transition is
complete.   These timing factors are partially offset by  generally higher growth rates in  developing
countries.

In addition, the methodology used to estimate global aerosol emissions includes an adjustment specific to
non-MDI aerosols. This adjustment was necessary because the ban on CFC use in aerosols caused the
United States to transition out of CFCs earlier than other countries.  Therefore,  the unweighted U.S.
consumption of non-MDI ODS substitutes (including a large market segment that transitioned into NIK
or hydrocarbon substitutes) was used as a proxy for U.S. 1990 non-MDI ODS consumption (see Step 3 in
Appendix A). For countries other than the United States, it was then assumed that 15 percent of the non-
MDI aerosols ODS consumption transitioned to HFCs,  while the remainder is assumed to transition to
NIK or hydrocarbon alternatives.


2.2.2  Baseline Emissions

Table 2-1 and Table 2-2 display total HFC emission estimates in million metric tons of carbon equivalent
(MMTCE) for the MDI and non-MDI aerosols sectors, respectively.  Both HFC-134a and HFC-227ea are
expected to be emitted from the use of MDIs in the future as substitutes for CFCs.  The MDI emissions
baseline accounts for all  emissions of HFC-227ea from the aerosols sector.  Non-MDI emissions are
responsible for the majority (approximately 60  percent) of the HFC-134a emissions from the aerosols
sector  (mainly  for specialty  applications) and all  of the HFC-152a emissions  (mostly  formulated
consumer products).

Table 2-1: Baseline HFC Emission Estimates from MDI Aerosols (MMTCE)
Region
United States
Non-U.S. Annex I
Non-Annex I
Total
2005
1.58
2.12
0.90
4.60
2010
1.70
2.35
1.47
5.52
2015
1.84
2.48
1.79
6.12
2020
1.98
2.63
2.21
6.82
 Note: Totals may not sum due to independent rounding.

Table 2-2: Baseline HFC Emission Estimates from Non-MDI Aerosols (MMTCE)
Region                  2005          2010         2015          2020
United States
Non-U.S. Annex I
Non-Annex I
3.00
4.65
0.02
3.31
5.59
0.03
3.65
6.12
0.03
4.03
6.71
0.04
Total	7.66	8.92	9.80	10.78
Note: Totals may not sum due to independent rounding.

2.3        Costs of HFC  Emission  Reductions for Aerosols

This section presents a cost analysis for achieving HFC emission reductions from the emission baselines
presented in Table 2-1 and Table 2-2 above.  All cost analyses for the non-MDI emission reduction
options assume a 10-year project lifetime; the cost analysis for the MDI option assumes a 15-year project
lifetime. Each abatement option is described below.
                                                                                          -51-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
2.3.1  Description and Cost Analysis of Abatement Options

Four potential mitigation options are analyzed in this report. The first mitigation option has the potential
to abate emissions from the MDI baseline (Table 2-1), while the other three options have the potential to
abate emissions from the non-MDI baseline (Table 2-2). The options are:

    •  MDI replacement with dry powder inhalers (DPIs);
    •  Non-MDI replacement with lower GWP HFCs;
    •  Non-MDI replacement with Not-In-Kind (NIK) alternatives; and
    •  Non-MDI replacement with hydrocarbon aerosol propellants.
DPIs have been authorized as a  substitute  for some HFC-propellant MDIs.  The non-MDI baseline
includes emissions from specialty  aerosol uses such as tire inflators, electronics cleaning products, dust
removal, freeze spray, signaling devices, and mold release agents as well as consumer products such as
hairsprays, mousse, deodorants and anti-perspirants, household products, and spray paints (ADL, 1999).
HFCs are  currently used when flammability  issues cannot easily be overcome, such as tire inflators and
air signaling horns that use HFC-134a to avoid potential explosivity associated with  highly flammable
propellants like propane or butane  (ADL, 1999).  HFC-152a has been used in dusters since 1993 (UNEP,
1999),  and continued substitution of HFC-134a with HFC-152a, or a lower GWP gas  in general aerosol
applications, is a reduction strategy that has had significant success thus far, and is expected to continue.
The other  options to reduce HFC emissions from non-MDI aerosol applications include NIK replacement
and hydrocarbon aerosol propellants.

The  remainder of this  section provides  a  description  of the economic  assumptions  for  these  four
abatement options.

MDI Replacement with Dry Powdered Inhalers (DPIs)

As MDIs transition away from CFC use, alternatives such as HFC propellants, DPIs, and oral medications
are  being  developed.  Although hydrocarbons have replaced CFCs as  propellants in many commercial
aerosols, they have been found to be unacceptable  for use in MDIs (IPAC, 1999).   Given the unique
medical requirements for developing MDIs,  and the fact that the industry has been investing heavily in
the  development of HFC technologies, an aerosol replacement for HFC-based MDIs is unlikely to be
developed within the time frame of this analysis. Globally, the number of HFC MDIs used has grown to
more than 100 million in 2001  (UNEP, 2002).  Rather than developing new alternatives that use HFCs,
some MDI manufacturers may turn to DPIs, oral medication, or other NIK alternatives.  In 2001, the
number of multi-dose DPIs used world-wide was estimated at 65 million (UNEP, 2002).30  This analysis
examines the option of further replacing HFC-based MDIs with DPIs because of its technical feasibility
and demonstrated success in the MDI market.

DPIs are a viable option with most anti-asthma drugs, although they are  not successful with all patients or
all drugs.  Micronised dry powder can be inhaled and deposited in the lungs from DPIs  as with MDIs, but
only in patients who are able to inhale robustly enough to transport the powder to the lungs. DPIs are not
suitable for persons with severe asthma or for young children.  Unlike MDIs,  powdered drug particles
contained  in DPIs tend  to aggregate and may cause problems in areas with hot and humid climates
(March Consulting Group, 1999; UNEP, 2002).  Other issues that doctors and patients consider when
30 Multiple-dose DPIs contain pre-measured doses that provide treatment for a day or up to one month. Single-dose
DPIs are also available where only one dose can be loaded at a time (UNEP, 2002).


                                                                                           -52-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
choosing a treatment device include the patient's manual dexterity, ability to adapt to a new device, and
their perception of the effectiveness of the medicine and taste of any added ingredients (Price et al.,
2004). It is important to note that the choice of treatment, including the type of propellant used in MDIs,
is a medical decision involving the pharmaceutical industry, the FDA or other regulatory authority, and
ultimately doctors and their patients.  Doctors and their patients will be involved in selecting the  method
of therapy, treatment regimen, and which type of device(s) and active ingredients(s) to prescribe that will
prove most effective for particular individuals (IPAC, 1999).

In 1998, DPI usage was estimated to represent 17 percent of all inhaled medication (i.e., inhaler units)
world-wide and has increased to 27 percent in 2002  (UNEP, 2002).  DPIs may represent a viable
alternative, as suggested by their increased use in Europe; for example, in Holland they account for more
than 65 percent of inhaled  medication (UNEP, 2002). The use  of newly available DPIs is on the rise in
the United States, where DPIs made up  14 percent of the total U.S. market share as of mid-2002 (UNEP,
2002).  There is also a trend toward the development of a broad range of oral treatments that would be
swallowed, rather than inhaled, and  may be introduced over  the next  10  to  20 years.   These new
medications may impact MDI use, although will likely not replace inhaled MDI therapy entirely.

This analysis assumes that  DPIs are technically applicable31 to all HFC emissions from MDIs. However,
due to the limitations in their use for severe  asthma patients and young children, and the difficulties
experienced in  hot and humid climates, this analysis assumes  a global incremental maximum  market
penetration into the HFC-based MDI market of 0 percent in 2005, increasing up to 50 percent in 2020
(see  Table 2-3).  Also assumed is a 100 percent reduction  efficiency.  To  the  extent that health  and
technical concerns are adequately met, a transition in inhalation therapy  away from propellant MDIs and
toward NIK alternatives may occur over the next 10 to 20 years.  The rapidity at which these changes will
occur is dependent upon product development cycles (generally about 10 years), cost-effectiveness, and
manufacturing capacity (UNEP, 1999).

Cost and Emission Reduction Analysis

Cost assumptions for this option were taken directly from Ecofys (2000).32 The following bullets describe
the cost and emission inputs used to  derive the final $/TCE  for the DPI option, the results of which are
presented in Section 2.4:

    •   One-time Costs. No one-time costs are assumed for implementing DPIs.
    •   Annual Costs.  The annual cost associated with using DPIs was estimated to be approximately
        €533,000 (in 1999  euros) per metric ton of substance (Enviros March, 2000), which translates to
        an annual cost of $599,625  dollars  based on an  exchange rate of $112.5/€100 (Universal
        Currency Converter Web site, 2003). According to the source cited by Ecofys (2000), this annual
        cost incurred by the industry takes into account the increase in cost of DPI treatment, the cost to
        market  the new treatment, and the cost to retrain the patients  in using the DPI (Enviros  March,
        2000).
31 In this report, the term technically applicable refers to the emissions to which an option can be applied.  Because
DPIs can eliminate emissions from MDIs, they are technically applicable to all MDI emissions but are not
technically applicable to non-MDI emissions. Other factors will affect their application and the market penetration
assumed in this analysis.  See Appendix I for a glossary of terminology.

32 The Ecofys (2000) report cites Enviros March (2000) costs, which were developed assuming a conversion to DPIs
from an MDI containing HFC-134a (Enviros March, 2000). For this analysis, these costs and the associated
emission reductions were applied to the total baseline MDI market, which consists of MDIs that use HFC-134a and
HFC-227ea.


                                                                                              -53-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •   Cost Savings. No cost savings are assumed for this option.
    •   Emission Reductions. This option is assumed to avoid 1,000 kilograms of HFC emissions under
       the cost scenario as provided by Ecofys (2000). This reduction equates to 354.55 TCE, calculated
       using the GWP of HFC-134a.

Non-MDI Replacement with  Lower GWP MFCs

Replacement of higher GWP HFCs, such as HFC-134a, with a lower GWP HFC, such as HFC-152a, has
the potential to greatly reduce emissions from the non-MDI aerosols sector. HFC-134a is the primary
non-flammable propellant  in certain  industrial  products.    HFC-152a  possesses  only moderate
flammability hazards and might therefore be  acceptable for  some applications (UNEP, 2002).   This
analysis assumes that converting to HFC-152a is technically  applicable to all emissions of HFC-134a
from the non-MDI baseline. Non-MDI emissions of HFC-134a are calculated by the Vintaging Model to
be  83  percent of total GWP-weighted non-MDI aerosol emissions.   As shown  in Table 2-3, the
incremental maximum market penetration  of this alternative is assumed to  increase from 10 percent in
2005 up to  50 percent in 2020. Because HFC-152a has a GWP of 140 (versus a GWP of 1,300 for HFC-
134a), this  substitution has an emission reduction efficiency of  89.2 percent (i.e., the  difference of the
GWPs divided by the GWP of HFC-134a).

Cost and Emission Reduction Analysis

The following  bullets describe the cost and emission inputs used to derive the final $/TCE for converting
to a lower GWP aerosol propellant, the results of which are presented in Section 2.4:

    •   One-time Costs.   Costs of converting a filling facility to  accept  HFC-152a may  range  from
       $400,000 to  $500,000 (Dupont, 2000).  To be conservative, this analysis assumes that a one-time
       cost of $500,000 dollars is required to achieve the assumed reduction scenario.
    •   Annual Costs. The cost per pound of HFC-134a ($1.70 per pound) is slightly lower than the cost
       per pound of HFC-152a ($1.93 per pound) (Atofma, 2003; Atofina 2004).  Thus, filling a can that
       requires  2 ounces  of propellant with HFC-134a  costs  $0.21,  versus  $0.24  with  HFC-152a;
       therefore, the difference in chemical costs is an additional  $0.03  per can.  An  annual cost of
       $281,250 was calculated by assuming  that a filling facility produces 10 million  units per  year,
       each requiring two ounces of aerosol propellant.
    •   Cost Savings. No costs savings are realized with this alternative since the HFC-152a is estimated
       to cost slightly more than HFC-134a.
    •   Emission Reductions. Assuming that 10 million eight ounce cans are converted from HFC-134a
       to HFC-152a, and the typical  quantity of propellant required per unit is two ounces (or 0.0567
       kilograms), the potential  quantity of HFC-134a avoided by the facility in one year is estimated to
       be 567,000 kilograms. Accounting for the reduction efficiency of 89.2 percent, this facility could
       avoid 0.18 MMTCE per year (i.e.,  567 metric tons of HFC-134a multiplied by the GWP of 1300,
       the reduction efficiency of  0.892,  and  the  conversion factor of carbon to carbon dioxide
       equivalents of 12/44).
Non-MDI Replacement with NIK Technology

Not-in-Kind (NIK) aerosol propellants include finger/trigger pumps, powder formulations, sticks, rollers,
brushes, nebulizers,  and bag-in-can/piston-can systems.  These systems often prove to be a better and
more cost-effective  option  than HFC-propelled  aerosols,  particularly  in  areas where a unique  HFC
property is  not specifically needed for  a certain end-use.  NIKs already occupy a sizable share of markets
                                                                                          -54-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
where they were introduced during the initial CFC phaseout. Since NIK products have already assumed
much of the available non-MDI HFC aerosol market share, an incremental maximum market penetration
of 5 percent was assumed in 2005 and 10 percent for years 2010, 2015, and 2020 (see Table 2-3).  The
analysis assumes that this option is technically applicable to all non-MDI emissions and has a reduction
efficiency of 100 percent.   The GWP of 538 was used to  represent both HFCs being abated and was
calculated using the weighted average of the HFC-134a and HFC-152a baseline emissions.

Cost and Emission Reduction Analysis

The following bullets describe the cost  and emission  inputs used to derive the final $/TCE for NIK
aerosol propellants, the results of which are presented in Section 2.4:

    •   One-time Costs.  Significant variability exists in financial components of projects targeting NIK
       replacements for HFC-containing aerosol products.  This variability is attributable to  the wide
       range of potential aerosol and NIK product types.  For this analysis, an incremental capital cost of
       $250,000 per facility was used.
    •   Annual  Costs.  In the case  of liquid  pumps and solid  applicators,  capital investments  are
       generally lower, but material costs will be higher than for HFCs (UNEP,  1999). To account for
       higher material  costs of the particular sticks, rollers, and pumps being used, the analysis assumes
       an estimated $500,000 in annual  costs for a facility that produces 10 million units (e.g., cans,
       pumps).
    •   Cost Savings.  Despite the costs  of this option, overall savings can be significant, due primarily
       to the avoidance of HFC costs. Filling a can that requires two ounces of propellant with an HFC
       was estimated to cost approximately $0.23 (based on the average price  per pound of HFC-134a
       and HFC-152a weighted by the mass percent of the baseline emissions comprised by each gas)
       versus no costs of chemical for an NIK formulated can, resulting in a savings of $2,309,703 per
       year for a  filling facility that produces  10 million total cans in one year. This cost savings is
       offset by the annual cost of $500,000, resulting in an annual savings of $1,809,703 for this option.
    •   Emission Reductions. Assuming that 10 million eight ounce units are converted from an HFC to
       an NIK process, each unit using approximately two ounces of propellant, the quantity of HFC
       avoided in one  year was  estimated at 567,000  kilograms, or 0.08 MMTCE using the weighted
       average GWP of 538.

Non MDI Replacement with Hydrocarbon Aerosol Propellants

Hydrocarbon (HC) aerosol propellants are usually mixtures of propane, butane, and isobutane.  Their
primary advantage lies  in their affordability; the price of HC propellants are less than one-tenth  that of
HFCs. The main disadvantages of hydrocarbon aerosol propellants are flammability and VOC  emission
concerns.  Hydrocarbons contribute to ground level ozone and smog and therefore may be regulated in
some  areas.  In markets where flammability and/or VOC emissions are less of a concern, hydrocarbons
already hold a sizable share.  Since hydrocarbon aerosol propellants have already penetrated a significant
amount of the market; further penetration is limited due to flammability and VOC concerns. Hence, this
analysis assumes an incremental  maximum market penetration of  5  percent in 2005, expanding to  10
percent in later years. The analysis also assumes that converting to hydrocarbons is technically applicable
to all non-MDI emissions, but that various factors including the flammability of hydrocarbons will limit
the market penetration of this option. The reduction efficiency of this abatement option is taken to be  100
percent, since the HFC  is completely replaced by a hydrocarbon  propellant with a very low GWP.  The
GWP of 538 was  used to represent both HFCs being abated and was calculated using the weighted
average of the HFC-134a and HFC-152a baseline emissions.
                                                                                           -55-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Cost and Emission Reduction Analysis

The following bullets describe the cost and emission inputs used to derive the final $/TCE for converting
to a hydrocarbon aerosol propellant, the results of which are presented in Section 2.4:

    •   One-time  Costs.   The one-time  cost of converting a filling facility to accept hydrocarbon
       propellants can range from $10,000 to $1.2 million, including the costs of installing safety control
       features (Nardini, 2002).  The high conversion cost accounts  for the fact that hydrocarbons are
       highly flammable gases  that require  stringent  safety precautions  in manufacturing,  storage,
       handling, transport, and customer use.  The range in one-time cost varies based on the need for
       investments in new equipment and  the need  to relocate to regions where the use of HCs is
       considered safe (Nardini,  2002).   One-time  costs are expected to be lower, for instance, for a
       facility converting from HFC-152a to a hydrocarbon propellant where  flammability controls are
       likely to already be in place.  This report assumes that a facility producing 10 million cans per
       year must invest $325,000 for this option.

    •   Annual Costs. Annual costs may be incurred to ensure good handling practices of hydrocarbons
       that are  considered  hazardous air pollutants (HAPs), regular maintenance on fire prevention
       devices such as fire detection systems,  sprinklers, and shut-off valves, and proper safety training
       for employees  (UNEP, 2002).  Such costs have not been quantified for this analysis; however,
       future work may be performed to investigate estimated annual costs.
    •   Cost Savings.  Hydrocarbon prices are  generally lower than those of HFCs, which lowers overall
       production costs and contributes to cost savings.  To represent savings for this  option, filling a
       can that requires two ounces of and  HFC propellant was estimated to cost $0.23 (based on the
       average price  per pound  of HFC-134a and HFC-152a weighted by  the  mass percent of the
       baseline emissions comprised by each gas) versus $0.04 for the cost of a HC (based on the price
       of a propane/isobutane blend of $0.30 per pound,  Diversified CPC, 2004), which yields a cost
       savings of approximately  $1,934,700 per year experienced by a filling facility that produces 10
       million cans in one year.

    •   Emission Reductions.  As with the scenario used  for the NIK option, the quantity of HFC
       avoided in one year by transitioning to hydrocarbons was estimated at 567,000 kilograms, or 0.08
       MMTCE.

2.3.2  Summary of Technical  Applicability and  Market Penetration of Abatement
       Options

Table 2-3 summarizes the technical  applicability and  incremental maximum market penetration of the
aerosol options presented in the discussions above.

Table 2-3: Technical Applicability and Incremental Maximum Market Penetration of Aerosol Options
(Percent)3
Option
DPI (MDI)b
HFC to HC (Non-MDI)
HFCtoNIK(Non-MDI)
HFC-134a to HFC-152a (Non-MDI)
Technical
Applicability
(All Years)
100%
100%
100%
83%'
2005
0%
5%
5%
10%
Maximum Market Penetration
2010 2015
5% 20%
10% 10%
10% 10%
25% 35%
2020
50%
10%
10%
50%
Assumed maximum market penetration of options is presented as a percentage of total sector emissions for which the options are applicable.
The baseline market penetration is assumed to be zero to assess the emission reductions possible due to increased use of each option.


                                                                                            -56-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
bAssumptions are separated by the line to reflect that the MDI option addresses different baseline emissions than the non-MDI options.
"Based on percent of non-MDI aerosol emissions as determined by the Vintaging Model.

To calculate the percent of emission reductions off the applicable (i.e., MDI or non-MDI) aerosols
baseline for each abatement option, the technical applicability (Table 2-3) is multiplied by the market
penetration value (Table 2-3), and by the reduction efficiency of the option.  For example, to determine
the percent reduction off the 2020 baseline for the conversion of HFC-134a aerosols to HFC-152a, the
following calculation is performed:

              Technical Applicability X Market Penetration in 2020 X Reduction Efficiency

                                     83% X 50% X 89.2% « 37.0%

Thus, using the assumptions in this analysis, converting from HFC-134a to HFC-152a could reduce over
one-third of the non-MDI emissions baseline in 2020. This value, along with the other emission
reduction potentials, is shown in Table 2-4.

Table 2-4: Emission Reductions off the Total Applicable Aerosols Baseline (Percent)
Option
DPI (MDI)a
HFC to HC (Non-MDI)
HFC to NIK (Non-MDI)
HFC-134a to HFC-152a (Non-MDI)
2005
0
5
5
7
2010
5
10
10
19
2015
20
10
10
26
2020
50
10
10
37
Calculated percentages are separated by the line to reflect that the MDI option addresses different baseline emissions than the non-MDI
options.

2.4         Results


Table 2-5 through Table 2-7 provide a summary of the potential emission reduction opportunities and
associated costs for United States, Non-U.S. Annex I countries, and Non-Annex I countries in 2020. The
costs, in 2000 U.S. dollars, to reduce one ton of carbon equivalent (TCE) are presented for two different
discount rate scenarios:  4 percent and 20 percent.  The tax rate associated with each cost scenario is 0
percent and 40 percent, respectively. Within the options that address non-MDI emissions, the results are
ordered by increasing costs per TCE.  Additionally, the emissions reduced by the option, in MMTCE and
percent of the regional aerosols (either MDI or non-MDI) baseline, are presented, as are cumulative totals
of these two figures.

Table 2-5: United States Emission Reductions in 2020 and Break-Even Costs for Aerosols
Reduction Option
DPI (MDI)a
HFC to HC (Non-MDI)
HFC to NIK (Non-MDI)
HFC-134a to 152a (Non-MDI)
Break-Even Cost (2000$/TCE) Percent
Discount Rate/ Tax Rate Emission Reduction
4%/0% 20%/40% Reduction of from 2020
Option (MMTCE) Baseline
$1,691.25
($22.76)
($21.37)
$1.91
$1,691.25
($21.95)
($20.75)
$2.49
0.99
0.40
0.40
1.49
50.0%
10.0%
10.0%
37.0%
Cumulative
Reductions
(MMTCE)
0.99
0.40
0.81
2.30
Cumulative
% Reduction
from 2020
Baseline
50.0%
10.0%
20.0%
57.0%
aResults are separated by the line to reflect that the MDI option addresses different baseline emissions than the non-MDI options.
                                                                                                 -57-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 2-6: Non-U.S. Annex I  Emission Reductions in 2020 and Break-Even Costs for Aerosols
Reduction Option
DPI (MDI)»
HFC to HC (Non-MDI)
HFCtoNIK(Non-MDI)
HFC-134a to 152a (Non-MDI)
Break-Even Cost (2000$/TCE) Percent
Discount Rate/ Tax Rate Emission Reduction
4%/0% 20%/40% Reduction of from 2020
Option (MMTCE) Baseline
$1,691.25
($22.76)
($21.37)
$1.91
$1,691.25
($21.95)
($20.75)
$2.49
1.32
0.67
0.67
2.48
50.0%
10.0%
10.0%
37.0%
Cumulative
Reductions
(MMTCE)
1.32
0.67
1.34
3.83
Cumulative
% Reduction
from 2020
Baseline
50.0%
10.0%
20.0%
57.0%
aResults are separated by the line to reflect that the MDI option addresses different baseline emissions than the non-MDI options

Table 2-7: Non Annex I Emission Reductions in 2020 and Break-Even Costs for Aerosols
Reduction Option
DPI (MDI)a
HFC to HC (Non-MDI)
HFC to NIK (Non-MDI)
HFC-134a to 152a (Non-MDI)
Break-Even Cost (2000$/TCE)
Discount Rate/ Tax Rate
4%/0% 20%/40%
$1,691.25
($22.76)
($21.37)
$1.91
$1,691.25
($21.95)
($20.75)
$2.49
Emission
Reduction of
Option
(MMTCE)
1.10
0.004
0.004
0.01
Percent
Reduction
from 2020
Baseline
50.0%
10.0%
10.0%
37.0%
Cumulative
Reductions
(MMTCE)
1.10
0.00
0.01
0.02
Cumulative
% Reduction
from 2020
Baseline
50.0%
10.0%
20.0%
57.0%
aResults are separated by the line to reflect that the MDI option addresses different baseline emissions than the non-MDI options

2.5        Summary


This analysis considers four mitigation options: 1) MDI Replacement with Dry Powdered Inhalers (DPIs);
2) Non-MDI Replacement with Lower GWP HFCs; 3) Non-MDI Replacement with Not-In-Kind (NIK)
Alternatives; and 4) Non-MDI Replacement with Hydrocarbon Aerosol Propellants.  The first option has
the potential to abate emissions from the MDI baseline; while the latter three options have the potential to
abate emissions from the non-MDI baseline.

MDI Aerosols

Global baseline HFC emissions from MDI aerosols are estimated to grow from 4.60 to 6.82  MMTCE
between the years 2005 and 2020.  In 2020,  the three  regions analyzed (the  United States, Non-U.S.
Annex I and Non-Annex I) are estimated to be responsible for approximately 29, 39 and 32 percent of the
baseline emissions, respectively (see Table 2-1). The highest emissions growth from the MDI baseline,
from 0.90 MMTCE in 2005 to 2.21 MMTCE in 2020, is expected to occur in Non-Annex I countries.

As Table 2-5 through Table 2-7 illustrate, converting from HFC MDIs to DPIs is not a cost-effective
abatement option—the estimated cost is more than $1,500 dollars per TCE for all regions—although the
option may  be popular for other reasons.   The option is assumed to abate 50 percent of global MDI
emissions, or 3.41 MMTCE, annually,  by 2020. The costs  per TCE for the three regions are equivalent
because available data on costs  for abatement technologies were not scaled to reflect potential differences
in the costs internationally.  Additional  research may be performed to determine  actual variability in costs
across regions.

Non-MDI Aerosols

Baseline HFC emissions from  non-MDI aerosols  are estimated  to grow  from 7.66 MMTCE to  10.78
MMTCE globally for years 2005 through 2020. In 2020, Non-U.S. Annex I emissions are assumed to

                                                                                           -58-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
account for approximately 62 percent of this total, while U.S. and Non-Annex I countries are assumed to
account for approximately 37 percent and 0.3 percent, respectively (see Table 2-1).  The highest absolute
emissions growth, from 4.65 MMTCE in 2005 to 6.71 MMTCE in 2020, is expected to occur in the Non-
11. S. Annex I region.

As  shown in Table  2-5 through Table 2-7, the greatest emission reduction opportunities in  all of the
regions analyzed may come from converting HFC-134a to HFC-152a, at a cost of $1.91 per TCE at a 4
percent  discount rate.  The other two options, converting to hydrocarbons and NIK, represent a  cost
savings of $22.76 and $21.37 per TCE at a 4 percent discount rate, respectively. Globally, 6.15  MMTCE,
or 57 percent of global  emissions from non-MDI aerosols, can  be reduced in 2020 at a cost below
$2.00/TCE.  As with MDI aerosols, costs per TCE for these three regions are equivalent because available
data on costs for abatement technologies were not scaled to reflect potential differences in the costs
internationally.  Additional research  may be performed to determine  actual variability in costs across
regions.


2.6        References

ADL. 1999.  Global Comparative Analysis of HFC and Alternative Technologies for Refrigeration, Air-
Conditioning, Foam, Solvent, Aerosol Propellant, and Fire Protection Applications. Final Report to the
Alliance for Responsible Atmospheric Policy. Reference Number 49648. Arthur D. Little, Inc. 1999.

Atofina. 2004. Personal communication between Dom Loconte of Atofma and Robert Russell of ICF
Consulting. March 2004.

Atofina. 2003. Personal communication between Dom Loconte of Atofina and Robert Russell of ICF
Consulting. October 2003.

Diversified  CPC, 2004. Personal communication between Bill Frauenheim of Diversified CPC and Mollie
Averyt of ICF Consulting, June 2004.

Dupont. 2000. Personal communication between John Lueszler of Dupont and ICF Consulting. July 2000.

Enviros March. 2000. Study on the Use of HFCs for Metered Dose Inhalers in the European
Union. Commissioned by the International Pharmaceutical Aerosol Consortium (IPAC). Enviros March.
September 2000.

Ecofys. 2000. Abatement of Emissions of Other Greenhouse Gases: Engineered Chemicals. Prepared for
the  IEA Greenhouse Gas R&D Programme. Ecofys. November 3, 2000.

IPAC. 1999. Ensuring Patient Care 2nd Edition. International Pharmaceutical Aerosol Consortium. 1999.
Available online at .

March Consulting Group. 1999. UK Emissions of HFCs, PFCs, and SF6 and Potential Emission
Reduction Options: Final Report. March Consulting Group.  1999.

Nardini, Geno. 2002. Personal communication between Geno Nardini and Iliriana Mushkolaj of ICF
Consulting. May 2002.
                                                                                          -59-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Price, David, Erkka Valovirta, and Jurgen Fischer. 2004. The importance of preserving choice in
inhalation therapy: the CFC transition and beyond. Journal of Drug Assessment. 7:45-61. May 2004.

UNEP. 2002. 2002 Report of the Aerosols, Sterilants, Miscellaneous Uses and Carbon Tetrachloride
Technical Options Committee: 2002 Assessment. United Nations Environment Programme. 2002.

UNEP. 1999. The Implications to the Montreal Protocol of the Inclusion ofHFCs andPFCs in the Kyoto
Protocol. United Nations Environment Programme, HFC and PFC Task Force of the Technology and
Economic Assessment Panel (TEAP). October 1999.

Universal Currency Converter Web site. 2003. Available online at . Accessed in
July 2003.
                                                                                           -60-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
3   Analysis of Costs to Abate International HFC Emissions
     from Foams
3.1        Introduction


Various hydrofluorocarbons (HFCs) are currently being used as blowing agents during the manufacture of
foams. These high GWP gases  are substitutes for ozone  depleting substances (ODSs) that were the
primary blowing agents historically used in the foams industry.  Parties to the Montreal Protocol  on
Substances  that  Deplete  the   Ozone  Layer  (Montreal  Protocol)  have  agreed  to  phase  out
chlorofluorocarbons  (CFCs)  and  many are  using  hydrochlorofluorocarbons (HCFCs)  as interim
substitutes.  Developed and developing countries  are at  different phases of replacing CFCs  with
alternatives. Developed countries such as the United States and the European Union have banned the sale
and distribution of most foam products  manufactured with HCFCs and have begun transitioning to HFC
use in foams where hydrocarbons and other alternatives are not already used.  For example, Denmark,
Austria, Finland, and Sweden phased out the use of HCFCs for foam blowing on January 1, 2002.

Developing countries have only recently  begun transitioning  from  CFC-11  to HCFCs  and other
alternatives.  Their rate of conversion to HFCs may be limited by the current availability of other ODS
substitutes and also by technical barriers and cost. For example, the main blowing agent alternatives for
CFC-11 in  rigid polyurethane (PU) insulating foams are hydrocarbons (HCs), such as pentanes, and
HCFCs. Applying alternative (i.e., HFC) technologies may require the use of higher density foam, which
will result in incremental operating costs.

The most commonly used HFCs are HFC-134a, HFC-152a, HFC-245fa, and HFC-365mfc in combination
with HFC-227ea.  These  blowing agents  can be released  into the atmosphere  during  the  foam
manufacturing  process, during on-site foam application,  while foams are in use, and when  foams are
discarded.  They have  100-year global warming potentials  (GWPs) of 1,300, 140,  950, 890, and 2900
respectively, and have replaced  historically used ODS blowing agents including  CFCs and HCFCs.
Foams studied in this analysis include the following:

    •   Appliance foams found in  various  commercial and domestic refrigerators, vending machines,
       freezers, water heaters, picnic boxes, flasks/thermoware and refrigerated containers (reefers);
    •   Spray polyurethane (PU)  foams found in roofing, wall insulation, plus various tank and vessel
       applications;
    •   PU continuous and discontinuous  sandwich panels foam found in cold storage, entrance and
       garage doors;
    •   PU one component foams found in sealing around  windows and doors, framing around pipes,
       cable holes, jointing insulating panels and around certain roof components; and
    •   Extruded polystyrene (XPS) boardstock foam used in building insulation.
                                                                                       -61-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
3.2        Baseline Emission Estimates


3.2.1   Emission Estimating Methodology

A full description of the emission model used to calculate ODS substitute emissions from all sectors of
the foam market is provided in Appendix A. Specific information on how the model calculates emissions
for this sector is described below.

EPA uses a detailed Vintaging Model of ODS-containing equipment and products to estimate the use and
emissions of various  ODSs and ODS  substitutes in the  United States,  including HFCs  and  PFCs.
Emission equations used by the model are presented below.

Emission Equations

Foams are given emission profiles depending on the foam type (open cell or closed cell).   Open cell
foams are assumed to be 100 percent emissive in the year of manufacture as described in the first equation
below.  Closed cell foams are assumed to emit a portion of their total HFC content upon manufacture, a
portion  at a constant rate over the lifetime of the foam, and a portion at disposal as described in the
second equation below.33

Open-Cell Foam

                             Ej = Qq

Where:

Ej  =   Emissions. Total emissions of a specific chemical in year/ used for open-cell foam blowing, by
        weight.
Qcj =   Quantity of Chemical. Total amount of a specific chemical used for open-cell foam blowing, in
        year j, by weight.
j   =  Year of emission.

Closed-Cell Foam

                             Ej = X (efj x QCJ-M)   fori=1—>k

Where:

Ej  =   Emissions. Total emissions of a specific  chemical in year j for closed-cell foam blowing, by
        weight.
ef  =   Emission Factor.  Percent of foam's original charge emitted in each year (1 —> k).  This emission
        factor is generally variable, including a rate for manufacturing emissions (occurs in the first year
        of foam life), annual emissions (every year throughout the foam lifetime), and disposal emissions
        (occurs during the final year of foam life).
Qc =   Quantity of Chemical. Total amount of a specific chemical used in closed-cell foams in year (/' - /
        + 1).
33 Emissions from foams may vary due to handling and disposal of the foam; shredding of foams may increase
emissions, while landfilling of foams may abate some emissions (Scheutz, et al.  2002, Scheutz, et al. 2003).
Average annual emissions are assumed in the model, which may not fully account for the range of foam handling
and disposal practices.
                                                                                          -62-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
i   =  Counter. Runs from 1 to lifetime (k).
j   =  Year of emission.
k   =  Lifetime. Average lifetime of foam product.

The emissions profile for foams assumed by the Vintaging Model is presented below in Table 3-1.

Table 3-1: U.S. EPA's Vintaging Model Emissions Profile for the Foams Sector
Foams Sector
Flexible PU
Polyisocyanurate Boardstock
Rigid PU Integral Skin
Rigid PU Appliance
Rigid PU Commercial Refrigeration
Rigid PU Spray
One Component
Rigid PU Slabstock and Other
Phenolic
Polyolefin
XPS Foam Sheet
XPS Boardstock
Sandwich Panel
Loss at
Manufacturing
(%)
100
6
95
4
6
25
100
37.5
23
95
40
25
5.5
Annual
Release Rate
(%)
0
1
2.5
0.25
0.25
1.5
0
0.75
0.875
2.5
2
2.5
0.5
Release
Lifetime
(years)
0
50
2
15
15
50
0
15
32
2
25
30
50
Loss at
Disposal
(%)
0
44
0
92.25
90.25
0
0
51.25
49
0
0
0
69.5
Total
Released
(%)
100
100
100
100
100
100
100
100
100
100
90
100
100
Regional Adjustments

All global emissions except the  U.S. and  Canada were  estimated  in the foam-blowing sector  by
developing Vintaging Model scenarios that were representative of country- or region-specific substitution
and consumption patterns. To estimate baseline emissions,  current and projected characterizations of
international total foams markets were used to create country or region-specific versions of the Vintaging
Model.  The market information was obtained from Ashford (2004), based on research conducted  on
global foam markets. Scenarios were developed for Japan, Europe (both  EU and non-EU  countries
combined), other developed countries (excluding Canada), CEITs, and  China. It was assumed that other
non-Annex I countries would not transition to HFCs during the scope of this analysis. Once the Vintaging
Model scenarios had been run, the emissions were disaggregated to a country-specific  level  based  on
estimated 1989  CFC consumption  for foams  developed for  this analysis. Emission estimates were
adjusted slightly to account for relative differences in countries' economic growth as compared to the
United States (USDA, 2002; and EIA, 2001).

Emission baselines  for Canada were  derived using country-specific  ODS consumption  estimates  as
reported under the Montreal Protocol in conjunction with U.S. Vintaging Model output  for each ODS-
consuming end-use sector. Preliminary estimates were calculated by assuming  that the transition from
ODSs to HFCs and other  substitutes follow the same general substitution patterns as observed in the
United States.  A detailed description of the Vintaging Model and the methodology used to estimate and
adjust emissions in Canada are presented in Appendix A.

Newly Manufactured Foam Emissions vs. Existing Foam Emissions

Technology options explored in the foams chapter are only applicable to new  (not existing) foams.
Therefore, the technical applicabilities of the technology options in this sector include only emissions


                                                                                          -63-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
from relevant end-uses that are from  "newly manufactured" foams, where "newly manufactured"  is
defined as foam manufactured in 2005 or later.

3.2.2  Baseline Emissions

Table 3-2 provides a summary of baseline emissions for the United States, Non-U.S. Annex I and Non-
Annex I countries through 2020.

Table 3-2: Baseline Emissions in MMTCE
Regions
United States
Non-US Annex 1
Non-Annex 1
Total
3.3 Costs
2005
0.53
2.40
0.00
2.94
of HFC
2010
1.90
3.40
0.01
5.30
Emission
2015
3.15
4.59
0.01
7.76
Reductions
2020
7.35
9.01
0.01
16.38
from Fi
This section presents a cost analysis of achieving HFC emission reductions from the emissions baseline
presented above.

Costs were based on the incremental  differences between using the HFC and switching to  an HFC
alternative. Financial information considered in this analysis includes: capital costs, which account for
equipment costs to modify existing plants  and to maintain production capacity; blowing agent costs,
which address the difference between costs  and the quantity of the  HFC and non-HFC  alternative
required; foam  costs, which  address changes in foam density, the amount of fire retardant used, the
quantity and type of polyol, etc.; testing, training or other costs associated with transitioning to non-HFC
alternatives; and costs to produce a thicker, denser foam to account for any energy efficiency differences.

In addition,  industry has indicated that there will be additional conversion or "learning curve" costs,
which are short-term costs incurred due to yield, rate, and density penalties associated with  conversion
uncertainties as  well as technical support costs. Such costs may be highly variable and are not addressed
in the analysis.


3.3.1  Description of Abatement Options

Specific opportunities to reduce HFC emissions from foams that were analyzed for this are:

    •  Replacing  HFC-134a,  HFC-245fa,  and  HFC-365mfc/HFC-227ea   with  hydrocarbons  in
       continuous and discontinuous panel foam;
    •  Replacing HFC-134a and HFC-152a with hydrocarbons in one component foam;
    •  Replacing HFC-134a/CO2(LCD) with CO2(LCD) in XPS foam;

    •  Replacing HFC-134a/CO2 (LCD) with CO2/Alcohol (LCD/Alcohol) in XPS foam;
    •  Replacing HFC-245fa/ CO2 (water) and  HFC-365mfc/HFC-227ea with CO2(water)  in spray
       foam;
                                                                                          -64-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •   Replacing HFC-245fa/CO2 (water)  and HFC-365mfc/HFC-227ea with hydrocarbons in spray
       foam;

    •   Replacing HFC-134a with hydrocarbons in appliance foam;
    •   Replacing HFC-245fa with hydrocarbons in appliance foam;
    •   End-of-life appliance foam practice: automated process with foam grinding and landfilling; and
    •   End-of-life appliance foam practice: manual process with incineration.
The remainder of this section provides a description of the economic assumptions for these  abatement
options. All abatement option cost analyses assume a 25-year project lifetime.

Replacement Options

The abatement options and costs of options to reduce HFC emissions from the foam sector are presented
by foam type: PU spray foams, PU appliance foams, other PU foams, continuous and discontinuous panel
foam, one component foam, and XPS foams. Each of the options includes the use of non-HFC blowing
agents such as hydrocarbons, water-blown CO2, and liquid CO2.  These foam technologies are described
below, followed by  Section 3.3.2, which gives specific analyses of the costs of applying these alternate
blowing agents to the particular foam types.

Hydrocarbons (HCs)

Hydrocarbons such  as propane, butane, isobutane, n-pentane, isopentane, cyclopentane, and isomers of
hexane are alternatives to HFCs in foam blowing appliances. HCs are inexpensive and have lower GWP
impacts relative to HFCs. However, key technical issues associated with the use of hydrocarbons exist, as
follows:

    •   Flammability. HCs  require stringent safety precautions in manufacturing, storage, handling,
       transport, and customer use. These factors necessitate factory upgrades and employee training.
       Specialized  equipment  that  might be  needed  includes  a  dedicated storage tank  for  the
       hydrocarbon, pre-mixers, adapted high-pressure dispensers,  suitable molds plus process exhaust,
       hydrocarbon detectors, and appropriate classification of electrical equipment. In order to reduce
       fire risks, some  applications might also require the use  of a larger quantity of flame retardants
       and/or the use of a more expensive fire retardant.
    •   Volatile Organic Compounds (VOCs). Because HCs contribute to ground level  ozone  and
       smog, they tend to be highly regulated. In many places, such as certain areas of the United States,
       HCs cannot be used without emission controls.  Implementation of these controls can  lead to
       significant increases in the costs of conversion.
    •   Performance.  Some HCs only yield approximately 85 percent of the insulating value of HCFC-
       14Ib and HFC-245fa (and HFC-365mfc/HFC-227ea).  Producing a thicker foam can compensate
       for this energy efficiency difference, but will increase the cost of production. This option might
       not be viable in  fixed thickness applications such as refrigerated trucks or in applications where
       an R-value  is prescribed by code such  as in PU spray  roofing insulation. Other performance
       considerations include dimensional stability and  solubility.  Addressing these factors might
       require a more expensive and more limited polyol formulation.
Costs of converting  to hydrocarbons and addressing technical considerations can be significant, but vary
according to factory-specific needs. In spite of these issues, hydrocarbons  are being considered in a wide
variety of applications (UNEP, 1998; Alliance, 2000; Alliance, 2001).
                                                                                           -65-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Liquid Carbon Dioxide (LCD)

The basic principle by which LCD blowing agents operate is the expansion of liquid CO2 to a gaseous
state.  Liquid CO2 is blended with other foam components under pressure  prior to the initiation of the
chemical reaction. When decompressed, the CO2 expands, resulting in froth  foam, which further expands
with the additional release  of CO2 from the water/isocyanate resin reaction that forms  the PU foam
matrix.  LCD might  require formulation changes  to  more readily  dissolve  the  CO2 and to prevent
deactivation of PU catalysts. When LCD is introduced at the head, often referred to  as third stream, the
metering equipment can be quite complicated and, to date, unreliable. Difficulties encountered in using
LCD include the limited solubility of the chemical mixture, controlled decompression, and distribution of
the unavoidable  froth  (UNEP,  1998).  Foams blown  with  CO2  may   suffer from lower  thermal
conductivity, lower dimensional stability, and higher density versus HCFC-blown foams.  To overcome
these limitations,  CO2 can be blended with hydrocarbons or HFCs (Williams et al., 1999; Honeywell,
2000; Alliance, 2001).

Water-Blown (in  situ) Carbon Dioxide (COi (water))

In this process, CO2 produced from a chemical reaction between water and polymeric isocyanate is used
as a blowing agent. During manufacture, no ODS or high-GWP gases are emitted, and there are limited
health and safety risks during processing. However, foams produced using  CO2/water are subject to the
same  performance limitations  discussed for  LCD-blown foams: lower  thermal  conductivity,  lower
dimensional stability, and higher density versus HCFC- and  HFC-  blown foams.   In some PU foam
applications, a major concern associated with using water-generated or LCD systems is  the increased
percentage of open cell content, which results in poorer water-proofing performance  and water-proofing
quality of the final product. Another consideration is that the polymeric  isocyanate content  must be
increased, which  cannot be accommodated  by  some  spray foam equipment.  To  overcome  these
limitations, CO2 can  be  blended with hydrocarbons or HFCs (Williams et  al., 1999; Honeywell,  2000;
Alliance, 2001).  In some other applications, e.g. PU Block, there can be problems with uncontrollable
exotherms using purely CO2 (water) systems.

Although LCD  and  CO2 generated  in situ have  similar performance  issues,  the  process limitations
associated with  each  differ. Fewer mechanical modifications are required to use in situ CO2, and the
foam manufacturer or spray foam applicator can be more certain of the final CO2 content and overall
foam properties, than  for LCD (Alliance, 2001).

End-of-Life Appliance Foam Practices

There are several methods for disposal of polyurethane foam, including landfilling and incineration, with
or without ODS recovery and recycling/destruction. Two of the methods are described below, followed
in Section 3.3.2 by specific analyses of the costs associated with each method.
Landfilling

Traditionally, most of the decommissioned foam  products have ended up in  landfills.  Although  the
regulations related to  the location and management of landfills have improved considerably, there is still
concern about the rate of release of blowing agent  from foam  in the first weeks of entering the landfill.
(UNEP, 2002b).
                                                                                          -66-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Incineration

Incineration of foams in municipal  solid waste incinerators (MSWIs) or waste-to-energy plants is  a
practical and highly competitive technique for destruction of polyurethane foam.  An advantage of this
technique is that the foam can be incinerated without separating the foam matrix from the blowing agent
prior to incineration, which makes the costs and potential risks of fugitive emissions lower.

3.3.2   Description of Abatement Options

The following section describes all of these options in greater detail and presents a cost analysis for those
options for which adequate cost data are available. The technology options explored in this chapter are
assumed to penetrate only the markets of new (not existing) foams.  "New" foam is defined as foams
manufactured in 2005 or later.

Continuous and Discontinuous  Panel Foam

The only abatement option that was  considered for this category is replacing HFC  with hydrocarbons.
The blowing agent constitutes approximately 9 percent of the foam, by weight.  This cost analysis
estimates the break-even carbon price for a hypothetical contractor to replace HFC  with hydrocarbons.
The  foams  manufactured with the alternative  are assumed to compensate  for lower insulating
performance relative to HFC-blown foams by increasing the thickness and density of the foam. Although
this end-use uses HFC-134a, HFC-245fa, and HFC-365mfc/HFC-227ea, the analysis was performed
based on a continuous and discontinuous foam contractor that  uses HFC-134a. A contractor that uses
HFC-245fa and HFC-365mfc/HFC-227ea would see higher cost savings for this replacement option since
these HFCs are more expensive than HFC-134a.  But, since they have lower GWPs, the option would
yield a lower TCE savings.  This analysis is based on a hypothetical continuous and  discontinuous foam
contractor that uses approximately 1 million pounds per year of HFC-134a (ICF Consulting, 2004).
Table 3-3: Base Case Assumptions for a Contractor  Using HFC-134a
Variable
Blowing Agent Component of Foam
Blowing Agent Consumption
Foam Produced
Foam Cost
Price of HFC-134a
Value
8.7%
1, 048,600 Ibs
1 2,052,874 Ibs
$1.02/lb
$1.70/lb
Source
Cannon, 2001
UNEP, 2002a
Calculation
Assumption
Atofina, 2004
Cost factors that are addressed include:

    •  capital equipment costs;
    •  increased cost of foam components (e.g., polyols, additives, etc.);
    •  increased consumption of foam components to compensate for increased foam density;
    •  increased use of fire retardant; and
    •  incremental differences in the costs of blowing agents and the quantity required.
                                                                                          -67-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Continuous and Discontinuous Panel: Replacing HFC with Hydrocarbons

This option is technically applicable34 to all emissions from the newly-produced continuous and
discontinuous panel foams. The technical applicability of this option from 2005 to 2020 is presented in
Table 3-20 and Table 3-20. This analysis assumes that the incremental maximum market penetration of
this option in 2010 into the newly-produced continuous and discontinuous panel market that uses HFC-
134a is 70 percent for the United States and 90 percent for the rest of the world, both rising to 100 percent
by 2020 (see Table 3-22), and its reduction efficiency is 100 percent. Assumptions specific to this
substitution are presented in Table 3-4.

Cost and Emission Reduction Analysis

    •  One-time Costs. According to industry experts, the one-time cost associated with this abatement
       option is the cost of installing safety equipment.  For this purpose, a one-time cost of $300,000
       was assumed for replacing HFC-134a with hydrocarbons.
    •  Annual  Costs. This analysis assumes increased operating costs of $2,242,777 for this abatement
       option.  These costs result  from costs associated with fire retardant use, costs associated with
       changes   in  foam  density,  and  worker  safety  training  costs associated  with the  use  of
       hydrocarbons.   Table  3-3  and  Table 3-4  summarize the assumptions  associated  with this
       abatement option. Costs associated with fire retardant use are $750,300, calculated as the amount
       of foam  produced (12,052,874 Ib) multiplied by the incremental  increase in fire retardant used in
       foam (3%)  and the fire retardant costs ($2.08/lb).  Worker training costs are estimated to be
       $11,429, calculated by multiplying the costs of training per day by the number of workers and the
       number of training days (i.e., costs per day = $4,000/14 days = $285.71/day; total cost per year =
       $286/day x 5 days * 8 workers = $11,429).  Costs associated  with foam density increases  are
       $1,481,057,  calculated  by multiplying the amount of foam produced by alternative foam  costs
       ($1.02/lb) and the increase in foam density (12%).
    •  Cost Savings. Because hydrocarbons are less expensive on a per kilogram basis than HFC-134a,
       there is a $l,153,460/year cost savings associated with this option.  Costs of the current blowing
       agent are $l,782,620/yr, while  costs of alternative blowing agent are $629,160/yr. Current costs
       are calculated by multiplying the per-pound cost of the blowing agent by the total amount of
       blowing  agent (l,048,6001b x  $1.70/lb = $1,782,620). The calculation of alternative blowing
       agent  cost  includes per-pound costs of  alternative blowing agent  ($0.60/lb), the amount of
       blowing  agent used (1,048,600 Ib), and the blowing agent substitution ratio (100%).

    •  Emission Reductions.  This analysis estimates that by replacing HFCs with hydrocarbons, 0.17
       MMTCE of the high GWP gas that would have been  emitted from foam manufactured by  the
       hypothetical contractor during the lifetime of the foam produced in a given year are eliminated.
Table 3-4: Assumptions and costs used  in the cost analysis to substitute HFC-134a Hydrocarbons
Variable                                          Value                   Source
Capital Costs                                     $300,000                Assumption
34 In this report, the term technically applicable refers to the emissions to which an option can be applied.  Because
this option examines the replacement of HFC-134a with hydrocarbons, and can not be retroactively applied to HFC-
134a foam that has already entered the market, the technical applicability is the percent of baseline foams emissions
that comes from HFC-134a from continuous and discontinuous panels placed on the market after 2004. Other
factors will affect the market penetration of the option assumed in this analysis. A glossary of terminology may be
found in Appendix I.
                                                                                              -68-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Variable
Value
Source
Incremental Increase in Fire Retardant Use
Cost of Fire Retardant
Blowing Agent Substitution Ratio
Increase in Foam Density
Assumed Increase in Cost of Foam
Price of Pentane
Training Class Costs
Days per Training Class
Number of Workers
Days of Training
Employee Training Costs
3%
$2.08/lb
100%
12%
$0.1 (Mb
$0.60/lb
$4,000
14 days
8 workers
5 days/yra
$11,429/yrb
Assumption
Exxon, 2004
Assumption
ICF Consulting, 2004
Assumption
Exxon, 2004
ICF Consulting, 2004
ICF Consulting, 2004
ICF Consulting, 2004
ICF Consulting, 2004
Calculation
a Assumes one-week per year of training per worker.
b Assumes a cost of approximately $3000 - 5000/training class, 6-10 employees, and 4 - 6 days of training. (Costs per day = $4000/14 days
= $285.71/day; Total cost per year = $286/day x 5 days x 8 workers = $11,429)
One Component Foam
Two blowing agent replacement abatement options were considered for this end-use:
    •   Replacing HFC-134a with propane/butane; and
    •   Replacing HFC-152a with propane/butane.
An analysis was performed  based  on a  hypothetical  one component foam contractor  that uses
approximately 300,000 pounds per year of HFC-134a or HFC-152a (ICF Consulting, 2004). The blowing
agent constitutes approximately 9 percent of the foam, by weight.  This cost analysis estimates the break-
even carbon price for this hypothetical contractor to replace HFC-134a or HFC-152a with hydrocarbons.
Table 3-5:  Base Case Assumptions for a Contractor Using HFC-134a and HFC-152a	
Variable
Blowing Agent Component of Foam
Blowing Agent Consumption
Foam Produced
Blowing Agent Substitution Ratio
Foam Cost
Price of HFC-134a
Price of HFC-152a
Price of propane
Price of butane
Value
8.7%
288,000 Ibs
3,31 0,345 Ibs
100%
$1.02/lb
$1.70/lb
$1.93/lb
$0.50/lb
$0.50/lb
Source
Cannon, 2001
UNEP, 2002a
Calculation
ICF, 2004
Assumption
Atofina, 2004
Atofina, 2004
Atofina, 2004
Atofina, 2004
Cost factors that are addressed include:
    •  capital equipment costs;
    •  increased cost of foam components (e.g., polyols, additives, etc.);
    •  increased consumption of foam components to compensate for increased foam density;
    •  increased use of fire retardant; and
    •  incremental differences in the costs of blowing agents and the quantity required.
                                                                                              -69-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
One Component:  Replacing HFC-134a with Hydrocarbons

This option is technically applicable to all HFC-134a emissions from the newly-produced one component
foams. The technical applicability of this option from 2005 to 2020 is presented in Table 3-20 and Table
3-21. This analysis assumes that the incremental maximum market penetration in 2010 for this option in
the newly-produced one component market that uses HFC-134a would be 70 percent for the United States
and 90 percent for the rest of the world, both increasing to 100 percent by 2020 (see Table 3-22), and its
reduction efficiency is 100.  Assumptions specific to this substitution are presented in Table 3-6.

Cost and Emission Reduction Analysis

    •   One-time Costs. According to industry experts, the one-time costs for replacing HFC-134a with
        hydrocarbons are $375,000, which includes the cost of installing safety equipment.
    •   Annual  Costs.  This analysis assumes operating  costs  of $321,103 for  this abatement option.
        These costs result from costs associated with fire retardant use and  worker safety training costs
        associated  with the use of hydrocarbons.  Table 3-5 and Table 3-6  summarize the assumptions
        associated  with this abatement option.  Costs associated  with fire retardant use are $309,103 and
        are calculated as the  amount  of foam produced multiplied by the  incremental increase in fire
        retardant used in foam and the fire retardant costs (3,310,345  Ib x 4.5% x $2.08/lb =  $309,103).
        Worker training costs are estimated to be $12,000,  and are calculated by multiplying the costs of
        training  per day by the number  of workers and  the number of training days ($285.7I/day x 14
        days x 3 workers = $12,000).  Costs per training  per day are calculated by dividing the total costs
        of training  class by the number of days ($285.7 I/day = $4,000/14 days).
    •   Cost  Savings.  Because propane and butane are less expensive on a per-pound basis  than HFC-
        134a, a $345,600 annual cost savings is associated with this option.  Costs of the current blowing
        agent are $489,600/yr, while costs of alternative blowing agent are  $144,000/yr.  Current costs
        are calculated by multiplying  the per-pound cost  of the blowing agent by the total  amount of
        blowing  agent used (288,000 Ib x $1.70/lb = $489,600/yr). The calculation of alternative blowing
        agent cost includes per-pound  costs of alternative blowing agent ($0.50/lb), the  amount of
        blowing  agent used (288,000 Ib), and the blowing agent substitution ratio (100%).

    •   Emission Reductions.  This analysis estimates that by replacing HFC-134a with hydrocarbons,
        0.046 MMTCE of the high GWP gas that would  have  been  emitted by this facility  during the
        lifetime of the one component foam produced in a given year are eliminated.
Table 3-6: Assumptions and costs used  in the cost analysis to substitute HFC-134a with Hydrocarbons
Variable                                            Value                   Source
Capital Costs
Incremental Increase in Fire Retardant Use
Cost of Fire Retardant
Training Costs
Days per Training Class
Number of Workers Trained
Number of Training Days Needed
    $375,000
      4.5%
      $2.08
$4,000/training class
     14 days
    14 workers
   3 days/year
 ICF, 2004
 ICF, 2004
Exxon, 2004
 ICF, 2004
 ICF, 2004
 ICF, 2004
 ICF, 2004
One Component:  Replacing HFC-152a with Hydrocarbons

This option is technically applicable to all HFC-152a emissions from the newly-produced one component
foams. The technical applicability of this option from 2005 to 2020 is presented in Table 3-20 and Table
3-21. This analysis assumes that the incremental maximum market penetration in 2010 for this option in
                                                                                             -70-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
the newly-produced one component foam market that uses HFC-152a would be 70 percent for the United
States and 90 percent for the rest of the world, both increasing to 100 percent by 2020 (see Table 3-22),
and its reduction efficiency is 100 percent. Assumptions specific to this substitution are presented in
Table 3-7.

Cost and Emission Reduction Analysis

    •   One-time Costs. According to industry experts, the one-time costs for replacing HFC-152a with
       hydrocarbons are $375,000,  which includes the cost of installing safety equipment.  Although
       some additional safety precautions already existed to handle the flammability of HFC-152a, they
       were assumed to be rather minor in comparison to a primary hydrocarbon blowing agent system;
       therefore,  the capital cost  is estimated  to be the same as the switch from  HFC-134a  to
       hydrocarbons.
    •   Annual Costs.  This  analysis assumes operating costs  of $321,103 for this abatement option.
       These costs result from  fire  retardant use and worker safety training associated with the use of
       hydrocarbons. Table 3-5 and Table 3-10 summarize assumptions associated with this abatement
       option.  Costs associated with fire retardant use are $309,103, and are calculated as the amount of
       foam produced multiplied by the incremental increase in fire retardant used in the foam and the
       fire  retardant costs.  Worker training costs  are estimated to be  $12,000 and are calculated by
       multiplying the costs of training  per day by the number of workers and the number of training
       days.
    •   Cost Savings. Because propane and butane  are less expensive on a per-pound basis than HFC-
       152a, a $410,400 annual cost savings is associated with this option. Costs of the current blowing
       agent are $554,400/yr, while costs of alternative blowing  agent are $144,000/yr.  Current costs
       are calculated by multiplying the per-pound cost of the blowing agent by the total amount of
       blowing agent used (288,0001b x  $1.925/lb =  $554,400/yr). The  calculation  of alternative
       blowing agent cost includes  per-pound costs of alternative blowing agent ($0.50/lb), the amount
       of blowing agent used (288,000 Ib), and the blowing  agent substitution ratio (100 percent).

    •   Emission  Reductions. This analysis estimates that by replacing HFC-152a with hydrocarbons,
       0.005 MMTCE of the high  GWP gas that would have been emitted by this facility during the
       lifetime of the one component foam produced in a given year are eliminated.
Table 3-7: Assumptions and costs used in the cost analysis to  substitute HFC-152a with Hydrocarbons
Variable
Capital Costs
Incremental Increase in Fire Retardant Use
Cost of Fire Retardant
Training Costs
Days per Training Class
Number of Workers Trained
Number of Training Days Needed
Value
$375,000
4.5%
$2
$4,000/training class
14 days
14 workers
3 days
Source
ICF, 2004
ICF, 2004
ICF, 2004
ICF, 2004
ICF, 2004
ICF, 2004
ICF, 2004
Extruded Polystyrene (XPS) Foams

Two blowing agent replacement options were considered for this end-use:

    •  Replacing HFC-134a/CO2 (LCD) with CO2 (LCD); and
    •  Replacing HFC-134a/CO2 (LCD) with CO2 (LCD)/Alcohol.
                                                                                           -71-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
An analysis was performed based on a hypothetical contractor that produces approximately 8.3 million ft3
of foam per year using an 80/20 blend of HFC-134a and CO2 as a blowing agent. The conversion to the
alternative  blowing agent is assumed to take place in only 10 percent of the line. Various "base case"
inputs and  assumptions are presented in Table 3-8.  This cost analysis estimates the break-even carbon
price for this hypothetical contractor to replace HFC-134a/CO2 with one of two non-HFC blowing agents:
1) CO2 or 2) CO2/Alcohol. The foams manufactured with the two alternatives are assumed to compensate
for lower insulating performance relative to HFC-blown foams by increasing the thickness and density of
the foam.  Thus, incremental differences in indirect emissions and  costs associated with energy penalties
are negligible.

Table 3-8:  Base Case Assumptions for a Hypothetical Contractor Using HFC-134a/C02 (LCD)	
Variable                                        Value                       Source
Blowing Agent Consumption                       1,500,000 Ib/yr           Estimated from Caleb (2001)
Polystyrene Consumption                            2.0 Ib/ft3                    Assumption
Foam Produced                                8,330,000 ft3/yr                 Assumption
Price of C02                                      $0.20/lb                   Airproducts, 2003
Price of HFC-134a                                 $170/lb                    Atofina, 2004
Price of Polystyrene	$0.40/lb	Purchasing.com, 2003

Cost factors that are addressed include:

    •   blowing agent costs;
    •   capital equipment costs;
    •   increased consumption of foam components to compensate  for increased foam density; and
    •   incremental differences in the costs of blowing agents and the quantity required.
XPS: Replacing HFC-134a/CO2 (LCD) with CO2 (LCD)

This option is technically applicable to all emissions from the newly-produced extruded polystyrene
foam, but the assumed market penetration is tempered by the existence of another feasible option (i.e.,
CO2/alcohol).  The technical applicability of this option from 2005 to 2020 is presented in Table 3-20 and
Table 3-21. The current blowing agent is assumed to be an 80/20 blend of HFC-134a and CO2. This
analysis assumes that the  incremental maximum market penetration of this option into the newly-
manufactured extruded polystyrene foam market would be 0 percent for the United States through 2020
and 35 percent for the rest of the world in 2010, rising to 45 percent by 2020 (see Table 3-22), and that its
reduction efficiency is 100 percent. Assumptions  specific to this substitution are  presented in Table 3-9.

Cost and Emission Reduction Analysis

    •   One-time Costs.  Because the conversion to a CO2 line requires high pressure equipment, based
       on  industry opinion, this analysis assumes a $3,500,000 one-time cost for replacing  HFC-134a
       with CO2.

    •   Annual Costs. This analysis assumes operating costs of $66,700 for this abatement option due to
       the increased consumption of polystyrene based on increased foam density. This cost is calculated
       by  multiplying the total amount of foam  produced by the amount of polystyrene needed and the
       per-pound  price  of polystyrene  (8,330,000 ft3  x  2.0 Ib/ft3 x $0.40/lb  = $6,664,000) and then
       accounting for the 10 percent foam density increase and the conversion of only 10 percent of the
       line. ($6,664,000 x 10% x  10% = $66,640).
                                                                                             -72-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •  Cost Savings.  Because the alternative blowing agent is less expensive on a per-pound basis than
       HFC-134a, there is a cost saving of $180,000/year associated with this option. This number is
       derived by subtracting the per pound costs of the alternative blowing agent used from the costs of
       current blowing agent blend ($1.40/lb - $0.20/lb = $1.20/lb).  The cost of the alternative blowing
       agent (CO2) is $0.20/lb, while the cost of the current blowing agent blend (80/20 blend of HFC-
       134a/CO2) is $1.40/lb. The resulting price difference is then multiplied by the total blowing agent
       used (1,500,000 Ib/yr) and by the converted 10 percent of the line.
    •  Emission Reductions. This analysis estimates that by replacing HFC-134a/CO2 (LCD) with CO2
       (LCD) in 10% of the line, 0.019 MMTCE of high GWP gas that would have been emitted by this
       contractor during the lifetime of the foam generated in a given year are eliminated.
Table 3-9: Assumptions and costs used in  the cost analysis to substitute HFC-134a/C02 with CCh	
Variable                                      Value                        Source
Capital Costs                                $3,500,000a                    Assumption
Increase in Foam Density                          10%                       Assumption
a Assumes that a conversion to a 100 percent C02 blowing agent requires high pressure equipment.

XPS: Replacing HFC-134a/CO2 (LCD) with CO2 (LCDVAlcohol

The current blowing agent for XPS is assumed to be an 80/20 blend of HFC-134a and CO2. This option
is technically applicable to all emissions from the newly-produced extruded polystyrene foam, but the
assumed market penetration is tempered by the existence of another feasible option (i.e., CO2).  The
technical applicability of this option from 2005 to 2020 is presented in Table 3-20 and Table 3-21. The
incremental  maximum market penetration of this option into the newly-produced extruded polystyrene
foam market is assumed to be 0 percent for the United States through 2020, and 35 percent in 2010, rising
to 45 percent by 2020 for China (see Table 3-22), with a reduction efficiency of 100 percent.
Assumptions specific to this substitution are  presented in Table 3-10.

Cost and Emission Reduction Analysis

    •  One-time Costs. According to industry experts, the one-time costs for replacing HFC-134a with
       a CO2 and alcohol blend are $800,000.  Blends of CO2 with alcohol (e.g., ethanol) require lower
       pressure extrusion. As a result,  capital costs associated  with this method are lower than  the
       conversion to pure CO2.
    •  Annual Costs.   This analysis assumes annual  operating costs of $33,300/year for this abatement
       option. Costs are calculated by multiplying the total amount of foam produced  in 10 percent of a
       line  by the amount of polystyrene needed and  the per-pound price of polystyrene (8,330,000ft3 x
       10% x 2.01b/ft3 x $0.40/lb = $666,500) and then accounting for the  5 percent foam density
       increase ($666,500 x 5% = $33,325).
    •  Cost Savings.  Because the alternative blowing agent is less expensive on a per-pound basis than
       HFC-134a, there is $166,500 annual cost savings associated with this option. This number is
       derived by subtracting the per-pound cost of the alternative  blowing agent used (an 80/20 blend
       of HFC-134a/ CO2) from the per-pound cost of the current blowing agent blend (an 80/20 blend
       of CO2/ethanol) ($1.40/lb -  $0.29/lb =  $1.1 Mb).  The resulting price is then  multiplied by  the
       total blowing agent used (1,500,000  Ib/yr) and by the 10 percent of the line converted.
    •  Emission Reductions. This analysis estimates that by replacing HFC-134a/CO2 (LCD) with CO2
       (LCD), 0.019 MMTCE of the high GWP  gas that would have been  emitted by this contractor
       during the lifetime of the foam generated in a given year are eliminated.
                                                                                            -73-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 3-10: Assumptions and costs used in the cost analysis to substitute HFC-134a/C02 with C02/Alcohol
Variable                                  Value                   Source
Capital Costs                              $800,000a                 Assumption
Increase in Foam Density                       5%                   Assumption
Price of Alcohol                             $0.65                Purchasing.com
a Blends with alcohol (e.g., ethanol) require lower pressure extrusion; therefore lower capital than pure C02.

Polyurethane (PU) Spray Foams

Two blowing agent replacement options were considered for this end-use:

    •   Replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with CO2 (water); and
    •   Replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with cyclopentane/isopentane.
An analysis was performed based on a hypothetical spray foam contractor that produces approximately
127,000 pounds of foam per year using a 75/25 blend of HFC-245fa35 and CO2 (water) as a blowing agent.
The blowing agent constitutes approximately  10 percent of the  foam, by weight.  Various "base case"
inputs and assumptions are presented in Table 3-11. The foams manufactured with the two alternatives
are assumed to compensate for lower insulating performance relative to HFC-blown foams by increasing
the thickness and density of the foam. Thus, there are no incremental differences in indirect emissions
and costs  associated with  energy  penalties.  Although  this  end-use  uses  HFC-245fa  and  HFC-
365mfc/HFC-227ea, the analysis was performed based on a spray foam contractor that uses HFC-245fa.

Table 3-11: Base Case Assumptions for a Hypothetical Spray Foam Contractor Using HFC-245fa/C02 (water)
Variable                                         Value                         Source
Blowing Agent Component of Foam                      10%                  Cannon, 2001; NCFI, 2001
HFC-245fa/C02 Ratio                                75/25                        Assumption
Blowing Agent Use                                12,735 Ib               Estimated from Caleb (2001)
Foam Produced                                  127,347 Ib                      Calculation
Original Foam Cost                                $0.92/lb                       Assumption
Price of Isocyanate                                $1.00/lb                      Cannon, 2001
Price of HFC-245fa	$4.00/lb	Honeywell, 2003	

Cost factors that are addressed include:

    •   fire testing costs incurred by system houses for various formulations;
    •   capital equipment costs;
    •   employee training costs (hydrocarbons only);
    •   increased cost of foam components (e.g., polyols, additives, etc.);
    •   increased consumption of foam components to compensate for increased foam density;
    •   increased use of fire retardant; and
    •   incremental differences in the costs of blowing agents and the quantity required.
35 TheEU countries use a blend of HFC-365mfc and HFC-227ea in ratios of 93:7 or 87:13, while Japan uses a blend
of HFC-245fa and HFC-365mfc in ratios of 80:20 or 70:30.  This report presents a cost analysis based on the 75/25
HFC-245fa/CO2 blend and applies it globally as a representative estimate.


                                                                                              -74-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Annual emission reductions were determined based on the estimated amount of blowing agent consumed
by the hypothetical contractor and from the emissions profile used in the Vintaging Model (see Table
3-1).

Sprav: Replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with CO2 (water)

This option is technically applicable36 to all emissions from the newly manufactured spray polyurethane
foam market, but the assumed market penetration is tempered by the existence of another feasible option
(i.e., hydrocarbons). The technical applicability of this option as well as other options from 2005 to 2020
is presented in Table 3-20 and Table 3-21. This analysis assumes that the incremental maximum market
penetration for this option in 2010 into the newly formulated polyurethane spray foam market is 5 percent
for the United  States and 7.5 percent for the rest of the world, both rising to 20 percent by 2020 (see Table
3-22), and that its reduction efficiency is 100 percent because the HFC blowing agent is completely
replaced (the GWP of CO2 is not included in the analysis).  For cost estimating purposes, this option
assumes that the current blowing agent is a 75/25 blend of HFC-245fa and CO2.  Assumptions specific to
this substitution are presented in Table 3-12.

Cost and Emission Reduction Analysis

    •  One-time  Costs.   According  to industry experts, contractors  that are  using HFC-245fa/CO2
       (water) can use the same equipment for CO2 (water) with only minimal modification  (Caleb,
       2001). This analysis assumes  that  a  one-time  cost of  $4,000 is  needed to convert to this
       alternative  blowing agent. This cost is associated  with fire testing, which is based on $250,000
       fire testing costs for new formulations, 20 systems houses that will need  to pass some tests, and
       approximately  1,250  spray foam contractors who equally share these costs ($250,000 x 20 /
       1,250) (Caleb, 2001).

    •  Annual Costs. This analysis assumes a unit annual operating  cost of $54,500 for this abatement
       option. These costs result from fire retardant use, costs due to increased density of foam, and the
       increased foam costs  (refer to  Table 3-11  and  Table 3-12  for detailed  assumptions). Costs
       associated with fire retardant use are $2,648 and they are calculated by multiplying the amount of
       foam produced (127,347 Ib) by the incremental percent of fire retardant in the foam (1%) and fire
       retardant costs ($2.08/lb). Increased foam costs are $12,735, and are calculated by multiplying the
       amount of foam produced by the increase in foam costs ($0.10/lb). Costs due to increased density
       are $39,121 and are calculated by multiplying the  amount of foam produced by alternative foam
       costs ($1.02/lb) and the increase in foam density (30%).
    •  Cost Savings.  Because the alternative blowing agent is less expensive on a per-pound basis than
       HFC-245fa, there  is an $11,461  annual cost savings associated with this  option.  Cost of the
       current blowing agent is $57,306, while the cost of alternative blowing agent is $45,845. Current
       costs are calculated by multiplying the per-pound cost of the blowing agent by the total amount of
       blowing agent used as follows:  total HFC-245fa blowing agent used x 75% x per-pound cost of
       the blowing agent  +  total  CO2 blowing agent used x  25%  x isocyanate/CO2 ratio x price  of
       isocyanate  ((12,7351b x 75% x  $4/lb) + (12,7351b x 25% x 6 x  $l/lb) = $57,306). The calculation
       of alternative  blowing  agent cost includes  the per-pound  costs of alternative blowing agent
       ($1.00), the amount of blowing  agent used  (12,7351b), the isocyanate/CO2 ratio (6), and  the
       blowing agent  component of foam ratio  (60%).   See  Table 3-11 for prices of the current and
36 In this report, the term technically applicable refers to the emissions to which an option can be applied. A
glossary of terminology may be found in Appendix I.
                                                                                            -75-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
       alternative blowing  agent  and Table 3-12 for other assumptions that are associated with cost
       savings.
    •  Emission Reductions.  This  analysis estimates that by replacing HFC-245fa/CO2 (water) with
       CO2 (water), 1,122 TCE (i.e., 12,735 Ib, or 5.78 metric tons, blowing agent multiplied  by 75
       percent content HFC-245fa, the GWP of 950, and the carbon to carbon-dioxide conversion ratio
       of 12/44) of the high  GWP gas that would have been  emitted by this hypothetical spray foam
       manufacturer during the lifetime of the foam produced in a given year are eliminated.
Table 3-12: Assumptions and costs used in the cost analysis to substitute HFC-245fa/C02 (water) with
(water)
Variable
Capital Costs
Fire Testing Costs
Incremental Increase in Fire Retardant Use
Cost of Fire Retardant
Isocyanate/C02 Ratio
C02/Water Component of Foam
Decrease Blowing Agent Component of Foam
Increase in Foam Density
Assumed Increase in Cost of Foam
Alternative Foam Costs
Value
Negligible3
$4,OOOb/Contractor
1%
$2.08/lb
6
6%
60%
30C%
$0.10/lbs
$1.02
Source
Caleb, 2001
Caleb (2001)
Assumption
Exxon, 2004
Cannon, 2001
Stepan, 2001
Stepan, 2001
Assumption
Assumption
Calculation
a Assumes that contractors that are using HFC-245fa/C02 (water) have equipment that can use C02 (water) with minimal modification.
b Based on $250,000 per systems house, 20 systems houses, and approximately 1,250 spray foam contractors (Caleb, 2001).
c Assumes that foam density increases from 2.5 Ib/ft3 to 3.25 Ib/ft3.

Spray:  Replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with Hydrocarbons

The difference in costs between this abatement option and replacing HFC-245fa/CO2 with CO2 is that this
abatement option has factored into the analysis the cost of training workers in handling, storing, and using
hydrocarbons. For cost estimating purposes, the current blowing agent is assumed to be a 75/25 blend of
HFC-245fa and CO2, while the alternative blowing agent is assumed to be an 80/20 blend of cyclopentane
and isopentane.  This option and the previous one are technically applicable to all emissions from the
newly-produced spray polyurethane foam market.  The technical applicability of this option from 2005 to
2020 is presented in Table 3-20 and Table 3-21.  This analysis assumes that the incremental maximum
market penetration of this option in 2010 in the newly-produced spray foam market would be 10 percent
for the United States and 5 percent for the rest of the world, rising in later years to 30 percent in the
United States and 15 percent in the  rest of the world (see Table 3-22), and its reduction efficiency is 100
percent. There could be some safety and liability concerns associated with this substitution, which could
lead to reduced market penetration or increased cost of this option. Assumptions specific to this
abatement opportunity are presented in Table 3-13.

Cost and Emission Reduction Analysis

    •    One-time Costs. According to industry experts, in the US, the one-time costs for replacing HFC-
        245fa/CO2 (water) with hydrocarbons are estimated to be $10,000 (Exxon, 2001).  Based  on
        $250,000 fire  testing costs for new formulations, fire testing  for  20 systems  houses, and
        approximately 1,250 spray foam contractors, one time costs associated with fire testing in the US
        are $4,000/contractor ($250,000 x  20/1,250) (Caleb, 2001).  Fire testing costs in the EU are
        $22,642, while in Japan these costs are $33,924 (BRE, 2004; JUFMA, 2004).
                                                                                              -76-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •   Annual Costs. This analysis assumes a unit annual operating cost for this abatement option of
        $40,437. These costs result from fire retardant use, costs due to increased density of foam, and
        worker safety training costs associated with the use of hydrocarbons.  Table 3-11 and Table 3-13
        summarize assumptions  associated  with this  abatement  option.   Costs associated  with fire
        retardant use are $7,927 and are calculated by multiplying the amount of foam produced (127,347
        Ib) by the incremental percent of fire retardant in foam (3%), and by the additional fire retardant
        costs  ($2.08/lb).   Worker training  costs are  estimated to  be $6,429  and are calculated by
        multiplying costs of training per day by the number of workers and the number of training days as
        shown in the table. Costs due to increased density are estimated to be $26,081, and are calculated
        by multiplying the amount of foam produced  by the per-pound foam costs ($1.02/lb) and the
        increase in foam density (20%).
    •   Cost Savings.  Because the alternative blowing agent is less expensive on a per kilogram basis
        than HFC-245fa, there is a  $49,398 annual cost savings associated with this option  and is
        calculated by subtracting current blowing agent costs ($57,306/yr) from the alternative blowing
        agent costs ($7,908/yr).  Current costs are calculated as shown in the previous  example.  The
        calculation of alternative blowing agent cost includes  per-pound costs of alternative blowing
        agent (i.e., (80% x $0.80/lb)  + (20%  x $0.25/lb) = $0.69/lb), the amount  of blowing agent used
        (12,7351b), and the blowing agent substitution ratio (90%).  See the  base case assumptions table
        (Table 3-11)  for prices of the current and alternative blowing agent and Table 3-13 for other
        assumptions that are associated with cost savings.
    •   Emission Reductions.  This analysis estimates that by replacing HFC-245fa/CO2 (water) with
        hydrocarbons, 1,122 TCE of the high GWP gas that would have been emitted by this hypothetical
        foam manufacturer during the lifetime of the foam produced in a given year are eliminated.
Table 3-13: Assumptions and costs used in the cost analysis to substitute HFC-245fa/C02 (water) with
Hydrocarbons
Variable
Capital Costs
Fire Testing Costs
Incremental Increase in Fire Retardant Use
Cost of Fire Retardant
Price of Cyclopentane
Price of Isopentane
Blowing Agent Component of Foam
Increase in Foam Density
Blowing Agent Substitution Ratio
Assumed Increase in Cost of Foam
Alternative Foam Cost
Employee Training Costs
Value
$10,000"
$4,OOOb/Contractor
3%
$2.08/lb
$0.80/lbc
$0.25/lbc
9%
20%d
90%
$0.10/lb
$1.02/lb
$6,450/yre
Source
Exxon, 2001
Caleb, 2001
Assumption
Exxon, 2004
Exxon, 2004
Exxon, 2004
Exxon, 2001
Assumption
Exxon, 2001
Assumption
Calculation
SPF, 2001
a Assumes that technical issues can be resolved.
b Based on $250,000 per systems house, 20 systems houses, and approximately 1,250 spray foam contractors (Caleb, 2001; Industry
Communication).
c HCs used as a BA in foams are approximately an 80/20 blend of cyclopentane and isopentane.
d Assumes that foam density increases from 2.5 Ib/ft3 to 3.0 Ib/ft3.
e Assumes a cost of approximately $215/employee/day, 2 crews of 3 employees (total of 6 employees), and 5 days of training.

PU Appliance Foams

Two blowing agent replacement abatement options were considered for this end-use:
                                                                                               -77-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •  Replacing HFC-134a with cyclopentane/isopentane; and
    •  Replacing HFC-245fa and HFC-365mfc/HCF-227ea and with cyclopentane/isopentane.
This scenario examines a hypothetical facility that manufactures approximately 536,000 refrigerators and
consumes about 1.68 million pounds of blowing agent annually.  The blowing agent was assumed to
constitute approximately 12 percent of the foam.   The costs of producing a refrigerator using each
blowing  agent (e.g.,  HFC-134a,  HFC-245fa,  and cyclopentane/isopentane)  were provided  by the
refrigeration industry. Data have been aggregated to protect confidential business information. This
scenario was developed for a facility manufacturing large appliances typically used in the United States.
While other markets may use different-sized refrigerators and hence per-appliance factors may differ, this
analysis assumes that the resulting cost per HFC emissions abated ($/TCE)  is approximately the same.
Factors considered in these data include:

    •  capital costs to convert;
    •  blowing agent costs;
    •  foam costs (including density considerations);
    •  High-Impact Polystyrene (HIPS) and Acrylonitrile-Butadiene-Styrene (ABS) liner costs; and
    •  additional costs  required to meet the  U.S.  2001  National Appliance Energy Conservation Act
       (NAECA) energy efficiency standards.

HFC emission reductions over time were derived from the emissions profile used in the Vintaging Model
(see Table 3-1), that account  for gases released from the manufacturing process, annual release, and
disposal.  Because the cost data are based on the assumption that the refrigerators manufactured using
various blowing agents meet the same energy efficiency standards, there are no incremental differences in
indirect emissions and costs due to energy efficiency.

PU Appliance: Replacing HFC-134a with Hydrocarbons

This option is technically applicable to all HFC-134a emissions from the newly manufactured appliance
foam. The technical applicability of this option from 2005 to 2020 is presented in Table 3-20 and Table
3-21. This analysis assumes that the incremental maximum market penetration in 2010 for this option in
the newly-manufactured appliance market that uses HFC-134a would be 25 percent for the United States
and 85 percent for the rest of Annex I, rising to 70 percent and 90 percent, respectively, by 2020 (see
Table 3-22). Because the HFC is completely replaced, the reduction efficiency is 100 percent.

Cost and Emission Reduction Analysis

    •  One-time Costs. According to industry experts, the one-time costs for replacing HFC-134a with
       hydrocarbons are $50,000,000, which includes the capital cost to convert.
    •  Annual Costs.  This analysis assumes that cyclopentane variable costs will be comparable to
       HFC-134a variable costs; therefore no increased  annual  unit operating costs are assumed to be
       associated with this abatement option.  Costs considered include  HIPS liner, ABS liner, foam
       density, and energy costs.

    •  Cost Savings.  Because cyclopentane  is less expensive on a per-pound basis than  HFC-134a, a
       $1,506,160  annual cost savings  is associated with this  option.   This  result derives  from
       incremental per-unit cost difference ($2.8 I/unit) multiplied by the number of units a hypothetical
       factory manufactures (536,000 units).
                                                                                            -78-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •   Emission Reductions. This analysis estimates that by replacing HFC-134a with hydrocarbons,
       0.27 MMTCE of the  high GWP gas that would have been emitted by this facility during the
       lifetime of the appliance foam produced in a given year are eliminated.
PU Appliance: Replacing HFC-245fa and HFC-365mfc/HFC-227ea with Hydrocarbons

Although this end-use uses HFC-245fa and HFC-365mfc/HFC-227ea, the analysis was performed based
on the cost to replace HFC-245fa in appliance foams. This option is technically applicable to all
emissions from the newly-produced appliance foams that use HFC-245fa.  The technical applicability of
this option from 2005 to 2020 is presented in Table 3-20 and Table 3-21.  This analysis assumes that the
incremental maximum market penetration of this option in 2010 into the newly-manufactured appliance
market that uses HFC-245fa is 15 percent for the United States and 85 percent for the rest of Annex 1,
rising to 50 percent and 90 percent, respectively, by 2020 (see Table 3-22).  The increase of market
penetration in China is 90 percent in 2020.  Because the HFC is completely replaced, the reduction
efficiency is 100 percent.

Cost and Emission Reduction Analysis

    •   One-time Costs.  According to industry experts, the one-time costs for replacing HFC-245fa with
       hydrocarbons are $50,000,000.
    •   Annual Costs. Due to costs associated with overcoming the energy gap between the foam blown
       with HFC and the foam blown with the alternative blowing agent, total annual costs for replacing
       HFC-245fa with hydrocarbons are estimated to be $11,202,400. Costs considered include  HIPS
       liner, ABS liner,  foam density, and energy.  Individual costs for each  of these elements  were
       provided from industry; however, because they are considered confidential, they are presented in
       aggregate.
    •   Cost Savings. No cost savings are associated with this abatement option.
    •   Emission Reductions. This analysis estimates that by replacing HFC-245fa with hydrocarbons,
       0.20 MMTCE of the high GWP gas that would have been emitted  by  a  representative facility
       during the lifetime of the appliance foam produced in a given year are eliminated.
PU Appliance: End-of-Life Options

In addition to the two blowing agent replacement options considered above, two  end-of-life abatement
options were considered for this end-use:

    •   Automated Process  with Foam Grinding, HFC Adsorption, and Foam Landfilling in Appliance
       Foam; and
    •   Manual Process with Foam Incineration in Appliance Foam.

The baseline emissions assume that the remainder of the blowing agent contained in the appliance foam is
released at the foam's end-of-life, as shown in Table  3-1. There are different technologies for abating
those end-of-life emissions  in  PU Appliance Foams.  These technologies include landfilling the  foam
after recovering the blowing agent (which  could either be destroyed or reclaimed and sold back to the
market) and incinerating  the foam in a municipal solid waste incinerator (MSWI) or waste-to-energy
plant. This analysis analyzes the landfilling after recovering HFC and the MSWI  options. This analysis
assumes that when the HFC is recovered,  it will still have value and hence contribute revenue to the
process. HFC-134a and HFC-245fa are used in appliance foam in some locations.  This analysis assumes
that half of the appliances processed use HFC-134a and the other half use HFC-245fa in order to account
for the  chemicals' different GWPs and costs. Further market research could refine this assumption.


                                                                                          -79-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
The remainder of this section presents cost estimates for each step involved in the removal  and/or
destruction of HFC contained in the foam, either through MSWI or grinding/adsorption/landfilling.  Costs
are presented in terms of dollars per refrigerator destroyed and in dollars per pound of HFC destroyed.
This analysis uses the best cost information available; however, the costs presented should be considered
illustrative rather than  definitive.  The analysis is done using the U.S. market as an example, recognizing
that the size of a U.S.  refrigerator/freezer is typically larger than those used in  other parts of the  world.
All assumptions where based on  a side-by-side refrigerator type.  The final results (i.e., cost per unit of
emissions abated) are applied to other regions because it is felt that the  relative costs and emissions abated
should scale roughly linearly to smaller appliances used elsewhere.

Cost factors that are addressed include:

    •   collection and  consolidation of appliances;

    •   transportation of appliances to disassembly location;
    •   disassembly and processing of appliances;
    •   transportation of foam to  landfilling or incineration location; and
    •   landfilling or incineration of foam.
                                                                                               -80-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
In general, two basic methods of handling appliances to abate blowing agent emissions exist:

•       Automated Process with Foam Grinding, HFC Adsorption, and Foam Landfilling.   This
        method involves purchasing a sophisticated  system where the appliance is brought into the
        system without much preparation work.   The  system  shreds  the  appliance  and  uses  various
        techniques such as magnets and eddy current to separate the metals, plastics and foams.  The
        blowing  agent (and the  refrigerant) are  collected by adsorption37 onto a  carbon  substrate.
        Typically, the absorbed gases are then incinerated, or they can  be reclaimed and sold back into
        the market.  These systems are capital-intensive, costing roughly  $4 million (JACO, 2004);
        however, once  established, the manual labor is reduced.  This type  of process is generally only
        cost-effective if a high flow of appliances (hundreds of thousands per year) is achieved.

•       Manual Process with Foam Incineration. This method uses mostly manual labor to evacuate
        and recycle the refrigerant, drain and recycle the compressor oil, and disassemble the appliances,
        recovering and recycling glass shelves, plastic interior parts, steel, aluminum and other valuable
        metals.  The foam is removed in large pieces,  which can be quickly sealed in plastic  bags to
        prevent further off-gassing of the blowing agent,  and sent for incineration.

Table 3-14  General Assumptions Applicable Both End-of-Life Options	
Variable
    Value
Source
Refrigerators per Truckload
Average quantity of foam per unit
Percent of BA remaining at disposal
Labor Rate
HFC-245fa content of PU Foam
HFC-134a content of PU Foam
Travel Distance to Disassembly Location
Truck Operating Rate
Volume
Density of Foam
Cost of steel
Steel content of the unit
77 refrigerators                   JACO, 2004
   22.87 Ib                     Whirlpool, 2004
   92.25%                     EPA estimates
 $12.00/hour             Bureau of Labor Statistics, 2004
    13%                      Whirlpool, 2004
    7.5%                      Whirlpool, 2004
  100 miles                     Assumption
  $1.75/mile                     JACO, 2004
  6,082 cu ft          Systems Transportation Equipment, 2004
  2 Ib/cu ft                       ICF, 2004
   0.04/lb                      JACO, 2004
   132.3lb                     Whirlpool, 2004
Table 3-15  Assumptions Applicable to the Manual Process with Foam Incineration
Variable
    Value
Source
Loading Labor Hours
Unload Labor Hours
Cost of Disassembly
Cost to Incinerate Foam
% Foam Recovered
% Blowing Agent Incinerated
4 hours
2 hours
$35/unit
$0.53/lb
92.5%
98%
JACO, 2004
JACO, 2004
JACO, 2004
JACO, 2004
JACO, 2004
UNEP, 2002
Table 3-16  Assumptions Applicable to the Automated Process with Foam Landfilling
Variable
    Value
Source
Refrigerators Disassembled per Hour                       6.00
Grinding/Absorption Labor Hours                      4 hours/truckload
Grinding/Absorption Equipment Operating Hours         4 hours/truckload
                               JACO, 2004
                               Assumption
                               Assumption
37 Other methods of blowing agent recovery are possible. For instance, some plants use liquid nitrogen to mitigate
explosion potential with hydrocarbon units. The nitrogen also serves to liquefy and collect the blowing agent.
                                                                                                 -81-

-------
Analysis of Costs to Abate International OPS Substitute Emissions

Variable
Grinding/Absorption Equipment Cost
% Blowing Agent Recovered
Recovery value of HFC-134a
Recovery value of HFC-245fa
Cost Per Container to Landfill
Container Volume
Value
$20/hour
90%
$1.70/lb
$2.88/lb
$250/container
540 cu. ft.
Source
Assumption
UNEP, 2002; JACO, 2004
Atofina, 2004
ICF, 2004
JACO, 2004
JACO, 2004
General Cost Analysis Applicable to Both Processes

    •    Collection  and  Consolidation. This  analysis assumes that appliances are  collected and
        consolidated over a 4-hour period into full truckloads  for shipment to a central  disassembly
        location.  The cost for collection and consolidation of appliances is $0.62 per refrigerator (i.e., 4
        hours x $12/hour / 77 refrigerators).  The collection and consolidation is roughly the same for
        both the automated and manual processes.
    •    Transportation  of Appliances to  Disassembly Location.    Transportation costs  to  the
        disassembly  location are  $2.58  per  unit.  The analysis  assumes the appliances are shipped a
        distance  of 100 miles from the  collection and  consolidation  location  to a central  disassembly
        location.  The operating cost of the truck is assumed to be $1.75 per truckload  mile. Truck
        unloading is  conducted by two people over a 1 hour period, and 77 refrigerators  per truckload is
        assumed.  The labor rate is $12.00/hour (ICF Estimate; JACO, 2004; Bureau of Labor Statistics,
        2004). The calculation is as follows: (($1.75 x  100) + (2 x $12.00)) / 77 = $2.58/refrigerator.
        Transportation of Appliances to Disassembly  Location  are  generally the same for both the
        automated and manual processes.
    •    Disassembly  of Appliances.   The appliances are then disassembled using  the  manual  or
        automated process. See details below.

    •    Recovery of HFC.  In the Automated Process with Foam Landfilling, the foam is ground and the
        HFC is recovered. See details below.

    •    Recovery of Metals. The steel content of the refrigerator is 132.3 Ib and steel prices are around
        $0.04/lb.  Benefits from salvaged steel are therefore $5.29/unit. The benefits of other recovered
        materials (other than the HFC) are not included in this analysis.

    •    Transportation of Foam to Disposal Location. The cost to transport and landfill the ground
        polyurethane foam (automated process) is assumed to be the  same as  the cost to transport and
        incinerate the  unground polyurethane  foam (manual  process).   This  analysis assumes that
        truckloads of ground or unground polyurethane foam are shipped a distance of 100 miles from the
        disassembly/grinding location to  an incineration/landfill location. The operating cost of the truck
        is assumed to be $1.75 per truckload mile (ICF estimate). The total weight of foam in the truck is
        the volume of that truck (6,082 ft3) multiplied by the packing efficiency (70%) and the density of
        the foam (2 Ib/ft), roughly 8,514.5 Ib. The cost to transport the polyurethane foam to the disposal
        location is equal to $0.47/refrigerator (i.e., 100 miles x $1.75/mile / 8,514.5 Ib foam x 22.87 Ib
        foam/refrigerator).
    •    Disposal  of Foam.  In the  Automated Process,  the ground foam is landfilled.   In the  Manual
        Process, the foam pieces are incinerated. See details below.
    •    Emission Reductions.  The HFC-134a blowing agent content at manufacture  is 1.72 pounds,
        determined by multiplying 22.87 Ib foam by 7.5% HFC-134a content. Likewise, 2.97 pounds of

                                                                                            -82-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
       HFC-245fa  are contained in each refrigerator at manufacture.  This value is  determined by
       multiplying  22.87 Ib foam by 13%. Because only 92.25 percent of the blowing agent remains at
       disposal, the maximum abatable emissions are 1.58 Ib (HFC-134a) or 2.74 Ib (HFC-245fa), which
       translates to 0.254 TCE and 0.322 TCE, respectively.  This analysis assumes that half of the
       refrigerators that are processed will  contain  HFC-134a foam, and the remainder will  contain
       HFC-245fa  foam.  Thus,  the average emissions potentially reduced per refrigerator are 0.288
       TCE.

Cost Analysis Applicable to the Automated Process with Foam Landfilling

    •   Disassembly of Appliances. For the automated  process, this analysis assumes that 12.83 labor
       hours  are required to  disassemble a full truckload of 77 refrigerators.   The refrigerators are
       assumed to be  disassembled at  a  rate of 6 refrigerators per  hour  (77/6 =  12.83).   Total
       disassembly costs are $2.00/refrigerator  (i.e.,   (12.83 hours  x $12/hour)/77  refrigerators  =
       $2.00/refrigerator).
    •   Automated Grinding  of  Foam for  Landfilling and Adsorption of HFC. In the automated
       process, the polyurethane foam is ground, the HFC is adsorbed onto  a carbon substrate,  and the
       ground polyurethane waste is transported to the landfill.   The non-labor operating cost of the
       grinding/adsorption equipment (e.g., electricity to operate the plant, periodic maintenance of the
       plant, etc.) is assumed to be $20 per hour. By grinding the foam, only an estimated 90 percent of
       the HFC is recovered, while the remaining 10 percent is lost to the atmosphere (UNEP, 2002b).
       Therefore, a total of 150 Ib of HFC  is recovered per truckload (i.e., 166.5  Ib of HFC prior to
       processing/truckload x  90%). The amount of HFC per truckload prior to processing is calculated
       by multiplying the amount of HFC in refrigerator foam prior to disposal (assuming 50% of the
       units have HFC-245fa foam and 50% of the units have HFC-134a foam) by  the total number of
       refrigerators per truckload.  The calculation is as follows: (0.5 x 2.74 + 0.5 x  1.58) x 77 = 166.5.
       The content-weighted recovery value of the HFC is $2.29/lb (i.e., 50% x $2.88 + 50% x  $1.70).
       However, there  is a cost  to recovery the HFC.  The  cost of HFC recovery is calculated by
       dividing  the grinding  and adsorption  costs by  the  total  amount of HFC  actually recovered
       ($128/truckload / 150 Ib/truckload =  $0.85/lb). Grinding and adsorption costs are calculated by
       summing the grinding and adsorption labor cost  (4 hours/truckload x $12/hour)  and the  cost of
       equipment used in grinding and adsorption ($20.00/hour x 4 hours of equipment/truckload).  The
       net recovery value of the HFC is therefore $1.43/lb (i.e., $2.29 - $0.85), or $215/truckload (i.e.,
       $1.43  x  150),  or $2.79/unit (i.e., $215/77).  Thus, grand total savings that include  disassembly,
       processing, and recovery of the HFC are equal to $0.79/unit (i.e., costs are $2.00 - $2.79).
    •   Landfilling of Foam. This analysis assumes that landfilling of the polyurethane foam occurs in a
       municipal solid waste landfill at a cost of $250.00 per pull  of foam for a 20 cubic yard cylinder.
       Since  the container volume is  540 ft3  (20  yd3),  the  packing efficiency is 70 percent, and the
       density of foam is 2 lb/ft3, the weight of foam loaded is 756 Ib/container.  Thus, landfilling foam
       will cost $0.33/lb. The total cost of landfilling per unit is  calculated by multiplying the  cost of
       landfilling per pound of foam with the foam content of the refrigerator (22.87  Ib of foam per
       unit), the cost of landfilling is estimated to be $7.56 per refrigerator.

Cost Analysis Applicable to the Manual Process with Foam Incineration

    •   Manual  Dismantling of Foam for Incineration. For the manual process, this analysis assumes
       that 90-95 percent of the  foam is recovered (for calculation purposes, 92.5% is used).  In the
       manual process, the large foam pieces are separated and sent for incineration.  The costs of
       disassembly are assumed to be $35.00/refrigerator (JACO, 2004).
                                                                                            -83-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •   Incineration of Foam.  The cost of incineration of polyurethane foam is estimated to be $ 11.11
       per refrigerator. This cost is derived by multiplying the foam content of the refrigerator (22.87 Ib)
       prior to processing by the percent of foam recovered (92.5%) and by the cost of incineration per
       pound ($0.53/lb).
PU Appliance: Automated Process with Foam Grinding, HFC Adsorption, and Foam Landfilling

The technical applicability of this option from year 2005 through 2020 is presented in Table 3-20  and
Table 3-21.  This analysis assumes that the incremental maximum market penetration of this option in
2020 in the  newly-manufactured appliance foam market would be 10 percent in the United States, 95
percent in the rest of the developed world (see Table 3-22).

Cost and Emission Reduction Analysis

    •   One-time Costs.  This analysis assumes a one-time capital cost of $4 million for an Adelmann or
       MeWa plant (JACO, 2004).
    •   Annual Costs.  The  capacity of the  facility is 250,000 refrigerators  per year (JACO, 2004),
       therefore, the annual  costs for this technology are  $1,290,000, calculated by multiplying the
       capacity of the facility with the net costs of this technology per unit (see Table 3-17).
    •   Cost Savings.   The  savings  from salvaging  the steel  ($5.29/unit) and  HFC blowing agent
       ($2.79/unit) are accounted for in the Annual Costs, above.
    •   Emission Reductions. This analysis estimates that by treating the appliance foam at the end of
       life with foam grinding,  HFC adsorption, and landfilling, 0.072 MMTCE (i.e., 0.288  TCE/unit x
       250,000 units) of the high GWP gas can potentially be eliminated annually.  Accounting for the
       reduction  efficiency of 90 percent (UNEP,  2002b), the actual annual emissions abated are 0.065
       MMTCE.
Table 3-17: Automated Process with Foam Landfilling
Method
Collection/Consolidation
Transportation to Disassembly Location
Disassembly & Processing of Refrigerators
Transportation to Disposal Location
Landfilling
Total Costs
Savings (salvaged steel)
Net Costs
Cost per Unit
$0.62
$2.58
-$0.79
$0.47
$7.56
$10.45
$5.29
$5.16
($)




PU Appliance: Manual Process with Foam Incineration

The technical applicability of this option from year 2005 through 2020 is presented in Table 3-20 and
Table 3-21. This analysis assumes that the incremental maximum market penetration of this option in
2020 in the newly-manufactured appliance foam market would be 30 percent in the United States, and 20
percent in China (see Table 3-22).

Cost and Emission Reduction Analysis

    •   One-time Costs.  This analysis assumes a one-time capital cost of $200,000 (JACO, 2004) for
       establishing offices, renting some equipment, leasing land for collection, etc.


                                                                                           -84-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •  Annual  Costs.   The capacity of the  facility is 10,000 refrigerators per year (JACO, 2004);
       therefore, the annual costs for this technology are  $388,292, calculated by  multiplying the
       capacity of the facility with the net costs of this technology per unit.  Table 3-18 summarizes
       costs associated with this specific technique.
    •  Cost Savings. The savings  from salvaged steel ($5.29/unit) are accounted for in the Annual
       Costs, above.
    •  Emission Reductions.  This analysis estimates that by treating the appliance foam at the end of
       life with incineration, 0.0029 MMTCE  (i.e., 0.288 TCE/unit x 10,000 units) of the high GWP gas
       can potentially be eliminated annually. Accounting for the assumption that 92.5 percent of the
       foam can be recovered from the appliance, and that incineration destroys 98 percent of the HFC
       in that foam, the actual annual emissions abated are 0.0026 MMTCE  (UNEP, 2000b).
Table 3-18: Manual Process with Foam Incineration
Method
Collection/Consolidation
Transportation to Disassembly Location
Disassembly & Processing of Refrigerators
Transportation to Disposal Location
Incineration
Total Costs
Savings (salvaged steel)
Net Costs
Cost per Unit
$0.62
$2.58
$35.00
$0.47
$11.11
$49.78
$5.29
$44.49
($)




3.3.3  Summary of Technical Applicability and Market Penetration of Abatement
       Options

Table 3-19 presents a summary of the assumed reduction efficiency, while tables Table 3-20 and Table
3-21 show the technical applicability of the abatement options. Technical applicability values are based
on the percent of total foam emissions from each end-use, and are derived from the baseline emissions
methodology described in Section 3.2.1.  The commercial technology options explored in this chapter are
assumed to penetrate only new (not existing) equipment, where "new" equipment is defined as equipment
manufactured in 2005 or later.

Table 3-19: Reduction Efficiency of Foam Options (Percent)	
 Option                                                          Reduction Efficiency
 Appliance: HFC-134a to HC                                                 100
 Appliance: HFC-245fa and HFC 365mfc/HCF-227ea to HC                          100
 Appliance: Automated Process with Foam Grinding, Landfilling                        90
 Appliance: Manual Process with Incineration                                      91
 Spray: HFC-245fa/C02 (water) and HFC-365mfc/HFC-227ea to HC                    100
 Spray: H FC-245fa/C02 (water) and H FC-365mfc/H FC-227ea  to C02 (water)             100
 XPS:HFC-134a/C02toC02                                                 100
 XPS: HFC-134a/C02 to C02/Alcohol                                           100
 PU One Component: HFC-134a to HC                                         100
 PU One Component: HFC-152a to HC                                         100
 PU Panels: HFC to HC                                                    100
                                                                                             -85-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 3-20: Technical Applicability of Foam Options for the US, Europe, and Japan (Percent)
Reduction Option



Appliance: HFC-134atoHC
Appliance: HFC-245faand HFC
365mfc/HCF-227ea to HC
Appliance: Automated Process with Foam
Grinding, Landfilling
Appliance: Manual Process with Incineration
Spray: HFC-245fa/C02 (water) and HFC-
365mfc/HFC-227ea to HC
Spray: HFC-245fa/C02 (water) and HFC-
365mfc/HFC-227ea to C02 (water)
XPS:HFC-134a/C02toC02
XPS: HFC-134a/C02 to C02/ Alcohol
PU One Component: HFC-134a to HC
PU One Component: HFC-152a to HC
PU Panels: HFC to HC
2005


(/)
3
26
29
29
51
51
0
0
0
0
0
SL
£
3
Ld
0
4
4
4
4
4
36
36
9
0
0
c
s.
to
— 1
0
2
2
2
13
13
45
45
1
0
0
2010


(/)
2
10
12
12
26
26
53
53
1
0
0
SL
£
3
Ld
0
4
4
4
6
6
36
36
3
0
0
c
s.
to
— 1
0
3
3
3
16
16
47
47
1
0
0
2015


(/)
6
8
14
14
25
25
54
54
0
0
0
SL
£
3
Ld
0
4
4
4
6
6
35
35
2
0
0
c
s.
to
— 1
0
3
3
3
17
17
49
49
1
0
0
2020


(/)
4
40
44
44
16
16
34
34
0
0
0
SL =
2 S.
5 3
0 0
18 11
18 11
18 11
4 17
4 17
21 46
21 46
1 1
0 0
0 0
*Assumed technical applicability of options is presented as a percentage of total foam sector baseline emissions.
                                                                                                          -86-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 3-21: Technical Applicability of Foam Options for the CEITs, China, and the Rest of the Developed
World (Percent)
Reduction Option



Appliance: HFC-134a to HC
Appliance: HFC-245fa and HFC
365mfc/HCF-227ea to HC
Appliance: Automated Process with
Foam Grinding, Landfilling
Appliance: Manual Process with
Incineration
Spray: HFC-245fa/C02 (water) and
HFC-365mfc/HFC-227ea to HC
Spray: HFC-245fa/C02 (water) and
HFC-365mfc/HFC-227ea to C02
(water)
XPS:HFC-134a/C02toC02
XPS: HFC-134a/C02 to C02/ Alcohol
PU One Component: HFC-134a to
HC
PU One Component: HFC-152a to
HC
PU Panels: HFC to HC
2005
H? C 0
LU !c O
O O Q£
0
0
0
0
0

0

0
0
100
0
0
0
0
0
0
0

0

0
0
100
0
0
0
7
7
7
6

6

0
0
0
0
0
2010
H? C 0
LLJ !c O
O O Q£
0
0
0
0
0

0

6
6
94
0
0
0
0
0
0
0

0

0
0
100
0
0
0
9
9
9
9

9

0
0
0
0
0
2015
H? C 0
LLJ i O
O O Q£
0
0
0
0
0

0

15
15
85
0
0
0
0
0
0
0

0

0
0
100
0
0
0
9
9
9
10

10

0
0
0
0
0
2020
H? C 0
LU !c O
O O Q£
0
0
0
0
0

0

17
17
83
0
0
0
0
0
0
0

0

0
0
100
0
0
0
18
18
18
7

7

0
0
0
0
0
A summary  of the  incremental  maximum market penetrations assumed for the abatement options
considered are presented Table 3-22 and Table 3-23.
                                                                                             -87-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 3-22: Incremental Maximum Market Penetration Expressed as Percent of New Emissions for which the
Options are Technically Applicable
Reduction Option








Appliance: HFC-134a to HC
Appliance: HFC-245fa and HFC
365mfc/HCF-227ea to HC
Appliance: Automated Process with Foam
Grinding, Landfilling
Appliance: Manual Process with Incineration
Spray: HFC-245fa/C02 (water) and HFC-
365mfc/HFC-227ea to HC
Spray: HFC-245fa/C02 (water) and HFC-
365mfc/HFC-227ea to C02 (water)
XPS:HFC-134a/C02toC02
XPS: HFC-134a/C02 to C02/ Alcohol
PU One Component: HFC-134a to HC
PU One Component: HFC-152a to HC
PU Panels: HFC to HC
2005






00
=
0
0
0
0
5
0
0
0
50
50
50


•o
c
to
a> ^

= O
1 1 1 rv*
65
65
0
0
0
5
25
0
60
60
60
2010






00
=
25
15
0
0
10
5
0
0
70
70
70
X
a>
c i
to <
a> ^
a. o
2 V
3 a>

85
85
0
0
5
8
35
0
90
90
90
2015






00
=
50
30
5
20
20
10
0
0
90
90
90
X
a>
c i
to <
a> ^
a. o
2 V
3 a>
LU a
90
90
90
0
10
15
35
0
95
95
95





to
IE
o
90
90
0
10
10
15
35
35
95
95
95
2020






00
=
70
50
10
30
30
20
0
0
100
100
100
X
a>
-a =
to <
a> ^
a. o
2 V
3 a>
LU o:
90
90
95
0
15
20
45
0
100
100
100





c
to
Q.
CO

90
90
95
0
15
20
45
0
100
100
100





to
IE
o
90
90
0
20
15
20
45
45
100
100
100
RODW: Rest of Developed World

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 3-23: Incremental Maximum Market Penetration Expressed as Percent of All Emissions for which the
Options are Technically Applicable
Reduction Option

Appliance: HFC-134ato
HC
Appliance: HFC-245faand
HFC 365mfc/HCF-227ea
toHC
Appliance: Automated
Process with Foam
Grinding, Landfilling
Appliance: Manual
Process with Incineration
Spray: HFC-245fa/C02
(water) and HFC-
365mfc/HFC-227ea to HC
Spray: HFC-245fa/C02
(water) and HFC-
365mfc/HFC-227ea to
C02 (water)
XPS:HFC-134a/C02to
C02
XPS:HFC-134a/C02to
C02/ Alcohol
PU One Component:
HFC-134atoHC
PU One Component:
HFC-152atoHC
PU Panels: HFC to HC
2005
00
=>
0

0


0

0

4


0

0
0
0
0
40
•o
to
111
3 a> c
L±J Qi <
47

54


0

0

0


4

0
0
0
0
48
§.
to
— 1
47

54


0

0

0


4

0
0
0
0
48
2010
SL
<* 1
=> LU
23 64

13 65


0 0

0 0

9 4


4 7

0 52
0 0
70 90
70 90
46 66
c *o x
lisi
-^ at <
64 64

59 58


0 0

0 0

4 3


6 5

51 30
0 0
90 90
90 90
59 59
2015
SL
<* 1
ID LU
12 16

24 68


2 28

6 0

18 8


9 13

0 57
0 0
90 95
90 95
51 67
c *o x
lisi
-^ at <
16 16

57 53


28 18

0 2

8 5


12 8

58 22
0 0
95 95
95 95
53 53
2020
00
=>
12

4


9

27

31


19

0
0
100
100
51
SL
|
LiJ
11

8


84

0

15


21

73
0
100
100
74
§.
to
— 1
11

6


84

0

13


18

76
0
100
100
50
to
c
!c
o
11

5


0

18

7


10

23
23
100
100
50
To calculate the percent of emission reductions off the total foams baseline for each abatement option, the
percent of baseline emissions from Table 3-20 and Table 3-21  is multiplied by the market penetration
values from table Table 3-23 and reduction efficiencies from Table 3-19. For example, to determine the
percent reduction off the 2020 baseline  for  end-of life option in appliance:  manual process with
incineration in the United States, the following calculation is used:

     Technical Applicability x Incremental Maximum Market Penetration x Reduction Efficiency =

                                     44%x27%x 91% ~ 11%

Thus, using the assumptions in this analysis, applying this end-of-life option could reduce U.S. baseline
emissions by approximately  11  percent in 2020.  This figure, along  with the other emission reduction
potentials, is shown in Table 3-24.
                                                                                              -89-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 3-24: Emission Reductions off Total Foams
Reduction Option




Appliance: HFC-134atoHC
Appliance: HFC-245fa and HFC 365mfc/HCF-
227ea to HC
Appliance: Automated Process with Foam
Grinding, Landfilling
Appliance: Manual Process with Incineration
Spray: HFC-245fa/C02 (water) and HFC-
365mfc/HFC-227ea to HC
Spray: HFC-245fa/C02 (water) and HFC-
365mfc/HFC-227ea to C02 (water)
XPS:HFC-134a/C02toC02
XPS: HFC-134a/C02 to C02/ Alcohol
PU One Component: HFC-134a to HC
PU One Component: HFC-152a to HC
PU Panels: HFC to HC
Total
Baseline for
2005


^0
ID
0

0

0
0

2

0
0
0
0
0
0
2
SL

3
LU
0

2

0
0

0

0
0
0
0
0
0
2
c
SL
to
— 1
0

1

0
0

0

1
0
0
0
0
0
1
the US,
Euro
2010


w>
ID
0

1

0
0

2

1
0
0
0
0
0
6
SL

3
LU
0

2

0
0

0

0
19
0
2
0
0
24
c
SL
to
— 1
0

2

0
0

1

1
24
0
1
0
0
28
pe, and
Japan (Percent)
2015
SL e
^ LU -^
1

2

0
1

4

2
0
0
0
0
0
11
0 0

2 1

1 1
0 0

1 1

1 2
20 29
0 0
2 1
0 0
0 0
27 35


2020


^0
ID
1

2

4
11

5

3
0
0
0
0
0
25
SL

3
LU
0

1

13
0

1

1
15
0
1
0
0
32
c
SL
to

0

1

9
0

2

3
35
0
1
0
0
50
Table 3-25: Emission Reductions off Total Foams Baseline for the CEITs, China, and the Rest of the
Developed World (Percent)
Reduction Option




Appliance: HFC-134atoHC
Appliance: HFC-245fa and HFC
365mfc/HCF-227ea to HC
Appliance: Automated Process with Foam
Grinding, Landfilling
Appliance: Manual Process with Incineration
Spray: HFC-245fa/C02 (water) and HFC-
365mfc/HFC-227ea to HC
Spray: HFC-245fa/C02 (water) and HFC-
365mfc/HFC-227ea to C02 (water)
XPS:HFC-134a/C02toC02
XPS: HFC-134a/C02 to C02/ Alcohol
PU One Component: HFC-134a to HC
PU One Component: HFC-152a to HC
PU Panels: HFC to HC
Total
2005

^~
LU
O
0
0
0
0
0
o

0
0
0
0
0
0
to
c
!c
o
0
0
0
0
0
o

0
0
0
0
0
0
*° 2
t/! c
a> c
at <
0
4
0
0
0
o

0
0
0
0
0
4
2010

^~
LU
O
0
0
0
0
0
o

2
0
85
0
0
87
to
c
!c
O
0
0
0
0
0
o

0
0
90
0
0
90
*° 2
t/! c
a> c
at <
0
5
0
0
0
o

0
0
0
0
0
6
2015

^~
LU
O
0
0
0
0
0
o

3
0
81
0
0
84
to 2 5
•— (75 c
.c a> c
o at <
0 0
0 5
0 1
0 0
0 1
0 1

0 0
0 0
95 0
0 0
0 0
95 8
2020

^~
LU
O
0
0
0
0
0
o

4
0
83
0
0
87

c
!c
o
0
0
0
0
0
o

0
0
100
0
0
100
*° 2
t/5 c
a> c

0
1
10
1
0
1

0
0
0
0
0
14
3.4
Results
Table 3-26 through Table 3-28 provide a summary of the potential emission reduction opportunities and
associated costs for United States, Non-U.S. Annex I and Non-Annex I countries in 2020.  The costs to
reduce one ton of carbon equivalent (TCE) data are presented for two different discount rate scenarios: 4
                                                                                            -90-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
percent and 20 percent. The tax
respectively.
Table 3-26: Emission Reductions in
rate associated

2020 and Costs
with each

cost scenario is 0 percent and 40 percent,

of Abatement for Foams
Break-Even Price
(2000$/TCE)
Discount /Tax Rate
Reduction Option
Spray HFC-245fa/C02 and HFC-
365mfc/HFC-227eatoHC
PU One Component HFC-152a to HC
XPS: HFC-134a/C02 to C02/Alcohol
PU One Component HFC-134a to HC
XPS:HFC-134a/C02toC02
Appliance HFC-134a to HC
Appliance: Automated Process with
Foam Grinding, HFC Adsorption, and
Foam Landfilling
PU Continuous and Discontinuous HFC
toHC
Spray HFC-245fa/C02 and HFC-
365mfc/HFC-227ea to C02
Appliance: Manual Process with Foam
Incineration
Appliance HFC-245fa and HFC
365mfc/HCF-227ea to HC
Table 3-27: Emission Reductions in
4%/0%
$(17.97)
$(9.40)
$(8.50)
$(0.07)
$11.48
$17.35


$23.82

$43.83

$96.45

$175.09

$201.80
2020 and Costs
20%/40%
$(13.98)
$2.77
$12.82
$14.08
$122.09
$404.45


$37.16

$56.26

$122.55

$191.65

$1,115.58
Emission
Reduction of
Option
(MMTCE)
0.37
0.0
0.0
0.02
0.0
0.04


0.26

0.0

0.22

0.78

0.12
of Abatement for Foams

in the US
Reduction off
Baseline
(Percent)

5.1%
0.0%
0.0%
0.2%
0.0%
0.5%


3.5%

0.0%

3.0%

10.7%

1.6%
in the non-US


Cumulative
Reductions
(MMTCE)

0.37
0.37
0.37
0.39
0.39
0.43


0.69

0.69

0.91

1.69

1.81


Cumulative %
Reduction from
2020 Baseline

5.1%
5.1%
5.1%
5.3%
5.3%
5.8%


9.4%

9.4%

12.4%

23.0%

24.6%
Annex I Countries
                                    Break-Even Price
                                      (2000$/TCE)
                                    Discount Rate/Tax
                                         Rate
   Break-Even Price    Emission  Reduction Cumulative Cumulative
     (2000$/TCE)

Discount Rate/Tax Rate
Reduction    off    Reductions    %
of Option  Baseline  (MMTCE)  Reduction
(MMTCE)  (Percent)            from 2020
Reduction Option
Spray HFC-245fa/C02 and HFC-
365mfc/HFC-227ea to HC
PU One Component HFC-152a to HC
XPS: HFC-134a/C02 to C02/Alcohol
PU One Component HFC-134a to HC
XPS:HFC-134a/C02toC02
ApplianceHFC-134atoHC
Appliance: Automated Process with
Foam Grinding, HFC Adsorption, and
Foam Landfilling
PU Continuous and Discontinuous
HFC to HC
Spray HFC-245fa/C02 and HFC-
365mfc/HFC-227ea to C02
Appliance: Manual Process with Foam
Incineration
4%/0%
Low High
$(17.97)
$(9.40)
$(8.50)
$(0.07)
$11.48
$17.35


$23.82

$43.83

$96.45

$175.09
$(15.13)
$(9.40)
$(8.50)
$(0.07)
$11.48
$17.35


$23.82

$43.83

$96.45

$175.09
20% /40% Baseline
Low High
$ (13.98)
$2.77
$12.82
$14.08
$122.09
$404.45


$37.16

$56.26

$122.55

$191.65
$1.53
$2.77
$12.82
$14.08
$122.09
$404.45


$37.16

$56.26

$122.55

$191.65
0.09
0.0
0.0
0.11
1.70
0.0


1.08

0.0

0.12

0.01
0.9%
0.0%
0.0%
1.2%
18.9%
0.0%


11.9%

0.0%

1.4%

0.1%
0.09
0.09
0.09
0.19
1.89
1.89


2.97

2.97

3.09

3.10
0.9%
0.9%
0.9%
2.1%
21.0%
21.0%


32.9%

32.9%

34.3%

34.4%
                                                                                                     -91-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Reduction Option
                      Break-Even Price
                        (2000$/TCE)
                      Discount Rate/Tax
                           Rate
                          4%/0%
                       Low     High
                                                   Break-Even Price    Emission Reduction Cumulative Cumulative
     (2000$/TCE)     Reduction    off
                   of Option  Baseline
Discount Rate/Tax Rate (MMTCE)  (percent)
     20%/40%
   Low      High
                                                                                      Reductions    %
                                                                                       (MMTCE)  Reduction
                                                                                                from 2020
                                                                                                Baseline
 Appliance HFC-245fa and HFC
 365mfc/HCF-227ea to HC
                     $201.80   $201.80  $1,115.58  $1,115.58    0.11
                                        3.20
35.6%
} Costs vary by country/region based on one-time or annual adjustment factors; therefore, the lowest and highest costs for the region are hown.
Table 3-28: Emission Reductions in 2020 and Costs of Abatement for Foams in non-Annex I Countries
Reduction Option
Spray HFC-245fa/C02 and HFC-
365mfc/HFC-227ea to HC
PU One Component HFC to HC
XPS: HFC-134a/C02 to C02/Alcohol
PU One Component HFC-134a to HC
XPS:HFC-134a/C02toC02
Appliance HFC-134a to HC
Appliance: Automated Process with Foam
Grinding, HFC Adsorption, and Foam
Landfilling
PU Continuous and Discontinuous HFC-
134atoHC
Spray HFC-245fa/C02 and HFC-
365mfc/HFC-227eatoC02
Appliance: Manual Process with Foam
Incineration
Appliance HFC-245fa and HFC
365mfc/HCF-227ea to HC
* Break-even costs do not vary by country within the
Break-Even Price Emission
(2000$/TCE) Reduction of
Discount /Tax Rate Option
4%/0% 20%/40% (MMTCE)
$(17.97)
$(9.40)
$(8.50)
$(0.07)
$11.48
$17.35
$23.82
$43.83
$96.45
$175.09
$201.80
Non Annex
$(13.98)
$2.77
$12.82
$14.08
$122.09
$404.45
$37.16
$56.26
$122.55
$191.65
$1,115.58
I region.
0.0
0.0
0.0
0.01
0.0
0.0
0.0
0.0
0.0
0.0
0.0

Reduction off
Baseline
(Percent)
0.0%
0.0%
0.0%
100.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%

Cumulative
Reductions
(MMTCE)
0.0
0.0
0.0
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01

Cumulative %
Reduction from
2020 Baseline
0.0%
0.0%
0.0%
100.0%
100.0%
100.0%
100.0%
100.0%
100.0%
100.0%
100.0%

3.5
Summary
Baseline emissions of HFCs from foams are estimated to grow from 2.94 to  16.38 MMTCE between
2005 and 2020. In 2020, Non-U.S. Annex I countries are assumed to account for about 55 percent of the
emissions, while  U.S. emissions  are assumed to account for 45 percent of this total.  Non-Annex I
countries contribute less than one percent to 2020 foam emissions. Similar emissions growth is expected
in the  United States (from 0.53 MMTCE in 2005 to 7.35 MMTCE  in 2020) and Non-U.S. Annex I
countries (from 2.40 MMTCE in 2005 to 9.01 MMTCE in 2020).

This analysis considers  nine  replacement emissions mitigation options for  spray,  appliance, XPS,
continuous panel, discontinuous  panel and  one-component  foams,  and two end-of-life  options for
appliance foams:

    •    Replacing HFC-134a,  HFC-245fa,  and  HFC-365mfc/HFC-227ea  with Hydrocarbons  in
        continuous and discontinuous panel foam;
                                                                                             -92-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •   Replacing HFC-134a with Hydrocarbons in one component foam;
    •   Replacing HFC-152a with Hydrocarbons in one component foam;
    •   Replacing HFC-134a/CO2 (LCD) with CO2 (LCD) in XPS foam;
    •   Replacing HFC-134a/CO2 (LCD) with CO2 (LCD)/Alcohol in XPS foam;
    •   Replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with CO2 (water) in spray foam;
    •   Replacing HFC-245fa/CO2 (water) and HFC-365mfc/HFC-227ea with Hydrocarbons  in spray
       foam;

    •   Replacing HFC-134a with Hydrocarbons in appliance foam;
    •   Replacing HFC-245fa and HFC 365mfc/HCF-227ea with Hydrocarbons in appliance foam;
    •   End of Life Appliance Foam Practice: Automated Process with Foam Grinding, HFC Adsorption,
       and Foam Landfilling in appliance foam; and

    •   End of Life Appliance Foam Practice: Manual Process with Foam Incineration in appliance foam.
The emission reduction benefits of each option were  compared in each region. For spray end-uses, the
costs associated with converting to alternative  blowing agent differ between the United States, Europe,
and Japan.  The costs per TCE of all other abatement options for these three regions are equivalent
because available data on costs for abatement technologies were not scaled to reflect potential differences
in the costs internationally. Additional research may  be required to determine actual variability in costs
across regions. This analysis shows that there are variety of options available at or below 25 $/TCE at 4
percent discount rate and 0 percent tax rate that may be used to eliminate the use of HFCs and reduce
HFC associated emissions from foams.


3.6        References

ADL. 1999. Global  Comparative Analysis of HFC and Alternative Technologies for Refrigeration, Air
Conditioning, Foam, Solvent, Aerosol Propellant, and Fire Protection Applications. Final Report to the
Alliance for Responsible Atmospheric Policy. Reference Number 49648. Arthur D. Little, Inc.

Alliance.  2000.  Comments of the Alliance for  Responsible  Atmospheric Policy on Draft  of "Cost and
Emission Reduction Analysis  of HFC Emissions from Foams in the United States." Fax sent from
Alliance to ICF Consulting on May  26, 2000.

Alliance.  2001.  Review of EPA  Draft Chapter 9  by Members  of the  Alliance for  Responsible
Atmospheric Policy. Alliance. May  16, 2001.

Ashford,  Paul. 2004. Personal  Communication between ICF  Consulting and Paul Ashford of Caleb
Group. April 8, 2004.

Atofina. 2004. Personal Communication  between Dom Loconte of Atofina  and Bob  Russell of ICF
Consulting. March 2004.

Caleb. 2001. Overall Review of the Challenges Facing the Polyurethane Spray Foam Industry and Other
Systems House Based Applications in the Light of Proposed Rulemaking [65 Fed. Reg. 42543,  July 11th
2000]. Caleb Management Services Limited. 2001.
                                                                                         -93-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
BRE. 2004.  European Reaction to Fire  Tests.  Proposal number  215905. Prepared by Paul Ashford,
Promat UK Ltd. 21st January 2004.

Bureau  of Labor  Statistics.  2004.  U.S.  Department of  Labor.  1999 National  Industry-Specific
Occupational Employment and Wage Estimates. SIC 421 - Trucking and Courier Services, Except Air.
Retrieved: 06/11/2004. 

Cannon. 2001. Personal Communication between Dick Werner of Cannon and ICF Consulting.

EIA. 2001. International Energy Outlook, Table 7. Comparison of Economic Growth Rates by Region,
1997-2020. United States Department of Energy, Energy Information Agency.

Exxon. 2001. Personal Communication between Bob Begbie of Exxon and ICF Consulting.

Exxon.  2004.  Personal Communication  between Ken Hobley of Exxon and  Bob  Russell of ICF
Consulting. October 2, 2003.

Honeywell. 2000. Comments of Honeywell Inc.  on U.S. Environmental Protection Agency Proposed
Listing of Certain HCFCs  and Blends as "Unacceptable" Substitutes for HCFC-141b - 65 Fed. Reg.
42653 (July  11, 2000). Personal  Communication  from  Richard Ayres of Howrey, Simon, Arnold, and
White to Anhar Karimjee of EPA on September 11, 2000. Available from EPA's Foams Docket A-200-
18, Document IV-D-41.

Honeywell. 2003. Personal Communication  between Carol Bib of Honeywell and Bob Russell of ICF
Consulting. October 2, 2003.

ICF Consulting. 2003. Evaluation of the  Energy and Environmental Effects of the California Appliance
Early Retirement and Recycling Program. Prepared for: California Public Utilities Commission and the
Appliance Recycling Centers of America. January 17, 2003.

ICF Consulting. 2004. Personal communication with Bob Russell, ICF Consulting, March 2004.

JACO. 2004. Personal Communication between Michael Dunham, Director, Energy  & Environmental
Programs, JACO Environmental, Inc. and Colm Kenny of EPA. May 13, 2004.

JUFMA. 2004. Personal Communication with Mr. Kara Kiyoshi of JUFMA (Japanese Urethane Foam
Manufacturer Association) on April 8, 2004.

NCFI. (North Carolina Foam Industry). 2001. Personal Communication between Gary Maechtle of NCFI
and ICF Consulting.

OC Landfills. 2002. "Gatefees." Retrieved January 7, 2003: 

Scheutz et al.  2002. Charlotte Scheutz  and Peter Kjeldsen.   2002.  Determination of the Fraction of
Blowing Agent  Released from  Refrigerator/Freezer  Foam  After  Decommissioning the Product.
Environment and Resources DTU, Technical  University of Denmark. April 2002.

Scheutz, et al.  2003. Charlotte Scheutz and  Peter Kjeldsen.  2003. Attenuation of Alternative Blowing
Agents in Landfills. Environment and Resources DTU, Technical University of Denmark. August 2003.


                                                                                        -94-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
SPFA (Spray Polyurethane Foam Alliance). 2001. Personal Communication between Mason Knowles of
SPF and ICF Consulting.

Stepan. 2001. Personal Communication between Rick Taber of Stepan and ICF Consulting. 2001.

Systems Transportation Equipment, 2004. Vehicle Selection and Specification. Retrieved May 2004:


UNEP (United Nations  Environment Programme). 1998.  1998 Report of the Flexible and Rigid Foams
Technical Options Committee. United Nations Environment Programme.

UNEP (United Nations Environment  Programme). 2002a. Report of the Technology and Economic
Assessment Panel. Progress Report. Montreal Protocol on the Substances that deplete the Ozone Layer.

UNEP (United Nations Environment  Programme), 2002b. Report of the Technology and Economic
Assessment Panel of the Montreal Protocol, Task Force on Destruction Technologies,  Volume 3b, April
2002.

USDA 2002. Real GDP (2000 dollars) Historical. International Macroeconomic  Data Set.  Available
online at: .

Whirlpool. 2004. Personal Communication between Robert W Johnson of Whirlpool and Colm Kenny of
EPA. April 19, 2004.

Williams et al. 1999. Williams, D.J., M.C. Bogdan, and  P.B. Logsdon.  1999.  Optimizing Performance
and Value: HFC-245fa and Blends of HFC-245fa for Insulating Foams. Conference Proceedings from the
Earth Technologies Forum TF 1999, pg 290-302.
                                                                                         -95-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
4   Analysis of Costs to Abate International HFC and PFC
     Emissions from Fire Extinguishing
4.1        Introduction


The principal greenhouse gases used in and potentially emitted from the fire extinguishing  sector are
hydrofluorocarbons (HFC-227ea, HFC-236fa, HFC-23), and blends containing perfluoromethane (CF4).
These gases have  100-year global warming potentials (GWPs) that range from 2,900 to  11,700 (IPCC,
1996).

These high GWP gases are substitutes for halons, ozone depleting substances (ODSs) that have been, and
in many countries are  still, widely used in fire-extinguishing applications.   Although halons were
produced in much lower volumes than other ODSs, they have extremely high ozone depletion potentials
(ODPs) due to the presence of bromine, which reacts more strongly with ozone than chlorine.  Halons
have been historically used in fire suppression and explosion protection applications because they are
electrically non-conductive, dissipate rapidly without residue, are safe for limited human exposure, and
are extremely efficient in extinguishing most types of fires (U.S. EPA, 1994).

Halon applications can  be divided into two categories: (1) portable fire extinguishers (e.g.,  streaming
applications) that originally used halon 1211, and (2) total flooding applications that originally used halon
1301 or halon 2402 (U.S. EPA, 2004; March Consulting Group, 1998 and  1999).  Historically,  SF6,
another high-GWP gas,  was used in select fire extinguishing systems uses, such as for system discharge
testing purposes by the U.S. Navy.  For the most part, however, SF6 is no longer used in any capacity in
the fire protection sector.

Portable  fire  extinguishers are most  frequently  used in  offices, manufacturing and retail facilities,
aerospace/marine applications,  and homes. Market penetration of HFCs in this sector has been limited,
and is unlikely to grow or even keep apace with the growth in portable extinguishers (Wickham, 2003a).
PFCs have had a very small penetration in the portable fire extinguisher market.  By 2020, only one HFC,
HFC-23 6fa, is expected to be used to  a limited extent as a halon replacement  in small segments of the
portable extinguishing sector.   Overall, portable applications represent a much smaller share of total fire
extinguishing emissions than do total flooding applications, and the U.S. EPA projects that their relative
share of emissions will decrease overtime, based on cost reasons outlined in Wickham (2002).

The majority of HFC emissions associated with fire extinguishing come from its use as a replacement for
some halon 1301 applications  in the total flooding market.  Total flooding systems are usually used to
protect a variety of spaces, including:

    •   Electronic and telecommunications equipment, such as tape storage areas, computer facilities,
       telecommunications gear, medical facilities, control rooms in nuclear power plants, and air traffic
       control towers;
    •   Military  applications,  including  aviation  engine nacelles38  and  dry  bays,  naval engine
       compartments, and engine compartments and occupied crew spaces of ground combat vehicles;
38 Nacelles are enclosed engine housings.
                                                                                         -96-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •   Oil production facilities;
    •   Flammable liquid storage areas;
    •   Engine nacelles and cargo bays of commercial aircraft;
    •   Cultural institutions and museums;
    •   Records storage areas;
    •   Bank vaults;
    •   Warehouses; and
    •   Special facilities, such as research laboratories and military facilities.
Other than in Russia, which used halon 2402, halon 1301 was widely used in total flooding applications
because of its unique features (Wickham, 2002). Halon 1301 is a clean agent, meaning that it does not
leave residue on equipment or in the protection enclosure after discharge. In addition, halon 1301 is safe
for limited,  acute human  exposure at the  concentration used for fire extinguishing.  It is also very
effective at extinguishing fires and works well over a broad temperature range.  Because halon 1301 was
economical and its  design and  installation  was relatively simple compared to other fire extinguishing
systems, these systems reached almost all segments of the total flooding fire extinguishing market.

The alternatives to halon 1301 in total flooding applications can be categorized as in-kind, gaseous agent
alternatives (i.e., halocarbons, CO2, inert gases, fluorinated  ketones) and  not-in-kind  alternatives (i.e.,
dispersed and condensed aerosol extinguishing systems, water sprinklers,  water mist,  foam,39 inert gas
generators).  In most Annex I countries, halocarbon HFC-227ea has emerged as the primary replacement
for halon  1301 in total flooding applications.  Other HFCs, such as HFC-23, HFC-236fa, and HFC-125,
as well as PFCs, such as C3F8 and C4Fi0, have been evaluated and determined to be safe for limited, acute
human exposure, but are used in smaller amounts as a result of environmental,40 technical, and economic
concerns. Based on confidential information collected for this report from members of UNEP's Halon
Technical Options Committee (HTOC), this analysis  projects that the  market share of HFC-23 will
increase in Russia as a total flooding agent over the next 20 years, although other information  (e.g.,
Wickham, 2003b) might point to a different conclusion.  Use of HFC-125, which has been limited to
normally unoccupied specialty applications (e.g., aviation engine nacelles), is also expected to increase in
Russia over time. A small number of telecommunications facilities use PFCs, with some of the highest
use of these chemicals being in Eastern Europe.  In the United States, PFC use in fire suppression will tail
off, as the U.S. manufacturer of PFCs  for fire suppression withdrew these agents from the market a
number of years  ago because  of concern about their high global warming potential.  In  addition,
hydrochlorofluorocarbons  (HCFCs)  have historically  also  been used  as halon  1301   replacements,
particularly in Eastern and Southern Europe.  Over time, the use of HCFCs and PFCs in total flooding
applications is expected to  be  phased  out, and replaced primarily with HFCs, in addition to  other
alternatives.

In particular, other available  in-kind,  non-halocarbon alternatives in total flooding applications include
carbon dioxide systems, used primarily in marine  and industrial applications,  fluorinated ketones, and
inert gas  systems, which contain nitrogen or argon or blends of these gases, sometimes incorporating
carbon  dioxide as  a third component.  Inert gas systems have become the  dominant halon  1301
replacement in many parts of Europe, most notably in northern European countries.
39 Foams can be protein-based or synthetic-based. It should be noted that some synthetic-based foams contain
fluorocarbons.
40 These gases have high GWPs, ranging from 2,800 to 11,700.


                                                                                             -97-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Not-in-kind (NIK) alternatives and technologies available include powdered aerosols, water sprinklers,
water mist systems, foams, and combinations of these systems, such as aerosols with a halocarbon agent,
or water mist with a gaseous agent or with foam.


4.2        Baseline Emission Estimates


4.2.1  Emission  Estimating Methodology

Description of Methodology

A full description of the emission model  used to calculate ODS substitute emissions from all sectors is
provided in Appendix A.  Specific information on how the model calculates fire protection emissions is
described below.

U.S. EPA uses a detailed Vintaging Model of ODS-containing equipment and products to estimate the use
and emissions of various ODS substitutes in the United States,  including HFCs and PFCs.  Emission
estimates for non-U.S. countries are derived  using country-specific ODS  consumption estimates as
reported  under the  Montreal  Protocol  in  conjunction with Vintaging Model  output for the  fire
extinguishing sector.

This analysis first incorporates estimates  of the consumption of ODSs by country, as provided by the
United Nations Environment Programme  (UNEP,  1999). Estimates for the European Union (EU) were
provided in aggregate  and GDP was used as a proxy to distribute  consumption among  the individual
member nations.

Emission Equations

This analysis assumes that total emissions from leakage, accidental discharges, and fire extinguishing, in
aggregate, equals  a  percentage of the total quantity of chemical in operation at a given time.  For
modeling purposes, fire extinguishing agent is  assumed to be released at  a constant rate for an average
equipment lifetime. This percentage varies for streaming and flooding equipment.

                             EJ< = r x Z QCj.j+1  fori=1—>k

Where:

E  =  Emissions. Total emissions of a specific chemical in year j for fire  extinguishing equipment, by
       weight.
r   =  Percent Released. The percentage of the total chemical in operation that is emitted to the
       atmosphere.
Qc =  Quantity of Chemical. Total amount of a specific chemical used in  new fire extinguishing
       equipment in a given year, j-i+1, by weight.
i   =  Counter. From 1 to lifetime  (k).
j   =  Year of emissions.
k   =  Lifetime. The average lifetime of the equipment.

Estimates used for the percent released  and lifetime  of equipment can have  a  significant effect on
resulting  emission estimates.  For this analysis,  the U.S.  Vintaging Model  assumes emission factors (i.e.,
                                                                                          -98-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
variable r in the equation above) of 1.5 percent in the total flooding sector (which lies between the range
cited in Verdonik and Robin (2004)), and 2 percent in the streaming sector. These estimates are chosen to
account,  on  average, for all emissions from  servicing,  leaks,  accidental/false  discharges, system
decommissioning, or intentional discharges to extinguish fires.  The U.S. Vintaging Model also assumes
equipment lifetime for streaming and flooding applications to be 10 and 20 years, respectively.

Regional Variations/A djustments

To estimate baseline emissions, information collected on current and projected market characterizations
of international total flooding sectors was used to create country-specific versions of the Vintaging Model
(i.e., country-specific ODS substitution patterns). Information for Australia, Brazil, China, India, Japan,
Russia, and the United Kingdom was obtained from HTOC members from those countries.41 General
information was  also  collected  on Northern,  Southern,  and  Eastern  Europe.   Baseline  emission
information from some  of these countries was used  to adjust the substitution patterns for all other non-
U.S. countries, as described below:

    •  Eastern Europe: used as a proxy for the countries in the Former Soviet Union and CEITs.
    •  Australia: used as  proxy for New Zealand.
    •  Brazil: used as a proxy for countries in Latin America and the Caribbean.

    •  India:  used as a proxy for all other developing countries.
For all other non-U.S. Annex I countries, the U.S. ODS substitution pattern (presented in Appendix A)
was  used as a proxy.42 In addition, an adjustment factor was applied to EU  countries to account for
European Regulation 2037/2000 on Substances that Deplete  the  Ozone  Layer, which mandated  the
decommissioning of all halon systems and extinguishers in the EU by the end of 2003 (with the exception
of those applications that  are defined as critical uses) (Europa, 2003).  To reflect this, the methodology
assumes that all halon systems in the EU will be decommissioned by 2004.43


4.2.2  Baseline Emissions

The resulting baseline estimates of GWP-weighted HFC and PFC emissions developed for this report are
summarized in Table 4-1.  The estimates of the global total flooding fire protection market developed for
this report are consistent with those in  the IPCC/TEAP (1999) report, which estimated that in the late
1990s, between 20 and 22 percent of systems that would formerly have used halons used HFCs, and that
less than 1 percent used PFCs.
41 Fire protection experts in these countries provided confidential information on the status of national halon
transition markets and average costs to install the substitute extinguishing systems in use (on a per volume of
protected space basis) for 2001 through 2020.
42 This analysis assumes that, of the new total flooding protection systems in which halons have been previously
used in the United States, the market is currently made up of approximately 33 percent HFC-227ea, 1 percent HFC-
23, 14 percent inert gas, and 52 percent other not-in-kind.
43 It should be noted that the use of halon in marine applications is unlikely to meet the 2004 phaseout deadline,
because these applications are  also governed by  regulations issued by the  International Maritime Organization
(IMO), and that many  EU ships still  contain halon  1301 fire suppression systems. However, due to a lack of
available data on emissions from marine-based  fire protection systems as a percentage of the total  EU fire
extinguishing sector, this analysis simply assumes full compliance with the EU regulation.


                                                                                              -99-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 4-1: Baseline HFC and PFC Emission Estimates from Fire Extinguishing (MMTCE)
 Region                  2005          2010          2015          2020
United States
Non-U.S. Annex I
Non-Annex I
0.43
1.51
0.31
0.65
1.00
0.61
0.80
0.87
0.98
0.89
0.94
1.37
 Total	2.26	2.26	2.65	3.19
Note: Totals may not sum due to independent rounding.

This analysis assumes that Class A surface fire hazards represent an estimated 95  percent of the total
flooding sector in all countries, and that the remaining 5  percent of the applications are for Class B
(flammable  liquids and gases) hazards.44  According to projected global  average  emission  estimates,
emissions from Class A fire hazards will account for approximately 71 percent of  the global total fire
extinguishing sector in 2005, 76 percent in 2010, 81 percent in 2015,  and 86 percent in 2020.  Table 4-2
presents the estimated global average breakout of total fire sector HFC and PFC emissions by application,
as estimated by U.S. EPA's Vintaging Model.

Table 4-2: Assumed Breakout of Total GWP-Weighted Baseline Fire Extinguishing Emissions3
                                           Annex I and  Non-Annex I Countries
                                        2005        2010       2015       2020
Flooding
Class A emissions (95% total flooding)
Class B emissions (5% of total flooding)
Streaming
Total
75.0%
71.3%
3.8%
25.0%
100%
80.0%
76.0%
4.0%
20.0%
100%
85.0%
80.8%
4.3%
15.0%
100%
90.0%
85.5%
4.5%
10.0%
100%
a Totals may not sum due to independent rounding.

4.3        Costs of HFC and PFC Emission  Reductions from Fire
        Extinguishing


This section presents a cost analysis for achieving HFC and PFC emission reductions from the baselines
presented in Table 4-1.  Each abatement option is described below, but costs are analyzed for only those
options not assumed to occur in the baseline and for which adequate cost data are available.  All cost
analyses assume  a 20-year project  lifetime.   To the extent possible, this analysis considered total
equivalent  warming impacts (TEWI), to  account for the cost and greenhouse gas emission impacts of
energy  consumption (i.e., indirect emissions) associated with the heating/cooling  of additional  space
needed to house alternative agents.  However, due to data limitations, a full life cycle analysis was not
possible.  For example, the cost and emission impacts associated with manufacturing alternative agents
and all system components were not assessed in this analysis, though they may potentially be significant.
44 Wickham (2002) estimates that over 90 percent of the halon 1301 systems ever installed in the United States were
designed to protect hazards where the anticipated fire type was primarily Class A in nature, and that approximately
10 percent of the U.S. applications served by halon 1301 had hazardous materials of the Class B type. However,
because much of the former halon 1301 Class  B applications have been replaced by non-HFC alternatives (e.g.,
CO2), this analysis assumes that only 5 percent of HFC emissions from the total flooding sector are from Class B
applications, and that the remaining 95 percent are from Class A applications.


                                                                                            -100-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
4.3.1   Description and Cost Analysis of Abatement Options

Because it is estimated that only a limited amount of fluorocarbon agents (e.g., HFC-236fa) will be used
in streaming applications, this cost  analysis focuses only on  abatement options in the total  flooding
sector.45 In 2005, the majority of emissions from the fire extinguishing sector will result from leaks and
discharges (both accidental and intended use to extinguish fires) from total flooding applications.  The
options for reducing HFC and PFC emissions from the fire protection sector include the use of alternative
fire protection agents and the use  of alternative technologies and practices.  Eight potential options are
identified, but only the first three are explored further in the cost analysis:

    •   Inert gas;
    •   Water mist;
    •   Fluorinated ketone (FK-5-1-12);
    •   Carbon dioxide;
    •   Recovery and reuse of HFCs;
    •   Improved detection systems;
    •   Aerosols; and
    •   Inert gas generators.
As described further below, available alternatives to reduce emissions in the fire protection sector may not
be technically or economically viable  for all end use applications.  For example, military applications
often have  very  specialized needs  that  do not exist  in other end use  applications.   In particular,
applications that are space- and/or weight-constrained, such as marine and aviation applications,  are more
limited in their choice  of alternative  agents.   Electronic and telecommunication applications,  which
represent the largest use of HFCs in the total flooding sector, offer the greatest opportunities to  consider
potential alternatives, although some economic penalties and technical challenges may exist.

The  remainder of  this  section provides an overview of each abatement option, and  presents  the
assumptions and results of cost analyses for inert gas, water mist, and fluorinated ketone options. For a
variety of reasons discussed further below, these options are assumed to be applicable only to new (not
existing) total flooding systems, where "new" is defined as systems installed in 2005 or later.

Inert Gas Systems

The  first option considered to displace HFC use is the  use of inert gas systems.   Inert gas systems  use
argon  or nitrogen or blends of these  two  gases, sometimes incorporating carbon dioxide as a third
component to extinguish fires (UNEP,  2001).  Inert gas systems provide an equivalent level of  both fire
protection and life safety/health protection in most Class A (ordinary combustible) fire hazards, including
electronics and telecommunications applications.
45 U.S. EPA estimates that more than 90 percent of the halon replacement market in the streaming sector currently
consists of not-in-kind alternatives, while HFCs account for less than 5 percent of this market.  By 2020, U.S. EPA
projects that HFCs will account for an even smaller portion of the halon replacement market in the streaming sector.
It is expected that the high cost of HFCs will ensure that they are only used where they are absolutely needed (i.e., in
areas where cleanliness is an absolute necessity) (Wickham, 2002).


                                                                                             -101-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Despite their effectiveness at extinguishing a fire, because the discharge time for inert gas systems is on
the order of 60 seconds or more—significantly slower than the discharge time for HFC systems, for
which discharge is typically 10 to 15 seconds (Kucnerowicz-Polak, 2002)—inert gas  systems are not
recommended for  areas where a rapidly developing fire can be expected (UNEP, 2001; Kucnerowicz-
Polak, 2002). Advances in  fire detection devices may help alleviate these concerns by recognizing and
extinguishing fires before they have the opportunity to develop and expand. Another factor impeding the
use of inert  gas systems in lieu of HFCs is that a substantially  larger volume of agent  is needed  to
extinguish fires.  The additional space and  weight needed to accommodate additional steel cylinders
containing inert gas may effectively prohibit the retrofit of many existing HFC systems. For example, the
retrofit of an HFC system to an inert gas system on small ships may be virtually impossible. The same
may hold true for many other applications for which the system infrastructure is fixed. Another factor to
consider with  this option  is  the  need  to  heat  and cool  this additional  space, leading  to  negative
implications for both cost and energy consumption.

This analysis assumes this option is technically applicable46 to the emissions from total flooding systems
designed for Class A fires. In addition, because of the additional floor space requirements associated with
inert gas  systems, it is  not  assumed to be economically feasible to retrofit existing HFC Class A fire
extinguishing systems to this option. This analysis therefore assumes that this option is only applicable to
new Class A applications (i.e., those installed in 2005 or beyond), as shown in Table 4-4. Because of the
additional  space and weight requirements and the slower discharge times associated with this option
relative to conventional HFC-227ea systems, market penetration  rates reflect the assumption that this
option cannot fully displace HFC use in new Class A total flooding applications.  Furthermore, because
this  option is  associated with additional costs (see discussion below),  it is  assumed that market
penetration estimates in non-Annex I countries are 50 percent less than in Annex I countries  for all
years—given the greater economic challenges faced by non-Annex I countries.  Table 4-4 and Table 4-5
present the assumed market penetration rates of this option in developed and developing countries.

Cost and Emission Reduction Analysis

Various data and assumptions about the costs and emission reductions associated with inert gas were used
to analyze this option. U.S. costs were determined  relative to conventional HFC-227ea  systems, as they
dominate the HFC flooding market in the  United States.   This analysis scales the costs of inert gas
systems in other countries to U.S. costs based on confidential country-specific cost information obtained
for this report from HTOC members, as described in more  detail below.  The following bullets describe
the inputs used to derive the  final $/TCE for this option, the results of which are presented in Section 4.4:

    •   One-Time  Costs.  This analysis bases average capital costs for inert gases on average  selling
        prices  to   distributors/installers, as  provided  in  Wickham  (2003b),  which   provides  a
        comprehensive  cost comparison of total flooding systems.47  Accordingly, in the United States,
        inert gas  systems  cost approximately $34 per cubic  meter  of protected space,  which  is
46 In this report, the term technically applicable refers to the emissions to which an option can be applied. Because
inert gas systems are assumed to be used only in Class A fire total flooding applications, the technical applicability
is 100 percent of the emissions associated with those types of systems. Other factors will affect the application of the
option, for example to new or existing systems, and the market penetration assumed in this analysis.  See Appendix I
for a glossary of terminology.
47 The cost estimates in Wickham (2003b) do not include agent distribution piping and fittings, pipe supports and
hangers, actuation tubing and fittings, electrical cables and junction boxes  or labor to install. Although the costs
identified in Wickham (2003) for inert gas and HFC-227ea systems will be higher for end users, the cost differential
between these two systems is assumed to be relatively comparable.


                                                                                             -102-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
       approximately  $6 more  than  conventional HFC-227ea systems, which are estimated to cost
       $28.05 per cubic meter of protected space.  In addition, because  inert gas systems require more
       space to house gas cylinders compared to conventional HFC systems, in some cases there will be
       additional one-time costs associated with constructing the additional  space for  storage of the
       system.  Specifically, an additional 0.023 square feet (0.0021 square meters) of floor space is
       needed per cubic meter of protected space (Wickham, 2003b). Assuming a construction cost of
       $150 per square foot, this additional space requirement translates into an incremental one-time
       cost of nearly  $3.45  per  cubic meter  of protected  space  (R.S. Means, 2001).   The  total
       incremental capital cost of this option in the United States, therefore, totals $9.45 per cubic meter
       of protected space.
       In all  developing countries, capital costs for this option were scaled based on  cost estimates
       provided by HTOC members from developing countries.  Specifically, incremental capital costs
       (relative  to conventional HFC-227ea systems)  were assumed to be  10 percent greater in
       developing countries (Hughes Associates, 2001).
       Annual Costs.   Depending on the  application, the space  required  to house  additional  gas
       cylinders (an additional 0.023 square feet per cubic meter of protected space) will need to be
       heated and  cooled.  To  be conservative, the  additional annual  heating and cooling costs are
       considered in this analysis.  Based on average U.S. electricity costs of $8 per square foot,  this
       option is associated with an annual cost of $0.18 per cubic meter of protected space in the United
       States (R.S. Means, 2001).  In all other countries, this annual  cost was adjusted  by average
       electricity prices (average of 1994-1999) based on EIA (2000).
       Cost Savings.   Annual  savings are  associated  with  the avoided HFC-227ea emissions  and
       associated replacement costs, which would have been incurred had a conventional HFC system
       been used  in place of inert gas (which  for this analysis is  assumed to have no agent  cost).
       Because on average approximately 0.633 kg of HFC-227ea are needed to protect one cubic meter
       of space (Wickham, 2003b), and assuming a release rate of 1.5 percent of the installed base,  it is
       assumed that the emission of approximately 0.009 kg of HFC-227ea is avoided each year per
       cubic meter of protected  space.  Based on an average HFC-227ea cost of $34.10/kg (Wickham,
       2002), this translates into an annual savings of $0.32 per cubic  meter of protected space (0.633 kg
       x 1.5%x $34.10/kg).

       Emission  Reductions.   Under the  inert gas systems described above, the direct emission of
       approximately  0.009 kg  (0.633 kg x  1.5%)  of HFC-227ea  can be replaced, resulting in the
       avoided emissions of nearly 0.008 TCE per cubic meter of protected space.  Indirect emission
       (energy) penalties associated with additional space requirements (of 0.023 ft2 per cubic meter of
       protected space  [Wickham, 2003b]) are calculated using the average electricity use to heat/cool
       the additional space required (assumed to be $8/ft2 [R.S. Means, 2001]  at an average electricity
       cost in the  United States  of $0.04/kWh [EIA,  2000]),  and the average  emission rate from
       electricity generation (assumed to be 0.606 kg CO2/kWh  [EIA, 2004]).   Thus, the  indirect
       emissions lower these  emission benefits  by nearly  10 percent (approximately  0.0007 TCE per
       cubic meter of protected space) in the United States.  Therefore, in the  United States, net annual
       emission reductions associated with this option are equal to approximately 0.0068 TCE per cubic
       meter of protected space.  In all other countries,  the indirect emission penalty was calculated by
       multiplying  the emission penalty assumed for the United States (0.0007 TCE) by a ratio of U.S.
       to regional national average CO2 emission rates  for electricity production, based on Sand et al.
       (1997).
                                                                                           -103-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Water Mist Systems

Water mist systems use relatively small droplet sprays under low, medium, or high pressure to extinguish
fires.  These systems use specially-designed nozzles to produce much smaller droplets than are produced
by traditional water-spray systems or conventional sprinklers, thereby requiring significantly less water to
achieve extinguishment (UNEP, 2001; Wickham, 2002). Another benefit of water mist systems is that, in
some  applications (e.g.,  marine applications), they can be brought into action  faster than HFC systems,
since  there is less concern in applying  water mist in situations where openings to the space are  not all
closed—which in turn leads to reduced  fire damage.  In addition, unlike HFC systems, which are usually
limited to a single discharge of agent, most water mist systems have an unlimited water supply in land-
based operations, and at least 30 minutes of potable water discharge followed by an unlimited amount of
seawater for marine applications (Wickham, 2003b).

To date, water mist systems have been used in shipboard accommodation, storage and machinery spaces,
combustion turbine enclosures, flammable  and combustible liquid machinery, and  light and ordinary
hazard sprinkler applications (UNEP, 2001). Water mist systems can provide equivalent fire protection
and life safety/health protection for Class B  fuel hazards, where low  temperature freezing is  not  a
concern. Class B (flammable liquid) fire hazards are estimated to account for approximately 5 percent of
the HFC total flooding market in the United States and were assumed to account for the same percent in
all non-U.S. countries for lack of data on those countries (Wickham, 2002). Water mist systems have also
found acceptance in Class A applications, but as replacements for water sprinklers, not HFCs. Therefore,
this report does not consider water mist as  an  option  for abating HFC emissions from  Class  A
applications.

Various difficulties have been identified with the use of water mist systems, impeding deployment of
these  systems and preventing broader market shares.  First, these systems have not proven effective in
extinguishing small fires in large spaces (volumes greater than 2,000 m3) (IMO, 2001; Wickham, 2002).
Additionally, because the relationship between the mechanism of extinguishment48 of water mist systems
is non-linear  and not well  understood, applications of water mist  systems  have  been  limited to those
where fire test  protocols  have been  developed,  based  on empirically-tested  system  performance.
Therefore,  new applications may require empirical  performance testing prior to the installation of such
systems, in order to ensure  safety and obtain approval  of the proper  regulatory or  standard  setting
authority.  Currently, an International  Maritime Organization (IMO) working group  is  studying this
situation and considering proposals that suggest an overhaul to the test methods and approval guidelines.
Should IMO change its water mist requirements to something more flexible regarding  the extinguishment
of small fires in large spaces, it will make a difference in the future cost and, thus, market acceptance of
water mist systems (Wickham, 2003b). In addition, the use of additives—such  as  salts  or foam or a
combination of these systems with gaseous agents—are other options under investigation to improve
system  performance  for specific applications.   Many researchers and industry experts believe that
solutions to these market barriers are well within reach (Wickham, 2002).

Other market barriers for this option include additional space requirements for system storage  compared
to conventional HFC-227ea systems. Indeed, water mist systems require  an  estimated seven times more
space than HFC-227ea (Wickham,  2003b).  In addition, water mist systems  used in marine applications
are cost prohibitive for protecting small  spaces (i.e., those less than 3,000 cubic meters in size).49
48 "Mechanism of extinguishment" refers to the amount of water mist needed to extinguish a given fire in a given
volume.
49 This cost information is based on water mist systems employed under the current IMO requirements for marine
systems, which are much more severe than the requirements for land based systems. The use of water mist systems

                                                                                           -104-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
This analysis assumes this option is technically applicable to the emissions from total flooding systems
designed for Class B fires.  Due to the additional space requirements associated with this option, it is
assumed that water mist systems could not feasibly replace any existing HFC systems in Class B fire
protection applications and, therefore, are only used in new Class B total flooding applications (i.e., those
installed in 2005 or beyond).

In terms of market penetration, this analysis assumes that the remaining technical constraints associated
with water mist systems will gradually be overcome, and that by 2020 in Annex I countries, water mist
systems will reach full market penetration in all new Class B fire suppression systems used to protect
large spaces.

Market penetration estimates for non-Annex I countries are assumed to be 50 percent less than those of
developed  countries, as  a result of economic considerations.   Table 4-4 and Table 4-5 present the
maximum market penetrations  assumed for this option.

Cost and Emission Reduction Analysis

The following bullets describe the cost and emission inputs used to derive the final $/TCE for the water
mist option, the results of which are presented in Section 4.4:

    •    One-Time Costs.  This analysis bases the average capital costs for water mist systems on the
        average selling prices to distributors for systems used in marine applications, as provided in
       Wickham (2003b), which provides a cost comparison of total flooding systems.50  According to
       that report, capital costs of water mist systems used in marine systems to protect large spaces are
        estimated to be  $29.67 per cubic  meter of protected space—or $4.07 more per cubic meter of
       protected space than conventional HFC-227ea systems in  large spaces (which are estimated to
        cost  $25.60  per  cubic meter  of protected  space).  It should be noted that for non-marine
        applications, costs are  more competitive than those presented here, as the requirements for land
       based systems are not as stringent as those currently required by the  IMO (Wickham, 2003a).
        This analysis uses the  costs for marine rather than land-based systems to obtain conservative
        results (i.e., higher costs). Other costs presented below are based on non-marine applications.
        In addition, because  water mist systems require more space  than conventional HFC systems, one-
       time costs associated with constructing additional  space are  also considered.  Specifically, an
        average of approximately 0.0472 square feet (0.0044 square meters) of additional floor space in
       the  building is needed per cubic meter of  protected space (Wickham, 2003b).   Assuming a
        construction cost of $150 per square foot in the United States (R.S. Means, 2001), this additional
        space requirement translates into an incremental one-time  cost of $7.08 per cubic  meter of
       protected space.  Therefore, the total incremental capital cost of this option in the United States
       totals $11.15 per cubic meter of protected space ($4.07 + $7.08).
        Reliable international cost information  on water mist systems was only obtained for India and
        Russia.  According to international experts, capital costs are the same in Russia as in the United
        States, and about 10 percent higher in India (Hughes Associates, 2001).  India is used as a proxy
        for estimating costs in all other developing countries (i.e., costs are assumed to be 10 percent
       greater than those in the United States).

in non-marine applications appear to be more cost competitive with other alternatives (Wickham, 2003a).
50 The cost estimates provided in Wickham (2003b) do not include feed water pipes, low pressure piping, electrical
cables and junction boxes or labor to install. Therefore, water mist and HFC-227ea systems costs will be higher than
presented here, but the cost differential between these two systems is assumed to be comparable.

                                                                                            -105-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •  Annual Costs. Because the additional space required to house water mist systems (0.0472 square
       feet per cubic meter of protected space) will need to be heated and cooled, annual heating and
       cooling costs  for this additional space are considered in this analysis.  Based on average U.S.
       electricity costs of $8 per square foot (R.S. Means, 2001), this option is associated with an annual
       cost of $0.38 per cubic meter of protected space in the United States.  In all other countries, this
       annual cost is adjusted by average electricity prices (average of 1994-1999) based on EIA (2000).
    •  Cost Savings.  Annual  savings are associated with the avoided HFC-227ea emissions and
       associated  replacement costs, which would have been incurred had a conventional HFC system
       been used in place of water (which for this analysis is assumed to have no agent cost).  Because
       on average approximately 0.630 kg of HFC-227ea are needed to protect one cubic meter of space
       (for large spaces) (Wickham, 2003b), and assuming a release rate of 1.5 percent of the installed
       base, it is assumed that the emission of approximately 0.009 kg  of HFC-227ea is avoided each
       year (0.630 kg x  1.5%).  Based on an average HFC-227ea cost of $34.10/kg, this translates into
       an annual savings of $0.32 per cubic meter of protected space (Wickham, 2002).
    •  Emission Reductions.  Under the water mist systems described above, the direct emission of
       approximately  0.009  kg  of  HFC-227ea   can  be  avoided, resulting  in the  reduction  of
       approximately  0.007 TCE per cubic  meter of protected  space.  Indirect emission (energy)
       penalties associated with additional space requirements (assumed to be 0.0472 ft2 per cubic meter
       of protected space [Wickham, 2003b]) are calculated using the average electricity use to heat/cool
       the additional space required (assumed to be $8/ft2 [R.S. Means, 2001]  at an average electricity
       cost in the United  States  of $0.04/kWh [EIA, 2000]), and the  average  emission rate  from
       electricity  generation (assumed to  be 0.606 kg  CO2/kWh  [EIA,  2004]).   Thus, the indirect
       emissions lower these emission benefits by roughly 20 percent (approximately 0.0015 TCE per
       cubic meter of protected space) in the United States.  Therefore, in the United States, net annual
       emission reductions associated with this option are equal to approximately 0.0060 TCE per cubic
       meter of protected space.  In all other countries, the indirect emission penalty was calculated by
       multiplying the emission penalty assumed for the United States (0.0015 TCE) by a ratio of U.S.
       to regional national average CO2 emission rates for electricity production, based on Sand et al.
       (1997).

Fluorinated Ketone (FK-5-1-12)

FK-5-l-12-mmy2  (also  known as  1,1,1,2,2,4,5,5,5-nonafluoro-4-(trifluoromethyl)-3-pentanone, and
commonly referred to as FK-5-1-12) is a fluorinated ketone with an atmospheric lifetime of up to two
weeks and a 100-year GWP of approximately  1 (ICF Consulting, 2003).  This alternative received U.S.
EPA Significant New Alternatives Policy (SNAP) approval as an acceptable replacement for halon 1301
in flooding applications at the end of 2002 and for halon 1211 in non-residential streaming applications in
early 2003.

Compared to HFC-227ea total flooding systems, FK-5-1-12 systems  are associated with slight space and
weight penalties—on the order of 10 and 15 percent, respectively—which could make its use  in confined
spaces (e.g., ships,  aircraft) less attractive, although some marine installations have already been reported
(Werner, 2004a).51   Moreover, in addition to its cost (see cost analysis  below) and its relatively recent
entry52 into this market as compared to the inert gas and water mist options, the extent of commercial
acceptance and future usage of this option is not yet known.
51 It has been reported that the space penalty is only associated with use in large systems, and that the weight penalty
has not proven to be an impediment (Werner, 2004b).
52 This agent is in the 2004 edition of the National Fire Protection Association (NFPA) Standard on Clean Agent

                                                                                           -106-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
While this option is not associated with major floor space penalties and appears not to suffer from any
significant technical barriers, it is only assumed to be applicable in new Class A flooding applications
(i.e., those installed in 2005 or beyond), because the cost analysis does not assess retrofit costs.  Table 4-4
and Table 4-5 present the maximum market penetration assumptions for this option, which project that
this option will gain a foothold in the marketplace and out-compete inert gas systems in new Class A total
flooding applications.  Due to  the reasons expressed above, this analysis conservatively assumes that
market  penetration will be low in early years,  although  others project higher sales (Werner, 2004b).
Market  penetration is assumed to be greater in Annex I countries than in non-Annex I countries, as a
result of economic considerations.

Cost and Emission Reduction Analysis

The following bullets  describe the cost and emission inputs used to derive the final $/TCE for FK-5-1-12
systems, the results of which are presented in Section 4.4:

    •   One-Time  Costs.   This analysis bases average  capital costs for FK-5-1-12 systems on the
        average selling prices to distributors/installers, as provided in Wickham (2003b), which provides
        a cost comparison of total flooding systems.53 According to  this report, in the United States,
        capital costs of FK-5-1-12 systems are estimated to be nearly  $36 per cubic meter of protected
        space—or nearly $7.94  more than conventional HFC systems (estimated to cost $28.05 per cubic
        meter  of protected space).  To be conservative, this analysis uses this figure, although others have
        reported increased costs  of only around  $3.60 per cubic meter of protected space (Werner,
        2004b).  Also, while the floor space requirements for this option are very similar to those of HFC
        systems, there is a slight increase in the floor space needed to protect each cubic meter of space
        (less than 0.001  square feet).  Assuming an average construction cost of $150 per square foot
        (R.S.  Means, 2001),  this  translates into  an  incremental  one-time  construction  cost  of
        approximately $0.07 per cubic meter of protected space. Therefore, the  total  incremental one-
        time cost of this option is $8.01 per cubic meter of protected space ($7.94 + $0.07). Due to a lack
        of available data on the costs of this option in other countries, no regional cost adjustments were
        made to this capital cost.
    •   Annual Costs.  Because the additional space requirement associated with this  option relative to
        conventional HFC systems is so small, the additional annual costs associated  with heating and
        cooling are also very small (less than $0.01 annually per cubic meter of protected space).  In all
        other countries, this annual cost was adjusted by average electricity prices (average of 1994-1999)
        based  on EIA (2000).  In addition,  it is assumed that an incremental annual cost of $0.04 per
        cubic meter of protected space is associated with annual emissions/agent replacement costs. This
        cost is based on the assumption that approximately 11 percent more FK-5-1-12 agent is required
        to protect the same amount of space as HFC-227ea for Class A fires, that approximately 0.009 kg
        of HFC-227ea is used per cubic meter of protected  space, and that the agent replacement costs are
        both equal (roughly $34/kg, although actual prices vary by region) (Werner, 2004a).

    •   Cost Savings. Because the agent is assumed to cost the same as HFC-227ea, there are no annual
        cost savings associated with this option.


Fire Extinguishing Systems (NFPA, 2004) and has been accepted for future addition to the ISO International
Standard on Gaseous Fire-Extinguishing  Systems (Wickham, 2003a).
53 The cost estimates in Wickham (2003b) do not include agent distribution piping and fittings, pipe supports and
hangers, actuation tubing and fittings, electrical cables and junction boxes or labor to install. Although the costs
identified in Wickham (2003) for FK-5-1-12 and HFC-227ea systems will be higher for end users, the cost
differential between these two systems is assumed to be relatively comparable.

                                                                                             -107-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
    •  Emission Reductions.  Under the FK-5-1-12 option described above, the direct emission  of
       approximately  0.009  kg of HFC-227ea  can be avoided per cubic meter of protected space,
       resulting in a reduction of nearly 0.008 TCE per cubic meter of protected space  (0.009 kg x 2,900
       x  12/44 x  1/1,000).   Indirect  emission (energy) penalties associated with additional space
       requirements (assumed to be 0.0005 ft2 per cubic meter of protected space [Wickham, 2003b]) are
       calculated using the average electricity use to heat/cool the additional space required (assumed to
       be $8/ft2 [R.S. Means, 2001] at an average electricity cost in the United States of $0.04/kWh
       [EIA, 2000]), and the average emission rate from electricity generation (assumed to be 0.606 kg
       CO2/kWh [EIA, 2004]).  Thus, the  indirect emissions reduce this environmental benefit by less
       than 0.25 percent (0.00002 TCE) in the United States. Therefore, in the United States, net annual
       emission reductions associated with this option are  equal to 0.0075 TCE per  cubic  meter  of
       protected  space.  In  all other countries, the indirect  emission  penalty was  calculated by
       multiplying the emission penalty assumed for the United States (0.00002 TCE) by a ratio of U.S.
       to regional national average CO2 emission rates for electricity production, based on Sand et al.
       (1997).

Carbon Dioxide

Carbon dioxide has been used for many decades in total flooding systems.  Some of the types of hazards
and equipment that carbon dioxide systems protect are flammable liquid materials;  electrical hazards,
such as transformers, switches, circuit breakers,  rotating equipment, and electronic equipment; engines
utilizing gasoline and other flammable liquid fuels;  ordinary combustibles, such as paper, wood, and
textiles; and hazardous solids (NFPA, 2000). Due to the lethal concentrations at which carbon dioxide is
required for use as a fire extinguishing agent, there have been concerns with incidences of deaths and
injuries attributed to exposure to this agent (U.S. EPA, 200b;  Wickham,  2003b).  The National Fire
Protection Association (NFPA) Technical  Committee on NFPA 12  Standard for Carbon Dioxide Fire
Extinguishing  Systems is currently reviewing a proposal to change the standard to prohibit use of these
systems in normally occupied areas (Wickham,  2003b).  The IMO's  Safety of Life at Sea (SOLAS)
standard does not prohibit the use of carbon dioxide in normally occupied areas, but calls for the use  of
suitable alarms and mandates against the use of automatic release of the fire-extinguishing medium,  as
noted in U.S. EPA (2000). IMO is also considering whether to prohibit use of carbon dioxide systems in
occupied areas as part of that organization's  broad review of the current performance testing requirements
for all shipboard fire extinguishing systems (IMO, 2003; Wickham, 2003b).

As one of the oldest fire extinguishing agents in use, and as a more economical option than HFCs, carbon
dioxide has generally developed its own niche market in narrow-use total flooding applications.  While
carbon dioxide could and does replace  some halon use where permitted by regulations, this analysis
assumes that carbon  dioxide would be selected as a first-choice replacement of halon, not as  a  second
transition, after more costly HFCs.  For example, the majority  of U.S. ship owners have  shifted from
halon 1301 to carbon dioxide for mandatory engine room protection for new ships (Wickham, 2002). For
this reason, any use of carbon dioxide is assumed  to occur in the baseline, and not as an option to replace
HFC systems.  It is therefore not considered  in the cost analysis.

Recovery and Reuse of HFCs

HFCs can be recovered for reuse at service and  decommissioning.  For several reasons, however, this
analysis does not incorporate this option into the cost analysis.  First, responsible halon management
practices are assumed to be standard convention in fire protection throughout the world.54  Second, given
54 Responsible use practices are currently being developed and endorsed worldwide.  For example, the halon
Recycling  Corporation  (HRC) published a Code of Practice for Halon Reclaiming Companies. Because the

                                                                                           -108-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
the high costs of HFCs, there is  a strong  financial incentive  for maximum recovery following the
decommissioning of large HFC systems.  While most HFC systems—with lifetimes ranging from 10 to
20 years—have not yet reached the end of their useful lifetimes and, therefore, widescale system recovery
and recycling at decommissioning  has not yet occurred, this analysis assumes that such practices will
occur in the baseline.

Improved Detection Systems

One effective way of reducing HFC emissions from the fire extinguishing sector is to install improved
detection and control systems to (1) prevent a false discharge  (e.g., high sensitivity smoke detection
systems that provide early warning to pre-empt the need for actual system discharge), or (2) minimize the
amount of agent discharged to extinguish a fire.

Since advanced detection systems have been available for the last decade or so, this analysis assumes that
total flooding HFC systems  have been and are  being equipped  with such  systems internationally.
Because improved detection systems are  assumed to be used in the baseline, this option is not considered
in the cost analysis.

Fine Aerosols

Aerosols are  being developed for use as extinguishing agents in niche markets in the United States, such
as aerospace, marine, and some military applications.  The NFPA has written a draft standard (NFPA
2010) for this agent (NFPA, 2003).  It is possible that if this agent is ever successfully brought to market,
that it may  be applicable in other end uses (Wickham, 2002).

Because fine aerosols are not currently a viable commercial option to HFCs in fire protection, and much
uncertainty exists as to whether or not the associated technical and economic barriers will be overcome to
enable them to become a viable alternative, this option is not considered in the cost analysis.

Inert Gas Generators

Inert gas generators use a solid material that oxidizes rapidly, producing large quantities of carbon dioxide
and/or nitrogen. While this technology  has  demonstrated space  and weight requirements equivalent to
halon 1301, it has thus far only been used in specialized applications in the United States (e.g., dry bays
on military aircraft) (Wickham, 2002).  Due  to insufficient data on these systems and to the uncertainty
associated with their applicability in other fire extinguishing applications, this option is not considered in
the cost analysis.
equipment and training needed to reclaim halons are also required to reclaim HFCs, the HRC  Code of Practice
establishes the necessary infrastructure and sets the practice of reclamation as the norm for how business is done in
the fire protection industry. While the HRC is a U.S. association, its membership consists  of multinational
corporations operating throughout the world. Similarly, the Halon Alternatives Research Corporation (HARC), U.S.
EPA,  and other organizations  have recently developed and endorsed the Voluntary Code of Practice for  the
Reduction of Emissions of HFC and PFC Fire Protection Agents. This Code of Practice will also have international
reach, since HARC members include multinational companies in the alternative agent manufacturing, equipment
manufacturing, and distribution sectors.


                                                                                              -109-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
4.3.2   Summary of Technical Applicability and Market Penetration of Abatement
        Options
Table 4-3 summarizes the technical applicability of each option, which is equal to the estimated global
average breakout of total fire sector HFC and PFC emissions for the application (i.e., total flooding, Class
A or B) addressed by the option. Technical applicability is used in conjunction with market penetration
assumptions to develop the emission reduction potentials for each option, as explained further below.
Table 4-4 provides the assumptions on  maximum market penetration into annual installations of total
flooding systems designed for the particular application (i.e., Class A or B fires) for each option in 2005,
2010, 2015  and 2020.  Market penetrations were  developed separately for Annex I and non-Annex I
countries, to best reflect region-specific qualitative information  and  possible future action.  Table 4-5
presents  the final maximum penetration into the installed  base  of equipment, taking into account the
percent of each applicable fire hazard market that is new (i.e., systems installed in 2005 or beyond) in all
preceding years.  Values from Table 4-5 are multiplied by technical applicabilities from Table 4-3 to
generate the percent reduction off baseline emissions, as presented in Table 4-6.

Table 4-3: Summary of Technical Applicability of Abatement Options3	
                                            Annex I and Non-Annex I Countries
                                        2005        2010        2015        2020
Inert Gas (Class A flooding)
Water Mist (Class B flooding)
FK-5-1 -12 (Class A flooding)
71.3%
3.8%
71.3%
76.0%
4.0%
76.0%
80.8%
4.3%
80.8%
85.5%
4.5%
85.5%
a Expressed as a percent of total fire extinguishing emissions.

Table 4-4: Incremental Maximum Market Penetration, Expressed as Percent of Annual Installation of New
Class A or Class B Systems	
                    Annex I Countries
               2005    2010   2015   2020
                              Non-Annex I Countries3
                           2005   2010   2015   2020
                                Considerations/Rationale
Inert Gas
(New Class A)
10%    20%    30%    30%
5%    10%    15%    15%
Can displace MFCs in new Class A
applications
Additional space and weight
requirements
Slower discharge times
Higher costs compared to baseline
HFC-227ea systems lead to lower
market penetration in developing
countries
Water Mist
(New Class B)
25%    50%    75%   100%
13%    25%    38%   50%
Can displace MFCs in new Class B
applications used to protect large
spaces
Technical constraints (assumed to
be gradually overcome)
Higher costs compared to baseline
HFC-227ea systems lead to lower
market penetration in developing
countries
                                                                                              -110-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
                     Annex I Countries

                2005    2010   2015   2020
                               Non-Annex I Countries3

                            2005   2010    2015   2020
                          Considerations/Rationale
FK-5-1-12
(New Class A)
4%     20%    40%    50%
10%    20%   25%
Can displace MFCs in new Class A
applications
No major additional space
requirements
Lowest up-front cost of all
alternatives considered in this
analysis
Newer player on market compared
to inert gas and water mist systems;
will take time to gain foothold in
market
Higher costs compared to baseline
HFC-227ea systems lead to lower
market penetration in developing
countries
3 To account for economic considerations, assumed market penetration values in developing countries are half of those assumed for developed
countries.

Table 4-5: Incremental Maximum Market Penetration Expressed as Percent of Entire Installed Base (Class A
or Class B)a

Inert Gas (Class A)
Water Mist (Class B)
FK-5-1-12 (Class A)
2005
0.5%
1.3%
0.2%
Annex
2010
4.5%
11.3%
3.6%
I Countries
2015
11.0%
27.5%
11.6%
2020
18.5%
50.0%
23.1%
2005
0.3%
0.7%
0.1%
Non-Annex
2010
2.3%
5.7%
1.8%
I Countries
2015
5.5%
13.9%
5.8%
2020
9.3%
25.2%
11.6%
a Expressed as a percentage of technical applicability (i.e., both new and existing Class A or Class B emissions).

To calculate the percent of emission reductions off the total fire extinguishing baseline for each abatement
option, the technical applicability (from Table 4-3) is multiplied by the market penetration values (from
Table  4-5). For example, to determine the percent reduction off the 2020 baseline for FK-5-1-12 in the
United States (or other Annex I countries), the following calculation is used:

                Technical Applicability * Incremental Maximum Market Penetration =

                85.5% x 23.l%~ 19.8%

Thus,  using the assumptions in this analysis, FK-5-1-12 could reduce approximately one-fifth of the
Annex I 2020 emissions baseline.   This figure, along with the other emission  reduction potentials, is
shown in Table 4-6.

Table 4-6: Emission Reductions off Total Fire Extinguishing  Baseline


Inert Gas
Water Mist
FK-5-1-12

2005
0.4%
0.0%
0.1%
Annex I
2010
3.4%
0.5%
2.7%
Countries
2015
8.9%
1.2%
9.4%

2020
15.8%
2.3%
19.8%

2005
0.2%
0.0%
0.1%
Non-Annex I
2010
1.7%
0.2%
1.4%
Countries
2015
4.4%
0.6%
4.7%

2020
7.9%
1.1%
9.9%
                                                                                                   -111-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
4.4
Results
Table 4-7 through Table 4-9 provide a summary of the potential emission reduction opportunities and
associated annualized costs for United States, Non-U.S. Annex I countries, and Non-Annex I countries in
2020.  The costs to reduce one ton of carbon equivalent (TCE) are presented for two different discount
rate scenarios: 4 percent and 20 percent.  The tax rate  associated with each cost scenario is 0 percent and
40 percent, respectively.  The results are ordered by increasing costs per TCE, using the highest cost in
the region under the 4 percent discount rate. The emissions reduced by the option and a cumulative total
of emissions reduced, in MMTCE and percent of the regional fire extinguishing baseline, are presented.

Table 4-7: United States Emission Reductions in 2020 and Break-Even Costs for Fire Extinguishing
                       Break-Even Cost      Emission
                         (2000$/TCE)        Reduction of     Percent     Cumulative   Cumulative %
                      Discount/Tax Rate       Option    Reduction from  Reductions  Reduction from
Reduction Option       4%/0%     20%/40%     (MMTCE)    2020 Baseline    (MMTCE)    2020 Baseline
Inert Gases
FK-5-1-12
Water Mist
$81.98
$83.96
$146.97
$410.54
$335.63
$587.31
0.14
0.17
0.02
15.8%
19.8%
2.3%
0.14
0.31
0.33
15.8%
35.6%
37.8%
Table 4-8: Non-U.S. Annex I Emission Reductions in 2020 and Break-Even Costs for Fire Extinguishing
                    Break-Even Cost (2000$/TCE)a       Emjssjon   Rercent              Cumulative
                         Discount/Tax Rate            Reduction  Reduction  Cumulative  % Reduction
Reduction           4%/0%            20%/40%       of Option  from 2020  Reductions   from 2020
Option           Low      High      Low      High    (MMTCE)   Baseline    (MMTCE)     Baseline
FK-5-1-12
Inert Gases
Water Mist
$83.72
$71.10
$111.76
$85.22
$136.29
$287.80
$334.91
$369.25
$467.17
$336.75
$483.77
$761.31
0.19
0.15
0.02
19.8%
15.8%
2.3%
0.19
0.33
0.36
19.8%
35.6%
37.8%
a Costs vary by country/region based on one-time or annual adjustment factors (e.g., electricity price); therefore, the lowest and highest costs
for the region are shown.

Table 4-9: Non Annex I Emission Reductions in 2020 and Break-Even Costs for Fire Extinguishing	
                   Break-Even Cost (2000$/TCE)a        Emissjon    Percent
                        Discount/Tax Rate           Reduction of  Reduction Cumulative Cumulative %
Reduction           4%/0%           20%/40%         Option     from 2020 Reductions Reduction from
Option            Low      High     Low     High     (MMTCE)    Baseline   (MMTCE)  2020 Baseline
FK-5-1-12
Inert Gases
Water Mist
$83.74
$81.29
$126.10
$84.36
$114.30
$219.85
$335.01
$414.31
$529.82
$336.18
$486.93
$740.97
0.13
0.11
0.02
9.9%
7.9%
1.1%
0.13
0.24
0.26
9.9%
17.8%
18.9%
a Costs vary by country/region based on one-time or annual adjustment factors (e.g., electricity price); therefore, the lowest and highest costs
for the region are shown.
4.5
Summary
Baseline HFC and PFC emissions from fire extinguishing are estimated to grow between the years 2005
and 2020, with the highest emissions growth expected to occur in Non-Annex I countries. It is estimated
that the vast majority of these emissions will come from total flooding applications; only a minor amount
will come from streaming applications.
                                                                                               -112-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Several alternatives to ozone-depleting halon 1301 for total flooding applications exist, including gaseous
alternatives such as halocarbons (HFCs and PFCs), carbon dioxide, inert gases and fluorinated ketones, as
well as non-in-kind alternatives such as dispersed and condensed aerosol systems, water sprinklers, water
mist, foam and inert gas generators.

This analysis reviewed these alternatives and analyzed in detail three mitigation options for total flooding
fire extinguishing applications: (1) substituting HFC systems used in new Class A fire hazards with inert
gas systems; (2) substituting HFCs used in new Class  B fire hazards with water mist systems; and (3)
substituting HFC systems used in new Class A fire hazards with FK-5-1-12 systems. Inert gas and FK-5-
1-12 systems may offer good opportunities to reduce emissions  in total flooding applications globally.
Water mist systems also have the potential to reduce global emissions from this sector, but to a lesser
extent.

This analysis demonstrates that there is a portfolio of alternatives  to HFCs and PFCs in the total flooding
sector that can be employed to reduce HFC and PFC use  and associated  emissions.  These alternatives
include FK-5-1-12, inert gases, water mist, and other agents and systems discussed qualitatively in this
report. The global implementation of each option through 2020 is based on a "best guess" scenario. With
more data, these forecasts can be improved.


4.6        References

EIA. 2004. Form EIA-1605, Long Form for Voluntary Reporting of Greenhouse Gases, Instruction, Data
through 2003 (Appendix C. Adjusted Electricity Emission Factors  by State and Region). Energy
Information Administration. 2004. Available online at
. Accessed pp.
50 on 7 June 2004.

EIA. 2000. Annual Energy Outlook 2000 (Electricity Prices for Industry 1994-1999). Energy Information
Administration. 2000. Available online at .
Accessed on 2 April 2002.

Europa. 2003. Regulation (EC) No 2037/2000 of the European Parliament and of the Council of 29 June
2000 on substances that deplete the ozone layer. Europa. 2003. Available online at
. Accessed on
October 13, 2003.

Hughes Associates. 2001. International Market Share Data for Total Flooding Sector: Current estimates
and future projections of total flooding sector market share by country (Brazil, Japan, India, Russia).
Confidential business information collected and compiled by Hughes Associates, Inc. for ICF Consulting.
November 2001.

ICF. 2003. Re-evaluation of a C-6 Oxyfluorocarbon (trade name Novec 1230) and References.
Memorandum delivered by ICF Consulting to Erin Birgfeld under EPA Contract Number 68-D-00-266,
Work Assignment 2-05 Task 03. September 10, 2003.

IMO. 2003. Performance Testing and Approval Standards for Fire Safety Systems - Report of the
Correspondence Group. Submitted by the United States  to the International Maritime Organization, Sub-
Committee on Fire Protection, 48th session, Agenda item 5, FP 48/5. October  10, 2003.
                                                                                          -113-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
IMO. 2001. Performance Testing and Approval Standards for Fire Safety Systems: Fire Test Protocols
for Fire-Extinguishing Systems. Submitted by Germany to the International Maritime Organization, Sub-
committee on Fire-Protection, 46th session, Agenda item 12. November 30, 2001.

IPCC. 1996. Climate Change 1995: The Science of Climate Change. Intergovernmental Panel on Climate
Change. J.T. Houghton, L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg, and K. Maskell, eds.
Cambridge, UK: Cambridge University Press. 1996.

IPCC/TEAP. 1999. Meeting Report of the Joint IPCC/TEAP Expert Meeting on Options for the
Limitation of Emissions ofHFCs andPFCs. Report jointly sponsored by the Intergovernmental Panel on
Climate Change Working Group III and the Technology and Economic Assessment Panel (TEAP) of the
Montreal Protocol (ECN-RX--99-029). July 1999. Available online at
. Accessed on November 20, 2003.

Kucnerowicz-Polak, B. 2002. Halon Sector Update. Presented at the 19th Meeting of the Ozone
Operations Resource Group (OORG), The World Bank, in Washington, DC. March 28, 2002.

March Consulting Group. 1999. UK Emissions ofHFCs, PFCs, and SF6 and Potential Emission
Reduction Options: Final Report. March Consulting Group. January 1999.

March Consulting Group. 1998. Opportunities to Minimize Emissions ofHydrofluorocarbons (HFCs)
from the European Union:  Final Report. March Consulting Group. September 1998.

NFPA. 2004. NFPA 2001: Standard on Clean Agent Fire Extinguishing Systems, 2004 Edition. National
Fire Protection Association. 2004.

NFPA. 2003. Proposed Draft of NFPA 2010: Standard for Fixed Aerosol Fire Extinguishing Systems,
2005 Edition. National Fire Protection Association. 21 July 2003. Available online at
. Accessed on October 24, 2003.

NFPA. 2000. NFPA 12: Standard on Carbon Dioxide  Extinguishing Systems, 2000 Edition. National Fire
Protection Association. 2000.

R.S. Means Company, Inc.  2001. Means Square Foot Costs, 22nd Annual Edition 2001. Kingston, MA:
R.S. Means Company, Inc.  2001.

Sand, J.R., S.K. Fischer, and V.D. Baxter. 1997. Energy and Global Warming Impacts of HFC
Refrigerants and Emerging Technologies. Prepared for Alternative Fluorocarbons Environmental
Acceptability Study and U.S. Department of Energy. Oak Ridge National Laboratory. 1997.

UNEP. 2001. Standards and Codes of Practice to Eliminate Dependency on Halons: Handbook of Good
Practices in  the Halon Sector. United Nations Publication ISBN 92-807-1988-1. United Nations
Environment Programme, Division of Technology, Industry and Economics (DTE) under the OzonAction
Programme under the Multilateral Fund for the Implementation of the Montreal Protocol, in cooperation
with The Fire Protection Research Foundation. 2001.

UNEP. 1999. Production and Consumption of Ozone Depleting Substances 1986-1998. United Nations
Environment Programme. October 1999.
                                                                                        -114-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
U.S. EPA. 2004. Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2002. EPA 430-R-04-
003. U.S. Environmental Protection Agency, Office of Atmospheric Programs. April 2004.

U.S. EPA. 2000. Carbon Dioxide as a Fire Suppressant: Examining the Risks. EPA 430-R-00-002. U.S.
Environmental Protection Agency, Office of Air and Radiation. February 2000.

U.S. EPA. 1994. SNAP Technical Background Document: Risk Screen on the Use of Substitutes for Class
I Ozone-Depleting Substances: Fire Suppression and Explosion Protection (Halon Substitutes). U.S.
Environmental Protection Agency. 1994.

Verdonik, Daniel P. and Mark L. Robin. 2004. Analysis of Emissions Data, Estimates, and Modeling of
Fire Protection Agents. Conference proceedings from the 15th Annual Earth Technologies Forum and
Mobile Air Conditioning Summit in Washington, DC. April 13-15, 2004.

Werner, Kurt. 2004a. Expert review comments on the Draft Analysis of International  Costs to Abate HFC
PFC Emissions from Fire Extinguishing. Comments received via email on January 26, 2004.

Werner, Kurt. 2004b. Expert Review Comments on the Draft Analysis of International Costs to Abate
HFC PFC Emissions from Fire Extinguishing. Comments received via email May 20-21, 2004.

Wickham, Robert. 2003a. Expert review comments on the Draft Analysis of International Costs to Abate
HFC PFC Emissions from Fire Extinguishing. Comments received in writing and by phone in October
2003.

Wickham, Robert. 2003b. Review of the Use of Carbon Dioxide Total Flooding Fire Extinguishing
Systems. Wickham Associates, 8 August 2003. Available online at
.

Wickham, Robert. 2002. Status of Industry Efforts to Replace Halon Fire Extinguishing Agents. Wickham
Associates. March 16, 2002. Available online at .
                                                                                         -115-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
5   Analysis  of Costs to Abate International HFC and PFC
     Emissions  from Solvents
5.1        Introduction


Ozone depleting substances (ODSs) have been used as solvents in a wide range of cleaning applications,
including precision, electronics, and metal cleaning (UNEP, 1999a).  Chlorofluorocarbons (in particular
CFC-113), methyl chloroform (1,1,1-trichloroethane or TCA), and to a lesser extent, carbon tetrachloride
(CC14), were  historically  used  as  solvents in the United  States.  Similar usage historically occurred
elsewhere, with the exception of India and China, where greater volumes of CCl4were consumed.

To  comply  with  the requirements  of the  Montreal  Protocol,55  many  countries  started  using
hydrochlorofluorocarbons (HCFCs), and  aqueous  and semi-aqueous  Not-in-Kind (NIK) solvents  as
substitutes for ODSs. For example, the majority of metal cleaning end-users and some of the electronics
and precision cleaning solvent end-users have already transitioned to no-clean, semi-aqueous cleaning,
and aqueous  cleaning  alternative  methods.   Many of the in-kind replacement solvents, including
hydrofluorocarbons (HFCs) and perfluorocarbons  (PFCs), have also  taken a  share of the substitute
market, as they have high reliability, excellent compatibility, good stability, low toxicity, and selective
solvency.  These HFCs and PFCs have  100-year global warming potentials (GWPs) ranging from 890 to
7,40056 times that of carbon dioxide and have relatively low boiling points (50°C to 90°C) that contribute
to their inadvertent release to the atmosphere. The replacement solvent technologies used globally are
summarized in Table 5-1.

HFC solvents include HFC-4310mee, HFC-365mfc, and HFC-245fa.  Of these HFCs, HFC-4310mee is
the  most common  solvent cleaner replacement.  HFC-365mfc is used as an additive to form solvent
blends with HFC-4310mee, helping to reduce the cost of these products (Micro Care, 2002). HFC-245fa
is used in the aerosol solvent  industry (Honeywell,  2003).  Certain solvent applications, particularly
precision cleaning end-uses,  will continue to use HCFCs,  especially HCFC-225ca/cb (until the HCFC
phaseout takes place), and to a much lesser extent, PFCs and perfluoropolyethers (PFPEs).

This report analyzes three solvent  end-uses: metal, precision, and electronics cleaning.  Metal cleaning
involves the removal of contamination  such as oils, greases, and particulate matter from metal surfaces
during the  production of metal parts,  and the  maintenance and repair of equipment and machinery.
Electronics cleaning,  or defluxing, consists mainly of the  removal of flux residue that remains after a
soldering operation for printed circuit boards and other contamination-sensitive electronics applications.
Precision cleaning may apply to either  electronic components or metal surfaces and is characterized by
products that require a high level of cleanliness and generally have complex shapes, small clearances, and
other  cleaning challenges (UNEP,  1999a).  Examples of applications and products requiring  precision
cleaning include disk drives, gyroscopes, medical devices, and optical components.  Based on current
  Parties to the Montreal Protocol on Substances that Deplete the Ozone Layer (Montreal Protocol) agreed to phase
out all ozone depleting substances, including those used as solvents. In developed countries, the solvent industry
has phased out its use of ODS. Developing countries are scheduled to phase out these substances (in particular
CFCs and 1,1,1-trichlorethane) between 2008 and 2010.
56 7,400 is the GWP of perfluorohexane (CgF^), and is used in this report for estimating purposes as the GWP for
PFC/PFPEs.


                                                                                         -116-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
understanding of market trends, HFC emissions from the precision and electronics cleaning end-uses
dominate the GWP-weighted emissions from the solvents  sector.   The metal cleaning  end-use  has
primarily transitioned away from ODSs directly into alternatives or processes that do not use high global
warming potential chemicals.
Table 5-1 : General Overview of Solvent Technologies Used Globally
Solvent Classes
Chlorinated Solvents
HCFC Solvents (HCFC-225 ca/cb and HCFC-141b)
HFC Solvents (primarily HFC-4310mee)
RFC Solvents
Hydrofluoroether (HFE) Solvents
Hydrocarbons
Alcohol Solvents
Brominated Solvents
Methyl Siloxanes
Metal
X




X
X
X
X
Electronics
X
X
X
X
X
X
X
X
X
Precision
X
X
X
X
X
X
X
X
X
Alternative Cleaning Technologies
Aqueous Cleaning
Semi-Aqueous Cleaning
No-Clean Processes3
X
X
X
X
X
X
X
X

al\lo-Clean processes include low-solids flux or paste and inert gas soldering.
5.2        Baseline Emission  Estimates


5.2.1  Emission Estimating Methodology

Description of Methodology

A full description of the emission model used to calculate ODS substitute emissions from all sectors is
provided in Appendix A.  Specific  information on how the model calculates solvent emissions is
described below.

EPA uses a detailed Vintaging Model of ODS-containing equipment and products to estimate the use and
emissions of various ODS substitutes in the United States, including HFCs and PFCs. Emission baselines
from non-U.S. countries were derived using country-specific ODS consumption estimates  as reported
under the Montreal Protocol in conjunction with Vintaging Model output for each ODS-consuming end-
use sector.   For sectors where detailed  information was available, these data were incorporated  into
country-specific versions of the Vintaging Model to customize emission estimates.  In the absence of
country level data, these preliminary estimates were calculated by assuming that the transition from ODSs
to HFCs and other substitutes follows the same general substitution patterns internationally as observed in
the United States.  From this preliminary assumption, emission estimates were then tailored to individual
countries or regions by applying adjustment factors to U.S. substitution scenarios,  based on relative
differences in (1) economic growth; (2) rates of ODS phaseout; and (3) the distribution of ODS use across
end-uses in each region or country.

For solvents, input into the model is based on estimates  of the market share  attributable to each  solvent
and projected growth for the precision and electronics applications, provided by industry experts.
                                                                                         -117-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Emission Equations

Generally, the emissions model assumes that some portion of solvent use remains in the liquid phase and
is not emitted as gas.  Thus, emissions are considered "incomplete," and are  set as a  fraction of the
amount of solvent consumed in a year. For solvent applications, a fixed percentage of the new chemical
used in equipment is  assumed emitted in that year with the  remainder  of the used solvent reused or
disposed without being released to the atmosphere. The following equation calculates emissions  from
solvent applications:

                                         Ej=L* Qq
Where:

Ej  =  Emissions. Total emissions of a specific chemical in a given year j from use in solvent
       applications, by weight.
L  =  Percent Leakage. The percentage of the total chemical that is lost to the atmosphere, assumed to
       be 90 percent.
Qcj =  Quantity of Chemical. Total quantity of a specific chemical sold for use in solvent applications in
       the given year j, by weight.
j    =  Year of emissions.

Many solvent users have retrofitted their equipment  by adding emission control features that  have
successfully  demonstrated  proper solvent  conservation  and resulted  in an overall  lower solvent
consumption. Eventually, nearly all of the solvent consumed in a given year is emitted, as the solvent is
continuously reused through a distilling and cleaning process or through recycling, while a small amount
of solvent is disposed with the sludge that remains.  The model used for this analysis assumes that 90
percent of the solvent consumed annually is emitted to the atmosphere.

Regional Variations/A djustments

The following adjustment factor assumptions, specific to the solvent sector, were used to customize the
global emissions estimating methodology, described above, for solvents:

    •   PFC/PFPE solvents are assumed to be used only in the United States.  Therefore,  U.S. emissions
       estimated for these compounds were removed prior to calculating ratios for other countries.
    •   Emissions in EU-15 countries are  assumed to equal only  80 percent of the preliminary estimate
       developed using the ratio methodology defined in Appendix A, to reflect that NIK technology has
       taken a more significant market share in European countries (ECCP, 2001).  Consequently, the
       resulting EU-15 emission estimate was reduced by 20 percent.
    •   A 50 percent adjustment factor was applied to Countries  with Economies  in Transition (CEIT),
       European countries that are not members of EU-15, and developing  (Non-Annex I) countries.
       For  these countries,  the primary  barriers to the transition from ODS solvents to fluorinated
       solvents has been the high cost of HFC-4310mee and lack of domestic production (UNEP, 1999a;
       UNEP, 1999b).

5.2.2  Baseline Emissions

Table 5-2 displays total  HFC and  PFC emission estimates for the solvent sector.  In the United States,
HFC-4310mee is responsible for the majority of the country's projected ODS substitute solvent emissions
while PFC/PFPE emissions are assumed to decline linearly until they are phased out completely in 2025.


                                                                                          -118-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
U.S. emissions reflect  the  continued decline  of PFC/PFPE consumption as a result of restrictions,
enforced by the U.S. EPA's Significant New Alternatives Policy program, that limit PFC and PFPE use to
those applications where  these solvents have  been deemed necessary to meet performance or safety
requirements. U.S. solvent end-users that have historically used PFC/PFPEs are turning to other solvents,
including HFC-4310mee.

Baseline emissions for countries other than the United States do not contain PFCs; rather, the baseline
solvent emissions from these countries are entirely composed of HFCs.

Table 5-2: Baseline HFC and PFC Emission Estimates from Solvents (MMTCE)
Region
United States
Non-U.S. Annex 1
Non-Annex 1
Total
2005
0.45
0.56
0.06
1.07
2010
0.49
0.57
0.09
1.14
2015
0.52
0.57
0.10
1.20
2020
0.57
0.58
0.12
1.26
Note: Totals may not sum due to independent rounding.

5.3        Cost of HFC and  PFC Emission Reductions for Solvents


This section presents a cost analysis for achieving FiFC and PFC emission reductions from the emission
baselines presented  in Table 5-2 above.  All cost analyses  for the solvent emission  reduction options
assume a 10-year project lifetime. Each abatement option is described below.


5.3.1  Description and Cost Analysis of Abatement Options

Some HFC and PFC emissions from the solvent sector can  be eliminated or mitigated through several
technologies and practices.  The emissions and uses of these compounds can be reduced by retrofitting
equipment and improving containment of the solvents, introducing carbon adsorption technologies, and
replacing outdated equipment with more modern technologies.  Additionally, NIK technologies and
processes already used in many solvent markets world-wide employ semi-aqueous, aqueous, or no-clean
methods in place of solvents.  Ongoing research continues to identify low-GWP alternatives, including
low-GWP HFCs and hydrofluoroethers (HFEs)  that could replace high-GWP PFCs and HFCs.  Some
alternative  solvent  cleaning approaches use other organic solvents  including chlorinated solvents,
alcohols, petroleum distillates, and aliphatic solvents.

Three potential mitigation options are identified and analyzed in this report.  These are:

    •  Conversion to HFE solvents;

    •  Improved equipment and cleaning processes using existing solvents (Retrofit); and

    •  Aqueous and semi-aqueous NIK replacement alternatives.

Because limited technical and cost information is  available  on  fire suppression equipment, explosion
proof wiring, and other workplace controls,  flammable  organic  solvent alternatives, such as ketones,
ethers, and alcohols—which can potentially replace HFCs and PFCs—are not addressed in this analysis.

The  remainder of Section  5.3 describes each of these options in detail, and provides a discussion of
associated cost and emission reduction estimates.
                                                                                         -119-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Conversion to HFE Solvents

HFC and PFC solvents can be replaced by alternative organic solvents with lower GWPs that are making
headway in the market.  These alternative  solvents include low-GWP HFCs  and  HFEs, hydrocarbons,
alcohols, volatile methyl siloxanes, brominated solvents, and non-ODS chlorinated solvents.  For the
purpose of this analysis, commercially available HFE-7100 and HFE-7200 are used as proxies for the
alternative  solvents abatement option because they display materials compatibility properties similar to
HFCs and PFCs, a prime factor that has  led to their current success in the market.  Specifically, HFEs
have replaced PFCs, CFC-113, 1,1,1-trichlorethane, HFCs, and HCFCs in certain precision  cleaning
operations.  Many solvent users have successfully transitioned from  PFC solvents to HFC-4310mee and
HFEs in cleaning applications such as computer disk lubrication, particulate  cleaning, and in  cleaning
electronic  assemblies  after soldering.   HFEs and azeotropes of HFEs are also  viable substitute
replacements for HFC-43 lOmee in certain precision and electronics cleaning operations.

Because PFCs are specific to a small portion of the U.S. market only, and they are likely more expensive
than HFCs, costs for this analysis are calculated based on a transition  from HFC-43 lOmee to HFEs, rather
than from PFCs to HFEs.  PFC solvent users in the United States that apply  this  option may therefore
experience a cost savings.

The average GWP of the alternative solvents is 222.5 (HFE-7100 has a GWP of 390 and HFE-7200 has a
GWP of 55), compared to the HFC-43 lOmee GWP of 1300.57 Due to the lower average  GWP, this option
has a reduction efficiency of 82.9 percent (i.e., the difference of the GWPs divided  by the GWP  of HFC-
43 lOmee).  This analysis assumes that the technical applicability58 of this option is 100 percent of total
solvent emissions for each region (see Table 5-4).

HFE solvents are gaining acceptance in U.S. industry due to their availability, safety,  and  effectiveness
(Salerno, 2001); however, some uncertainty exists regarding the likelihood and ease with  which HFC-
43 lOmee users will convert to an HFE alternative solvent because of application-specific  requirements
(UNEP, 1999b).  The  incremental maximum market penetration of this option in the United  States is
assumed to increase from 10 percent  in  2005 to 60 percent in  2020, which addresses all  PFC solvent
emissions and more than half of the HFC solvent emissions in the United States, as shown in Table 5-4.

For all other countries, the incremental  maximum market penetration is  assumed to increase from 5
percent in 2005 to 25 percent in 2020, representing a slower adoption of this option and less reliance on
the use of fluorinated compounds compared to the assumed scenario for the United States (see Table 5-4).
This assumption is based on current market data that indicates that HFE solvents are available and being
used in the same regions where HFC solvents are being used (3M Performance Materials, 2003).
57 Although the GWP value for HFC-43 lOmee was taken from the IPCC Second Assessment Report (1996), that
report did not provide GWP values for either HFE; consequently this analysis uses the GWP values listed in the
IPCC Third Assessment Report (2001) for both HFEs.  The GWPs of HFEs are still being studied; for instance,
some analyses show the GWP of HFE-7100 to be approximately 300 (3M Performance Materials, 2003).
58 In this report, the term technically applicable refers to the emissions to which an option can be applied. Because
HFEs can be substituted for both HFCs and PFCs, they are technically applicable to all HFC and PFC solvent
emissions. Other factors will affect their application and the market penetration assumed in this analysis.  See
Appendix I for a glossary of terminology.


                                                                                            -120-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Cost and Emission Reduction Analysis

The following bullets describe the cost and emission inputs used to derive the final $/TCE for converting
to an HFE solvent, the results of which are presented in Section 5.4:

    •   One-time  Costs.   HFE solvents are  very similar to HFC-4310mee  in their key chemical
       properties such that existing equipment designed with low emission features can still be used.
       Although the need for minor adjustments, such as resetting the heat balance of the equipment,
       might arise, these modifications are  not  likely to amount to a substantial one-time  cost (ICF
       Consulting, 2003; 3M Performance Materials,  2003).  This analysis therefore assumes no one-
       time costs for converting to an HFE solvent.
    •   Annual  Costs.   According  to  industry experts, HFE  solvents have  roughly equal pricing
       structures compared to HFCs (3M  Performance Materials, 2003).  Therefore, this analysis
       assumes no annual costs are incurred when transitioning to an HFE solvent.
    •   Cost Savings.  Currently, this analysis does not assume a cost savings. A net cost savings may
       occasionally be experienced by end-users that choose HFE solvents  that are lower in density
       relative to HFC-4310mee.  For example, since  the same volume of solvent is used and solvents
       are sold on a mass basis, formulations blended with HFE-7200 may be lower in cost relative to
       formulations containing HFC-4310mee (3M Performance Materials, 2003).  Long-term savings
       may be realized when this option is combined with  retrofitting the  equipment, consequently
       reducing the costs per item cleaned. These potential cost savings are not analyzed here.
    •   Emission Reductions. For the purpose of this analysis, a one to  one  mass ratio between HFC-
       4310mee to an HFE alternative solvent is assumed. Therefore, for every one kilogram of HFC-
       4310mee avoided, the use of HFE solvents is estimated to reduce  emissions by 0.29 metric tons
       of carbon equivalent (TCE) (i.e., 1 kg  or 0.001 metric tons of solvent, multiplied by the difference
       in GWPs and the conversion factor of carbon to  carbon dioxide equivalents of 12/44).

Improved Equipment and Cleaning Processes Using Existing Solvents (Retrofit)

HFC and PFC prices  are much higher than those of historically used solvents, such as CFC-113 and
HCFC-141b. Attempts to reduce emissions, and hence save costs, have led to significant improvements
in the  containment technology of degreasing,  defluxing, and  other cleaning  equipment. Engineering
controls (such as  increasing freeboard height, installing freeboard chillers, and using automatic hoists),
improved containment, and other abatement technologies can reduce emissions of HFCs and PFCs used
in solvent cleaning (UNEP, 1999a; ICF Consulting, 1992).  For example,  some cleaning equipment that
uses HFC solvents is being retrofitted with higher freeboard height and  low-temperature secondary
cooling coils.  Likewise, it is possible to reduce emissions to a minimal level by implementing good
handling practices, such as reducing drag-out loses of solvent from systems by keeping the workload in
the vapor zone long  enough to drain and  dry  any entrapped or remaining solvent (UNEP,  1999a;
Petroferm, 2000).  It is also possible to minimize evaporative losses by improving the design of solvent
bath enclosures and of vapor recovery condensing systems (March Consulting Group, 1998 and  1999).

As shown in Table 5-3, retrofitting a vapor degreaser with an open-top area of 13 square feet,  combined
with proper operation and maintenance, can reduce emissions from a solvent process by as much as 46 to
70 percent, depending on the specific retrofit  methods chosen (Durkee, 1997).  For example, installing a
freeboard refrigeration device, sometimes referred to as a chiller (a set of secondary coils mounted in the
freeboard), and maintaining a freeboard ratio of 1.0 (to minimize diffusional solvent losses) can  reduce
emissions by 46 percent, while installing heating coils to produce superheated vapor along with installing
a chiller can reduce emissions by 70 percent.  For the purpose of this analysis, the reduction efficiency of
                                                                                          -121-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
the retrofit option is assumed to equal 70 percent, which can be achieved at a one-time cost of $18,000
dollars (see Table 5-3).

Table 5-3: Retrofit Techniques for Batch Vapor Cleaning Machine (Less than 13 Square Feet)	
 Retrofit Technique                                          Reduction Efficiency       One-Time Cost
 Freeboard Ratio of 1.0, Freeboard Refrigeration Device                     46%                $12,000
 Working Mode Cover, Freeboard Refrigeration Device                      64%                $17,000
 Superheated Vapor, Freeboard Refrigeration Device                        70%                $18,000
Source: Durkee, 1997.

For vapor degreasing machines that are larger than 13 square feet, the cost to retrofit increases; however,
the emission reduction efficiencies achieved are as high as 85 percent.  Furthermore, for larger operations
where there is more than one vapor degreaser, retrofit methods,  such as installing a carbon adsorber, can
be implemented to capture solvent vapor from the air for the entire facility. The reduction efficiency of a
carbon adsorber combined with installing heating coils and chillers has been estimated at 88 percent for
larger (i.e., greater than 13 square feet) vapor degreasers (Durkee, 1997).

In the United States, many enterprises have bought new equipment or retrofitted aging equipment to be in
compliance with the National Emissions Standard for Hazardous Air Pollutants (NESFfAP), which limits
emissions from degreasers using traditional  chlorinated solvents such as trichloroethylene.  Fluorinated
solvents such as HFCs are not covered by this  regulation; but, nonetheless, a number of companies using
HFCs and other non-chlorinated solvents have adopted NESFfAP -compliant solvent cleaning machines
because  of  the  economic,  occupational,  and environmental benefits  associated with  NESFfAP
specifications  (Durkee,  1997).   Although  solvent  users  are  likely to perform regularly  scheduled
maintenance and good handling practices to ensure the quality and lifetime of their investment,  previously
retrofitted or new equipment is assumed to already have a combination of emission reduction features that
retrofitting  offers, such as those techniques listed above.  Consequently, end users in the United States are
not expected  to benefit from this option in the future.  Thus, this analysis assumes that the incremental
maximum market penetration  will drop from  5 percent in 2005 to  zero in 2010 through 2020  (i.e., by
2010 and beyond, the solvent equipment in use will either already have been retrofitted or will not require
retrofitting, and the resulting lower emissions  are already incorporated into the baseline). The resulting
maximum market penetrations  are shown in Table 5-4.

Likewise, many countries in Europe have imposed stringent environmental and safety regulations  that
require the lowest level  of emissions attainable by solvent degreasing equipment.  Retrofit techniques
were either already implemented or simply  not required if the user had purchased new emission-tight
vapor degreasers.  Therefore, for Non-U.S. Annex I countries, the maximum market penetration for this
option is also assumed to be 5 percent in 2005 dropping to zero by 2010.

This analysis assumes that most solvent users in Non-Annex I  (developing) countries may consider the
equipment  retrofit option, as updating their equipment may be  preferred over investing in entirely new
units. Consequently, this region is assumed to slowly adopt these techniques such that in 2005, 5 percent
of the market adopts this  option, increasing at a slow,  steady rate to 15 percent in 2020 (see Table  5-4).

Cost and Emission Reduction Analysis

The following bullets describe the cost and emission inputs used to derive the final $/TCE for the retrofit
option, the results of which are presented in Section 5.4:

    •   One-time  Costs.  The costs of retrofitting  modern  or high quality  batch cleaning  equipment
        ranges from $12,000 to $18,000 for small units to approximately $20,000 or more for larger units.

                                                                                            -122-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
       To be conservative, this report assumes that an $18,000 investment is required to implement a
       retrofit on a vapor degreaser with an open-top area 13 square feet in size (Durkee, 1997).
    •  Annual Costs. This analysis assumes no annual labor costs. However, good handling practices,
       such as employee training and regular maintenance to reduce the risk of leaks, are encouraged on
       equipment that is retrofitted, and would likely lead  to an increase in operating  costs  (UNEP,
       2003).
    •  Cost  Savings.  According  to experts, HFC users that retrofit their equipment will most likely
       experience a drop in solvent costs per operating hour.  The cost savings will depend on the size of
       the cleaner and the specific  solvent conservation features installed. To account for cost savings in
       the model, the estimated cost of HFC-4310mee ($48.00 per kilogram) was used to represent the
       value of the solvent per kilogram  saved through retrofitting (ICF Consulting, 2003). The example
       from  Durkee (1997) that reports the emission  reduction of 70 percent estimates that 4,898.4
       pounds, or 2,221.9 kg, of solvent were still being emitted from the retrofitted cleaner.  Thus, it is
       estimated that 5,184 kilograms (i.e., 2,221.9 kg divided by 30% yields the total cleaner's mass of
       7,406.3 kg; a reduction of 70% is thus equal to 5,184 kg) of solvent per year can be avoided for a
       13 square foot unit.  This savings equates to close to $250,000 per year. Hence, investments in
       retrofit options frequently provide a profitable return for the end-user.
    •  Emission Reductions.  Assuming a retrofit on a vapor degreaser with an open-top area 13 square
       feet in size, annual emissions avoided are  estimated to be  5,184 kilograms of HFC-4310mee
       (Durkee, 1997), or 1,838 TCE (i.e., 5.184 metric tons HFC avoided, multiplied by the GWP of
       1300  and the conversion factor of carbon to carbon dioxide  equivalents of 12/44). The Durkee
       (1997) data used for this analysis are  consistent with industry experience.  For example, one
       anonymous user reduced annual  emissions  usage by close to  3,500 kilograms after retrofitting
       two of their solvent cleaners of 27 and  42  gallon capacities  (roughly equivalent to  vapor
       degreasers with an open-top area 5 to  7 square feet in size) (Ultrasonics, 2002).

Aqueous and Semi-Aqueous Not-In-Kind (NIK) Replacement Alternatives

In addition to the  emission reduction  approaches that use a combination of improved equipment and
cleaning practices, NIK technology  processes and solvent replacements can be used  to substitute for PFC-
and HFC-containing  systems.  In the  aqueous  process, a water-based cleaning solution is used as the
primary solvent and is usually combined  with a detergent to  remove  contaminants. In the semi-aqueous
process, the cleaning solution is an organic solvent that is blended  with  a  surfactant, making it water-
soluble.  An example of a solvent/surfactant blend would be a terpene/water combination blended with
glycol ethers  (UNEP, 1999a). The  reduction efficiency of NIK abatement options  is assumed to be 100
percent, since the HFC is completely replaced  by water and an organic solvent, combinations of which
have low to no global warming potential.

Many enterprises in the electronics, metal, and precision cleaning  end-uses  have already switched to
aqueous and semi-aqueous NIK cleaning  methods.  Both NIK processes have proven very successful for
large-scale metal cleaning where equipment and wastewater treatment  costs  are less of a concern due to
the large volumes  processed (UNEP, 1999a).  Aqueous cleaning technologies have  been available and
widely used for over 25 years and have replaced many electronics cleaning solvent  systems in developed
countries (Chaneski, 1997; UNEP, 1999a).  Semi-aqueous cleaning has also been available for years as a
popular option; however, it has since lost much of its initial promise in many developed nations for the
cleaning of electronic assemblies due to the expenses associated with the technology (UNEP, 1999a).

Because the NIK options are applicable to both the electronic and precision cleaning end-uses, the NIK
options are assumed to be  applicable to  100  percent of HFC solvent emissions, resulting in a technical
                                                                                          -123-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
applicability of 100 percent, for all regions except for the United States, where the technical applicability
is equivalent to the percent of HFC emissions out of total solvent emissions (see Table 5-4).

Technical limitations in the use of NIK technologies arising from issues such as  substrate corrosion, or
inadequate performance for applications with complex parts, can lead to reduced market acceptability.
The U.S. incremental maximum market penetrations for these options are assumed to be smaller than in
other regions to reflect that the U.S. market will likely prefer fluorinated solvents such as HFCs and HFEs
(see Table 5-4), and because a large percentage of the operations that can use aqueous and semi-aqueous
technologies already are doing so.   For Non-U.S. Annex  I and Non-Annex I regions, the maximum
market penetrations of these two NIK options are assumed to be similar to one-another from 2005 to
2020. NIK alternatives are  currently gaining market  share in European countries, a trend that is assumed
to continue for this region (ECCP, 2001).

NIK technologies  also  are assumed to be preferred  by some developing countries because  of their
perceived low costs.  For instance, aqueous cleaning is popular in China because of the very small cost
per kilogram of the non-fluorinated cleaning chemicals used, despite the high costs per unit cleaned that
result from newly introduced costs such as wastewater treatment.  Conversely, the availability of water,
the costs  associated with energy to dry the product, and  local wastewater treatment  regulations can
discourage companies in developing regions of the  world from considering this  option (UNEP, 2003).
For all regions, the semi-aqueous option is assumed to have slightly smaller market penetrations relative
to the aqueous cleaning option.

Cost and Emission Reduction Analysis

The following bullets describe the cost and emission inputs used to derive the final $/TCE for the aqueous
and semi-aqueous cleaning options, the results of which are presented in Section 5.4:

    •  One-time Costs. This cost analysis assumes that the incremental investment required to convert
       to a typical NIK process  is approximately $67,000 for aqueous cleaning and $8,000 for semi-
       aqueous cleaning (ICF Consulting, 1992).59

    •  Annual Costs. Annual increased operating costs are dependent on a variety of factors, including
       the cost of aqueous cleaning chemicals, electricity and/or other utilities, and the specific parts
       being cleaned.   In  general, the drying process is more energy-intensive for aqueous processes.
       Similarly, wastewater treatment and effluent monitoring can add to the cost of aqueous processes.
       In addition, as the geometry of the  parts increases in  complexity  (i.e., blind holes or other
       cleaning agent entrapment  sites), so too does  the drying operation,  which in  turn leads to
       increased costs.  Because cost savings have yet to be quantified for this analysis, which may
       offset  increased operating costs, this analysis does not assume annual  costs for this option;
       however, future work could be performed to investigate annual costs and cost savings realized.
    •  Cost Savings.  Due to the potential complexity of the process, this report assumes no cost savings
       for NIK technologies; however, many companies that have installed NIK systems have realized
       long-term cost  savings because annual solvent expenditures are significantly reduced (Chaneski,
       1997).
    •  Emission Reductions.  This analysis assumes that when converting to either NIK process, annual
       emissions avoided for a standard unit are 3,494 kilograms (ICF Consulting, 1992).  Assuming the
       HFC avoided is HFC-4310mee, this reduction equates to 1,239 TCE for both semi-aqueous and
       aqueous cleaning.
59 Costs taken from the 1992 study were updated to 2000 dollars.
                                                                                           -124-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
5.3.2  Summary of Technical Applicability and Market Penetration of Abatement
        Options
Table 5-4 summarizes the technical applicability and the maximum market penetration of the  solvent
options presented in the discussions above.  By 2020, it is assumed that these options can be applied to 75
percent of the baseline solvent emissions in  the  United  States,  55 percent of the  baseline  solvent
emissions in the Annex I countries, and 70 percent of the  baseline solvent emissions in Non-Annex I
countries.

Table 5-4: Technical Applicability and Incremental Maximum Market Penetration of Solvent Options
(Percent)3





Option
Conversion to HFE
Solvents
Retrofit
NIK Replacements
Semi-Aqueous
Aqueous
X
CU
§ CU
2 1
00 O O
^ z z
Technical
Applicability
(All Years)

100 100 100
90-98" 100 100
-
90-98 100 100
90-98 100 100
X
O
< g
00 C
^ <
00 O O
^ z z


2005

10 5 5
555
488
7 3 3
355
X
O
< g
00 C
^ <
00 O O
"^ Z. Z.


2010

30 10 10
008
8 15 15
355
5 10 10
X
CU
i x
< £
00 C
^ <
00 O O
"^ Z. Z.


2015

45 15 15
0 0 12
12 23 23
488
8 15 15
X
O
1 s
IS 1
00 O O
"^ Z. Z.


2020

60 25 25
0 0 15
15 30 30
5 W W
JO 20 20
aAssumed maximum market penetration of options is presented as a percentage of total sector emissions for which the options are technically
applicable. The baseline market penetration is assumed to be zero to assess the emission reductions possible due to increased use of each
option.
bThe percent of total emissions represented by MFCs varies by year. The technical applicability is 90 percent in 2005, 93 percent in 2010,96
percent in 2015, and 98 percent in 2020.

To calculate the percent of emission reductions off the total solvent baseline for each abatement option,
the technical applicability (Table 5-4) is multiplied by the market penetration value (Table 5-4), and by
the reduction efficiency of the option.  For example, to determine the percent reduction off the 2020
baseline for the conversion to HFE solvents in the United States, the following calculation is performed:

             Technical  Applicability X Market Penetration in 2020 X Reduction Efficiency =

                                    100% X 60% X 82.9% « 49.7%

Thus, using the assumptions in this analysis, converting to an HFE solvent could reduce approximately 50
percent of the U.S. emissions baseline in 2020. This figure, along with the other emission reduction
potentials, is shown in Table 5-5.
                                                                                               -125-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Table 5-5: Emission Reductions off the Total Solvent Baseline (Percent)




Option
Conversion to HFE Solvents
Retrofit
NIK Replacements
Semi-Aqueous
Aqueous



to
X
CD
<
to
^
|

X

-------
Analysis of Costs to Abate International OPS Substitute Emissions
5.5        Summary


Baseline global HFC and PFC emissions from solvents are estimated to grow from 1.07 to 1.26 MMTCE
between the years 2005 and 2020.  In 2020, U.S. emissions are assumed to account for approximately 45
percent of this  total, while  Non-U.S. Annex I and Non-Annex I countries are assumed to account for
approximately 46 percent and 9 percent, respectively. The highest absolute emissions growth, from 0.45
MMTCE in 2005 to 0.57 MMTCE in 2020, is expected to occur in  the United States. In relative terms,
however, the largest percent growth occurs  in Non-Annex I countries from  0.06 to 0.12  MMTCE, an
increase of 100  percent.

This analysis considers  three emissions mitigation  options for solvent use:  1) adoption of alternative
(HFE-7100 or HFE-7200) partially fluorinated solvents; 2) improved  system  design through retrofitting
solvent processes; and 3) conversion to NIK (aqueous and semi-aqueous replacements). The costs and
emission reduction benefits of each option were compared in each region.  For all regions, retrofitting
represents the most cost-effective option for reducing HFC emissions from the solvent sector with a cost
savings of $134.18 dollars per TCE at a 4 percent discount rate and zero percent tax rate, though by 2020
this option is assumed to be fully implemented in the baseline in all  Annex I countries. Converting to an
HFE solvent is  a cost-neutral option  for all regions.  Costs per TCE for the three regions are equivalent
because available data on costs for abatement technologies were not  scaled to reflect potential differences
in the  costs internationally.  Actual costs for abatement options for specific countries may vary and
subsequently  affect  these estimates.   Additional research is required to  determine actual  variability in
costs across regions.

In terms  of greatest emission  reductions,  converting to alternative  solvents represents the greatest
opportunity in the United States.  NIK aqueous technology and alternative solvents equally represent the
greatest emission reduction opportunities  for both Non-U.S. Annex I and Non-Annex  I countries.  By
2020, 0.73 MMTCE, or 58 percent of global baseline emissions from solvents, can be reduced at a cost
under $20/TCE.


5.6        References

3M Performance Materials. 2003. Written correspondence between industry technical expert John G.
Owens, P.E. of  3M Performance  Materials  and Mollie Averyt and Marian Martin Van Pelt of ICF
Consulting. October 27, 2003.

Chaneski, Wayne. 1997. Competing Ideas:  Aqueous Cleaning ~ The  Cost-Friendly Solution. Modern
Machine Shop. November 1997.

Durkee, J.B. 1997. Chlorinated Solvents NESHAP ~ Results to Date, Recommendations and
Conclusions. Presented at the International Conference on Ozone Layer Protection Technologies in
Baltimore, MD. November 12-13, 1997.

ECCP. 2001. Annex I to the Final Report on European Climate Change Programme Working Group
Industry Work Item Fluorinated Gases: ECCP Solvents Position paper provided by European
Fluorocarbon Technical Committee (EFTC).  European Climate Change Program. February 2001.
                                                                                          -127-

-------
Analysis of Costs to Abate International OPS Substitute Emissions
Honeywell. 2003. Genesolv® S-T: A new HFC-Trans Blend Based Solvent for Industrial Aerosol,
Specialty Cleaning, Flushing and Deposition. Honeywell Technical Bulletin. 2003. BJ-6108-3/03-XXXX.
Available online at .

ICF Consulting. 2003. Personal communication between solvent industry experts and William Kenyon of
ICF Consulting. October 7, 2003.

ICF Consulting. 1992. Cost of Alternatives to CFC-113 and Methyl Chloroform Solvent Cleaning for the
Safe Alternatives Analysis.  ICF Consulting. March 12, 1992.

March Consulting Group. 1998. Opportunities to Minimize Emissions ofHydrofluorocarbons (HFCs)
from the European Union: Final Report. March Consulting Group. 1998.

March Consulting Group. 1999. UK Emissions of HFCs,  PFC, and SF6 and Potential Emission Reduction
Options: Final Report. March Consulting Group. 1999.

Microcare. 2002. What is HFC-365 and What Does It Do? Microcare Marketing Services, Vertrel®.
2002. Available online at .

Petroferm. 2000. Solvent Loss Control. Petroferm Technical Bulletin. January 2000.

Salerno, Cheryl. 2001. The New Generation of Solvents:  Developmental Challenges Inspire Creative
Solutions. CleanTech. January 2001. Available online at  .

Ultrasonics. 2002. Personal communication between Carl Wolf of Ultrasonics and William Kenyon of
ICF Consulting. April 1, 2002.

UNEP. 2003. UNEP 2002 Report of the Solvents, Coatings, andAdhesives Technical Options Committee
(STOC): 2002 Assessment.  United Nations Environment  Programme, Ozone Secretariat.  January 2003.

UNEP. 1999a 1998 Report of the Solvents, Coatings, and Adhesives Technical Options Committee
(STOC): 1998 Assessment.  United Nations Environment  Programme, Ozone Secretariat.  April 1999.

UNEP. 1999b. The Implications to the Montreal Protocol of the Inclusion of HFCs andPFCs in the
Kyoto Protocol. United Nations Environment Programme, HFC and PFC Task Force of the Technology
and Economic Assessment  Panel (TEAP). October 1999.
                                                                                         -128-

-------