DRAFT REPORT

THE U.S. PHASEOUT OF HCFCS:

PROJECTED SERVICING DEMAND IN THE U.S.
AIR-CONDITIONING, REFRIGERATION, AND FIRE
SUPPRESSION SECTORS FOR 2020-2030

U.S. Environmental Protection Agency
Office of Air and Radiation
Stratospheric Protection Division
1200 Pennsylvania Avenue, NW
Washington, DC 20460

April 2018


-------
Questions concerning this report should be directed to:

Katherine Sleasman
Stratospheric Protection Division
U.S. Environmental Protection Agency
1200 Pennsylvania Avenue, NW (6205T)

Washington, D.C. 20460
1-202-564-7716 (phone)

1-202-343-2338 (fax)
sleasman.katherine@epa.gov

This report was prepared with support from ICF International under contract number
EP-BPA-16-H-0021.


-------
Background

Hydrochlorofluorocarbons (HCFCs) are a class of chemical compounds that deplete the stratospheric
ozone layer, increasing the chances of overexposure to ultraviolet (UV) radiation at the earth's surface.
Excessive UV radiation damages biological systems and causes malignant melanoma and other skin
cancers, cataracts and other eye damage, and harm to certain crops and marine organisms. Reversing
the course of stratospheric ozone depletion is crucial to protecting human and environmental health
worldwide. Under the Montreal Protocol on Substances that Deplete the Ozone Layer (Montreal
Protocol), a global agreement to protect the stratospheric ozone layer, all countries including the United
States have agreed to phase out HCFC consumption and production on set schedules. Full
implementation of the Montreal Protocol globally is expected to avoid more than 280 million cases of
skin cancer, approximately 1.6 million skin cancer deaths, and more than 45 million cases of cataracts in
the United States among individuals born between 1890 and 2100 (EPA 2015).

The HCFC Phaseout Schedule

Table 1 shows the HCFC consumption1 phaseout schedule that applies to the United States under the
Montreal Protocol. The upcoming milestone is the commitment to reduce HCFC consumption by 99.5%
below the baseline by January 1, 2020, with consumption from 2020-2029 restricted to the servicing of
air-conditioning (AC) and refrigeration equipment existing on January 1, 2020.

Table 1: U.S. HCFC Consumption Phaseout Schedule under the Montreal Protocol

Date

Control Measure

Maximum Consumption (ODP
Weighted MT)

January 1, 1996

Baseline set at 2.8% of the 1989 ODP-weighted
CFC consumption plus 100% of the 1989 ODP-
weighted HCFC consumption

15,240

January 1, 2004

35% reduction from the baseline

9,906

January 1, 2010

75% reduction from the baseline

3,810

January 1, 2015

90% reduction from the baseline

1,524

January 1, 2020

99.5% reduction from the baseline; consumption
restricted to the servicing of AC and refrigeration
equipment existing on January 1, 2020

76.2

January 1, 2030

100% reduction from the baseline

0

a An ozone depletion potential (ODP-)weighted metric ton (MT) takes into account each ozone depleting substance's relative contribution to
ozone depletion. One MT equals approximately 2,205 pounds (lbs.).

Table 2 shows the HCFC phaseout schedule contained in EPA regulations (40 CFR Part 82 subpart A),
which takes into account both Montreal Protocol and Clean Air Act requirements and adopts a "worst-
first" approach to phasing out HCFCs, restricting the most-ozone depleting HCFCs first.

1 Consumption is defined as production plus imports minus exports; production is defined as the manufacture of a controlled substance minus
amounts destroyed and amounts completely used as feedstock in the manufacture of other chemicals. Neither consumption nor production
includes amounts that are reused or recycled.

2


-------
Table 2: Detailed Regulatory HCFC Phaseout Schedule

Date

Restriction

January 1, 2003

• Ban on production and import of HCFC-141b.

January 1, 2010

• Ban on production and import of HCFC-22 and HCFC-142b except for on-
going servicing demand in equipment manufactured before January 1,
2010.a

January 1, 2015

• Ban on the production, import, and introduction into interstate
commerce or use of HCFCs except where the HCFCs are used as a
refrigerant in equipment manufactured prior to January 1, 2020, or
where HCFCs are used as a fire suppression agent for non-residential
applications.a

January 1, 2020

•	Ban on remaining production and import of HCFC-22 and HCFC-142b.a

•	Ban on production and consumption of all other HCFCs except for use in
servicing AC and refrigeration equipment manufactured before January
1, 2020.

January 1, 2030

• Ban on production and import of all HCFCs.3

a Exemptions apply, including exemptions for (1) HCFCs used in processes resulting in their transformation or destruction, or (2) HCFCs

that are recovered and either recycled or reclaimed.

Report Objective

This report evaluates the amount of HCFCs that would allow U.S. AC and refrigeration equipment
owners to service equipment using HCFCs between 2020-2029, consistent with the phaseout schedule
as summarized in Table 2. As such, the report only considers the markets for HCFC-123 and HCFC-124.
The report presents quantitative estimates of:

•	the projected number of units of equipment using HCFC-123 between 2020 and 2030 in the
United States;

•	HCFC-123 and HCFC-124 servicing demand for AC and refrigeration equipment that will be in use
after 2019; and

•	possible future recovery scenarios and estimates of future reclamation between 2020 - 2030.

EPA understands that the fire suppression market will likely continue to use HCFC-123 during at least
part of the period covered in the report and this sector cannot be sourced from new production or
consumption. Under current regulations at 40 CFR Part 82 Subpart A, only used, recovered and recycled
material, or HCFC-123 imported prior to January 1, 2020, can be used for this purpose in 2020 and
beyond. As such, the report also estimates the market demand for HCFCs for charging and servicing non-
residential fire suppression applications.

The remainder of the report is organized as follows:

•	Section 2 provides an overview of HCFC-123 and HCFC-124 use in the United States in AC,
refrigeration, and fire suppression.

•	Section 3 provides the projected market demand for HCFC-123 and HCFC-124.

3


-------
•	Section 4 provides projected recovery and reclamation scenarios for HCFC-123.

•	Appendix A: Methodology Used to Calculate Projected Servicing Demand

Section 2. HCFC-123 and HCFC-124 Uses in the United States

Historically, HCFCs have a variety of applications in the foam, aerosol, solvent cleaning, fire protection,
and sterilization industry sectors; however, their largest use is in the AC and refrigeration sector.

2.1	HCFC-123: AC and Refrigeration

Chillers regulate the temperature and humidity in offices, hotels, shopping centers, and other buildings.
There are two major categories of chillers—centrifugal and positive displacement. Centrifugal chillers
are centralized AC systems typically used in larger buildings. Positive displacement chillers are smaller
and may be water-cooled or air-cooled. There are three principle types of positive displacement chillers:
reciprocating, screw, and scroll, each of which is named for the type of compressor employed.
Historically, these chillers used CFC-11, CFC-12, CFC-114, and R-500 as refrigerants. Today, HCFC-22 and
HCFC-123 chillers are still in use, as are chillers using HFC-134a, R-407C, and R-410A. More recently,
newer alternatives, such as R-513A, R-514A, and HCFO-1233zd(E), have been listed by the Significant
New Alternatives Policy (SNAP) program as acceptable.

Industrial process refrigeration (IPR) systems are complex, customized systems used to cool process
streams in industrial applications in the chemical, food processing, pharmaceutical, petrochemical, and
manufacturing industries. The choice of refrigerant for a specific application depends on ambient and
required operating temperatures and pressures. IPR systems historically used CFC-11 and CFC-12. Today,
IPR systems with HCFC-22, HCFC-123, and R-401A (a component of which is HCFC-124) are still in use,
though this market may be transitioning to R-717 (ammonia), R-744 (carbon dioxide), R-448A, R-449A,
R-450A, R-513A, R-514A, and HCFO-1233zd(E), in addition to using HFC-134a, R-404A, R-407A, R-407C,
and R-410A.

2.2	HCFC-123: Fire Suppression

HCFC-123 is also used in fire-suppression applications, which can be divided into two categories:
streaming applications (i.e., portable fire extinguishers) that historically used halon 1211, and total
flooding applications that historically used halon 1301, or halon 2402 in limited applications (EPA 2006).
Streaming applications are used to protect against fires of limited size in commercial and industrial
facilities, (e.g., computer rooms, data and telecommunications centers, ship control rooms, art galleries,
libraries, warehouses, clean rooms, motor control rooms, and nuclear and oil/gas facilities), aerospace
and aviation (e.g., spacecraft, onboard aircraft, aircraft rescue and firefighting (ARFF) vehicles, and areas
on aircraft fields and the hangers where airplanes are parked and serviced, which are also known as
aircraft flightlines), and the military (e.g., ship control rooms and aircraft flightlines). To replace halon
1211 in streaming applications, the SNAP program has listed the use of in-kind clean agent substitutes
(e.g., C02, Halotron I, HFC-236fa, and other halocarbons such as the fluoroketones FK-5-1-12 (Novec
1230) and FK-6-1-14 (C7 Fluoroketone)) and not-in-kind substitutes (e.g., water, dry chemical, foam) as
acceptable. Approximately 75% of streaming applications in the United States in 2015 were using C02,
water, or dry chemicals; however, water and dry chemicals are not considered clean agents (i.e., does
not leave residue on equipment or in the protected enclosure after discharge). Where clean agent

4


-------
alternatives are preferred, there is still a sizeable installed base of halon 1211. Recycled halon 1211 is
also used to service these portable fire extinguishers (EPA 2018).

HCFC-123 is the primary constituent in Halotron I (also referred to as HCFC Blend B in EPA regulations).
Specifically, the blend Halotron I is used in commercial/industrial, maritime, and military applications in
the United States (Halotron 2008). After January 1, 2020, recovered and recycled or reclaimed HCFC-
123, as well as stockpiled material imported prior to 2020, can be used to meet fire suppression
demand. There are no known technical reasons that would prevent the recycling or reclaiming of HCFC-
123 to appropriate purity levels sufficient for use in fire suppression applications (UNEP 2016).
Furthermore, imports of used and/or recycled HCFC-123 may also meet demand similar to how the
United States manages used halon through the import petition process at 40 CFR Part 82.13(g)(2).

2.3 HCFC-124: AC and Refrigeration

HCFC-124 is minimally used as a refrigerant; its primary use as a refrigerant is in blends, mainly R-401A
which is used in industrial process and transport refrigeration equipment, as well as the blends R-409A
and R-414B which are used in commercial refrigeration applications. As a stand-alone refrigerant, it is
used in some niche applications that reach high condensing temperatures and as an alternative for CFC-
114 in some naval chillers. Its use in blends R-401A, R-401B, R-409A, R-414A, R-414B and others is
constrained because those blends also contain HCFC-22 and in some cases HCFC-142b. Starting in 2020,
HCFC-22 and HCFC-142b will not be produced or imported for use in these blends, although reclaimed
and previously produced or imported HCFCs will still be available for use. HCFC-124 is also a component
of R-416A, which does not contain any other ODS.

Section 3. Estimated Market Demand for HCFC-123 and HCFC-124

3.1. HCFC-123

As discussed above, HCFC-123 is used in the AC and refrigeration sector mainly in centrifugal chillers and
IPR as well as in the fire suppression sector. Average consumption of HCFC-123 over the past 5 years has
been stable at approximately 1,200 MT per year.2 This number is lower than the approximately 2,000
MT of annual consumption allowances provided during the same period.

As presented in Table 3, approximately 47,000 units of AC equipment using HCFC-123 are estimated to
be in use in 2020. The number of HCFC-123 chillers is projected to decrease by 42 percent between 2020
and 2030. Approximately 14,000 units of HCFC-123 IPR systems are estimated to be in use in 2020,
decreasing by approximately 37 percent by 2030. EPA's Vintaging Model (see Appendix A) does not
provide a disaggregated estimate of fire suppression equipment by specific product type, so this
information is not provided in Tables 3 and 4 below.

Table 3: Projected Number of HCFC-123 AC and Refrigeration Units in Operation (1000s of Units), 2020-2030

Equipment Type

2020

2021

2022

2023

2024

2025

2026

2027

2028

2029

2030

Chillers (AC)

47

45

43

41

39

37

35

33

31

29

27

IPR

14

13

13

12

12

11

11

10

10

9

9

Source: EPA (2018).

2 Based on average consumption in 2012-2016.

5


-------
The Vintaging Model assumes the average charge of HCFC-123 in centrifugal chillers to be 445 kg (980
lbs.). IPR has a modeled charge of approximately 618 kg (1,360 lbs.) of HCFC-123, an average to
represent the wide variety of such systems.

Table 4 presents the projected installed base of HCFC-123 in chillers and IPR in metric tons (MT).

Table 4: Projected Installed Base of HCFC-123 in AC, Refrigeration, and Fire Suppression Units (MT), 2020-2030

Equipment
Type

2020

2021

2022

2023

2024

2025

2026

2027

2028

2029

2030

Chillers (AC)

20,900

20,000

19,100

18,300

17,400

16,500

15,600

14,700

13,800

13,000

12,100

IPR

8,400

8,100

7,800

7,500

7,200

6,900

6,600

6,300

6,000

5,700

5,300

Total

29,300

28,100

26,900

25,800

24,600

23,400

22,200

21,000

19,800

18,700

17,400

Source: EPA (2018).

Projected AC and refrigeration demand for HCFC-123 was approximately 1,990 MT in 2015, decreasing
to 1,010 MT in 2019. Table 5 presents projected HCFC-123 demand for servicing equipment from 2020
through 2030. In 2020, approximately 560 MT of HCFC-123 is estimated for servicing AC and
refrigeration equipment. This estimate is projected to decrease to 320 MT by 2030. To estimate future
market demand for HCFC-123 fire extinguishers, EPA consulted with industry. Over the past several
years, demand has varied. The average is approximately 260 MT per year.3 For purposes of this report,
EPA is assuming that current fire suppression demand remains at this average level through 2030, and
the demand for fire suppression can be met with pre-2020 inventory of HCFC-123 and/or recycled or
reclaimed material.4

Table 5: Projected HCFC-123 Demand for AC, Refrigeration, and Fire Suppression Equipment (MT), 2020-2030

Equipment
Type

2020

2021

2022

2023

2024

2025

2026

2027

2028

2029

2030

Chillers (AC)

210

200

190

180

170

160

150

150

140

130

120

IPR

350

330

320

310

290

280

260

240

230

210

200

Fire

Suppression3

260

260

260

260

260

260

260

260

260

260

260

Total

820

790

770

750

720

700

670

650

630

600

580

Source: EPA (2018) and stakeholder input.

a Assumes no change from current demand. Is inclusive of charging new fire suppression equipment and servicing existing equipment.
Note: Numbers may not sum due to independent rounding.

3.2

HCFC-124

HCFC-124 is used in the refrigeration sector mainly in IPR and medium temperature retail food
refrigeration (small condensing units). Average consumption of HCFC-124 over the past 5 years was
250 MT per year.5 That number has varied significantly, but is reasonably consistent with the 2015-
2019 consumption allocation of approximately 200 MT. Reclamation of HCFC-124 is minimal
(averaging 2 MT per year), so servicing demand for equipment operating on HCFC-124 and blends
thereof is primarily met from consumption.

3	EPA has some information that suggests there could be stockpiling of HCFC-123 underway for fire suppression or for other uses and will
continue to examine the 2018 data.

4	This projection does not consider the role that alternatives could play in affecting future market demand for HCFC-123 in fire suppression.

5	Based on average consumption in 2012-2016.

6


-------
Because the market for refrigerants containing HCFC-124 is so small and is primarily for use in
equipment that was retrofitted from a CFC refrigerant, EPA is relying on recent refrigerant sales data
collected by the California Air Resources Board (CARB). The CARB data summary displayed in Table 6
shows that in California, the amount of HCFC-124 included in blends and as neat refrigerant sold in
2013 totaled 18 MT, which decreased to 12 MT in 2015, and further decreased to 8 MT in 2016. If use
were proportional to population, the California values would imply approximately 150 MT of HCFC-
124 for the entire United States in 2013, decreasing to 98 MT in 2015, and further decreasing to 65
MT in 2016 (CARB 2018).6 The 2013-2016 trend provides insight into the transition away from HCFCs
and the retirement of legacy CFC systems that were retrofitted to use an HCFC blend, and is
consistent with feedback from stakeholders.

Table 6: Summary of California Sales Data for HCFC-124 in Blends and Refrigerant (MT), 2012-2016

HCFC

2012

2013

2014

2015

2016

HCFC-124

43

18

18

12

8

Based on recent sales data in California, recent consumption, and feedback from stakeholders, EPA
estimates that annual demand for HCFC-124 is 100 to 200 MT for servicing AC and refrigeration
equipment in 2020, with a lower amount estimated for servicing in 2030.

Section 4. Projected Recovery and Reclamation Scenarios for HCFC-123

The Vintaging Model provides an estimate of how much HCFC-123 can be recovered from retiring
equipment for reuse. The modeled recovery scenario for HCFC-123 assumes a 90 percent recovery rate
for HCFC-123 chillers and IPR systems that have reached end of life. The Vintaging Model does not
project whether the recovered HCFC-123 is reclaimed (it could be stockpiled and/or used by the same
owner in other equipment); however, the amount recovered from retiring equipment provides an
estimate of the amount that may be available for use in other equipment, after it is reclaimed.

Figure 1 below shows how reported HCFC-123 reclamation, which could include material recovered
from retiring equipment or recovered during servicing events of equipment still in operation, compares
to the Vintaging Model's modeled recovery. Based on the comparison, the Vintaging Model projection
and actual amounts of reclaimed HCFC-123 both increase at a similar rate. However, the model projects
higher estimates of recovered HCFC-123, particularly after 2010, than the amount that was reclaimed.7
An increase in recovered HCFC-123 is expected in the future, because HCFC-123 systems have such long
lifetimes and many have not reached their end of life yet. As these appliances are retired, more HCFC-
123 will likely be available for recovery and reclamation given the low leak rates and large charge sizes.

6	Population data from the United States Census Bureau, available at

https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=PEP_2017_PEPANNRES&src=pt. Accessed on March 13,
2018.

7	Annual reclaim totals disaggregated by ODS are available at www.epa.gov/section608/summary-refrigerant-reclamation.

7


-------
Figure 1: Comparison of HCFC-123 Reported Reclamation and Modeled Recovery (MT), 2000-2017

400
350
300
250
200

•ITTjNM

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017
Reported Reclaimed	VM Modeled Recovery

Table 7 provides annual HCFC-123 reclamation data since 2009.

Table 7. Reported HCFC-123 Reclamation (MT), 2009 - 2017



2009

2010

2011

2012

2013

2014

2015

2016

2017

Reclamation

198

145

152

145

202

170

181

188

271

As noted above, while EPA expects the amount of HCFC-123 that is recovered for reuse to rise, that
number is higher than what is reclaimed in most years. Table 8 below shows an estimate of future
reclamation values based on recent trends in HCFC-123 reclamation shown in Table 7. The range in each
year is calculated based on average growth in HCFC-123 reclamation between 2009 and 2017 on the low
end of the range, and 2012 and 2017 on the high end.

Table 8: HCFC-123 Expected Reclamation (MT), 2020-2030



2020

2025

2030

Expected

Reclamation

(Extrapolated)3

300-350

340-480

390-610

a Linearly extrapolated HCFC-123 reclamation data from 2009 to 2017 on the low end of the range, and 2012-2017 on the high end.

Table 9 shows the total estimated market demand for HCFC-123 for all AC, refrigeration, and fire
suppression uses, and the expected recovery and reclamation of HCFC-123.

8


-------
Table 9: HCFC-123 Modeled Supply and Demand (MT), 2020-2030



2020

2025

2030

Estimated Demand

820

700

580

Modeled Recovery3

970b

1,060

1,120

Expected

Reclamation

(Extrapolated)0

300-350

340-480

390-610

Source: EPA (2018).

a HCFC-123 recovery estimates are modeled from the Vintaging Model. These estimates do not reflect potential increases in recovery
associated with the increased equipment estimates for HCFC-123 fire suppression systems.

b There is a large modeled increase in HCFC-123 recovery in 2018 when the first vintages of HCFC-123 chillers (assumed to have entered the
market in 1992) reach end-of-life. Prior to 2018, there was a minimal amount of recovery from older HCFC-22 units that were replaced with
HCFC-123. For 2015-2017, EPA estimated 330 MT of HCFC-123 was available for recovery and reclamation.

c Linearly extrapolated HCFC-123 reclamation data from 2009 to 2017 on the low end of the range, and 2012-2017 on the high end.


-------
Appendix A: Methodology Used to Calculate Projected Servicing Demand

This appendix outlines the methodology used to calculate the projected servicing demand and new
chemical demand of refrigeration, AC, and fire suppression equipment using HCFC-123 and HCFC-124.
This appendix contains two sections:

•	Section 1 provides an overview of
EPA's Vintaging Model (EPA
2018), which was used to
estimate the units of equipment
using HCFC-124 for servicing
demand and HCFC-123 for new
equipment demand and servicing
demand beyond 2020.

•	Section 2 discusses the limitations
to the servicing projections
presented in this report.

1. EPA's Vintaging Model

EPA's Vintaging Model was developed as
a tool for estimating the annual chemical
emissions from industrial sectors that
have historically used ODS in their
products. Emissions are estimated from
the following end-use sectors: 1) Air-
Conditioning and Refrigeration; 2) Foams;
3) Aerosols; 4) Solvents; 5) Fire
Suppression; and 6) Sterilants. Within
these sectors, there are over 65
independently modeled end-uses. The
model requires information on the
market growth for each of the end-uses,
as well as a history and projection of the
market transition from ODS to
alternatives. As ODS are phased out, a
percentage of the market share originally
filled by the ODS is allocated to
substitutes.

BoxA-1: Developing and Maintaining EPA's Vintaging Model

The Vintaging Model synthesizes data from a variety of sources,
including:

•	EPA's ODS Tracking System and submissions to the SNAP
program, both maintained by the U.S. EPA Stratospheric
Protection Division;

•	Published literature from the United Nations UNEP Technical
Options Committees, the Alternative Fluorocarbons
Environmental Acceptability Study (AFEAS), and those
provided in industry-related and EPA conference proceedings;
and

•	Numerous representatives at companies and trade
associations, such as the Alliance for Responsible Atmospheric
Policy, the Air-Conditioning Heating and Refrigeration Institute
(AHRI), the Association of Home Appliance Manufacturers
(AHAM), and the Alliance of Automobile Manufacturers.

In some instances, the unpublished information that EPA uses in the
model is claimed as Confidential Business Information (CBI). The
annual emissions inventories of chemicals are aggregated in such a
way that CBI cannot be inferred.

The Vintaging Model is continually updated to improve assumptions
and modeling techniques and refine inputs based on information
received from these sources. Since 2014, the AC and refrigeration
sectors of the Vintaging Model have been updated multiple times. In
2017, an update to the centrifugal chillers end-uses had an impact on
the projected service, leak, and new chemical demand, as well as
disposal recovery of HCFC-123.

EPA shares the revised assumptions and results of model
improvements with industry through the annual Inventory of U.S.
Greenhouse Gas Emissions and Sinks and other publications such as
this one as well as through presentations, such as those given at the
March 2014 Spring Meeting of the Halon Alternatives Research
Corporation (HARC). In 2017, a peer review was conducted on end-
uses within the refrigeration/AC and fire suppression sectors.

The model, named for its method of

tracking the emissions of annual "vintages" of new equipment that enter into service, is a "bottom-up"
model. This means it models the consumption of controlled ozone-depleting substances and their
substitutes based on:

• Estimates of the quantity of equipment or products sold, serviced, and retired or retrofitted
each year, and

10


-------
• The quantity of the chemical required to manufacture and/or maintain the equipment.

The model makes use of this market information to build an inventory of in-use stocks of equipment and
quantities of ODS/ODS substitutes in each of the end-uses.

Emissions are estimated by applying annual leak rates, service emission rates, and disposal emission
rates to consumption data for each vintage of equipment. Emissions from AC and refrigeration
equipment are split into two categories: emissions during equipment lifetime and disposal emissions.
The first category includes the amount of chemical leaked during equipment operation and the amount
of chemical emitted during service. Consumption required to service or refill equipment is driven by the
need to replace such losses, and therefore, emissions during the lifetime of equipment are equal to
consumption for servicing (since it is assumed that all leaked refrigerant is replaced). Emissions, and
therefore, consumption from leakage and servicing can be expressed as follows:

ESj = (la + Is) x I QCj i+i for i=l->k

Where:

ESJ = Emissions from Equipment Serviced. Emissions in year j from normal leakage and

servicing of equipment.

Ia = Annual Leak Rate. Average annual leak rate during normal equipment operation

(expressed as a percentage of total chemical charge).

Is =	Service Leak Rate. Average leakage during 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,
i =	Counter, runs from 1 to lifetime (k).

j=	Year of emission.

k =	Lifetime. The average lifetime of the equipment.

The assumptions used in this calculation range by equipment, refrigerant, and vintage, reflecting that as
new technologies replace older ones, improvements in their leak, service, and disposal emission rates
are assumed to occur.

Estimates from EPA's Vintaging Model are often cross-checked with actual historical data from EPA's
ODS Tracking System, which tracks actual ODS production and consumption (including import and
export) by U.S. companies, and to data reported under the Greenhouse Gas Reporting Program, which
implements 40 CFR Part 98.

2. Limitations and Caveats

This analysis utilized the best data available from various sources. Nonetheless, when making
projections several assumptions are required. The following caveats are noted below.

a. Recycled or Reclaimed Material

EPA's Vintaging Model was used to inform the quantities of HCFC-123 from existing (recycled or
reclaimed) sources that can meet post-2020 servicing demand. Throughout an equipment's lifetime, the
Vintaging Model assumes a loss rate that represents the percent of the total charge that leaks in a given

11


-------
year plus the amount, on an annual basis, emitted at service. Because the amount lost from leaks and
servicing is annualized, equipment is assumed to reach the end of its lifetime with a full charge.8 EPA's
Vintaging Model then applies a "recovery rate," which refers to the percent of total charge of the
equipment that is recovered and reused at the time of disposal. These recovery rates represent
averages, intended to capture the range of possible practices occurring at disposal.

Differences between modeled and actual equipment lifetimes, recovery rates, and charge sizes can
affect estimated recovery. For example, the Vintaging Model predicts a significant increase in modeled
recovery of HCFC-123 in 2018, because that is the first year the Vintaging Model assumes HCFC-123
chillers are retired (i.e., 25 years after installation). A similar increase in actual HCFC-123 recovery is
expected in the future as HCFC-123 systems are retired.

The model aggregates the quantities recovered but does not distinguish the "pool" of available material
(e.g., refrigerant) between quantities that are reclaimed versus those that are possibly recycled and
reused. The model then assumes that the entire pool of recovered material re-enters the market within
the same year. The model assumes that any additional demand for material above the estimated
amount recovered is met by virgin manufacture. The recovery pool and the remaining virgin
manufacture are evaluated only at the most aggregate level, across all end-uses, and not at the end-use
level, as the model does not differentiate between virgin and recycled material when calculating
demand for each end-use. This model attribute reflects a more realistic scenario in that reclaimers are
not likely to only sell back to the end-use market sector from which the used material originated (e.g.,
chillers); rather, reclaimed material can be retailed to the overall market for each specific HCFC.

The model does not consider the quantity of material that companies send off for destruction after
equipment is decommissioned. Although the quantities of destroyed material are very small, they are
not subtracted from the recovered pool, so the quantity available for reuse may be slightly
overestimated in the model. Additionally, the model does not account for any stockpiling of recovered
material beyond a one-year timeframe. To the extent that stockpiling has occurred over the last few
years, the quantity of recovered material modeled as re-entering the market may be overestimated in
earlier years (i.e., when material is banked), and the quantity modeled as re-entering the market in later
years may be underestimated (i.e., when the accumulated stockpile is accessed as a source).

To address these potential limitations in modeled demand and recovery, EPA has presented both a
modeled recovery number, as well as an estimate based on recent reclamation trends to support HCFC-
123 demand assumptions.

b. Date of Retirement and Date of Installation of HCFC-123 and HCFC-124 Units

The model assumes that the entire vintage of HCFC-123 and HCFC-124 units within a given end-use sold
or shipped in a given year are being installed and retired in the same year, whereas, in reality some of
these units may instead be installed a year or two after shipment, starting their lifetime and servicing
demand a year or two after the shipment date. Given the use of an average lifetime in the model,
variations of a year or two between shipment and installation date should fall within "normal"
fluctuations in lifetime seen in actual equipment use. Furthermore, equipment operating conditions and

8 For the majority of equipment types, the assumption that equipment contains a full charge at the end of life is theoretically applicable. To
ensure proper and continual functioning of equipment, homeowners and businesses typically have their AC and refrigeration systems serviced
regularly. Technicians will check these systems for leaks using proper techniques and if a loss of refrigerant is found, will refill the system to
ensure it is functioning efficiently after repairing any leaking seals or damaged components.

12


-------
owner decisions may limit or extend the actual time from equipment installation until retirement. For
these reasons, the model is not necessarily accurately reflecting the actual start and end date for
servicing demand for this subset of equipment. Instead, the model uses an average lifetime for each
unique equipment type. This average is intended to account for units that are used for a few years less
than the assumed useful life as well as those that are used well beyond their assumed useful life.

13


-------
References

California Air Resources Board. 2018. Memorandum: Preliminary 2013, 2014, 2015, and 2016 Sales and
Distribution Data from the California Air Resources Board's Refrigerant Management Program.

EPA. 2003. Federal Register. "Protection of Stratospheric Ozone: Allowance

System for Controlling HCFC Production, Import and Export." Volume 68, Number 13. January 23, 2003.
40 CFR, Part 82. Available at http://www.epa.gov/ozone/title6/phaseout/68fr2819.pdf.

EPA. 2006. Global Mitigation of Non-C02 Greenhouse Gases. EPA Report 430-R-06-005. Washington,
DC: U.S. Environmental Protection Agency, Office of Atmospheric Programs.

EPA. 2009. Federal Register. "Protection of Stratospheric Ozone: Adjustments to the Allowance System
for Controlling HCFC Production, Import, and Export." Vol. 74, No. 239. Tuesday, December 15, 2009. 40
CFR Part 82. Available at http://www.gpo.gov/fdsys/pkg/FR-2009-12-15/pdf/E9-29569.pdf

EPA. 2011. Federal Register. "Protection of Stratospheric Ozone: Adjustments to the Allowance System
for Controlling HCFC Production, Import, and Export." Vol. 76, No. 151. Friday, August 5, 2011. 40 CFR
Part 82. Available at http://www.gpo.gov/fdsys/pkg/FR-2011-08-05/pdf/2011-19896.pdf.

EPA. 2014. "Servicing Tail Report: The U.S. Phaseout of HCFCS: Projected Servicing Need in the U.S. Air-
Conditioning, Refrigeration, and Fire Suppression Sectors Updated for 2015 to 2025," Available at
https://www.regulations.gov/document?D=EPA-HQ-QAR-2013-0263-0124

EPA. 2015. "Updating Ozone Calculations and Emissions Profiles for Use in the Atmospheric and Health
Effects Framework Model," available at https://www.epa.gov/ozone-layer-protection/updating-ozone-
calculations-and-emissions-profiles-use-atmospheric-and-health

EPA. 2018. U.S. EPA Vintaging Model. Version VM IO_v4.4_03.16.18.

UNEP. 2016. Technology and Economic Assessment Panel. Volume 1. TEAP Decision XXVI1/5 Working
Group Report: Issues Related to the Phase-out of HCFCs. Available at: http://conf.montreal-
protocol.org/meeti ng/oewg/oewg-

38/presession/Background%20Documents%20%20TEAP%20Reports/BDN WG XXVII-
5 report June2016.pdf

UNEP. 2018. Technology and Economic Assessment Panel. Volume 1. TEAP Decision XXIX/9 Working
Group Report on Hydrochlorofluorocarbons and Decision XXVI1/5. Available at: http://conf.montreal-
protocol.org/meeti ng/oewg/oewg-

40/presession/Background%20Documents%20are%20available%20in%20English%20only/TEAP-
DecXXIX9-WG-Report-March2018.pdf

14


-------