Refrigeration system performance using alternative refrigerants

Refrigeration system performance using alternative refrigerants

MJ145X Degree project in the sustainable energy engineering

PETER ERIKSSON

Bachelor’s Thesis at SCI Supervisor: Pavel Makhnatch

Examiner: Catharina Erlich

München, 2015-06

Abstract

The European Parliament has shown its legislative capabil- ities by adopting a tough approach on the reduction of fluo- rinated gases (F-gases). The EU Regulation No 517/2014, will phase out F-gases with high global warming poten- tial (GWP) within a well-defined step down time schedule.

This will affect refrigerants which are commonly used in commercial cooling applications.

While the time schedule implies existence of replace- ment refrigerants, the market continues to develop alter- natives for the refrigerants already in deployment. Though the R404A,which is a commonly used refrigerant in station- ary cooling applications, will be prohibited of use 1 Jan- uary 2020. In this paper, the two alternatives R448A from Honeywell and R449A from DuPont were compared to the R404A baseline, within a theoretical model with empirical compressor and system input data.

The key points of comparison were cooling capacity,

compressor discharge temperature, coefficient of performance

(COP) and total equivalent warming impact (TEWI). The

outcome of the model showed a decrease in cooling capacity

for both alternatives, as well as an overall increase in com-

pressor discharge temperature. For low evaporator temper-

atures, a decrease in COP was present and vice versa for

high evaporator temperatures, both for medium and high

condenser temperature. However, the TEWI for the both

refrigerants, showed a decrease in carbon dioxide (CO 2 )

equivalent emissions during the refrigerant system lifespan,

running on both R448A and R449A, regardless of which

European country the electricity was produced in.

Kylsystems prestanda vid alternativa kylmedier

Det Europeiska Parlamentet har visat sin lagstiftande för- måga i och med antagandet av en tuff förordning om redu- cering av F-gaser. EU Förordningen Nr 517/2014 kommer att fasa ut flourinerade gaser med högt GWP värde och det med en väl definierad nedtrappningsperiod. Detta kommer bland annat att påverka köldmedier som ofta förekommer i kommersiella kylapplikationer.

Medan tidsschemat förutsätter redan existerande ersät- tare fortsätter marknaden att utveckla alternativ för de köldmedier som idag är i bruk, samtidigt som användan- det av det vanligt förekommande kylmediet R404A kom- mer att förbjudas i och med 1 januari 2020. I den här rap- porten jämförs de två alternativa kylmedierna R448A från Honeywell och R449A från DuPont i en teoretiskt modell, med empiriska data för kompressor- och systemvariabler, där kylmediet R404A används som referens.

Huvudsakligen jämfördes kyleffekt, kompressorns ut- strömningstemperatur, COP och TEWI. Resultaten från modellen visade en minskad kyleffekt för båda de båda al- ternativen, såväl som en ökad utströmningstemperatur för kompressorn. Låga evaporatortemperaturer resulterade i en minskning av COP och vice versa för höga evaporatortem- peraturer, både för mellan och hög kondensortemperatur.

Dock konstaterades för TEWI, en minskning av de sam-

manlagda CO 2 ekvivalenta utsläpp under kylsystemets livs-

längd, under drift på både R448A och R449A oavsett i vil-

ket Europeiskt land elektriciteten producerats i.

Contents

List of Tables 4

List of Figures 4

1 Introduction 7

1.1 Background . . . . 9

1.1.1 First generation 1830 - 1930 . . . . 9

1.1.2 Second generation 1930 - 1990 . . . . 9

1.1.3 Third generation 1990 - 2010 . . . . 10

1.1.4 Fourth generation 2010 - . . . . 11

2 The vapor compression cycle 13 2.1 Refrigerant theory . . . . 14

2.1.1 Thermophysical properties . . . . 14

2.1.2 Toxicity . . . . 15

2.1.3 Flammability . . . . 15

2.1.4 Classification . . . . 16

2.1.5 Saturated liquid and vapor . . . . 17

2.1.6 Temperature glide . . . . 18

2.1.7 Ozone depleting potential . . . . 19

2.1.8 Global warming potential . . . . 20

2.1.9 Total equivalent warming impact . . . . 21

2.2 New regulations and previous research . . . . 22

2.2.1 R404A . . . . 22

2.2.2 R448A . . . . 23

2.2.3 R449A . . . . 24

3 Problem definition and goals 25 3.1 Focus . . . . 26

3.2 Expectations . . . . 26

3.3 Goals . . . . 26

4 Method 27

4.1 Limits . . . . 27

4.4 Concept . . . . 29

4.4.1 Condenser . . . . 30

4.4.2 Evaporator . . . . 31

4.4.3 Compressor . . . . 32

4.4.4 Excecution . . . . 33

4.5 Outputs . . . . 33

4.5.1 Isentropic efficiency . . . . 33

4.5.2 Flow rates . . . . 34

4.5.3 Discharge temperature . . . . 34

4.5.4 Cooling capacity . . . . 34

4.5.5 Volumetric cooling capacity . . . . 34

4.5.6 Coefficient of performance . . . . 34

4.5.7 Total equivalent warming impact . . . . 35

5 Results and discussion 37 5.1 Isentropic efficiency . . . . 37

5.2 Flow rates . . . . 39

5.3 Discharge temperature . . . . 40

5.4 Cooling capacity . . . . 40

5.5 Volumetric cooling capacity . . . . 41

5.6 Coefficient of performance . . . . 42

5.7 Sub cooling & Super heating . . . . 44

5.8 Reference system effects . . . . 45

5.9 System modifications . . . . 47

5.10 Inaccuracies . . . . 47

5.11 Improvements . . . . 47

6 Conclusions 48

Bibliography 50

Acronyms

AR1 first assessment report. 20 AR5 fifth assessment report. 20, 23

ASHRAE American Society of Heating, Refrigerating and Air-conditioning Engi- neers. 15, 16, 22–24

ASHRAE 34 ASHRAE standard designation and safety classification of refriger- ants. 16, 17, 24

CFC chlorofluorocarbon. 10, 11, 18, 19

CO 2 carbon dioxide. 8, 11, 12, 15, 20–22, 36, 46, 47 COP coefficient of performance. 16, 23, 24, 35, 45, 48

COP ₂ cooling coefficient of performance. 5, 34, 35, 42–46, 48, 49 F-gas fluorinated gas. 7, 12

GHG greenhouse gas. 7, 11, 25

GWP global warming potential. 7, 8, 11, 16, 18, 20–26, 35, 48 GWP ₁₀₀ GWP integrated over 100-years. 18, 20, 22–24, 26, 35 HC hydrocarbon. 10–12

HCFC hydrochlorofluorocarbon. 10, 11, 26 HFC hydrofluorocarbon. 10, 11, 16, 22 HFO hydroflouroolefine. 11, 12, 16, 23

IPCC Intergovernmental Panel on Climate Change. 20

ISO international organization for standardization. 15

LFL lower flammability limit. 15, 17 MAC mobile air conditioner. 11

ODP ozone depleting potential. 7, 11, 18, 19, 22, 23 ODS ozone depleting substance. 10

RF radioactive forcing. 20

TEWI total equivalent warming impact. 21, 22, 35, 46, 47, 49 UFL upper flammability limit. 15

UN United Nations. 10, 11

UNEP United Nations Environment Programme. 20

UNFCC United Nations Framework on Climate Change. 11, 20

Glossary

E _annual annual energy consumption. 36, 46 L _annual annual leakage rate. 35

T c condenser temperature. 5, 31, 37–45 T _dis compressor discharge temperature. 34, 46 T _e evaporator temperature. 5, 32, 37, 44, 45 α recovery destruction recovery rate. 35 β indirect emission factor. 36

E ˙ t,c compensated compressor power. 36, 46 E ˙ _t compressor power. 35, 36, 46

Q ˙ 2 cooling capacity. 34, 36, 46 V ˙ ₂ volumetric flow rate. 34 m mass flow rate. 34, 35 ˙

η _is compressor isentropic efficienct. 33, 34, 46

h _1k,is compressor outlet enthalpy at isentropic compression. 34 h 1k compressor outlet enthalpy. 34, 35, 45

h _2k compressor inlet enthalpy. 34, 35, 45 h s expansion valve enthalpy. 34, 35, 45 m refrigerant charge mass. 35

n system life expectancy. 35, 36

q _vol volumetric cooling capacity. 34

v _2k compressor inlet specific volume. 34

EN12900 Standardized polynomial for calculation of compressor consumption and mass flow rate. 29, 33, 34, 39, 44, 48

evaporator A component in which the refrigerant enters as mixture of vapor and liquid and exits as pure vapor.. 13, 14

volatility The measure of a substance tendency of evaporation. 16, 18, 19

List of Tables

2.1 Refrigerant classification . . . . 16

2.2 European Parliament regulation 517 Annex III . . . . 22

2.3 Refrigerant properties . . . . 23

2.4 Refrigerant mixture in percentage by weight . . . . 24

4.1 Input variables for system components . . . . 28

4.2 Refrigerant mole fraction mixture properties . . . . 29

4.3 The component input data from a reference system in deployment. . . 30

5.1 Results for the reference system . . . . 46

List of Figures 2.1 The vapor compression cycle, enthalpy to pressure . . . . 13

2.2 Refrigerant enthalpy by pressure diagram . . . . 17

2.3 Azeotropic behavior of a mixture . . . . 18

2.4 How decay times affect GWP . . . . 21

4.1 Nonlinear temperature glide of R448A at p = 400kP a . . . . 32

4.2 Model code workflow . . . . 33

4.3 Emissions electricity production 2009, CO ₂ /kWh in grams (EEA 2011). 36 5.1 Compressor isentropic efficiency, T _c = 35 ^◦ C. . . . 37

5.2 Compressor isentropic efficiency, T _c = 55 ^◦ C. . . . 38

5.3 Mass flow rate, T _c = 35 ^◦ C. . . . . 39

5.4 Volumetric flow rate, T _c = 35 ^◦ C. . . . 39

5.5 Discharge temperature, T _c = 35 ^◦ C. . . . 40

5.6 Cooling capacity, T _c = 35 ^◦ C. . . . 41

5.7 Cooling capacity, T _c = 55 ^◦ C. . . . 41

5.8 Volumetric cooling capacity, T _c = 35 ^◦ C. . . . 42

5.9 Coefficient of performance, T _c = 35 ^◦ C. . . . 42

5.10 Relative coefficient of performance, T _c = 35 ^◦ C. . . . 43

5.11 Coefficient of performance, T _c = 55 ^◦ C. . . . 43

5.12 Relative coefficient of performance, T _c = 55 ^◦ C. . . . 44

5.13 Sub cooling effects on cooling coefficient of performance (COP ₂ ), T _c = 35 ^◦ C T _e = −5 ^◦ C. . . . 44

5.14 Super heating effects on COP ₂ , T _c = 35 ^◦ C T e = −5 ^◦ C . . . . 45

5.15 Impact on TEWI after change to alternative refrigerants. . . . . 46

Pavel Makhnatch Sara Knight Carl Berglund

and

The CoolProp community

Chapter 1

Introduction

In all natural processes, heat moves from a warm source to a colder ambient; it’s one of the basic laws in thermodynamics. To move heat in the opposite direction, from a cold source to a warmer ambient, we need to utilize a refrigerating system.

By moving heat we can extend the storage time of our groceries and control the temperature in the spaces we live and work in.

The most common type of refrigerating system is the vapor compression cycle, a system that in its most simple setup is built up by four main parts: a compressor, two heat exchangers and an expansion valve. Through the systems parts flows a compound called refrigerant, which allows for heat to be transported between the two heat exchangers (Granryd et al. 2005).

The many parts of the vapor compressor cycle, results in a lot of joints where the different parts must be tightly connected to prevent the refrigerant from leaking out. The leakage can be reduced, but very seldom to no leakage at all. This means that the refrigerant successively exits the refrigeration system.

The leaked refrigerants had historically questionable properties such as toxicity, flammability or both. These properties combined with refrigeration domestication, were the incentives for the refrigerant development. After nearly 100 years we are now entering what’s called the fourth generation (Calm 2008). A generation that inherits the knowledge of toxicity, ozone depleting potential (ODP), GWP and a certain future of tough regulations.

A new refrigerant mustn’t be put on the market without sufficient knowledge to assure that it wont make way for a future crisis. Looking at the track record, the first generation was directly dangerous, the second generation damaged the ozone layer, and the third gave us very powerful greenhouse gases (GHGs).

On 16 April 2014, the European Union released its regulation No 517/2014 (EU

2014) taking an aggressive approach on reducing the production of F-gases. It’s

replacing its predecessor released in 2006 which turned out to be less effective than

expected (EIA 2011). The new regulation is equipped with a step down schedule

for the new fluorinated refrigerants put to the market, and within the near future

most of the existing synthetic refrigerants will be restricted by legislative means and

ultimately prohibited. What this means in other words is that a significant part of the stationary cooling systems in Europe, soon will be prohibited of service. These systems are running on a refrigerant called R404A, which has been the main choice in stationary cooling equipment since the mid 1990’s, and is estimated to be the refrigerant of choice in 46% of stationary cooling equipment in supermarkets world wide (Coolingpost 2014).

The regulation may at first seem unnecessary, but it is estimated that if we left the high GWP refrigerants unregulated and only focused on direct CO ₂ emis- sions, they could become responsible for between 28% and 45% of CO ₂ equivalent emissions by 2050 (Carbajal and Kanter 2009).

Systems already on the market and systems placed on the market before the prohibition in 2020 will, before their life cycle end need a conversion to alterna- tive refrigerants with a lower GWP. The ideal would be a refrigerant with identical properties but with a substantially lower GWP. As of now, no such product ex- ists, and therefore efficiency change and eventual practical problems caused by the refrigerants new properties must be evaluated.

However, research has been conducted, where screening of suitable candidates resulted in a few plausible retrofit refrigerants. Also examinations of economizer compressor performance running on these alternatives has been conducted. The combined result, proposes the refrigerants named N40 and DR-33 as suitable candi- dates as step in refrigerants, which are the development names for the refrigerants R448A and R449A manufactured by Honeywell and DuPoint respectively.

This paper will focus on the suggested refrigerants, which were developed with minor system adjustments for systems today running on R404A. To evaluate the refrigerants performances, a model will be developed with real system and com- pressor data as inputs. The performance data will be compared with the R404A as baseline. The results will also present the environmental impact during a fictive system lifetime and the expected new efficiencies running on the alternatives.

The refrigerants are relatively new, and therefore accurate mixture properties and compressor data may include some inaccuracies. However, the results is ex- pected to closer reflect the reality than idealizations and hopefully the results for the model can be used as guidelines for the retrofit future for the R404A.

As an introduction the background of refrigerant development is summarized in

section 1.1 Background. Further, more contemporary work and theory is discussed

in chapter 2 The vapor compression cycle. With the basis laid out, the problems

and challenges of this paper are described in chapter 3 Problem definition and

goals. The problem solving and explanations of concept in chapter 4 Method. The

chapter 5 Results and discussion presents and discusses the data received from the

model. And finally chapter 6 Conclusions.

1.1. BACKGROUND

1.1 Background

Domestic refrigeration in the early 1800’s mainly consisted of cavities chilled by ice blocks, which had to be extracted in the winter, stored in haystacks through the summer and transported to the end location when needed. The inconveniences led many to the conclusion that the issue of refrigeration had to have a modern technical solution, as did many other problems in the midst of the industrial revolution.

1.1.1 First generation 1830 - 1930

In 1805, the American inventor Oliver Evans described a system which evaporated ether under vacuum. The ether vapor was then pumped through a condenser unit, liquifying it and making the ether re-useable, thus making it a closed system able to create ice. It is the first documented system of its kind, but there is no evidence that Evans ever built one.

It wasn’t until the 1830’s that Jacob Perkins went to action by building the first working vapor compression cycle system. With his patent in 1834, Perkins got to name the refrigeration process behind his creation: The Perkins cycle. The design utilized diethyl ether as refrigerant, which was only one of the many natural refrigerants used at the time. Significant for the first generation of refrigerants were properties like flammability, explosiveness or both and some were also highly reactive.

1.1.2 Second generation 1930 - 1990

After 100 years, the vapor compression refrigeration systems still utilized refrig- erants with questionable properties exposing the end users to great risks. The technology was at the time unreliable and had led to numerous accidents due to its refrigerants. Instead of conquering the domestic market, many vapor compression refrigerators ended up on the backyard.

Ultimately, the disaster at the Cleveland Clinic in 1929 where methyl chloride fumes escaped from an explosion in a x-ray clinic received national publicity in the United States. The ignition of nitrocellulose x-ray film resulted in over 100 fatalities from poisonous gases. The explosion had nothing to do with refrigeration, but it suffered such bad publicity that the refrigerant could no longer be used in refrigeration. The American Medical Association published a report later the same year rating methyl chloride as the most dangerous for delayed toxic effects (Giunta 2006). With the market unwilling to adopt the new technological advantage of the vapor compression cycle, refrigerant development was the only answer. Charles Kettering who at the time was in charge of General Motors research department said to his head of research Thomas Midgley "...the refrigeration industry needs a new refrigerant if they ever expect to get anywhere...". (Midgley 1937)

Midgleys team began by looking at the periodic table of elements and eliminated

those that were known to contribute to toxicity or instability. Further eliminations

were made because of inappropriate boiling points and for substances known to lead to low tendencies of vaporization and low boiling points. The eight substances which passed the rigorous selecting process were carbon, nitrogen, oxygen, sulfur, hydrogen, fluorine, chlorine, and bromine. (Midgley 1937)

They noted that among the already known refrigerants, fluorine was the only substance not combined with the other eight. It was also the least toxic and at the same time least flammable, substance which was advantageous to achieve the goal of a pure refrigerant. By chlorination and fluorination of hydrocarbons (HCs), they created a compound, which was to be called R-12 or Freon-12. The chlorofluorocar- bon (CFC) refrigerant family was born, and with it the refrigerant development en- tered its second generation with properties such as being non-toxic, non-flammable, colorless, odorless and low reactive (Midgley 1937).

Nearly 50 years after the product placement of CFCs, being widely adopted for example as refrigerant or aerosol propellant, questions regarding the environmental effect caused by its emission into the atmosphere arose. The scientist, environmen- talist and inventor of the electron capture detector, James Lovelock, was the first to detect high levels of CFC in the air over Ireland, with his own invention. After Lovelock had taken measurements in both the Arctic and Antarctic, the scientists Mario Molina and Sherry Rowland suggested a connection between ozone depletion and CFC emissions (Molina and Rowland 1974).

What they discovered was that the CFCs low reactivity, makes it possible for the substance to make its way to the upper stratosphere. Once in the upper stratosphere ,the sun’s ultraviolet radiation is strong enough to break down the CFC molecules, releasing a chlorine atom, thus causing a chain reaction breaking down the ozone molecules.

As a result of the report released in 1974 and the proven connection between released CFCs and ozone layer depletion, the Vienna convention was held in 1985 and subsequently the Montreal protocol was signed in 1987. The protocol focused on the CFCs and demanded that the refrigerant group be abandoned with a step down process of new CFC put on the market, with an ultimate ban in new equipment as of 1996.

All members of the United Nations (UN) signed this treaty, and the Montreal protocol is regarded as the most successful environmental agreement ever since.

But the narrow focus on only CFCs made way for the hydrochlorofluorocarbons (HCFCs), which first entered their step down process as of 1996 and will reach their total phase-out in 2030. Exceptions were also made for developing countries stating complete phase-out of CFCs in 2010, and specific military applications are still authorized to use CFCs .

1.1.3 Third generation 1990 - 2010

The refrigerants which led the transition from the ozone depleting substances (ODSs)

were the HCFCs and as long term refrigerant the hydrofluorocarbons (HFCs) were

predicted. Although the HCFCs were no new invention, with the HCFC-22 first

1.1. BACKGROUND

produced in 1936 (Calm and P.E. 1994). The transition also gave new interest in natural refrigerants such as ammonia, carbon dioxide, HCs, and water (Calm 2008).

The generation three refrigerants were not as stable as the CFCs, denying most of the emissions to enter the stratosphere. With a lower ODP than the previous gen- eration, the transitional HCFCs were still depleting the ozone layer, which was why HFC without chlorine making the refrigerant zero ODP was intended as the final replacement.

As the refrigerant industry focused on reducing the ODP, the resulting genera- tion three refrigerants came to be big contributors of GHGs.

In 1992 the United Nations Framework on Climate Changes (UNFCCs) provided a framework on which future agreements could be designed. All members of the UN signed this treaty, which in it self is non-binding but is created with the intention of reducing GHG emissions in general. Protocols, in which levels of reductions are negotiated, are developed using the existing framework of the treaty.

The most famous one signed in Kyoto in 1997, hence the name the Kyoto Pro- tocol. The treaty requires reduction of CO ₂ equivalent emissions of at least 5% of the levels measured in 1990 and does not take in account where the emissions take place. This allows for parties to invest in projects abroad, where reduction is easier to achieve, to reduce their own footprint (UNFCCC 1998).

Along the development in the mid 1990’s GreenPeace launched a campaign for a higher usage of natural refrigerants. With their concept of GreenFreeze for refrigeration purpose, the synthetic refrigerants can be completely excluded. The refrigerant family proclaimed consists of HCs such as propane R290, which was sold as the safe and odorless alternative to ammonia back in 1920’s (Calm 2008) but never made a brake through. According to a GreenPeace estimate, 40% of the global domestic refrigeration runs on GreenFreeze, except for North America where it’s prohibited to use HCs for domestic use (Carbajal and Kanter 2009).

In 2006, the European Parliament aimed to reduce the fluorinated refrigerant emissions from mobile air conditioner (MAC), with the MAC Directive, which states that new car models placed on the market as of 2011 must use refrigerants with a GWP below 150, as do all manufactured cars as of 2017 (European Regulation 842/2006 2006).

1.1.4 Fourth generation 2010 -

The fourth and most current generation of refrigerants focuses on efficiency and short lifetime in atmosphere as well as zero ODP and low GWP. The focus on short lifetime is a tool for minimizing the GWP value and the ODP since a fast decay limits the amount of refrigerant that enters the upper atmosphere. The lifetime must be in the right span where it does not decay too fast and thereby contribute to urban smog (Calm 2008).

The hydroflouroolefine (HFO) are therefore gaining new interest since they have

high efficiency and zero ODP as well as low GWP. The fluorinated olefins generally

have a GWP under 10 and are mainly intended for air-conditioning applications,

although some are low flammable (M. O. McLinden et al. 2014) . A trending refrigerant is the R1234yf, which can be used as substitute in system running on R134a without any changes to the system or to the lubricants. The R1234yf is a HFO performing with marginally lower efficiency compared to its reference R134a (Jarall 2012).

In May 2014, the European Parliament adopted its regulation No 517/2014 (EU 2014) taking a new turn on reducing F-gases being produced. This regulation is the ultimate whip for the european refrigerant development. Either the industry will have to turn to the new synthetic compounds or return to its origins of natural refrigerants.

HC, CO ₂ and ammonia are natural refrigerants of increasing interest. With the

new regulations, CO ₂ is being promised to be the best alternative when building

new systems (Mota-Babiloni, Navarro-Esbrí, Barragán-Cervera, et al. 2015). With

lower efficiency in conventional system design, CO ₂ looks promising since it’s non-

toxic and non-flammable and not harmful to the environment and need sophisticated

systems to reach high efficiency (Bansal 2012). Ammonia has been around for a long

time and is mainly used in industrial applications where it out performes syntethic

refrigerants (Bansal 2012). GreenPeace is still promoting an extended usage of

natural refrigerants (Carbajal and Kanter 2009).

Chapter 2

The vapor compression cycle

The principle of this thermodynamic cycle is to manipulate the refrigerant temper- ature throughout the cycle, thus utilizing the flow of heat from a warm source to a colder ambient. By a supply of external energy the compressor raises the pressure of the refrigerant and thereby also its temperature. As the refrigerant is propelled through the first heat exchanger, called condenser, it enters as vapor with a higher temperature relative to the heat exchanger, thus allowing for heat to be transferred from the refrigerant before it exits the condenser as liquid.

In the next stage the refrigerant passes the expansion valve, and the refrigerant is forced by simultaneous suction and pressure of the compressor through a nozzle, evaporating the refrigerant, which result in decrease of its temperature. The evap- orated refrigerant then enters the second heat exchanger, called evaporator, where

h

h

h

h

p

p

10 1

3 2 4 5

6

7

8

9

Enthalphy [ kJ/kg ]

Pressure [ bar ]

Figure 2.1. The vapor compression cycle, enthalpy to pressure, 1-2.Compressor,

2-3.Discharge line, 3-6.Condenser with its Dew point at 4 and its Bubble point at 5,

6-7.Liquid line, 7-8.Expansion valve, 8-10.Evaporator with its Dew point at 9, and

10-1.Suction line.

the refrigerant has a lower temperature relative to the heat exchanger. This allows for heat to be transferred to the refrigerant from the evaporator. Between entry and exit of the evaporator the refrigerant gradually returns to a vapor state and is ultimately facing the compressor making it a closed system, see Figure 2.1.

2.1 Refrigerant theory

A modern refrigerant is the perfect example of something which cannot be described as ideal, in fact its come to be the very opposite. However, if a refrigerant were ideal, it should have all the thermodynamic properties suitable for its application.

It should be non-toxic, non-flammable, completely stable under deployment and environmentally friendly including all its decomposition products.

To extend the properties of the ideal refrigerant, it should be compatible with all the materials within the system, also the lubricants. The production should be possible within the already built up infrastructure. It should be easy to detect and not pose any risk while being handled, served or destructed. The ideal refrig- erant should also work under reasonable pressure allowing for price efficient heat exchangers, tubing and easy sealing at tube connections.

The many properties a refrigerant should and must have, makes the the alter- natives very few. With a new tough regulation recently adopted, the alternatives have become even fewer. At the same time as the refrigerants need to take in more properties in account the demand for energy efficiency also increases.

The efficiency enhancements achieved have been primarily through improve- ments and optimization in system design rather than in refrigerant performance.

Since every new refrigerant inherits its predecessors work environment and also needs to fulfill new requirements, they all tend to be equally or less efficient. The ideal refrigerant will almost certainly remain ideal and a discovery is extremely unlikely (Calm and Didion 1997).

2.1.1 Thermophysical properties

The described cycle in Figure 2.1, presumes that the refrigerant is in a vapor state once exiting the evaporator. This is ensured by the boiling point and the refriger- ant’s volatility. The boiling point can be lowered through a reduction of pressure, which is generated by the compressor. With a boiling point suitable for the evap- orator temperature, in combination with a sufficient volatility, a fully vaporized refrigerant can be obtained at the end of the heat exchanger.

As the fundamental function of the refrigerant is to move heat, the thermal

conductivity of the refrigerant is of great importance, where a compound with a

higher thermal conductivity can transfer heat at a higher rate.

2.1. REFRIGERANT THEORY

2.1.2 Toxicity

Just as relevant today as when the vapor compression cycle refrigerator was domes- ticated is for the application to be non-toxic. For an application to be safe, either the refrigerant must be non-toxic or the application must be designed in such way that with the right safety precautions it becomes safe.

Toxicity can be ambiguous since the definition is based on both quantity and exposure time. Substances that with a small quantity or during a short time of exposure pose high risk, are considered highly toxic. Therefore is the air we breathe considered not toxic even though CO ₂ itself causes suffocation in higher concentra- tions (Calm and P.E. 1994).

In the human body, toxicity causes decomposition or malfunction of tissue either through corrosion or by decay of the toxic substance that disables key functions in the cells. The substances stable enough to enter vital organs in the body and then decompose are regarded as most toxic.

A convenient way to detect leakage is either by odor or color. An example is ammonia which is a very toxic substance but detectable through smell even at low concentrations where it poses no harm. It outperforms the synthetic refrigerants, but causes panic at nontoxic levels and is therefore avoided in public areas (Bo- laji and Huan 2013). The synthetic refrigerants, on the other hand, are almost exclusively non-smelling which requires specific detection equipment for every com- pound. Such equipment is also needed when the refrigerants density is higher than the breathable layer of air which could cause toxicity in case of release in closed compartments.

2.1.3 Flammability

How likely a substance is to ignite, burn and combust is quantified by the flamma- bility. In terms of concentration needed for ignition, ignition energy, heat of com- bustion and flame propagation the compounds are labeled in a scale from 1 for non-flammable to 3 for highly flammable. The flammability is among many stan- dardized by the American Society of Heating, Refrigerating and Air-conditioning Engineers (ASHRAE) and the international organization for standardization (ISO).

The concentration range a compound can ignite within is bound by the lower flammability limit (LFL) and upper flammability limit (UFL). The deployed re- frigerant experiences concentrations much higher that the UFL and is therefore regarded as safe. But due to the nature of leakage the concentration in the ambient must be held below the LFL, otherwise the leaking refrigerant poses a high risk.

The limits are measured in grams per cubic meter.

If the right mixture for combustion is reached, the energy required for ignition is relevant. The more energy a compound needs for ignition, the safer it is regarded.

Many refrigerants applicable under todays regulations are flammable, thus mak-

ing it interesting to quantify how fast the flames spread. For this matter, the flame

propagation property has emerged when in the case of a fire a slow spread is desired.

In case the compound were to ignite and propagate, it is interesting to measure the heat generated through by the combustion in order to chose the right materials in the design process. To lower the flammability of a refrigerant, it’s common to blend a flammable with a non-flammable refrigerant and thus create a low flammable compound (Pham and Technologies 2010).

However, components in a mixture with different volatility often tend to evap- orate at different rates. Therefore certain components could account for the most of the leakage in different parts of the system. After some time it may be that the initial mixture ratio had changed and altered the flammability of the refrigerant (Girodroux, Kusmierz, and Dahn 2000).

For HFC and HFO used as alternatives for high GWP refrigerants studies show that non flammable refrigerants often tend not to perform as well in terms of COP as low flammable alternatives. The low flammable refrigerants may be used in low charge systems such as air-conditioning or domestic refrigeration but not in large expansion systems with high charges such as supermarkets (Mota-Babiloni, Navarro-Esbrí, Barragán-Cervera, et al. 2015).

2.1.4 Classification

For the toxicity and flammability the ASHRAE has a standard of numbering and classifying refrigerants. The standard is objective, meaning that no endorsements or applicable recommendations are implied (Walter, Kennoy, and Brock 2008).

With the ASHRAE standard designation and safety classification of refrigerants (ASHRAE 34) and ISO refrigerants designation and safety classification (ISO 817), the range of refrigerants available on the market are classified in tabular form.

The standard also includes naming refrigerants with the significant R followed by numerical combinations describing the chemical structure.

Refrigerants are classified based on their long term toxicity with the prefix A or B. The first indicates a refrigerant with a permissible exposure limit of 400 ppm or greater, meaning that it does not pose a risk during exposure below the set limit and is labeled with lower toxicity. The latter is labeled with higher toxicity, and the compound with such prefix poses a risk below the set limit of 400 ppm (Watson 2013). The toxicity divides the categorization into two columns as seen in Table 2.1.

Following the toxicity prefix is the flammability ranging from 1 non-flammable to 3 highly flammable. With an index of 1 the compound has no flame propagation.

Table 2.1. Refrigerant classification, regarding toxicity and flammability according to ISO 817 and ASHRAE 34.

Lower toxicity Higher toxicity

No flame propagation A1 B1

Lower flammability A2L B2L

Flammable A2 B2

Higher flammability A3 B3

2.1. REFRIGERANT THEORY

The following index 2 is a flammable compound with a LFL of more than 0.10kg/m ³ and a heat of combustion of less than 19M J/kg. The last index 3 indicates a higher flammability with a LFL of less than 0.10kg/m ³ and a heat of combustion or more than 19M J/kg (Watson 2013).

Refrigerants with low environmental impact are often slightly flammable, which has turned the industry towards a new view on flammability. The new class A2L is a product of this turn since it refines the categorization of flammable refrigerants with a lower flammability prefix L with the same boundaries as for index 2 but with burning velocity of less than 10cm/sec. The addition is in line with the ISO 817, making ASHRAE 34 classification identical (UNEP 2010).

2.1.5 Saturated liquid and vapor

Within the closed system the refrigerant circulates and by doing so, constantly changes state from liquid to vapor and back. During its transition, the refrigerant vapor coexists with liquid refrigerant, which is illustrated by the hatched surface in Figure 2.2, called the two-phase region. This region is wrapped by two boundaries together called the envelope. To define the relative existence of refrigerant vapor, the vapor mass quality property is introduced. It’s a value between 0 and 1 describing the left bound as fully saturated liquid and the the right as fully saturated vapor.

Outside the envelope the refrigerant is either sub cooled liquid to the left or super heated vapor to the right. And the upper point of the envelope, marks the highest pressure under which vapor and liquid can coexist.

150 200 250 300 350 400

1 10 100

Enthalpy [ kJ/kg ]

Pressure [ bar ]

Figure 2.2. Refrigerant enthalpy by pressure diagram, the crosshatched area indi-

cates where the refrigerant appears in both vapor and liquid state simultaneously, the

dot on the top shows the maximum pressure in which the refrigerant is both vapor

and liquid. Outside the cross hatched area, to the left the refrigerant is in liquid and

to the right in vapor state.

Within the vapor compression cycle, the refrigerant transits from high enthalpy to lower, while heat is transferred to the condenser. The refrigerant enters the two- phase region as the first liquid drop is formed. Moving further lower in enthalpy, the refrigerant successively becomes more liquified, before at the end of the condenser, in a fully liquid state exits the two-phase region. For pure compounds, the transition from saturated vapor to saturated liquid and vice versa proceeds under constant pressure and temperature, called azeotropic behavior (Granryd et al. 2005).

2.1.6 Temperature glide

Since the study which led to the production of CFCs was conducted, major efforts have been made to find new refrigerants. The latest includes a screening of 56.000 candidate molecules which were narrowed down through several processes to 62.

None of the reduced clique were ideal in regards thermodynamic properties, toxicity and flammability etc. (M. O. McLinden et al. 2014).

Nevertheless, mixtures with the refrigerants R1234yf and R1234ze(E), which resulted from the study, have become common. The two brings properties of GWP integrated over 100-years (GWP ₁₀₀ ) values below 10 and are combinable in mixtures with refrigerants with a higher GWP. The mixtures result in non-flammable and zero ODP refrigerants with lower GWP, which can be used as retrofit a for current high GWP refrigerants in deployment (DuPont 2014; Honeywell 2014)

When different pure compounds are mixed, new properties emerge that neither held alone. It is natural that different compounds have their specific properties such as boiling point, volatilities, and critical temperatures. As the refrigerant transits within the two-phase region, different components will be vaporized or liquified

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

−45

−40

−35

−30

Mixture relation [ mol/mol ] T emp erature [

C ]

Figure 2.3. Azeotropic behavior of a mixture between R125 and R1234yf, illustrates

how the difference in temperature between bubble point and dew point separates as

the mixture relation changes. At 0 and 1 the mixture is either pure R125 or R1234yf.

2.1. REFRIGERANT THEORY

at different stages, due to various volatility. Mixtures with such properties shows non-azeotropic behavior.

For non-azeotropic refrigerant’s, saturated vapor and saturated liquid are not ad- equate to describe the transition in the two-phase region. Therefore non-azeotropic refrigerants have bubble points and dew points translating to saturated liquid and saturated vapor respectively.

In Figure 2.3, an arbitrary mixture with non-azeotropic behavior is illustrated, in which the molecular fractions of its components are changed. The upper boundary is limited by the dew points and the lower boundary is restricted by the bubble points evaluated for molecular fractions from 0 to 1. The figure illustrates that when the components are pure, the dew and bubble temperature are equal. But once the compounds are mixed, the temperature difference increases. This phenomenom is called temperature glide. The bars height in Figure 2.3 indicates the glide with an equal molecular mixture ratio.

The effect of the temperature glide, can negatively affect the heat transfer in the systems heat exchangers. Where refrigerants with high temperature glide may require larger heat exchangers (UNEP 2010).

2.1.7 Ozone depleting potential

The CFCs was designed for maximum stability with low reactivity in mind. This allowed the chlorinated molecules to slowly populate the upper stratosphere with very little decay on the way up. The breakthrough in ozone hole research was the conclusion that once the CFCs reached the stratosphere, the suns strong unfiltered UV-light brakes down the CFC molecules, which causes the release of one chlorine atom. The lone atom reacts with the ozone

Cl + O ₃ → ClO + O ₂ (2.1)

resulting in chlorine monoxide and and dioxygen. The chlorine monoxide then reacts with the free oxygen that exist in the stratosphere, as the ozone layer is under constant regeneration. The reaction

ClO + O → Cl + O ₂ (2.2)

creates a free chlorine atom and more dioxygen allowing the chlorine atom to begin reacting with the ozone once more. The constant return to a single atom allows for it to decompose huge amounts of ozone. The overall reaction is that ozone and free oxygen react to two oxygen molecules.

O ₃ + O → 2O ₂ (2.3)

For a refrigerant to possess zero ODP, it has to either lack chlorine or have a

speed of decay such that it never reaches the stratosphere.

2.1.8 Global warming potential

In the first assessment report (AR1) by the Intergovernmental Panel on Climate Change (IPCC), a new measurement of a refrigerant’s long term effects on the environment was introduced. The GWP was intended for categorization and to allow for regulations to refer to a value rather than adress every single refrigerant.

It was also used to illustrate the difficulties in comparing different physical properties within one metric (Pachauri et al. 2014). The GWP metric has been adopted by UNFCC as well as the Kyoto protocol.

The concept of GWP is to integrate the contribution to global warming under a specific time interval with CO ₂ as reference. In other words, the GWP is an index that quantifies a substance’s contribution to global warming in relation to the effects of an equal amount of CO ₂ , within a chosen timespan, as seen in Equation 2.4. A high GWP correlates with the a long atmospheric lifetime as well as a large infrared absorption.

GWP(x) = Z H

0

RF x (t)dt Z H

0

RF _r (t)dt

(2.4)

The calculation is based on an instantaneous release of a substance which is integrated over the time interval H. The radioactive forcing (RF) is the difference between the absorbed infrared energy and the energy radiated back to space. A positive RF means that the substance absorbs more than it reflects. The change over time is dependent on substance decay.

The integration interval of 100 years has become industry standard but spans of 20 and 50 years are also used. For a refrigerant with a fast decay, the long integration time results in a low GWP ₁₀₀ , see Figure 2.4.

In the fifth assessment report (AR5) released by IPCC it’s pointed out that there is no scientific evidence to support the 100 years integration time (Pachauri et al. 2014). With a value that can be compared, it’s natural that relative prefixes evolve such as "high" and "low". A high GWP refrigerant has a GWP ₁₀₀ of 2500 or more, and a low GWP ₁₀₀ refrigerant has less than 150. The question about what really is a low GWP refrigerant has arisen, and the confusion is not unjustified since the effects of the assumption that low GWP ₁₀₀ is a refrigerant with a value of 150 or limits the range of refrigerants significantly. The fact that there is no clear definition of the thresholds was emphasized by the United Nations Environment Programme (UNEP) report (UNEP 2010).

Nonetheless, the new EU regulation 517 refers to a GWP ₁₀₀ of 150 as a threshold

for fluorinated refrigerants after 1 January 2022 (EU 2014), and this is the current

future for the European refrigeration market.

2.1. REFRIGERANT THEORY

0 20 100

Years after emission

Radioactiv e forcing

Figure 2.4. How decay times affect GWP, the fast decaying arbitrary refrigerant is compared with the slower decaying CO

. Illustrating the difficulty in choise of time horizon while comparing environmental effects.

2.1.9 Total equivalent warming impact

The single metric of GWP does not take indirect effects in account, meaning that the GWP is only a metric for the direct effect of fugitive emissions. To broaden the view of the summarized effects on the environment, the TEWI was introduced.

It’s a metric taking in account both direct and indirect emission caused by the refrigerant once emitted and the emissions of keeping the system running.

The TEWI firstly sums both the total release of refrigerant during deployment and final disposal, and secondly the impact of electricity production throughout the systems lifetime (see Equation 2.5).

TEWI = GWP _direct + GWP _indirect (2.5) The direct emissions are caused by the annual leakage rate L _annual multiplied by the charge mass m, years of service n and the GWP. Some emissions also oc- cur during emptying of the system in the disposal procedure, which causes direct emissions where the residual of the recovery factor α _recovery is multiplied by charge mass and the GWP (see Equation 2.6).

GWP _direct = GWP ∗ m ∗ L _annual ∗ n + GWP ∗ m ∗ (1 − a _recovery ) (2.6) The electricity production causes indirect emissions which are calculated by energy consumption per year E _annual multiplied by the indirect emission factor β and system operating time in years, see Equation 2.7.

GWP _indirect = E _annual ∗ β ∗ n (2.7)

The metric is measured in mass of CO ₂ equivalent. With TEWI it becomes clear, that the direct effects of GWP are minor compared to the indirect effects with the current production methods of electricity (Airah 2012).

2.2 New regulations and previous research

The recent progress in refrigerant development has been forced by the many regula- tions, which are narrowing the range of possible compounds. Recently, the European parliament adopted the regulation No 517/2014 (EU 2014), which has acquired le- gal force and begun its regulative actions as of 1 January 2015. The regulation is design to ultimately prohibit fluorinated high GWP compounds of production, with the practical implementation of a step down procedure (see Table 2.2).

As of January 1 2020, the regulation prohibits refrigeration systems running on HFC refrigerants with a GWP of 2500 or more to be placed on the market and as of 1 January 2022 systems filled with HFC refrigerants with a GWP of 150 or more.

It further states that as of 1 January 2030 systems running on HFC refrigerants with a GWP of 2500 or more are banned from service (EU 2014).

Table 2.2. European Parliament regulation 517 Annex III (EU 2014).

Products and equipment

Where relevant, the GWP of mixtures containing fluorinated greenhouse gases shall be calculated in accordance with Annex IV, as provided for in point 6 of

Article 2

Date of prohibition

11. Refrigerators and freezers for commercial use

(hermetically sealed equipment)

that contain HFCs with GWP of 2 500 or more

1 January 2020 that contain HFCs with

GWP of 150 or more

1 January 2022 12. Stationary refrigeration equipment, that contains, or

whose functioning relies upon, HFCs with GWP of 2 500 or more except equipment intended for application designed to cool products to temperatures below -50 C

1 January 2020

2.2.1 R404A

The R404A is a HFC mixture refrigerant well known for its performance in medium and low temperature applications suitable for supermarket use. It has low toxicity and is non-flammable making it an ASHRAE A1 refrigerant. The refrigerant came on the market as a replacement with zero ODP but was equipped with a high GWP.

It’s a near azeotropic blend meaning that it has a negligible temperature glide. But

because of the new regulation 517/2014 it will turn R404A obsolete within short

time due to its GWP ₁₀₀ of 3922 (Pachauri et al. 2014).

2.2. NEW REGULATIONS AND PREVIOUS RESEARCH

2.2.2 R448A

As an intermediate solution, Honeywell has developed a refrigerant with lower GWP and near step in properties, meaning that very few adjustments have to made to the system. The Solstice N40 is an ASHRAE R448A refrigerant, with low toxicity and non-flammability, classifying it as ASHRAE A1.

The mixture’s HFO components R1234yf and R1234ze(E), see Table 2.4, have low GWP values resulting in a total GWP of 1273 (Pachauri et al. 2014). With the wide spread of mixture components comes a high temperature glide, which may result in need for system modifications.

A previous comparison has been conducted, proposing an increase in discharge temperature, although below the set temperature limit of 130 ^◦ C. The system changes predicted were little to no modifications while yielding of up to 9% over- all system efficiency improvement. In the area of risk assessment, the new mixture poses no danger of shifts in composition during leakage, while the mixture remained within experimental uncertainty of ±5%. The environmental improvements are due to a great reduction in GWP and a significantly higher efficiency (Motta, Becerra, and Spatz 2012).

Another study indicates lower volumetric flow during high condensing temper- atures as well as marginally higher volumetric flow under low condensing tempera- tures. The lower volumetric flow, may force the system of changing its tubing and heat exchangers to larger cross section, in order to obtain an overall performance gain, under both high and low condensing temperatures, compared to the R404A.

The best performance gains were seen in applications with high temperature differ- ence (Mota-Babiloni, Navarro-Esbrí, Barragán, et al. 2014).

Previous findings were confirmed by a study about economizer style compressors, which exhibited a rise in discharge temperature of approximately 22%. Also a lower cooling capacity near 7% was predicted as well as a gain near 4% in COP (Sjoholm, Kleinboehl, and Ma 2014).

Recently, a study praised N40 to be the most promising replacement for R404A, due to an overall gain in COP and a peak gain during high temperature differences.

The cooling capacity experienced a gain under high condensing temperatures but

Table 2.3. Refrigerant properties, (DuPont 2005; DuPont 2014; Honeywell 2014).

R404A R448A R449A

ASHRAE A1 A1 A1

ODP 0 0 0

GWP ₁₀₀ AR5 3922 1273 1282

GWP ₁₀₀ as % of R404A 100 32.5 32.7

Boiling point at 1 atm, ^◦ C -46.5 -45.9 -46.0

Critical temperature, ^◦ C 72.1 87.3 81.5

Critical pressure, kP a 3732 4660 4457

Temperature glide at 1 atm, K 0.8 6.1 6.0

a drop under low condensing temperatures. The study also empathized the lower energy consumption achieved while running on R448A. (Mota-Babiloni, Navarro- Esbrí, Barragán-Cervera, et al. 2015)

2.2.3 R449A

The second retrofit refrigerant is the by DuPont developed and manufactured Opteon XP40, which is an ASHRAE R449A refrigerant. It has been on the market quite some time under the development name DR-33. It’s a low toxic and non-flammable compound classified according to ASHRAE 34 as A1. This low GWP alternative with a GWP ₁₀₀ of 1282 (Pachauri et al. 2014), utilizes the R1234yf as a large com- ponent to lower it’s GWP, see Table 2.4. The refrigerant shows the same tendency towards temperature glide as the R448A, see Table 2.3.

The manufacturer has released test results on systems running on DR-33 as opposed to R404A. For freezer systems, energy savings between 0 and 4% were observed. As well as for display cases, a decrease of energy consumption between 3 and 4% for low condensation temperatures was recorded and 8 to 12% for medium condensation temperatures (Minor and Leck 2013).

During a study of economizer style compressors a discharge temperature rise of 20% for the R449A was predicted, as well as a slightly lower cooling capacity with a drop of near 1%, with a total COP gain near 4% (Sjoholm, Kleinboehl, and Ma 2014).

Another comparison conducted with a theoretical model yielded promising re- sults in terms of COP gain, which spanned between 6 and 25% depending on the total temperature difference between the heat exchangers. The highest gain in COP was achieved with the highest temperature difference. Also a gain in volumetric flow under low condenser temperatures was predicted as well as the contrary for high condenser temperatures (Mota-Babiloni, Navarro-Esbrí, Barragán, et al. 2014).

The same team of scientist later released results in line with previous results, concluding that an energy consumption between 8 and 12% lower at medium tem- perature conditions and between 3 and 4% lower at low temperature conditions is to be expected (Mota-Babiloni, Navarro-Esbrí, Barragán-Cervera, et al. 2015).

Table 2.4. Refrigerant mixture in percentage by weight, from REFPROP (Lemmon, Huber, and M. McLinden 2013).

R32 R125 R134a R143a R1234yf R1234ze(E)

R404A 44 4 52

R448A 26 26 21 20 7

R449A 24.3 24.7 25.7 25.3

Chapter 3

Problem definition and goals

Over the last decades the knowledge about human-caused climate change has in- creased, making it clear that the current situation is unsustainable. The effects of measured climate changes such as the increasing average temperatures, intensity of storms or draught are notable and only caused by a fraction of the earth’s pop- ulation (IPCC 2007). With agreements signed in both Montreal and Kyoto, the industry is forced to make major changes, but still our lifestyle is continuing in a very much unaffected way towards a more energy consuming nature.

In our daily life we interact largely with refrigeration systems. This becomes most obvious when entering a supermarket; products from all over the world are made available through a vast number of refrigeration systems. With examples from the fishing industry, where some of the trawling takes place in the Barents Sea, processing in China for consumption in Europe. The number of refrigerating systems that the product comes in direct and indirect contact with are astonishing.

Indisputably the refrigeration systems are many, and they are all associated with leakage due to inaccuracies in components and sealings. The refrigerant seeps out during deployment, contributing to the global warming due to being potent GHG with high GWP values.

A lot of action has been taken, but the signed legislative agreements excluded or delayed actions for developing countries, meaning that the developed countries technologies could continue to yield profits a while longer in an other part of the world (Buxton 1988). With this in mind the near future is a rather critical time as we have to reduce our emissions drastically simultaneously, while the interest of refrigeration is rising.

As of 1 January 2015 the belt has been tightened one more notch by the EU

Regulation 517, demanding quick action of reducing and ultimately prohibiting

fluorinated gases. This forces a large number of refrigeration systems already in

deployment to change their refrigerants within short.

3.1 Focus

The stationary refrigeration systems are in need of new retrofit refrigerants to com- ply with the new regulations in the European Union. The HCFC R404A has been in service since the beginning of the 1990’s and will be prohibited due to its high GWP ₁₀₀ value of 3922 (Pachauri et al. 2014). It is estimated that the R404A re- frigerant is deployed in 46% of the stationary refrigeration systems in supermarkets (Coolingpost 2014), and is therefore urgently in need of replacements.

As retrofit refrigerants, the R448A and R449A produced by Honeywell and DuPont respectively are available with lower GWP and proposed higher efficiency.

The performance of the two refrigerants will be compared under real system data and the results will be presented with the R404A as baseline on the following:

• Isentropic compressor efficiency

• Flow rates

• Discharge temperature

• Cooling capacity

• Volumetric cooling capacity

• Coefficient of performance

• Total equivalent warming impact

Both of the retrofits are blend refrigerants with a relatively high temperature glide and intended to be step-in solutions for system already in deployment.

3.2 Expectations

A lot of research has already been done with much data pointing in the same direction. The following outcome from the model is likely to be expected:

• Discharge temperature rise

• Lower volumetric flow

• Higher efficiency

• Lower total environmental warming impact

3.3 Goals

With the results from the model, recommendations of when and why the different

refrigerants are suitable as replacements will be reached. This will serve as guide

line for the retrofit industry’s future work of supermarket refrigerant transition.

Chapter 4

Method

In order to conduct a comparison between the refrigerants a basic vapor compression cycle model was built. Basic, in this context, means that the system consists of four main components: Evaporator, Compressor, Condenser and Expansion valve, and they are connected by a Suction line, Discharge line and Liquid line respectively.

Note that the expansion valve and the evaporator are directly connected. The model is a computer-based software that reads input data, calculates points of interest and presents output data.

The purpose of writing a new model, lies in the softwares released by the man- ufacturers, which only are able to calculate their own refrigerants. Therefore a comparison between the R448A and R449A wouldn’t be possible, since they are produced by different manufacturers. Also the existing softwares cannot be ex- tended with real compressor data from compressor manufacturers and the batch capabilities of large data are poor.

4.1 Limits

To narrow down the scope of the model some idealizations are necessary:

• No heat transfer to the surrounding

• Isenthalpic process at the expansion valve

In other words, as the refrigerant flows through components and tubing, heat transfer via the material is not taken in account. This means that heat transfer in the evaporator and the condenser includes the overall system transfer in form of thermal conductivity.

In the expansion valve the refrigerant is evaporated from its liquid state, causing

the pressure to drop which results in a proportional temperature drop, according

to the ideal gas law. The process of evaporation is modeled as isenthalpic, which

simplifies the process by assuming that the enthalpy is preserved.

4.2 System parameters

The first limit regarding heat transfer simplifies the model in such a way that the input data can be specified as absolute temperatures, relative pressure drops and relative temperature changes, see Table 4.1. This also allows for the input data to be measured on a real system.

For the two heat exchangers, the absolute temperature, relative pressure drop and outlet temperature change are defined by input data. The temperature change at the outlet translates to sub cooling for the condenser and super heating for the evaporator, which is the refrigerants’ temperature difference when exiting the heat exchanger compared with its temperature once fully saturated. For the case of the condenser it describes how much colder the refrigerant has become compared to its saturated liquid temperature. This applies analogously for the evaporators super heating which shows how much warmer the fully vaporized refrigerant is at the outlet compared to its saturated vapor temperature.

Table 4.1. Input variables for system components

A verage temp

erature

Sub co oling

Sup

er heating T emp

erature change

Pressure drop

Isen tropic

effic iency

Ph ysical

sp ecifications

Compressor • •

Discharge line • •

Condenser • • •

Liquid line • •

Expansion valve

Evaporator • • •

Suction line • •

The interconnecting tubings called Discharge line, Liquid line and Suction line are all defined by pressure drop and temperature change, see Table 4.1. This allows for modeling of systems where the components are spread, forcing the refrigerant to travel long distances. The temperature changes are present because of non-ideal isolation along the tubing.

The second limit of the model is the idealization of an isenthalpic process at the expansion valve, which requires no input data since the process will always be ideal.

The system compressor inlet and outlet pressure are results of the system design.

This results in the compressor being the last calculation step in the model connecting the suction line and discharge line.

From the compressor developer and manufacturer, Bitzer Kühlmaschinenbau

GmbH, files containing coefficients for a third degree polynomial were downloaded.

4.3. REFERENCE SYSTEM DATA

The physical specifications of the compressor in the input, will be matched with one from Bitzer. According to the EN12900, the polynomial describes compressor power consumption and mass flow rate as functions of the evaporator and condenser temperatures, which were combined to calculate the enthalpy rise in the compressor.

The polynomial coefficients are refrigerant specific and only valid within a specified temperature interval.

The isentropic efficiency input, is a value calculated from the real system, and will not be active in the calculations. But will serve as a relative value to the calculation results.

With the system design defined, the next step was to define the refrigerants, see Table 4.2. The mixture definition was made through mole fractions, which are relative measurements of the number of molecules of every component in the mixture. The composition differs from the weight per centage described Table 2.4 due to highly differing mole masses. The mixture data was taken from REFPROP (Lemmon, Huber, and M. McLinden 2013).

Table 4.2. Refrigerant mole fraction mixture properties, from REFPROP (Lemmon, Huber, and M. McLinden 2013).

R32 R125 R134a R143a R1234yf R1234ze(E)

R404A 0,357817 0,038264 0,603919

R448A 0,431218 0,186914 0,177587 0,151319 0,052962 R449A 0,407365 0,179481 0,219673 0,193481

4.3 Reference system data

Parallel with this paper, a super market located in the region of Stockholm, Sweden underwent a change of refrigerant from the R404A to the retrofit R449A. From this site data before the change was extracted with 30 second interval samples. With the system in active deployment, pressure data could not be extracted since required sensors would disturb service.

From the reference site, a fridge system was chosen since it at the time con- tained the most data points. The selected data set was chosen where a leveling in discharge temperature and compressor power was seen. Meaning a sample where the compressor had a long run time just before shutting of, see Table 4.3.

4.4 Concept

The thermodynamic process modeling is intimately connected with equations of

state, which describes the relationship between state variables. An equation of

state makes it possible to acquire all possible information about a thermodynamic

state with a set of just two input data. The relationships can be formed with

two inputs from either Pressure, temperature, enthalpy, entropy, specific volume or