Heat Pumps: Solving Energy and Environmental Challenges

It has long been recognized that realizing the potential for energy conservation and diversification by using heat pumps offers considerable benefits to the environment. Important work on more efficient and ozone-friendly working fluids will further enhance the case for greater support of heat pump research. This book contains the Proceedings of the Third International Energy Agency Conference held in Tokyo in March 1990. The main theme of the Conference, 'Heat Pumps - Solving Energy and Environmental Challenges', is explained in great depth, covering not only technical characteristics but economic factors and the role of government and other bodies in promoting research, and the uses of all types of heat pumps are also fully considered. As well as publishing the papers presented at the meeting, the book also contains the extensive complementary poster sessions from the Conference.

• Language: English
• Publisher: Elsevier Science
• Release date: Oct 22, 2013
• ISBN: 9781483287317

Book preview

Japan

SESSION I

Opening Plenary Session

Outline

Chapter 1: Welcoming Address

Chapter 2: Opening Address

Welcoming Address

Mizuo Iwasa,     

Chairman of 1990 IEA Heat Pump Conference, Organizing Committee of Japan; The Tokyo Electric Power Co., Inc., Uchisaiwai-cho 1-1-3, Chiyoda-ku, Tokyo 100, Japan

Publisher Summary

This chapter focuses on various worldwide developments going on energy efficient technologies. Sustainable development refers to a process where the existing energy resources are used as efficiently as possible to make them last longer, and while the global environment, and each nations economic development are being maintained, time is gained to develop new no fossil energy resources, and accelerate their commercialization. The important role that heat pumps can play in achieving this purpose has long been emphasized by the experts, and can be substantiated through discussions of different viewpoints. With funding contributions from the government of each member nation and the private sector, the IEA Heat Pump Conference in Tokyo has been organized to exchange information on the development and diffusion of heat pump technologies among member nations, to accelerate this development and diffusion, and to create its greater acceptance among those in many different fields. It is important to approach the government officials responsible for energy policy in each nation.

On the occasion of the 1990 IEA Heat Pump Conference in Japan, I would like to greet you on behalf of the Organizing Committee of Japan.

In recent years, general interest in energy conservation and development of alternative energy sources waned as oil demand weakened. This has led to reduced investment worldwide in the development of energy efficient technologies. And the number of researchers in these fields has also decreased. It seemed that the enthusiasm for technological development and diffusion in these fields had faded.

Very recently, however, environmental problems represented by the greenhouse effect and acid rain have been rapidly thrust upon us, and we are now observing growing interest in energy efficiency.

It is indeed significant that the 1990 IEA Heat Pump Conference is being held in Tokyo at this time, emphasizing worldwide cooperation which is based on pooling our knowledge to solve energy and environmental problems simultaneously. On behalf of the Organizing Committee of Japan, I would like to express my heartfelt welcome to all of you from many different countries who have gathered together here to attend this conference. Since the current conference is taking place in Asia, we are fortunate to have specialists from the People’s Republic of China and South Korea.

The Summit held in Paris this last July indicated that our efforts can yield more fruitful results in developing or diffusing heat pump technologies. The Summit on global energy resources and environmental preservation concluded that sustainable development is necessary.

Sustainable development refers to a process where the existing energy resources are used as efficiently as possible to make them last longer, and while the global environment and each nation’s economic development are being maintained, time is gained to develop new nonfossil energy resources and accelerate their commercialization. I believe that a worldwide consensus can be reached on this idea.

The important role that heat pumps can play in achieving this purpose has long been emphasized by the experts, and can be substantiated through discussions of different viewpoints and data accumulated at the 1st Graz Conference, the previous Orlando Conference, and the current Tokyo Conference.

Details of these discussions at the Tokyo Conference will tremendously benefit many people involved in the energy and environmental fields. On this occasion, however, I would like to request all of you to avoid holding back your data on possible roles for heat pumps as a global energy resource and in environmental preservation, but instead make this data widely available through your own contacts or through other agencies or institutes. It is important to approach the government officials responsible for energy policy in each nation. You are respected by governments, institutes and industries, and your expertise will provide fresh insights and carry considerable weight in influencing those who have so far been indifferent to the use of heat pumps. You are the driving force for protecting our planet.

Under the main theme Solving Energy and Environmental Challenges, we sincerely hope that all of you who are participating in the Third IEA Heat Pump Conference will engage in productive discussions and informational exchanges beneficial to each other in a friendly and familiar atmosphere. The preservation of global energy resources and the environment is one of the most important issues facing us today, and we hope that this conference will make major contributions in seeking solutions.

With funding contributions from the government of each member nation and the private sector, the IEA Heat Pump Conference in Tokyo has been organized to exchange information on the development and diffusion of heat pump technologies among member nations, accelerate this development and diffusion, and create its greater acceptance among those in many different fields.

At this point, I would like to thank the government of each nation and the enterprises which have made funding contributions to this conference. Many thanks also to the members of the IEA Organizing Committee, the IEA Secretariat, the Organizing Committee of Japan, and many other people who have helped in the preparations for this conference.

Not only advanced technologies, but also time-honored traditions and culture have been preserved in Japan, including a wonderful custom of entertaining guests from faraway countries.

We at the Organizing Committee of Japan are more than pleased when you take part directly in our traditions while inspecting facilities which demonstrate current Japanese heat pump technologies, or when you participate in various other events.

In closing, I hope that this conference will be of value to the participants and they will retain many pleasant memories of Japan from it.

Opening Address

M. Sugiura,     Director General, Agency of Industrial Science and Technology, Ministry of International Trade and Industry, Kasumigaseki 1-3-1, Chiyoda-ku, Tokyo 100, Japan

Publisher Summary

This chapter describes the various developments in heat pump technology. In the area of heat pump technology, one hopes to find a way to use it without causing environmental destruction and while utilizing wasted heat energy to the maximum extent possible. The task facing today is to improve the present heat pump system to the point where it does not harm the environment. One should aggressively search for new kinds of refrigerants for working fluids. Apart from the environmental problems, one still has a lot to do on the technology side. It is necessary to perform R&D to increase heat pump performance, especially in the area of temperature range, and energy efficiency. This is essential because of its broader application. In commercialization, economic principles always apply. The heat pump is no exception. The economic considerations have impeded the advancement of heat pump technology as a whole.

PREFACE

Ladies and Gentlemen, and honored Guests,

It has been three years since our last Conference in Orlando, Florida. As you recall, our two previous conferences in Graz and Orlando were very successful. And I believe everyone here is seeking a way to promote heat pump technology in the future. If we can achieve this, we will then be able to look back upon the third IEA Heat Pump Conference in Tokyo as equally successful.

PRESENT SITUATION

In the last three years we have been confronted with important changes. The first remarkable change affecting us has been the development of political stability in the Middle East. While this is a desirable development, it is, however, bringing about a rapid increase in energy prices. We are therefore again reminded of the oil shocks in the 1970s and must recognize just how limited our energy resources are, in today’s world.

Of course, the present situation does not necessarily mean that market application of heat pump technology is favorable at this time. But it does suggest that it is necessary to prepare for overall deployment of R&D in this technology. If we do not prepare, I believe that this prospective technology may not be utilized to save energy despite its considerable potential.

The second change involves global environmental problems, which all of you here are well aware of. This new situation has both advantages and disadvantages for promoting heat pump technology. From a positive standpoint is the problem of the so-called greenhouse effect. Although there may be some countermeasures under consideration, including a reliance on nuclear energy or renewable energy, social attitudes do not make them feasible at the present time. Therefore, because we recognize the important role of energy-saving technology, utilization of heat pump technology can be seen in a more favorable light.

Of course heat pump technology has disadvantages as well. Five years ago, no one could foresee the seriousness of the CFC problem as viewed by our society today. And since we cannot escape this problem, I would like to stress that success in solving it will be the key for future expansion of heat pump technology.

From the abovementioned, we can see just how rapidly and drastically our society is changing. Scientific knowledge, human attitudes, politics and so on are advancing so fast that they are causing dramatic social changes. And recently, technological progress has been lagging behind these other advancements. If such a slow pace is hampering our ability to solve these problems, we must work harder, and I hope everyone here is ready to take on a greater role in these efforts.

PERSPECTIVES TOWARD THE FUTURE

Taking into account the present situation, let us turn now our attention to the future, looking at it from two perspectives heat pump technology and marketing.

In the area of heat pump technology, we hope to find a way to use it without causing environmental destruction and while utilizing wasted heat energy to the maximum extent possible. The task facing us today is to improve the present heat pump system to the point where it does not harm the environment. In other words, we should aggressively search for new kinds of refrigerants for working fluids.

But these are only temporary means. For a long-term solution, we might have to wait for the development of a new heat-transfer cycle which is different from the conventional compression or absorption type. I am pleased to hear that there are some signs of such an innovative new development. In this effort, international collaboration will play an indispensable role.

Moreover, IEA was established for this purpose, and its work has been very useful. Above all, I believe that the Implementing Agreement on Advanced Heat Pump Systems in IEA is one of the best examples of international collaboration.

Apart from the environmental problems, we still have a lot to do on the technology side. We need to perform R&D to increase heat pump performance, especially in the area of temperature range and energy efficiency. This is essential because of its broader application.

In commercialization, economic principles always apply. The heat pump is no exception. That is, it is true that the economic considerations have impeded the advancement of heat pump technology as a whole. But as we have already seen, the situation is changing. We now need to take a broader perspective. As a matter of fact, in some countries heat pump type air-conditioning has become indispensable, even though oil prices have been kept very low. The cause of its success seems to be its excellent operability and convenience. In this regard, these advantages have surpassed in importance any economical disadvantages. Moreover, from the standpoint of commercialization, any new environmental problems will also have an effect. Those in the industry should always be aware of any changes in the current situation.

CONCLUSION

Ladies and Gentlemen,

As all of you know, collaboration based on the IEA Implementing Agreement on Advanced Heat Pump Systems has worked quite well. The importance of this collaborative work is becoming greater and greater, as shown by the large number of people participating in this conference. Through this Conference, we hope to exchange information from a variety of sources, including academia and private industries, that will help us confront future challenges in the area of heat pump technology.

In conclusion, I would like to introduce an old and familiar saying of the ancient Chinese philosopher, Confucius, to express our feelings about this conference.

It is very pleasant to review occasionally after learning,

It is very pleasant to welcome friends from a distance.

Thank you for your attention.

SESSION II

Advances in Electric Heat Pumps

Outline

Chapter 3: Heat Pump Operated with Refrigerants of Non-Azeotropic Mixture

Chapter 4: Compression-Absorption Heat Pumps

Chapter 5: Inverter-Aided Multisystem Air Conditioner (Air to Air Heat Pump) for Conditioning Use in Buildings

Chapter 6: Heat Pump with Thermal Storage

Heat Pump Operated with Refrigerants of Non-Azeotropic Mixture

T. Saito and E. Hihara,     Faculty of Engineering, The University of Tokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo-113, Japan

ABSTRACT

Mixed refrigerants have been pursued for their improved characteristics for use in heat pumps and refrigerating machines. Recently, the problem of ozone layer depletion in the stratosphere has also directed attention to mixed refrigerants to help reduce use of CFC such as R11 or R12. For mixtures, as well as the proper estimation of the equation of state, and thermal properties, the heat transfer characteristics were studied. The mixture of R11+R22 is promising from the thermodynamic standpoint but falls behind in thermal stability. A heat pump using R114+R22 was simulated. The advantage of non-azeotropic mixtures is recognized in the temperature distribution in the heat exchanger. The reduction of CFC may be evaluated in relation to the COP and the HCFC fraction in the mixture.

KEYWORDS

COP

heat pump

heat transfer

mixed refrigerant

simulation

INTRODUCTION

The high efficiency and the low compression ratio of a heat pump or a refrigerating machine have been pursued for saving energy. Mixed refrigerants have been studied for this purpose. Recently, the depletion of the ozone layer in the stratosphere has become an issue of great concern. According to assessment this depletion is related to the diffusion of CFC (R11, R12, R13, R113, R114) into the stratosphere. Regulation of CFC use has already been put into effect. Consequently, the mixing of HCFC into CFC is expected to be helpful in reducing the use of CFC, as well as improving the performance of the machine. To find a promising combination of refrigerants, it is useful to establish a simple method of calculation. The equation of state, and thermal properties are calculated, based on the mixing rules. From thermodynamic calculations, some kind of mixture may be anticipated to give higher values of COP than that of pure components. However, the heat transfer coefficient of phase change generally tends to become worse by mixing compared with that of pure components. The COP should be estimated by the simulation, which takes the heat transfer characteristics into consideration. In this report the effect of mixing of HCFC (R22) into CFC (R11, R12, R114) is examined.

EQUATION OF STATE

Equations of state may be divided into two types of Virial equation and Van der Waals equation. For the mixture of R12+R22, Kagawa(1983) used BWR equation of the former type.

For our simulation study of cycle using R114+R22, and R11+R22, the Pen-Robinson equation of the latter type is mainly used, because of its easier computation with 3 parameters. Referred to the research by Ototake (1985) the parameters were selected.

The mixture of refrigerants is treated as a sort of fluid which is assumed to have a pseudo critical point, and certain mixing rules are applied. Figure 1 is the Pressure-Enthalpy Diagram of 60mol%R114-40mol%R22.

Fig. 1 Pressure-Enthalpy Diagramm.

TRANSPORT PROPERTIES

The estimation methods of the transport properties of mixture are collected in the book by Reid et al. (1977).

Viscosity

Viscosity of gas under normal pressure is calculated by the Sutherland equation. Under high pressure it is estimated by using the pseudo-critical properties given by the Prausnitz-Gunn rules.

Viscosity of liquid is estimated by the method of averaging the logarithmic function of viscosity of each component. (Reid 1977 Chapter 9)

Thermal conductivity

Thermal conductivity is estimated by the Wassiljewa equation for gas and by the Filippov method for liquid. (Reid 1977 Chapter 10)

Diffusion coefficient

Diffusion coefficient is given by the Brokaw method for gas and by the Vignes correlation for liquid. (Reid 1977 Chapter 11)

HEAT PUMP CYCLE

The heat pump cycle is a fundamental one, as shown in Fig. 2.

Fig. 2 Heat pump cycle

The refrigerant changes in the process of a cycle as follows:

1-2: isentropic compression

2-3: isobaric cooling (condenser)

3-4: eisenthalpic expansion

4-1: isobaric heating (evaporator)

For non-azeotropic mixtures, the temperature during phase change does not remain uniform under constant pressure in the evaporator as well as in the condenser. The temperature rises with the increase of quality.

HEAT TRANSFER

The heat transfer characteristics of mixed refrigerant are described. (Hihara 1989) In the region of superheated vapor the Dittes-Boelter relation holds.

(1)

Forced convective boiling

Heat transfer analyses of mixtures have been tested by many investigators from various standpoints. (Miyara 1988) For mixture Bennett & Chen (1980) modified the Chen correlation by taking the mass transfer effect into account. In the heating process of mixed refrigerants, the LBP (lower boiling point) refrigerant first evaporates at the higher rate, and the boundary layer with a concentaration gradient is formed. As the boiling proceeds, the concentration of the LBP refrigerant decreases near the heating surface and that of the HBP (higher boiling point) refrigerant increases. Accordingly, the saturation temperature of the mixture increases. Therefore, the superheating to drive the boiling and evaporation decreases, causing the heat transfer to decline. The evaporative heat transfer coefficient of the mixture does not exceed that of any of the pure components, and is often lower. Forced convective boiling is composed of nucleate flow boiling, and forced convection vaporization.

For pure-component fluid, the heat transfer of two phase flow versus that of liquid flow is given by the equation

(2)

(3)

where

(4)

For mixtures the heat transfer is expressed by Eq. (2) with modification, making reference to the studies by Scriven (1959) and Van Strahlen (1979). The first term in Eq. (2) is modified by the modeling of bubble behavior enclosed in superheated liquid mixture, and is multiplied by the resulting factor ΔTe/(Te-Ts). (Fig. 3a, 3b)

Fig. 3 Modeling of vaporization of mixture.

According to the modeling of vaporization from a liquid film of a mixture, the second term is multiplied by the factor Δ/ΔTs. (Fig. 3b, 3c) Consequently,

(5)

Condensation

A great concern in condenser design is that the condensation is considerably restricted by the mixing of a small amount of non-condensible gas. This tendency must be considered in the condensation process of the mixture of different boiling point components. A number of studies on the condensation of vapor of binary mixture were reported. (Hijikata (1986), Koyama (1986))

For pure-component refrigerants, the condensation heat transfer is calculated by the Cavallini- Zecchin Equation.

(6)

(7)

The annular flow of a mixed refrigerant is schematically described in Fig. 4. As the vapor concentration boundary layer of the HBP component occurs, the interfacial temperature becomes Ti. Accordingly, the right hand term in Eq. (6) is multiplied by the factor ΔT/ΔTs.

Fig. 4 Modeling of condensation.

(8)

THERMODYNAMIC ANALYSIS OF CYCLE CHARACTERISTICS

To compare a cycle using a pure-component refrigerant with one using a binary mixture, it is important to establish standard condition.

Case 1 establishes the initial condensation temperature (T2′) (Fig. 2) and the evaporation-completed temperature (T1′) for both of the cycles to be compared.

Case 2 establishes T2’ and the evaporator inlet temperature T4 for both cycles. Case 2 gives the cycle using mixture more advantageous condition for comparison than Case 1.

R12-R22

Figure 5 shows the COP of the refrigeration cycle under the condition that the initial condensation temperature is 30° C and the compressor inlet temperature is -15° C. The maximum COP corresponds to 5 – 20% R22 mole fraction, which varies among the investigators. (Hihara 1986)

Fig. 5 COP of a refrigeration cycle (R12+R22).

For reference, it is reported that the mixture of R12 and R22 shows azeotropic characteristics (azeotropic point -70° C at 75 wt%, -18° C at nearly 0 wt% of R22). (Kriebel 1967)

R11+R22, R114+R22

Consider the effect of mixing R22 into R11 or R114. The cycle condition is given below.

Condensation initiation temperature: 360 K

evaporator inlet temperature: 268 K

Subcooling: 5 K

Superheating: 10 K

For R114+R22, the ratio of the exit pressure to the inlet pressure of the compressor is almost constant below 50 mole % of R22.

The COP characteristics are described in Fig. 6.

Fig. 6 COP of heat pump cycles (R11+R22, R114+R22).

Under the condition of the Case 1, the cycle using R114 is not improved by adding R22, though it has a maximum around 30 mole fraction of R22. The COP of the cycle using R11 is considerably improved by adding of R22 and the COP reaches a maximum at 50% mole fraction of R22. According to the note by refrigerant manufacturer, R11 is thermally stable at 80° C but decomposes itself at 120° C under actual circumstance. Therefore this kind of mixture seems more adequate for refrigerating machines than heat pumps.

SIMULATION

The experimental apparatus of a heat pump has a capacity of a car air condir-tioner.

The refrigerant is a mixture of R114 and R22.

The compressor has a capacity of 81.6 cc/rev with 10 cylinders.

The lubrication oil is separated out by a series of two oil separators. The cooler is installed between the compressor and the condenser, so that the inlet state of the condenser is regulated to be isentropic to the inlet state of compressor.

The superheating at the exit of the evaporator is 5 K.

The subcooling at the exit of the condenser is 0 K.

The evaporator or condenser is composed of concentric tubes. The outer tube contains the water which exchanges heat with the refrigerant in the inner tube. The length of the heat transfer tube and the water conditions (temperature and flow rate) are adjusted.

For water which exchanges heat with refrigerant the heat transfer is expressed in the the following equations based on the measurement.

In the evaporator

(9)

and in the condenser

(10)

where de is the diameter gap between the outer tube and the inner one. The heat transfer relations of Eq.(1)–(10) are taken into consideration for the simulation, as well as thermodynamic relations. The calculated total coefficient of performance (COP) is compared with the experimental results in Fig. 7. The calculation shows that the COP reaches a maximum at 10 mole % R22. The heat transfer area of Case B is adjusted to be 1.67 times of that of Case A. The COP of Case B is larger than that of Case A, and the ratio of COPmax to COPpure (R114) shows the same tendency between the two cases.

Fig. 7 Simulation of a heat pump cycle (R114+R22)

As the temperature of the pure refrigerant is held constant during the phase change there occurs a pinch point in the temperature difference for heat transfer. (Fig. 8) For non-azeotropic refrigerant mixtures, the temperature difference can be regulated to be nearly uniform, as shown in Fig. 9. It becomes smaller if there is a large heat transfer area. These characteristics provide advantages for efficient use of available energy. Consequently, it is important for use of mixed refrigerants to make advantage of the small uniform temperature difference at heat exchangers as much as possible. In the calculation of COP, Fujii (1987) evaluated the effect of heat transfer by changing the FK-value (heat transfer area times average overall heat transfer coefficient). He noticed that the COP increase by augmentation of FK-value for mixed refrigerants is larger than that for pure refrigerant.

Fig. 8 Temperature distribution in condenser (R114).

Fig. 9 Temperature distribution in condenser (R114+R22).

EXAMPLES OF PRACTICAL USE

One of the typical examples of using a non-azeotropic mixture of refrigerants is the liquefaction installation of natural gas.

The examples of using the mixture of CFC and HCFC in Japan are given in Table 1. Please note that the Japanese national project of the Superheat Pump is not included in this table.

Table 1

Examples of practical use of mixed refrigerants (Japan)

CONCLUSION

It is important to discuss the usefulness of mixed refrigerants for improving the performance of heat pumps, and reducing the consumption of CFC. However, there are a great number of combinations of refrigerants and compositions. The formation of equation of state for mixtures was described, as well as the estimation of thermal properties. The equation of forced convection vaporization of mixtures was derived, based on experimental results. Condensation heat transfer was expressed in similar form.

For the mixture of CFC+HCFC the improvement is defined by the ratio of [COP(mixture)/COP(pure CFC)]. It is noteworthy that this factor is influenced by the temperature condition. When the evaporation completed temperature (T1′) and the initial condensation temperature (T2′) are set respectively equal between the cycles to be compared, the thermodynamic calculation shows that the improvement factors for R12+R22 and R114+R22 remain almost 1. For R11+R22, the improvement factor is 1.6 around 50% (R22) mole fraction. But because R11 begins to decompose around 100° C, it is not appropriate for heat pumps. It is adequate rather for refrigerating machines. By a simulation that includes heat transfer characteristics, the cycles using R114+R22 were evaluated under given heat source condition.

The total COP is influenced by the heat transfer area. If the heat transfer area is large enough, the improvement factor is 1.18.

The CFC consumption reduction rate by mixing increases with COP, and with the mixed fraction of HCFC. For actual use, the compression ratio, the influence of oil on the heat transfer, and other factors should be considered.

REFERENCES

Bennett, D. L., Chen, J. C. Forced convective boiling in vertical tube for pure components and binary mixtures. AIChE J. 1980; 26:454–461.

Fujii, T., Koyama, S., Miyara, A. Theoretical consideration on the characteristics and the performance evaluation for a heat pump cycle of non-azeotropic refrigerant mixtures. JAR Trans. 1987; 4:27–34.

Hihara, E., Muneta, Y., Saito, T. Characteristics of a mixed refrigerant vapor compression cycle. JAR Trans. 1986; 3:115–122.

Hihara, E., Tanida, K., Saito, T. Forced Convective Boiling Experiments of binary Mixtures. JSME International J. 1989; 32:98–106.

Hijikata, K., Mori, Y., Himeno, N., Inagawa, M., Takahasi, K. Film condensation of a binary mixture of vapors. JSME Trans.(B). 1986; 52:2195–2201.

Kagawa, N., Takaishi, Y., Uematu, M., Watanabe, K. JSME Trans. (B). 1983; 49:2811–2820.

Koyama, S., Miyara, A., Fujii, T., Takamatsu, H., Yonemoto, K. Condensation of refrigerant mixtures R22+R114 inside a Horizontal Tube. JSME Trans. (B). 1988; 54:1447–1452.

Kriebel, M. Phasengleichgewichte zwischen Fluessigkeit und Dampf im binaeren System R22-R12. Kaeltetechnik-Klimatisierung. 1967; 19:8–14.

Miyara, A., Takamatsu, H., Koyama, S., Yonemoto, K., Fujii, T. Forced convective bnoiling of nonazeotropic refrigerant mixture of R22 and R114 inside a horizontal tube. JSME Trans. (B). 1988; 54:2523–2528.

Ototake, N. (1985). Inquiry into selection of optimum working fluid from simple and compound systems. Research on effective use of thermal energy. 5.

Reid, R. C., Prausnitz, J. M., Sherwood, T. K. Properties of gases and liquids, 3rd Ed. McGraw-Hill Inc., 1977.

Scriven, L. E. On the dynamics of Phase Growth. Chem. Eng. Sci. 1959; 10:1.

Van Strahlen, S. J., Cole, R. Boiling Phenomena. Hemisphere. 1979.

Nomenclature

Bo: Boiling number (q/(Ghfg)

d: Inside diameter

G: Mass velocity

hfg: Latent heat of vaporization

l: heat transfer length

Pr: Prandtl number

q: Heat flux

Re: Reynolds number

T: Temperature

W: Fow rate

W: Concentration

X: quality

Xtt: Lockhart and Martinelli parameter

Z: coordinate along heat tube

α: Heat transfer coefficient

δ: Boundary layer thickness

μ: Viscosity

λ: Thermal conductivity

ρ: Density

Subscripts

b: Bulk

c: Condenser

e: Evaporator

g: Vapor

i: Interface

l: Liquid

m: Mixture, Mass

s: saturated

t: Thermal

tp: Two-phase

Compression-Absorption Heat Pumps

L. AHLBY* and D.L. HODGETT**,     *Chalmers University of Technology, Gothenburg, S-41296 Sweden; **Electricity Research Centre, Capenhurst, Chester, CH1 6ES, UK

SUMMARY

This paper describes the recent work worldwide on compression-absorption heat pumps, sometimes called the compression heat pump with solution loop. Most workers have concentrated on the use of NH3/N2O as working pair, a choice justified both in terms of performance and environment. Work has also tended to concentrate on higher temperature applications, an area for which the pair NH3/H2O is well suited. Although several large scale experimental plants have been built and operated there is still considerable work to be done before such systems can be considered practical.

INTRODUCTION

The idea of combining the compression and absorption cycles to produce a hybrid machine has been attributed to Altenkirch (1950) who first published it in 1950. More recently evidence has come to light that the idea had been put forward much earlier by Nicholle (Higham, 1980) and Sloan and Roncin (1920). However, apart from some theoretical analysis no real effort was put into what was for long considered a scientific curiosity. Only with the energy crisis of the 1970’s did the idea begin to be explored Nowotny (1979a, b), Lotz (1981) and Mucic and Scheuermann (1984). All of these considered the simplest cycle (Fig. 1) with NH2/H2O as working pair for low temperature applications such as refrigeration or space-heating. More recently laboratory scale equipment has been constructed to obtain experimental data on such systems (Stokar, 1986; Stokar and Trepp, 1987, Rane et al, 1989) and one pilot-scale plant was built which was incorporated in a distinct heating network (Mucic and Scheuermann, 1984). However, later work has shown that the cycle with NH3/H2O is more suitable for higher temperature applications and could be used as an alternative to compression heat pumps with R114, H2O, cascaded R12/R114 systems or absorption systems based on H2O/LiBr. 11,12. In the recent past Malewski (1987,1989), has constructed a 500kW heat output machine in the laboratory which was capable of achieving an output temperature of 120°C while Mucic (1989) has built a 1MW installation within a chemical plant for operation at up to 115°C.

Fig. 1 Simple single stage compression-absorption cycle

Meanwhile, theoretical studies (Alefeld, 1982; Rademacher and Howe, 1988; Ahlby, 1987), have shown the diversity of cycle combinations that are possible and how these could be applied over a wide range of operating conditions, Ahlby et al (1989) has shown that optimisation of the cycle parameters can give significant gains in cycle performance for a given set of external operating conditions.

Several ideas for improving the design or operation of the cycle have been put forward such as using the compressor lubricant as the solvent in the solution loop (Lourdudoss and Stymne, 1986), lubricating the compressor by the solution (Malewski, 1981); Patnode and Leonard, 1976) replacing NH3/H2O by other fluids such as R22 with organic solvents (Hodgett and Ahlby, 1987; Pourreza-Djourshari and Rademacher, 1986) or to use a pair of refrigerants with very widely differing boiling points (such as R13b1/R11 (Rademacher, 1988)). These activities illustrate the fact that the cycle does show significant advantages over the single fluid compression heat pump but that there are still serious problems to be resolved before it can be considered a realistic option. The fall in fossil fuel prices since 1984 has seriously affected the development of the cycle as governmental agencies have cut back funding but the current concern with environmental considerations (such as the role of CFC’s in the depletion of the ozone layer and that of energy conservation in reducing CO2 emissions) may give fresh impetus to its development.

POSSIBLE CYCLES

It is not the purpose of this paper to describe in detail the myriad possibilities for combined cycles which result from a theoretical analysis. These result from the number of possible pressure and temperature levels, the arrangement of absorbers and desorbers in relation to each other and the possibilities for internal heat exchange etc. Alefeld (1982) has published extensively on this topic and has identified several combinations which provide particularly good characteristics for certain applications, while Ahlby (1987) has analysed several of these in detail. Rademacher (Rademacher and Howe, 1988; Rademacher, 1988) has, in particular, emphasised the benefits which can be achieved by judicious internal heat exchange. However, all of these depend on the successful demonstration of the simple cycle as a practical device before the more complex cycles (and no doubt their more complex control aspects) can be demonstrated.

One variant which has been demonstrated is the use of the solution as lubricant for the compressor (Malewski, 1981; Patnode and Leonard, 1976), which results in a modification of the simple (i.e single stage) cycle (see below).

WORKING FLUIDS

All the experimental work so far has concentrated on using NH3/H2O as the working pair. This is because the cycle requires pressures above atmospheric to reduce the compressor size and to avoid excessive compression ratios. On these grounds H2O/LiBr is unsuitable. Another option would be to use a CFC or HCFC with an organic solvent, but the evidence from work on absorption heat pumps and small amount of theoretical work done (Hodgett and Ahlby, 1987; Pourreza-Djourshari and Rademacher, 1986) show that such pairs are likely to give inferior performance to NH3/H2O and have significant penalties with regard to the solution heat exchanger (low heat transfer coefficients) and solution pump (high pressure differences, low flow and aggressiveness). Thus, although NH3 has not been favoured by the refrigeration industry because of its flammability, and to a lesser extent its toxicity, it remains the best refrigerant for this cycle. There are strong arguments that the NH3/H2O compression/absorption machine can be relatively safe. For instance, the mixture has low flammability and NH3 only exists in high concentrations in the vapour phase (and then in small quantities) while a well designed and maintained machine with no, or few, flanged connections will be sufficiently safe for industrial installations. Indeed there is a trend back to NH3 in the industrial refrigeration market as a result of the restrictions on the use of R12. However, Ahlby et al (1989) has shown that the cycle is best at high temperatures so that there could be a need for an alternative working fluid pair for lower temperature applications ie spaceheating and distinct heating. This market could possibly be satisfied by zeotropic mixtures operating within conventional compression machines with suitably designed heat exchangers so that it may not be necessary to search for a fluid pair for compression-absorption heat pumps for this niche.

EXPERIMENTAL RESULTS

Six experimental rigs and one commercial machine are known to have been constructed. Table 1 summarises their operational data (results have not yet been published by the group at the Technical University of Munich). The first experimental plant was constructed by Mucic in a district heating system in Heidelberg. The reciprocating compressor was driven by a gas engine. This plant operated successfully under somewhat artificial conditions (cooling district heating return water to reheat some of it for local use) for several hundred hours showing a high COP (11.3). However a considerable part of the heat supply could have been obtained by heat exchange between the source and sink. The machine included rather complex heat exchangers in order to obtain counterflow of the solution without back mixing, and several systems for oil separation and recovery.

Table 1

Basic data for the seven plants

The experimental system at ETH Zurich (Fig. 2) provided much useful data on heat transfer coefficients in the desorber, absorber and internal heat exchanger (Stokar, 1986; Stokar and Trepp, 1987). The test plant showed a significant improvement in performance, needing 22% less motive power than a system using a pure refrigerant and if the desorber had behaved as anticipated then this would have been 32%. The desorber was constructed as a falling film device but this exhibited wavy flow resulting in uneven distribution of liquid over the coil and therefore a much lower heat transfer coefficient than anticipated.

Fig. 2 Experimental system at ETH Zurich

The experimental device at the University of Maryland (Rane et al, 1989) showed that the performance increased with increasing NH3 concentration in the solution loop. This is attributed to the increase in the desborber pressure which reduces the volume flow and hence the pressure drop in the desorber. Thus the compressor experienced a smaller compression ratio and its work is decreased. An increase in strong solution flow rate at reduced capacity also reduced the COP. This was attributed to lower solution heat exchanger effectiveness. This effect was counterbalanced at high desorber capacity at small soltuion flow rates by the loss of efficiency of the compressor due to increased compression ratio.

The major conclusion of the results so far from this rig is that there is an optimum size of solution heat exchanger for a given operating condition and that the desorber and absorber temperature glides should first be optimised before the solution heat exchanger is designed. This way a maximum COP can be achieved.

The large scale (500 kW heat output) plant at Borsig (Malewski, 1987, 1989) differed from the other plants in that the weak solution from the desorber was injected into the single-screw compressor after passing through the solution heat exchanger (fig. 3). This was done to avoid the large superheats produced by the compression of NH3, which at suction is already superheated relative to saturated NH3 (but in equilibrium with the desorber solution). The solution is separated from the vapour after compression and distributed over the absorber tubes. A Hall single-screw compressor was chosen because it has much lower bearing loads than the twinscrew and the alternative design by Grasso needed oil injection. Considerable problems were met with the compressor bearings, these failing after 20 hours operation. Replacement of the metal cages by synthetic ones gave satisfactory operation for 150 hrs, including many start/stop sequences, before the main bearing on the control side failed. It was concluded that sealed grease packed bearings would have to be used to overcome this problem.

Fig. 3 Schematic diagram of Borsig experimental heat pump

Difficulties were also experienced with the solution pump during capacity control. Rapid falls in desorber pressure resulting from reduced compressor capacity caused the solution pump suction head to fall below that which it could accommodate. The pump had therefore to be placed in a pit. The machine was also found to be sensitive to fluctuations in loading resulting from variations in the cooling water loop used for heating the desorber and cooling the absorber.

The results showed that the machine showed an improvement in COP (over a single fluid machine) for concentration differences between strong and weak solutions of greater than 10% and that this improvement increased with increasing concentration difference. The conclusion was that the machine was best suited for high temperature applications using heat sources between 40 and 80°C.

The only commercial installation of a compression-absorption heat pump has been in a chemical plant producing sodium carbonate (Mucic, 1989). Brine at 110°C is flashed to 95°C (0.85 bar) and the vapour produced is condensed to heat the desorber. The resorber takes 115°C feed water and produces steam at 1.7 bar (115°C). Figure 4 is a simple flow diagram of the system. A COP of 9.1 was claimed for the system.

Fig. 4 Steam raising system by Thermo-Consulting - Heidelberg

However after 2 weeks operation the system stopped because of failure of oil seals in the compressor. The seals were replaced and the system restarted (this was in 1987). No experimental results are known to have been published and there is no further information available on the reliability of the plant.

The COP of 9.1 operating between 95°C and 115°C is not outstanding since this process could have been achieved by the direct compression of water vapour (either produced by flashing the brine or by heat exchange to produce clean steam in a reboiler). This would have resulted in a lower capital cost plant with a COP equal to or higher than that achieved by the compression/absorption heat pump. However this was intended to be the first stage of a larger plant in which the brine would have been flashed through several pressures giving a much larger, and hence more economically viable plant, operating over a temperature difference too large for simple steam compression.

THEORETICAL RESULTS

Stokar and Trepp (1987) conducted a theoretical sensitivity analysis by fitting experimental data to empirical equations and then varied many of the operating parameters to test the sensitivity of the system behaviour to them. From this he concluded that the COP was most sensitive to the overall heat transfer coefficients in the resorber and desorber (which was lower in their experimental rig).

When the overall heat transfer coefficient in the desorber equals that in the absorber (200 W/m²k) then the COP would be improved from 4.39 to 4.89 for heating water from 40 to 70 while cooling water from 40 to 15°C. This was compared to a single fluid heat pump operating with the same conditions which, with NH3 as working fluid, gave a COP of 3.3. However, these conditions are very favourable to the compression-absorption heat pump. The single stage compression heat pump operates between 7 and 30.5 bar so that the addition of an economiser or a flash intercooler would normally be used to enhance its performance.

The University of Maryland group have also conducted a theoretical analysis (Amrane et al, 1989) of a single stage compression-absorption machine using NH3/H2O. Results were obtained for the heat source (water) being cooled from 10°C to 0°C, the heat sink (water) being heated from 60°C to 70°C, a compressor and pump isentropic efficiency of 70% and fixed UA values for the desorber, absorber and solution heat exchangers. The weak solution concentration was initially fixed at 0.5 kg/kg but was later varied to test the sensitivity of the system performance.

The results show (fig. 5) that increasing the NH3 concentration increases the capacity (because of the increased vapour pressure and hence density) while the COP shows a maximum at about 30% concentration. The latter effect is due to the conflicting effects of increased absorber heat output (the increased capacity) and the increased compressor work due to the higher mass flow of NH3 (this does not include the effect of changing compression ratio and its influence on compressor isentropic and volumetric efficiencies). The maximum COP coincides with the best match between the temperature changes of the solution within both the absorber and desborber to th 10K charges in the source and sink. Varying the UA values in the three heat exchangers showed a strong sensitivity, particularly for the absorber, and that a UA value of 1kW/K in the solution heat exchanger and 4 kW/K in the absorber and desorber were necessary to ensure optimum performance. The authors also found that there was an optimum solution volumetric flow rate relative to COP and the optimum showed an 18% improvement in COP over a NH3 heat pump.

Fig. 5 COP and heat capacity versus the weak concentration (Amrane, K., et al)

Ahlby (1987, Ahlby et al, 1989) has made an exhaustive analysis of the cycle both to determine optimum operating conditions for given external conditions and to compare its performance to conventional systems for several applications. In this work NH3/H2O was taken as the working fluid, the compressor was assumed to be a lubricated twin-screw compressor with isentropic and volumetric efficiencies fitted to experimental data for NH3. The cycle operating conditions were first calculated (Ahlby, 1987) for a fixed temperature lift (60K) over the cycle, a maximum cycle pressure of 25 bar and a maximum concentration of 80% by weight NH3. The results showed that for a fixed temperature glide in the absorber the COP was not very sensitive to the absorber pressure but that the compressor displacement decreased with pressure and with increasing sink outlet temperature. The cycle seemed to give best results for the sink temperatures above 60°C and was superior to R12 and R114 in both COP and compressor displacement for all operating conditions when the sink is heated to at least 60°C. Results were also obtained for two variations of the cycle, one with two absorbers at different pressure levels (to give a lower temperature glide in the heat sink than in the heat source) and one with the addition of an evaporator and condenser (the latter transferring heat internally to the desorber, give a high lift).

Later (Ahlby et al, 1989) the work included the heat transfer within the heat exchangers by assuming UA values, and the cycle is optimised by varying the temperature glide in the absorber and also its pressure to obtain the maximum COP. The losses in the cycle (the losses due to imperfect heat exchange in the solution heat exchanger, the compressor inefficiency and the expansion loss) were then analysed to find which most strongly influences the cycle performance.

Figure 6 shows the results obtained for one case considered in which the sink is heated by 1MW from 75 to 80°C and the source cooled from 15 to 10°C. This represents a district heating case with unfavourable conditions for the cycle (low temperature glide). The cycle is compared to R12 without economiser with two different conditions, one with a high heat condenser UA value (200kW/K) and one with a low UA value (100kW/K). The evaporator UA value is assumed to be 60% of these values. The results show that there is an optimum temperature glide in the absorber of about 14K (compared to 5K in the sink) and that maximum COP coincides with the minimum SCD (compressor displacement) which is a very favourable situation. This optimum is superior to both R12 cases. Figure 7 shows the results for an industrial case in which steam at 1.7 bar (117°C) is raised by the absorber and heat is extracted from a source by cooling it from 85 to 80°C. The optimum COP of 4.6 compares to the experimental results of between 4 and 4.5 obtained by a heat pump using steam and an oil free screw compressor (Chalmers et al. 1987). Again the compressor displacement has a minimum, but this occurs at an absorber temperature glide of about 15K compared to the optimum COP at 18K.

Fig. 6 COP and specific compressor displacement against absorber temperature gradient for a district heating case (Ahlby et al, 1989).

Fig. 7 COP and specific compressor displacement against absorber temperature gradient for a steam raising case (Ahlby et al, 1989)

The optimum value of ΔTabs, i.e that for which the cycle performance is maximised, is a compromise between the ΔTabs that minimise the losses caused by the finite solution heat exchanger area, Q1, and the pump work, Wp, and the value that minimises the pressure ratio across the cycle. Depending on the size of the external temperature gradients, the optimum ΔTabs is found close to the ΔTabs giving the minimum compression ratio or further away depending on the influence of Q1 and Wp. In those cases where one or both external temperature gradients are large the optimal Tabs is found closer to the larger gradient.

When both external gradients are small the choice of optimal ΔTabs is more important. Even though a small ΔTabs minimises the pressure ratio across the cycle the resulting Q1 and Wp are very large. Since these decline with increasing ΔTabs the COP is maximised for a ΔTabs considerably larger than any of the external temperature gradients.

However, the location of the maximum COP also depends on the size of the solution heat exchanger. As the UA value of the solution heat exchanger increases, the optimum ΔTabs would be exactly equal to the ΔTabs giving the minimum compression ratio, since the heat losses in the solution heat exchanger would then be completely eliminated. Thus, as the influence of the solution heat exchanger decreases, the characteristics of a compression/absorption cycle approaches those of a compression heat pump. This situation cannot be obtained, even with an infinitely large heat exchanger area, since the solution streams differ in both mass flow and specific heat capacities.

This work is now being extended to test the sensitivity to total heat exchanger area and its distribution within the system.

CONCLUSIONS

The work so far on the compression-absorption cycle shows that it can achieve better performance than the simple fluid cycle and this is particularly so for high output temperatures. Although the COP enhancement is greater at larger source and sink temperature glides the cycle still shows superior performance, when optimised, over conventional systems even for glides as low as 5K. However, there are still many technical problems which have to be overcome, such as the design of heat exchangers (absorbers and desorbers) which provide high heat transfer coefficients with no back mixing. A major unsolved problem is the development of a suitable compressor. Either an unlubricated compressor, which can operate over moderately high compression ratios (>4) with high efficiency, or alternatively a lubricated compressor could be used if a satisfactory lubricant can be found which will not foam, or form an emulsion (with the small amount of H2O present in the vapour).

The cycle also shows a capability of providing a wide degree of capacity control, but the experimental work so far has shown that the machine is very sensitive to system variations (such as changes in source or sink conditions) and so control strategies need to be carefully researched.

Finally, the cycle is capable of replacing R12 at (higher temperatures) and R114 systems so that it could make a useful contribution to reducing the dependancy on CFC’s for higher temperature applications.

There is therefore need for more work on the cycle, particularly in the investigation of absorber and desorber design, compressor development, and control before further commercial installations should be considered.

ACKNOWLEDGEMENTS

The authors are grateful to the Swedish National Board for Technical Development (STU), the Swedish Council for Building Research (BFR) and the Swedish National Energy Administration (STEV) for supporting this work financially. They also wish to thank the other workers in the field for providing data and for giving permission to republish their results.

REFERENCES

Ahlby, L. (1987). Compression/Absorption cycles for large heat pumps-system simulations. Licentiate Thesis, Department of Heat and Power Technology, Chalmers University of Technology, Gothenburg, Sweden

Ahlby, L., Berntsson, T. and Hodgett, D. L. (1989). Kompressions/Absorptions - Drivna Varmepumpar-systemaspekter., Foredrag vid 13 Nordiske Kjolemote och 4. Nordiske Varmepumperdager i Loen, Norway, 30 May-3 June

Ahlby, L. and Hodgett, D. L. (1987). Compression/Absorption systems simulation of two cycles for different applications. XVII Int. Congress of Refrigeration, Vienna, Volume B, pp