lable at ScienceDirect

Energy 35 (2010) 1403–1411

Contents lists avai

Energy

journal homepage: www.elsevier .com/locate/energy

A study of organic working fluids on system efficiency of an ORC using low-gradeenergy sources

T.C. Hung a, S.K. Wang a,*, C.H. Kuo b, B.S. Pei c, K.F. Tsai d

a Department of Mechanical Engineering, I-Shou University, Dashu Township, kaohsiung county 84001, Taiwanb Department of Mold & Die Engineering, National Kaohsiung University of Applied Sciences, Taiwanc Department of Engineering and System Science, Tsing-Hua University, Taiwand RiteKom Photonics Corp., Taiwan

a r t i c l e i n f o

Article history:Received 20 April 2009Received in revised form23 November 2009Accepted 24 November 2009Available online 29 December 2009

Keywords:Organic fluidsOrganic rankine cycleLow-grade energy sourcesSystem efficiencyOverall irreversibility

* Corresponding author. Tel.: þ886 7 6577711; fax:E-mail address: skwang@isu.edu.tw (S.K. Wang).

0360-5442/$ – see front matter � 2009 Elsevier Ltd.doi:10.1016/j.energy.2009.11.025

a b s t r a c t

Rankine cycles using organic fluids (as categorized into three groups: wet, dry, and isentropic fluids) asworking fluids in converting low-grade energy are investigated in this study. The main purpose is toidentify suitable working fluids which may yield high system efficiencies in an organic Rankine cycle(ORC) system. Efficiencies of ORC systems are calculated based on an assumption that the inlet conditionof the working fluid entering turbine is in saturated vapor phase. Parameters under investigation areturbine inlet temperature, turbine inlet pressure, condenser exit temperature, turbine exit quality, overallirrversibility, and system efficiency. The low-grade energy source can be obtained from a solar pond or/and an ocean thermal energy conversion (OTEC) system. Results indicate that wet fluids with very steepsaturated vapor curves in T-s diagram have a better overall performance in energy conversion efficienciesthan that of dry fluids. It can also be shown that all the working fluids have a similar behavior of theefficiency-condenser exit temperature relationship. Furthermore, an appropriate combination of solarenergy and an ORC system with a higher turbine inlet temperature and a lower condenser temperature(as operated deeply under sea level) would provide an economically feasible and environment-friendlyrenewable energy conversion system.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

In heavily industrialized countries like Taiwan, a steady supplyof electricity has become an important issue faced by both thegovernment and the industries. This problem is even more aggra-vated by an accelerated consumption of fossil fuels which havecaused many environment-related concerns. Power generated fromlow-grade energy resources has become an inevitable option forcountries like Taiwan. Low-grade heat from renewable energysources is considered to be a good candidate to generate electricity.Among those sources, OTEC and solar energy are typically utilizedin converting low-grade heat into power generation and otherapplications [1–9].

Power generation using organic fluids in recovering low-gradeenergy sources has increasingly attracted attention. Several ORCsystems have been installed for recovering waste heat and widelyused for converting renewable energy into power. Hung et al. [10]

þ886 7 6578853.

All rights reserved.

used some cryogens as working fluids in an ORC operated betweentwo isobaric curves, and they found that the system efficiencyincreases and decreases for wet and dry fluids, respectively, whenturbine inlet temperature – the main parameter under consider-ation – increases. Liu et al. [11] investigated the effects of severalworking fluids on an ORC for waste heat recovery. They found thatthe presence of hydrogen bond in certain molecules such asammonia, water, and ethanol may cause these fluids behave likewet fluids due to their large vaporization enthalpies, and thesefluids are regarded as inappropriate for ORC systems. They alsofound that the maximum waste heat recovery efficiency occurs atan appropriate evaporator operation temperature between theinlet temperature of waste heat and the condensing temperature ofheat sink. OTEC systems use the ocean’s natural thermal gra-dientdthe fact that the ocean’s layers of water have differenttemperaturesdto drive a power-producing cycle. Madhawa Het-tiarachchi et al. [12] proposed a cost-effective optimum designcriterion for ORCs. Regardless of using geothermal heat sourceinstead of OTEC or solar energy, they found that the choice ofworking fluid can greatly affect the objective function which isa measure of power plant cost – in some instances the differences

Nomenclature

cp specific heath enthalpyh4a turbine exit enthalpy for an irreversible processT temperature_I irreversibility ratev specific volumew12 work done by pumpw34a work done by turbine for an irreversible processhfg latent heatk thermal conductivity_m mass flow rate

p pressureq23 heat added to evaporator

s entropyx qualityhth thermal efficiencyw34 work done by turbinef availability ratio

Subscripts1w4 locations of stateso ambientsys systemt turbineH HighL Lowtot total

T.C. Hung et al. / Energy 35 (2010) 1403–14111404

could be more than twice. Saleh et al. [13] used alkanes, fluorinatedalkanes, ether and fluorinated ethers as working fluids in ORCs forgeothermal power plants at high pressures up to 20 bars. Theyfound the highest thermal efficiency was 0.13 for the high boilingsubstances with overhanging saturated vapor line in subcriticalprocesses with n-butane. Manolako et al. [14] combined an ORCengine with a reverse osmosis (RO) desalination unit. They foundthat ORC can be effectively used to expolit low temperature thermalsources (i.e., in the range from 40 to 70 �C). Such low temperaturesources can be available from solar collectors and geothermal fields.Desai and Bandyopadhyay [15] found that the basic ORC can bemodified by incorporating both regeneration and turbine bleedingto improve thermal efficiency. They propsed a methodology forappropriate integration and optimization of an ORC as a cogenera-tion process with the background process to generate shaft-work.

As stated in the previously mentioned studies, ORCs arepotentially feasible in recovering low-grade energy and generatingpower if adequate working fluids are used. The efficiency canfurther be improved if the high temperature end of the cycle isboosted by solar energy. The objective of this study is to gaina comprehensive understanding of the thermodynamic perfor-mances of an ORC using various working fluids. System efficienciesare calculated for an ORC using OTEC as the heat source and sinkwith and without the boost of solar energy. The following analysesfocus on thermodynamic performances of the ORCs as scopingcalculations without considering detailed system integration, e.g.,the solar thermal pond or solar energy collector served asa boundary condition of the inlet temperature of the evaporator.Detailed calculations of pressure losses and heat transfer in

Fig. 1. A schematic flow diagram of an ORC system.

evaporator and condenser are also ignored since they dependstrongly on materials and configurations of the system compo-nents. Instead, irreveribilities of the working fluids in various majorcomponents of the cycle are calculated to evaluate the effects ofthose losses. Parameters under consideration are turbine inlettemperature, turbine inlet pressure, condenser exit temperature,turbine exit quality, overall irrversibility, and system efficiency.

An ORC system using low-grade energy sources is depicted inFig. 1. The system is composed of an evaporator (waste heat boiler),a turbine expander, a condenser, and a pump. A working fluid flowsinto the evaporator in which the high-temperature heat source(which may come from the warm seawater or a solar pond) isutilized. The vapor of the boiling fluid enters the turbine expanderand generates power. The exit fluid from the turbine expander thenenters the condenser in which the low-temperature cooling water(i.e., the cold seawater) is utilized to condense the fluid. Finally,a fluid pump raises fluid pressure and feeds the fluid into theevaporator to complete the cycle. So long as a temperature differ-ence between the high- and low-temperature ends is large enough,the cycle will continue to operate and generate power. The objec-tive of this study is focused on thermodynamic analyses of theworking fluids and the overall system efficiency rather than hard-ware arrangements such as the system integration of solar energyand OTEC. Therefore, issues regarding material selections, compo-nent configurations, frictional losses, heat transfer performances ofthe evaporator and condenser, and cost analysis are not consideredin this study.

2. Selection of working fluids

In an ORC, a suitable selection of the working fluids is a criticalfactor for achieving an efficient and a safe operation. Thermo-physical properties of various working fluids used in this study arelisted in Tables 1 and 2.

Each working fluid has its own range of applicability accordingto its thermophysical properties under the considerations of a high

Table 1Thermophysical properties of wet working fluids.

Working fluid R–11 R–12 R–152a R–500 R–502

Turbine exit condition Two-phase

Two-phase

Two-phase

Two-phase

Two-phase

Critical temperature (K) 471 111.8 389.4 378.6 355.2Critical pressure (MPa) 4.41 4.125 4.45 4.43 4.075Normal boiling

temperature (K)296.2 243.2 248.0 239.5 227.6

hfg @ 1 atm (kJ/kg) 178.8 166.0 318.4 200.8 172.5

Table 2Thermophysical properties of dry working fluids.

Working fluid R–113 R–114 R–123 C6H6 C7H8 C8H10

Turbine exit condition S-heateda S-heateda S-heateda Satb Satb Satb

Critical temperature(K) 487.3 419 456.9 562 591 616.2Critical pressure (MPa) 3.41 3.261 3.67 4.89 4.119 3.51Normal boiling temperature (K) 296.3 276.7 411.5 353 383.6 411hfg @ 1 atm (kJ/kg) 182 131.8 171.5 395.4 362.5 339.9

a S-heated: Super-heated.b Sat: Saturated.

T.C. Hung et al. / Energy 35 (2010) 1403–1411 1405

efficiency and a safe operation. Important factors of the workingfluids needed to be considered are listed below:

1. Toxicity of working fluid: All organic fluids are inevitably toxic.A working fluid with a low toxicity should be used to protectthe personnel from the threat of contamination in case ofa fluid leakage.

2. Chemical stability: Under a high pressure and temperature,organic fluids tend to decompose, resulting in material corro-sion and possible detonation and ignition. Therefore a chemi-cally-stable working fluid operated under working conditionsshould be selected.

3. Boiling temperature: Some of the organic fluids have a very lowboiling temperature under atmospheric pressure. For thosefluids, the temperature of cooling water in the condensershould be reduced. This can result in a more stringentrequirement for the selection of the condenser.

4. Flash point: A working fluid with a high flash point should beused in order to avoid flammability.

5. Specific heat: A high value of specific heat represents a highload for the condenser. Hence a working fluid with a lowspecific heat should be used.

6. Latent heat: A working fluid with a high latent heat should beused in order to raise the efficiency of heat recovery.

7. Thermal conductivity: A high conductivity represents a betterheat transfer in heat-exchange components.

The fluids under consideration in this study are refrigerant-seriesfluids such as R–11, R–12, R–113, R–114, R–123, R–152a, R–500, andR–502; and benzene-series fluids such as C6H6, C7H8, and C8H10. Thetemperature-dependent thermophysical properties of these fluidsare plotted in Fig. 2. The functional dependences of temperature willbe used in analyzing the system efficiency in this study.

3. Mathematical analysis

The following mathematical model is used to analyze thermo-dynamic behavior of ORC systems. The slopes of saturated vaporcurves in the T-s diagrams are used to identify the types of theworking fluids (i.e., wet fluids and dry fluids) as shown in Fig. 3.Pressure drops occurred in various components and pipes are notconsidered in this model.

Fig. 4 shows the conditions of working fluids at various locationsand paths of power generation in an ORC. The mathematical modelis analyzed as follows:

Pump:

w12 ¼ ðp2 � p1Þv1 ðv1wv2 since State 2 is liquidÞ (1)

h2 ¼ h1 þw12;h3 ¼ f ðT3; x3Þ (2)

Heat exchanger:

q23 ¼ ðh3 � h2Þ (3)

h4 ¼ f ðp4; s4Þ (4)

Turbine expander:

w34 ¼ ðh3 � h4Þ (5)

ht ¼ 1 Overall efficiency : hth ¼w34 �w12

q23(6)

Turbine efficiency:

x4awx4aw1 h4a ¼ f ðp3; x4aÞ (7)

ht ¼ ðh3 � h4Þ=ðh3 � h4aÞ (8)

w34a ¼ ðh3 � h4aÞ (9)

Overall efficiency

h ¼ w34a �w12

q23(10)

Practically, due to irreversibility in an actual thermodynamicsystem, it is impossible to convert all the available thermal energyinto useful work. Furthermore, irreversibility provides an addi-tional means of estimating the system efficiency of a thermody-namic cycle. From the second law of thermodynamics, the equationof irreversibility rate can be expressed for uniform flow as follows:

_I ¼ Todstot

dt¼ _mTo

24X

exit

s�Xinlet

sþ dssys

dtþX

j

qj

Tj

35 (11)

Assuming that the system reaches a steady state, and there is onlyone exit and one inlet for any component, Equation (11) becomes

_I ¼ _mTo

hðsexit � sinletÞ þ

qT

i(12)

where To is the ambient temperature. Since the major contributionsof the irreversibilities are from the processes 1–2 and 3–4, the totalirreversibility rate becomes

_Itot ¼X

j

_Ijy _mTo

�� h3 � h4

THþ h4a � h1

TL

�(13)

From Eq. (13), one can see that the heat transfer rates in theevaporator and condenser associated with the ambient tempera-ture are the key factors affecting the overall irreversibility; andaccordingly, the system efficiency. For the sake of a better under-standing of the effects of pressure on irreversibility, the availabilityratio, f is defined as follows

f ¼_mq23 � _Itot

_mq23(14)

The factor is the ratio of the available energy to the total energyobtained from the heat source.

Fig. 2. a. Temperature-dependent thermophysical properties of wet organic fluids used in ORC Systems.

T.C. Hung et al. / Energy 35 (2010) 1403–14111406

4. Results and discussion

A computer program employing MATHCAD was developed tosimulate the thermodynamic performances of the working fluidsunder various working conditions. As shown in Fig. 4, the turbineinlet condition is assumed to fall on State 3 which will be in satu-rated or superheated region. Material requirements of the evapo-rator are more stringent as the working fluid becomes superheated,and a lower thermal conductivity of the superheated vapor wouldresult in a lower heat transfer rate as compared with the saturatedvapor. Also, a heat source with a higher temperature for theevaporator is required if the fluid entering the turbine is super-heated. Therefore, the working fluid is assumed to be saturatedwith its corresponding saturation pressure at the inlet of theturbine. The following analyses will be based on two types ofenergy resources: OTEC and solar energy.

4.1. Case 1: using OTEC as energy source

An ORC system using OTEC operates between a high tempera-ture (supplied by the warm seawater) in the evaporator and a lowtemperature (supplied by the cold seawater in deep sea) in thecondenser. Based on the slopes of saturated vapor curves and theinlet temperatures of the turbine, system efficiencies using various

working fluids can be estimated. The inlet temperature of the pumpis fixed at 5 �C, i.e., heat transfer in the condenser is assumed to bevery efficient, and the inlet temperature of the turbine is variedfrom 20 to 40 �C to simulate the heat source in the calculations. Dueto a more amount of energy is received in the evaporator, systemefficiency increases nearly linearly as the turbine inlet temperatureincreases for every working fluid under investigation as shown inFigs. 5 and 6. For dry fluids as shown in Fig. 5, refrigerant- andbenzene-series have almost the same efficiencies when the turbineinlet temperature is low. As the turbine inlet temperature increases,benzene-series in general have a better performance in systemefficiency since their saturated vapor curves become almost iden-tical to those of the isentropic wet fluids. This can be seen in Fig. 3;C8H10, for example, has a saturated vapor curve with a positiveslope changing to a negative slope as temperature decreases. Alsoindicated in Fig. 5, efficiency curves for R-113 and R-123 intersect atw30 �C; below 30 �C R–123 has a slightly higher efficiency than R-113 and a reverse performance occurs when temperature is above30 �C. The intersection of efficiency curves of R–123 and R–113 canbe explained by their thermophysical properties. As shown inFig. 2b, R–123 has higher values of thermal conductivity and latentheat than those of R–113. This means that R–123 has a better heattransfer performance than that of R-113. On the other hand, R–123has a higher specific heat than that of R–113. Since a regenerator is

Fig. 2. b. Temperature-dependent thermophysical properties of dry organic fluids used in ORC systems.

T.C. Hung et al. / Energy 35 (2010) 1403–1411 1407

not considered in this study, the inlet condition of the working fluidwould become superheated before entering the condenser. Thisimposes a higher load on the condenser and reduces the systemefficiency as a dry fluid goes through an isentropic expansion at the

Fig. 3. T/TC-s/sC Diagram of working fluids.

exit of the turbine. Furthermore, the latent heat ‘‘bandwidth’’ at slow pressure on the T-s diagram also affects the system efficiency. Abroader latent heat bandwidth represents a greater amount of heatmust be taken away from the condenser. This effect is not signifi-cant since the temperature is very low as the condenser is operatedunder a low pressure. However, this effect becomes important as

Fig. 4. T-s Diagram of working fluids in turbine under a fixed TH.

Fig. 5. Case 1: System efficiency-turbine inlet temperature dependence for dry fluids.

T.C. Hung et al. / Energy 35 (2010) 1403–14111408

the inlet temperature of the turbine increases. Therefore, a carefulselection of the working fluid under different working tempera-tures is crucial in optimizing the performances of ORCs.

In general, benzene-series fluids perform better than refrig-erant-series in system efficiency among the dry fluids. Under thesame temperature of the low temperature reservoir (i.e., coldseawater), refrigerant-series fluids usually have a lower boilingpoint, and accordingly, a higher vapor pressure. Thus, the areawhich represents a net work of the cycle in the T-s diagram isinevitably reduced; and so is the system efficiency. Therefore, asindicated in Fig. 5 for the dry fluids with a low operation temper-ature, a suitable working fluid of benzene-series can be selectedbased on two considerations: its saturation curve at a low operationtemperature, and the temperature difference between the evapo-rator and the condenser, i.e., temperature difference in seawater.

As for the wet fluids, both R–11 and R–12 have a rather betterperformance in system efficiency than that of the other wet fluidsexcept for R–152a as shown in Fig. 6. This is because that both R–11and R–12 are nearly identical to isentropic wet fluids, i.e., theirsaturation vapor curves are virtually vertical in this region. Inpractical operation with a careful arrangement, the fluid conditionsat the turbine exit can be adjusted to fall on the saturation vaporcurve. This has two obvious advantages: no occurrence of moisture

Fig. 6. Case 1: System efficiency-turbine inlet temperature dependence for wet fluids.

during isentropic expansion in the turbine, and no need for usinga regenerator to reduce condensation load.

However, isentropic fluids are not always suitable for ORCsystems when other thermophysical properties are taken intoconsideration. As indicated in Fig. 2a, relatively, R-12 has a lowerthermal conductivity and a lower latent heat than those of theother wet fluids; resulting in a less efficient heat transfer as well ashear recovery; and consequentially, a lower overall system effi-ciency. Based on all the thermophysical properties, R–502 exhibitsthe lowest efficiency. When the turbine exit temperature is low, theefficiency curves of R–152a and R–500 are roughly identical. As thetemperature exceeds w25 �C, R–500 shows a clear drop in effi-ciency. This is because that the latent heat of R-500 decreases‘‘faster’’ than that of R–152a as the turbine inlet temperatureincreases, and this has a direct impact on heat recovery of thesystem. As the system efficiency is considered, R–152a has a betterperformance based on a fixed turbine inlet temperature.

Figs. 7 and 8 show the dependence of the system efficiency onthe condenser exit (i.e., pump inlet) temperature as the turbineinlet temperature is fixed. From the T-s diagram, a rise in thecondenser outlet temperature represents a smaller net work of thecycle; thus, resulting in a lower system efficiency as can be clearlyseen in Figs. 7 and 8. All the fluids appeared in these figures showa similar temperature dependence.

As compared with Figs. 5 and 6 for the efficiency at variousturbine temperatures, virtually all the fluids exhibit a similar effi-ciency for a fixed condenser exit temperature. In other words, if theturbine inlet temperature is fixed, there is a broader range ofselection of the working fluids since the efficiency may vary fromfluid to fluid. On the other hand, only factors which are required tobe considered are stability and safety of the fluids and their impactson environment if the condenser exit temperature varies. Hence,the condenser exit temperature is an important factor which mustbe taken into consideration in designing an ORC system.

Based on the analysis stated above, three factors are believed tohave major impacts on the system efficiency according to thethermophysical properties of the fluids under investigation: theslope of saturation curve, specific heat, and latent heat. Undera broader range of working temperature between TH and TL, latentheat can vary significantly. It has a direct influence on the energy

Fig. 7. Case 1: System efficiency-condenser exit temperature dependence (at a Fixedturbine inlet temperature of 30 �C).

Fig. 8. Case 1: System efficiency-condenser exit temperature dependence (at a Fixedturbine inlet temperature of 40 �C).

Fig. 10. Turbine efficiencies for wet fluids.

T.C. Hung et al. / Energy 35 (2010) 1403–1411 1409

absorbed in the evaporator, and the ‘‘bandwidth’’ of the saturationcurve in the T-s diagram at the low temperature end; and conse-quentially, the net work done by the cycle. Fig. 9 shows thetemperature dependence of latent heat for various fluids. It can beseen that the working fluids with a high latent heat usually performbetter in efficiency, and this trend gradually disappears as thetemperature increases. Despite the fact that R-152a has the highestefficiency among all the fluids, however, its latent heat drops fasterthan other fluids as the working temperature rises, resulting ina lower system efficiency. As a conclusion, latent heat is a majorfactor to be considered in selecting a proper working fluid inaddition to examining other thermophysical properties.

As stated above, isentropic fluids in general have a betterperformance in efficiency. Wet fluids also perform well in systemefficiency. However, a major problem with the wet fluids is thepossible damage to the turbine blades due to corrosion from themoisture of the two-phase state of the working fluids as they do notfollow a saturation curve during expansion in the turbine. There-fore, emphasis is also focused on the influence of the ‘‘quality’’ ofthe wet fluids on turbine efficiency as shown in Fig. 10. Here, theworking conditions are: 5 �C for the low temperature end, 40 �C for

Fig. 9. Temperature dependence of latent heat.

the high temperature end, and a saturation state for the turbineinlet. Fig. 11 shows the dependence of the system efficiency andturbine efficiency on the quality of the fluids at turbine exit. As theexit quality reaches 1, i.e., a high-efficiency turbine is not necessary,R-152a has the lowest system efficiency of 6.7% while R-11 and C7H8

have the highest system efficiency around 8.5%. An isentropic fluidsuch as R-11 expands in the turbine isentropically to a low pressurewith a quality very close to 1. Therefore, design constraints are lessstringent for such fluids. When the quality at the turbine exit drops,turbine efficiency must be reduced accordingly in order to have anexit quality close to 1. R-152a on the other hand has a rather slowly-varying positive slope of the saturation vapor curve; this results ina very low quality at turbine exit. The situation can be improved byoperating the fluid in superheated region before it entering theturbine (see curve 3w3a of Fig. 4). In practice, it can be designedthat the fluid at turbine exit is saturated vapor in order for R–152a

to operate under a low temperature range.Since the main focus of this study is on the thermodynamic

behavior of the ORCs, detailed calculations of pressure losses andheat transfer in evaporator and condenser are ignored since theydepend strongly on materials and configurations of the systemcomponents. Instead, irreveribilities of the working fluids occurred

Fig. 11. Case 1: System efficiency and turbine efficiency on quality.

Fig. 12. Distributions of irreversibilities in the major components of a rankine cycle forfluid C7H8 at TH ¼ T þ 15 �C.

Fig. 14. System Efficiency versus turbine inlet pressure for fluids at x3 ¼1.0 andht ¼ 0.8.

T.C. Hung et al. / Energy 35 (2010) 1403–14111410

in the major components of an ORC are calculated to evaluate theeffects of those losses. Also, the effects of the working pressures ofthe evaporator are also evaluated. As stated above, heat transferrates in the evaporator and condenser are the key contributions tothe system irreversibility. The following calculations are based ona 10 MW heat source. The irreversibility in a system differs fromcomponent to component. Fig. 12 shows the irreversibilitiescontributed from the major components of an ORC under variousturbine inlet pressures for the fluid C7H8. It shows that thecondenser has the major contribution of irreversibility; and thus,improvement of the condenser performance is an importantconcern when the upper end of the cycle is operated under a highpressure. Fig. 13 shows the irreversibilities for various workingfluids under a wide range of turbine inlet pressures at a constantturbine inlet temperature. It is clear that the enthalpies of the fluidsat point 3, i.e., the turbine inlet, increase as the pressure increases,and so do their respective irreversibilities. This effect is moreprofound for benzene-series than for refrigerant-series. However, itis a totally different story when it comes to system efficiency. Forinstance, when C7H8 is used as the working fluid and the pressurevaries from 500 to 1000 kPa, there is a 4% increase in Dhcondenser

compared with 7.4% increase inDhevaporator. The correspondingchanges in efficiency for C7H8 are 16.4% and 19.4% as shown in

Fig. 13. Irreversibility versus turbine inlet pressure at TH ¼ T3þ15 �C.

Fig. 14, while refrigerant fluids in general have a lower efficiencydue to their low boiling point. In other words, for a given temper-ature, refrigerants have a relatively higher vaporization pressure;and hence, a smaller area representing cycle work on the T-sdiagram.

For the sake of a better discussion, the availability ratio definedin Eq. (14) is also used as a guideline in selecting a suitable workingfluid. Fig. 15 shows the variations of the system efficiency andavailability ratio versus turbine inlet pressure for the fluid C8H10. Asexpected, the availability ratio decreases as the turbine inlet pres-sure increases. Therefore, an optimal choice of the working fluidmay be achieved by taking the intersection of the efficiency curveand the availability ratio curve. From Fig. 15, it seems that a systempressure below 500 kPa is adequate for recovering low-gradeenergy if C8H10 is used as the working fluid.

4.2. Case 2: solar energy as high-temperature reservoir

In this case the energy source of an ORC system is supplied bysolar energy in the evaporator. The high temperature source may befrom a solar pond or a solar collector. As a scoping calculation, theoperation conditions of the ORC are: 20 �C for pump inlettemperature and 40–60 �C for turbine inlet temperature. The actualturbine inlet temperature can be much higher depending on thedesign of the solar energy collection systems. The study shows thatbenzene-series fluids have a better performance in efficiency thanthat of the refrigerant-series fluids. Among the wet fluids, R-11 hasthe best performance. This result is similar to Case 1. Fig. 16 shows

Fig. 15. System efficiency and availability ratio versus turbine inlet temperature forfluid C8H10.

Fig. 16. Case 2: System efficiency and turbine efficiency on quality.

T.C. Hung et al. / Energy 35 (2010) 1403–1411 1411

the dependences of the system efficiency and turbine efficiency onturbine exit quality. Compared with Case 1, the difference is theturbine exit quality; resulting from different slopes of the satura-tion vapor curves at a new temperature range (between TH and TL).A suitable working fluid can thus be selected based on a consider-ation of the system efficiency in order to meet the required effi-ciency of the turbine.

Based on the two cases studied here, benzene-series fluids havea better performance among the dry fluids. In spite of the proper-ties of dry fluids, benzene-series fluids behave like isentropic fluidsin saturation vapor region, and their qualities at turbine exit arevery close to 1 after isentropic expansion at the turbine exit. Theefficiencies of the systems using benzene-series fluids range from 5to 10% for Case 1; and 6.2–11% for Case 2. When using wet fluids asthe working fluids, the slope of the saturation vapor curve for eachfluid is the major factor to be considered. The effect of changing theturbine inlet temperature on the change of latent heat; andaccordingly, the system efficiency also needs to be taken intoconsideration. A composite system which combines OTEC and solarenergy to serve as an ORC system with a higher temperaturedifference between TH and TL, i.e., an evaporator with a higheroperation temperature supplied by solar energy and a condensercooled by cold water in deep sea, is believed to yield an even betterperformance. In this case, the system efficiency can be as high as11–15% if the operation temperature of the evaporator is between40 and 60 �C. As shown in Fig. 16, isentropic (or nearly isentropic)fluids have higher efficiencies. By comparing R-11 and R-113, R-11has a better efficiency performance than that of R-113. This can beexplained from the T-s diagram in which R-113, a dry fluid, becomessuperheated at the turbine exit, and this reduces the area of network in the T-s diagram. Furthermore, R-11 has a higher thermalconductivity and latent heat than those of R-113. Therefore, anoptimal operation of high efficiency can be achieved with a properselection of the working fluid in the combined system.

5. Conclusion

ORC systems using refrigerant- and benzene-series fluids asworking fluids to convert energy from renewable energy sources,such as solar energy and ocean thermal energy, are investigated inthis study. Eleven fluids include R–11, R–12, R–113, R–114, R–123,R–152a, R–500, R–502, C7H8, C7H8, and C8H10 are analyzed based ontheir thermophysical properties. It can be seen from the casesunder investigation, suitable working conditions of various fluidscan be identified based on their saturation vapor curves and theirresponses to the temperature of energy source. System efficiencycan be optimized by selecting a proper working fluid operated atsuitable working conditions. In terms of disadvantages, dry fluids ingeneral generate superheated vapor at the turbine exit, and thisreduces the area of net work in the T-s diagram. A generator may beneeded in order to relieve the cooling load of the condenser. Themajor disadvantage for the wet fluids is their moisture contentduring expansion process in the turbine. Isentropic (or nearlyisentropic) fluids in general is free from the concern of the moisturecontent, and does not need a regenerator to relieve the cooling loadof the condenser. The only issues of concern for the isentropic fluidsare their cost, chemical stability, and safety. Therefore, they areconsidered to be the best candidates of the working fluids for ORCs.A combination of OTEC and solar energy to yield a highertemperature difference in an ORC system is believed to give an evenbetter performance in efficiency.

References

[1] Lennard DE. The viability and best locations for ocean thermal energyconversion systems around the world. Renewable Energy 1995;6(3):359–65.

[2] Tanner D. Ocean thermal energy conversion: current overview and futureoutlook. Renewable Energy 1995;6(3):367–73.

[3] Tseng CH, Kao KY, Yang JC. Optimal design of a pilot OTEC power plant inTaiwan. Journal of Energy Resources Technology, Transactions of ASME1991;113:294–9.

[4] Takazawa H, Amano M, Tanaka T. Performance characteristics of barome-tric-type open-cycle OTEC system. Heat Transfer-Japanese Research 1996;25(4):226–37.

[5] Odum HT. Emergy evaluation of an OTEC electrical power system. Energy2000;25:389–93.

[6] Yeh RH, Su TZ, Yang MS. Maximum output of an OTEC power plant. OceanEngineering 2005;32:685–700.

[7] Wu C, Burke TJ. Intelligent computer aided optimization on specific power ofan OTEC Rankine power plant. Applied Thermal Engineering 1998;18(3):295–300.

[8] Straatman PJT, Wilfried GJHM, Sark V. A new hybrid ocean thermalconversion – offshore solar pond (OTEC-OSP) design: a cost optimizationapproach. Solar Energy 2008;82:520–7.

[9] Yamada N, Hoshi A, Ikegami Y. Performance simulation of solar-boosted oceanthermal energy conversion plant. Renewable Energy 2009;34:1752–8.

[10] Hung TC, Shai TY, Wang SK. A review of organic Rankine cycles (ORCs) for therecovery of low-grade waste heat. Energy 1997;22(7):661–7.

[11] Liu BT, Chien KH, Wang CC. Effect of working fluids on organic Rankine cyclefor waste heat recovery. Energy 2004;29:1207–17.

[12] Madhawa Hettiarachchi HD, Golubovic M, Worek WM, Ikegami Y. Optimumdesign criteria for an Organic Rankine cycle using low-temperaturegeothermal heat sources. Energy 2007;32:1698–706.

[13] Saleh B, Koglbauer G, Wendland M, Fischer J. Working fluids for low-temperature organic Rankine cycles. Energy 2007;32:1210–21.

[14] Manolakos D, Kosmadakis G, Kyritsis S, Papadakis G. Identification of behaviorand evaluation of performance of small scale, low-temperature organicRankine cycle system coupled with a RO desalination unit. Energy 2009;34:767–74.

[15] Desai NB, Bandyopadhyay S. Process integration of organic Rankine cycle.Energy 2009;34:1674–86.