Thermodynamic analysis of an organic Rankine cycle for waste heat recovery from gas turbines

Energy 65 (2014) 91e100

Contents lists available at ScienceDirect

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

Thermodynamic analysis of an organic Rankine cycle for waste heat recovery from gas turbines Carlo Carcasci*, Riccardo Ferraro, Edoardo Miliotti DIEF: Department of Industrial Engineering, University of Florence, Via Santa Marta, 3, 50139 Florence, Italy

a r t i c l e i n f o

a b s t r a c t

Article history: Received 11 April 2013 Received in revised form 18 October 2013 Accepted 29 November 2013 Available online 6 January 2014

The use of an organic Rankine cycle is a promising choice for the recovery of waste heat at low/medium temperatures. In fact, the low temperature heat discharged in several industrial applications cannot be recovered with a traditional bottomer steam cycle but, using an organic Rankine cycle, this waste heat can be converted into electrical energy. The choice of the fluid is fundamental for a good cycle performance because the optimal thermophysical properties depend on the source temperature. This study illustrates the results of the simulations of an organic Rankine cycle combined with a gas turbine in order to convert the gas turbine waste heat into electrical power. A diathermic oil circuits interposed between these two plants for safety reasons. This paper presents a comparison between four different working fluids in order to identify the best choice. The selected fluids are: toluene, benzene, cyclopentane and cyclohexane. The design is performed by means of a sensitivity analysis of the main process parameters and the organic Rankine cycle is optimized by varying the main pressure of the fluid at different temperatures of the oil circuit; moreover, the possible use of a superheater is investigated for each fluid in order to increase electrical power. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Organic Rankine cycle Gas turbine Combined cycle Waste heat recovery

1. Introduction In the last decades, the global demand for energy has grown steadily and consequently the consumption of fossil fuels has increased. This, in turn, has led to several environmental problems such as air pollution, global warming and the depletion of the ozone layer. Furthermore, according to several studies [1,2], industrial applications waste more than 50% of the total heat generated at a low/medium temperature. In recent years the number of installations of small/medium-sized power plants has increased due to the outsourcing of generation systems. Aeroderivative gas turbines present a high thermodynamic efficiency and a low/medium temperature waste heat and the conversion of low grade waste heat into electrical energy results in a reduction of fossil fuels consumption. The traditional steam bottoming cycles do not perform satisfactorily when waste heat at low/medium temperature is used due to its low thermal efficiency and large volume flows. Other power plant configurations were studied, i.e. Carcasci et al. [3,4] have studied the CRGT (Chemical Recuperated Gas Turbine) cycle. The ORC (organic Rankine cycle) power plant has

* Corresponding author. Tel.: þ39 055 4796245. E-mail addresses: carlo.carcasci@unifi.it, (C. Carcasci).

[email protected]fi.it

0360-5442/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.energy.2013.11.080

proved to be an attractive solution, indeed it is one of the most promising technologies for converting low/medium grade heat into electrical power. Thermodynamic analysis and working fluid selection have become the main topics in recent years. Waste heat recovery ORCs have been studied in a number of previous works: Badr et al. [5], Gu et al. [6], Dai et al. [7] used simple thermodynamic models comparing different working fluids. These studies illustrate the reliance of efficiency on the evaporating pressure and some present a parametric optimization and performance analysis of waste heat recovery from low grade sources [7e10]. Advanced cycle configurations have been studied as well: Gnutek et al. [11] proposed an ORC cycle with multiple pressure levels and sliding vane expansion machines in order to maximize the efficiency of the heat source. Chen et al. [12] studied a transcritical CO2 power cycle. The different use of several ORC bottoming cycles has been analyzed: solar applications [13e15], geothermal heat sources [16,17], high temperature fuel cells [18] and heat recovery from gas turbines. Chacartegui et al. [19,20] showed a parametric optimization of a combined cycle with a gas turbine topping cycle and an ORC low temperature bottoming cycle in order to achieve better integration between these two technologies and they presented a part-load analysis. Organic fluids are classified into three different categories depending on the slope of their saturation vapor curves in the T-s diagram: fluids with a negative slope are called “wet fluids”, fluids

92

C. Carcasci et al. / Energy 65 (2014) 91e100

Nomenclature c L m Mm P Q s T W Wsp

specific heat [kJ/kg K] specific work [kJ/kg] mass flow rate [kg/s] molar mass pressure [bar] heat [kW] entropy [kJ/kg K] temperature [ C] power [kW] specific Power [kJ/kg]

Greek symbols efficiency density [kg/m3]

h r

Subscripts air air in ambient condition amb ambient app approach con condenser cr fluid critical condition ec economizer el electric ex expander exh exhaust from gas turbine fl organic fluid

with a positive slope are called “dry fluids” and fluids whose slope of the saturated vapor curve tends to be infinitive are called “isentropic fluids”. Lai et al. [21] and Sahleh et al. [22] presented a detailed review of working fluids for a low and high temperature organic Rankine cycle and cyclopentane seems to be the best answer for a high temperature organic Rankine cycle. Vankeirsbilck et al. [23] showed a high efficiency regenerative cycle based on toluene. According to Chacartegui et al. [19], toluene and cyclohexane present high global efficiency in a gas turbine combined cycle and a low purchase cost of the fluid. Chacartegui et al. [19] analyzed toluene and cyclohexane and concluded these working fluids are a good solution to replace steam in small sized combined cycles. They highlight how an increase in combined cycle efficiency with respect to the steam cycle would compensate for the fluid acquisition. Other works presented some applications based on the use of toluene [24e26]. Victor et al. [27] studied organic Rankine cycles and Kalina cycles for the heat sources at temperatures between 100  C and 250  C. In particular, Carcasci et Ferraro [28] showed that a gas turbine cycle combined with an organic Rankine cycle based on toluene offers the best conditions in a regenerative non-superheated layout and also that a higher value of the maximum oil temperature allows to recover more heat and to produce more power, although this may lead to the formation of acid condensate due to the very low stack temperature. Bianchi et al. [28] described benzene as one of the working fluids with the highest bottoming cycle power and total heat recovery efficiency. Tchanche et al. [29] studied lowtemperature solar organic Rankine cycle systems. He et al. [30] studied a subcritical organic Rankine cycle using 22 working fluids for a 150  C temperature heat source. In the present paper, four different working fluids have been used to simulate an organic Rankine cycle: toluene, benzene, cyclopentane and cyclohexane. A thermodynamic cycle with a

gb GT in is lim max oil opt pp pump rec sat s,max st sub

gearbox gas turbine inlet isentropic limit maximum oil optimized pinch point pump reuperator saturation saturation curve with maximum entropy stack subcooling

Acronyms CON condenser CRGT chemical recuperated gas turbine ECO economizer EV evaporator EX expander GT gas turbine HRB hot gaseoil heat recovery boiler HRSG oil-fluid heat recovery steam generator ORC organic Rankine cycle REC recuperator/regenerator SH superheater

regenerator has been selected and the use of a superheater has been evaluated for each working fluid. Generally, working fluids used for an organic Rankine cycle are highly flammable, therefore a diathermic oil circuit between the heat source and the ORC is needed to prevent explosions. Common commercial diathermic oils can reach up to 400  C [32]; thus, in order to determine the influence of the heat transfer recovery on the cycle power output, the cycle is simulated at three different maximum temperatures of the diathermic oil: Toil,max ¼ TB ¼ 360  C, 380  C and 400  C. Finally, a cycle optimization is presented by varying the expander inlet pressure. The gas turbine considered is the GE10-1 from General Electric e Nuovo Pignone which is a heavy duty single shaft turbine used for oil and gas or power generation applications [33]. The main GE10-1 specifications are shown in Table 1. 2. Working fluids Four different dry working fluids have been tested: toluene, benzene, cyclopentane and cyclohexane. NIST (National Institute of Standards and Technology) software has been used to simulate the behavior of these working fluids. The thermodynamic properties of the selected fluids are obtained at the critical point values in terms of both pressure and temperature, which are shown in Table 2. The last column illustrates the pressure where the vapor saturation entropy has reached the maximum value. In a plant layout without Table 1 GE10-1 main specification [33]. WGT

hGT

mGT

TGT

11,250 kW

31.4%

47.5 kg/s

482  C

C. Carcasci et al. / Energy 65 (2014) 91e100

93

Table 2 Thermodynamic properties of selected working fluids. Working fluid

Tcr [ C]

Pcr [bar]

Mm [g mol1]

Ps,max [bar]

Toluene Benzene Cyclopentane Cyclohexane

318.6 288.8 238.6 281.0

41.1 48.9 44.3 40.7

92.14 78.10 70.13 84.16

36.0 37.5 34.7 35.9

superheater, if the working pressure exceeds this value, the expansion line can present wet vapor, causing problems to the expander. Toluene has the highest critical temperature and molecular weight, while cyclopentane has the lowest. The maximum critical pressure is reached by benzene and the minimum by cyclohexane. Cyclohexane has the highest value with regard to the slope of the saturation vapor curve. Toluene and cyclopentane show approximately the same gradients, while benzene has the lowest values. Although cyclohexane is more curved than toluene, the latter presents the highest entropy variation, considering a range between the entropy value which corresponds to the saturation temperature (Tsat ¼ 50  C) and the maximum entropy value. The slope of the saturation vapor curve is an important parameter because it influences the behavior of the recuperator. Another important parameter which influences the behavior of the condenser is the saturation pressure; Fig. 1 shows its dependence on the saturation temperature. At a fixed temperature, cyclopentane shows the highest saturation pressure, while toluene shows the lowest. Benzene and cyclohexane present intermediate values, but cyclohexane has the lowest limit pressure, while benzene presents the highest. These considerations are significant not only because the maximum oil temperature can be used as a reference for the maximum organic fluid temperature but also because its pressure level can be determined (see Fig. 1). Moreover, for the same reasons, the condenser pressure changes for each fluid.

3. Power plant layout The power plant considered is a combined gas turbine topping cycle and a subcritical organic Rankine bottoming cycle. Fig. 2 shows the power plant layout of the bottoming cycle.

320 280 240

Tsat [°C]

200 160

Toluene Benzene Cyclopentane Cyclohexane

120 80 40 0 0

4

8

12

16

20

24

28

32

36

40

P [bar] Fig. 1. Saturation temperature of organic fluid versus pressure.

44

48

Fig. 2. ORC power plant layout.

The heat transfer of the hot gas from the gas turbine to the organic fluid occurs through an intermediary diathermic oil circuit, interposed for safety reasons. The plant layout presents one pressure level boiler with a SH (superheater) and an internal heat exchanger (recuperator REC) in order to increase the system efficiency [21]. The hot exhaust gas heats the diathermic oil in the first heat recovery unit (HRB). In the second loop, the hot oil passes through the second heat recovery unit, composed by a SH (superheater), an EV (evaporator) and an ECO (economizer), where the organic fluid is heated and enters in an EX (expander). The exhaust fluid exchanges heat in the REC (recuperator), thus it heats the condensed fluid. Finally, the organic fluid is cooled in an air-cooled CON (condenser) and pressurized in a pump. This particular type of condenser has been chosen considering that the plant location is a waterless area. 4. Thermodynamic and design approach In the HRB (hot gaseoil heat recovery boiler) recovery unit the temperature and the flow rate of hot gas (mGT, TGT. See Table 1) from the gas turbine are fixed as well as the pinch point and the maximum temperature oil (TB ¼ Toil,max). Thanks to the energy balance in the HRB, the return oil temperature is initialized, thus the oil mass flow rate (moil) and the stack hot gas temperature (Tst) are determined using the pinch point temperature difference (DTpp,HRB: Tst ¼ Toil,out þ DTpp,HRB ¼ TA þ DTpp,HRB / moil/ mGT,exh ¼ cg$(TGT,exh  Tst)/[coil$(TB  TA)]). Fixing the inlet expander pressure (P1) and the approach temperature difference (DTapp ¼ TB  T1 e if SH is present), both the working fluid maximum temperature (Tex,in ¼ T1: without SH / T1 ¼ Tsat(P1); with SH / T1 ¼ Toil,max  DTapp) and its mass flow rate (mfl) can be calculated once the pinch point temperature difference (DTpp ¼ TD(T7 þ DTsub)) by exploiting the energy balance in the SH and EV. When considering the air condenser, the discharge pressure at the expander exit (P2) can be determined: the saturated temperature of the organic fluid (Tsat) and consequently its pressure (Pcon ¼ P3) are evaluated by fixing the pinch point temperature difference (DTpp,con ¼ Tsat  Tair,out) and the ambient air temperature (Tair,in). By fixing the difference of the cooling air temperatures in the condenser (DTair,con ¼ Tair,out  Tair,in), the air mass flow rate is determined and thus the fan absorbed power (Wfan) can be calculated. Considering the expander inlet conditions, the discharge pressure (Pcon ¼ P3) and fixing the expander isentropic efficiency (hex ¼ Lex/Lex,is), both the output power (Wex ¼ mfl$Lex) and the working fluid exhaust conditions can be evaluated.

C. Carcasci et al. / Energy 65 (2014) 91e100


DPcon; air Wel; tot ¼ hgb $hel $mfl $ Lex  Lpump  mcon; air $

rair

5.1. Superheater evaluation The use of an ORC plant scheme with or without a superheater is a significant choice and depends on the selected working fluid and on the thermal source temperature. According to Bianchi et al. [31], the selection of adequate organic fluids represents a key factor to maximize the ORC thermodynamic performance, which can be significant, as shown in this study, even if a small sized turbine is used. Thus, in some papers [34e36] the ORC is shown with a superheater, while in others it is presented as a non-superheated cycle [19,20]. Carcasci et Ferraro [28] show an organic Rankine cycle without a superheater as the best configuration for a toluene based cycle. For this reason, a sensibility analysis has been conducted evaluating the possibility of whether or not to use a superheater for each fluid, by fixing the maximum oil temperature (Toil,max ¼ TB ¼ 380  C) and the approach temperature difference in the superheater (DTapp ¼ 40  C) and consequently by varying the fluid pressure. Table 3 shows the values imposed for the thermodynamic analysis. Fig. 3 shows the net electrical power with and without a superheater, varying the maximum pressure of the different organic fluids. By analyzing the layout with the superheater and fixing DTapp, an increase in the fluid pressure leads to a rise in output power. Sun et al. [38] present similar results: higher expander inlet pressure generates, on average, more net power, but the expander inlet pressure depends on the working fluid property. To maximize the net power generation, both the optimal relative working fluid mass flow rate and the optimal condenser fan air mass flow rate increase with the heat source temperature increasing. Only toluene Table 3 Power plant parameters used for the thermodynamic analysis. Parameter

Value

Parameter

Value

Toil,max ¼ TB Pex,in ¼ P1

380.0  C 36.0 bar 0.98 10.0  C 30.0  C 8.0  C 0.85 0.98

hel

0.96 15.0  C 20.0  C 20.0  C 2.0% 15.0  C 15.0  C 0.70

hpump

3000

2800

2600 ΔT

=40°C - No SH Toluene Benzene Cyclopentane Cyclohexane

2400

2200 15

5. Simulation results

hHRB DTpp,HRB DTapp DTpp hex hgb

3200

(1)

The authors developed an in-house code able to perform thermo-dynamical simulations of the proposed power plant. The code is developed in ANSI Standard of the FORTRAN 90 programming language and the elementary energy balances are validated with a commercial code.

Tair,in DTair,con DTpp,con DPcon,air DTpp,rec DTsub

3400

Wel tot [kW]

The energy balance equations are used in the recuperator, and the pinch point temperature difference (DTpp,rec) is fixed in order to evaluate the inlet conditions of the condenser and of the organic fluid at the HRSG. The no-boiling phenomena must be verified inside the recuperator. If this event occurs, the pinch point temperature difference of the recuperator (DTpp,rec) must be increased. By exploiting the energy balance in the economizer of the HRSG, the oil TE (exhaust temperature) is determined (a DTsub is considered in order to prevent premature boiling in the economizer). This data is necessary in order to calculate the energy balance in the oil HRB (see above). Finally, using the fluid mass flow rate and the specific work of the expander, the output power can be determined:

20

25

30

35

40

Pex in [bar] Fig. 3. Net electrical power with and without superheater versus inlet expander pressure.

presents a maximum point (about Pex,in,max ¼ 23.0 bar). Victor et al. [27] and Tchanche et al. [29] obtained similar results, but the temperature of their analysis was lower (maximum 250  C). The non-superheater configuration is the most appropriate, (except with cyclopentane where the use of the superheater is convenient in terms of power); this can be explained considering the saturation curve of toluene: the low critical temperature and the low value of the slope gradient allow to improve the performance without a superheater. However, the range of operability is very limited and the produced power is the lowest among the selected fluids (15% lower than cyclohexane). Only in the case of toluene, the maximum net power does not correspond to the maximum operability pressure (Pex,in ¼ 26.13 bar); however, using the maximum inlet expander pressure (Ps,max ¼ 36.0 bar), the net power decreases by only about 1% with respect to the optimum value. The expander power is due to the contribution of the specific work of the expander and the fluid mass flow rate. The specific work of the expander increases with the working fluid pressure (Fig. 4), in fact the expander pressure ratio grows. In the non-superheater configuration, the maximum values are reached by toluene, while benzene and cyclohexane present rather equal values and cyclopentane the lowest. These values can be

200

180

Lex [kJ/kg]

94

160

140

120

ΔT

=40°C - No SH Toluene Benzene CycloPentane CycloHexane

100 15

20

25

30

35

40

Pex in [bar] Fig. 4. Specific power produced by expander versus inlet expander pressure.

C. Carcasci et al. / Energy 65 (2014) 91e100

95

8000

40

7000 6000

QREC [kW]

mfl [kg/s]

36

32

ΔT

28

=40°C - No SH Toluene Benzene CycloPentane CycloHexane

24

5000 4000 3000 2000

ΔT

=40°C - No SH Toluene Benzene CycloPentane CycloHexane

1000

20 0 15

15

20

25

30

35

40

20

25

30

35

40

Pex,in [bar]

Pex in [bar] Fig. 5. Organic fluid mass flow rate versus expander inlet pressure with and without superheater.

understood when considering the saturation pressure (Fig. 1), in fact with the same condenser temperature the minimum condenser pressure is obtained using toluene. When the superheater is present, the same trends are present; obviously the specific work increases with respect to the configuration without a superheater, because the expander uses a superheated fluid and so the inlet expander enthalpy grows. Moreover, in the superheater layout the effects of the condenser pressure are influenced, so toluene shows to be the fluid leading to the maximum expander specific work. The organic fluid mass flow rate (Fig. 5) tends to decrease with pressure and, contrary to the specific work of the expander, greater values are shown in the layout without a superheater. In fact, due to the superheater the fluid enthalpy increases, but the mass flow rate decreases because the heat exchanged across the diathermic oil circuit is concentrated into an evaporator whose pinch point is fixed. Cyclopentane presents the highest values and toluene the lowest, while benzene and cyclohexane show about the same value which lies between the other two organic fluids. The heat exchanged in the superheater (Fig. 6) decreases by raising the pressure, in fact the dry vapor enthalpy on the

Fig. 7. Recuperator heat exchanged versus inlet expander pressure with and without superheater.

saturation vapor curve increases, too. The heat exchanged in the superheater is the highest when cyclopentane is used, because, when the saturation temperature is fixed, both the pressure (Fig. 1) and the critical point (Table 2) are low. Fig. 7 shows that the heat exchanged in the recuperator is greater with a superheater; in fact, the exhaust organic fluid temperature from the expander is higher in this case. The approach difference temperature (DTapp, so the inlet expander temperature) and the conditions at the condenser are both fixed. A rise in pressure leads to a decrease in the recuperator inlet temperature and therefore the heat recovery is lower at least until the maximum entropy value is reached. In a nonsuperheated layout, a rise in pressure leads to an increase in recuperator inlet temperature, so the barely increasing heat exchanged is explained. In Fig. 8, the Tes diagram of benzene thermodynamic cycle is shown as an example to observe the variations of the REC (recuperator) inlet temperature. Cyclopentane and cyclohexane present a greater energy recovery capacity when the superheater is considered. In a non-superheater layout, the recovered heat is lower for cyclopentane and benzene and higher for toluene and cyclohexane. In fact, the saturation curve gradient

8000 7000 6000

QSH [kW]

5000 4000 3000 ΔT

2000

=40°C Toluene Benzene CycloPentane CycloHexane

1000 0 15

20

25

30

35

40

45

Pex in [bar] Fig. 6. Superheater heat exchanged between oil and organic flow versus inlet expander pressure.

Fig. 8. Benzene thermodynamic cycle with and without superheater and for two different pressures.

96

C. Carcasci et al. / Energy 65 (2014) 91e100 10000

260 240

8000

220

ΔT

6000

=40°C - No SH Toluene Benzene CycloPentane CycloHexane

4000

Tst [°C]

QECO [kW]

ΔT

=40°C - No SH Toluene Benzene CycloPentane CycloHexane

200 180 160 140

2000

120 100

0 15

20

25

30

35

40

15

20

25

30

Pex,in [bar]

35

40

Pex in [bar]

Fig. 9. Economizer heat exchanged versus inlet expander pressure with and without superheater.

Fig. 11. Stack temperature versus inlet expander pressure with and without superheater.

of cyclohexane is higher and so the expansion line from the dry saturation curve ends in the superheated zone. Fig. 9 shows the heat exchanged in the economizer, which tends to increase with pressure because due to the higher saturation conditions. The heat exchanged is higher for toluene; it is obvious considering its saturation conditions (Fig. 1). With the superheater, the heat exchanged decreases because the recovered heat inside the recuperator increases (Fig. 8). In the case of low inlet pressure, the exhaust organic fluid from the expander presents a high enthalpy level, so the heat recovered in the recuperator can reach a condition of saturation in the cold side and the heat exchanged in the economizer tends to become zero; in this case, the recuperator pinch point (DTpp,rec) must be increased, otherwise the boiling phenomenon in the recuperator occurs. Thus, with a superheater, an inferior limit in terms of working pressure should be fixed which, however, reduces its operability range. When a superheater is used, the oil mass flow rate is higher (Fig. 10), because the heat recovery in the recuperator increases and therefore, the heat transferred by the oil into the economizer decreases, and the TE (exhaust oil temperature) grows. In the no-

superheater layout (lower heat exchanged in the recuperator), TE is lower and thus the hot gas exchanges a greater amount of heat with the oil in the HRB. However, the stack temperature (Fig. 11) decreases and therefore, acid corrosion may happen. Without a superheater, the cycle needs an upper pressure limit (Plim ¼ Ps,max), because if the fluid pressure is higher than the pressure where the entropy of the saturation condition is maximum, the organic fluid enters in the wet vapor region during expansion which leads to a two-phase flow. The maximum working pressure for each liquid is shown in Table 2. Table 4 shows the optimal working pressure and the maximum net power for each fluid when the maximum oil temperature is

Table 4 Optimized condition for oil maximum temperature of 380  C. Fluid

SH

Wmax [kW]

Popt [bar]

Toluene Benzene Cyclopentane Cyclohexane

No No Yes No

3432.4 3397.6 2935.7 3273.1

26.13 37.5 45.0 35.90

35

3800

34

3600

33

3400 ΔT

Wel tot [kW]

moil [kg/s]

360 - 380 - 400 <= T

=40°C - No SH

32

Toluene Benzene CycloPentane CycloHexane

31

Toluene (No SH) Benzene (No SH) CycloPentane (SH) CycloHexane (No SH)

3200

3000

2800

30

2600

29 15

20

25

30

35

40

Pex,in [bar] Fig. 10. Oil mass flow rate versus inlet expander pressure with and without superheater.

20

25

30

35

40

45

Pex,in [bar] Fig. 12. Net electrical power versus inlet expander pressure of fluid varying maximum oil temperature and fluid.

C. Carcasci et al. / Energy 65 (2014) 91e100 200

97

38 36

190

34 180

32

ΔTsub [°C]

Wsp ex [kJ/kg]

30 170 160 150

28

360 - 380 - 400 <= T Toluene (No SH) Benzene (No SH) CycloPentane (SH) CycloHexane (No SH)

26 24 22

360 - 380 - 400 <= T

140

20

Toluene (No SH) Benzene (No SH) CycloPentane (SH) CycloHexane (No SH)

130

18 16

120

14 20

25

30

35

40

45

20

25

Pex in [bar]

30

35

40

45

Pex,in [bar]

Fig. 13. Expander specific power versus inlet expander pressure of fluid varying maximum oil temperature and fluid.

Toil,max ¼ 380  C. Toluene is the best choice with the lowest optimal pressure. With Benzene, the maximum net power is about 1% lower than with toluene, but the pressure level increases sharply. The worst choice is cyclopentane, with the minimum net power (about 14.5% lower than toluene), a high pressure level and with a superheater. 5.2. Maximum oil temperature analysis In Section 5.1 the effect of the presence of the superheater has been studied. Toluene, benzene and cyclohexane work best without a superheater, while cyclopentane works best with a superheater. Quoilin et al. [37] illustrated the influence of the maximum oil temperature and research the best condition for each fluid. Fig. 12 presents the influence of the inlet expander pressure on the net electrical power for several maximum oil temperatures for each organic fluid. Generally, power increases with pressure; for toluene, considering that the oil temperature is lower than 380  C, a maximum point is present. The net power produced using toluene is very sensitive to oil temperature variations, while its influence is much less marked for the other fluids because they reach the

Fig. 15. Subcooling temperature difference in economizer versus inlet expander pressure of fluid varying maximum oil temperature and fluid.

temperature limit earlier. In fact, benzene shows about the same net power values for maximum oil temperatures higher than 370  C and cyclohexane is not affected by oil temperature variations. Dai et al. [7] show how it is not always true that the higher the turbine inlet temperature, the greater the turbine power output. For the working fluids with a positive slope of the saturation vapor curves, the turbine inlet temperature should be kept as low as possible above the boiling point. The expander power output is given by the organic fluid mass flow rate and by the enthalpy difference. As shown in Fig. 13, the specific power of the expander reaches higher values for higher pressure values. Using cyclopentane, where the superheater is present, the specific power increases with the maximum oil temperature and thus with the superheater exit temperature. For the other fluids (non-superheater configuration) the enthalpy difference in the expander is constant with the oil temperature variations because the temperature inlet expander is fixed by the saturation temperature which depends on the pressure level. In the case without a superheater, when increasing the oil maximum temperature, the organic fluid mass flow rate tends to grow (Fig. 14), until the maximum mass flow rate value is reached

34

240

32

220

30

200

360 - 380 - 400 <= T

Tst [°C]

mfl [kg/s]

36

28 26 24

Toluene (No SH) Benzene (No SH) CycloPentane (SH) CycloHexane (No SH)

20 20

25

180 160 140

360 - 380 - 400 <= T

22

Toluene (No SH) Benzene (No SH) CycloPentane (SH) CycloHexane (No SH)

30

120

35

40

45

Pex,in [bar] Fig. 14. Organic fluid mass flow rate versus inlet expander pressure of fluid varying maximum oil temperature and fluid.

100 20

25

30

35

40

45

Pex,in [bar] Fig. 16. Stack temperature versus inlet expander pressure of fluid varying maximum oil temperature and fluid.

98

C. Carcasci et al. / Energy 65 (2014) 91e100 12000

500 450

Pex,in =35. bar

10000

400. | 360. | T oil,max

400

HotGas Oil Fluid Fluid (REC)

350 360 - 380 - 400 <= T

6000

T [°C]

QECO [kW]

8000

Toluene (No SH) Benzene (No SH) CycloPentane (SH) CycloHexane (No SH)

4000

300 250 200 150

2000

100 0 20

25

30

35

40

45

50 0

Pex,in [bar]

5000

10000

15000

20000

Q [kW] Fig. 17. Heat exchanged in economizer versus inlet expander pressure of fluid varying maximum oil temperature and fluid.

due to the oil thermal capacity. Using toluene, the maximum oil capacity is not reached and so the fluid mass flow rate is sensible to the maximum oil temperature, whereas using cyclohexane for each case the limit is reached. The fluid mass flow rate decreases when the pressure grows because the saturation temperature rises and the recovery heat in the evaporator decreases. Unlike toluene, the cyclopentane mass flow rate is quite constant under pressure variations and decreases when the oil temperature increases. The mass flow rates of benzene and cyclohexane, on the other hand, are independent from maximum oil temperature variations, because, by executing an energy balance in the economizer, the exhaust oil from the evaporator does not have the sufficient heat for the organic fluid to meet the evaporator imposed inlet conditions (a DTsub ¼ 15  C was imposed, see Table 3). Thus, the subcooling temperature difference increases (Fig. 15) and, consequently, the exchanged heat in the economizer decreases. The evaporator has to process a colder fluid and, for this reason, the mass flow rate is fixed at the maximum value the hot oil can generate. In the present study, the economizer subcooling temperature difference DTsub has been modified, but equivalent results can be obtained by increasing

Fig. 19. Temperature of fluid versus exchanged heat in heat exchangers for benzene (without superheater) for two different maximum oil temperatures.

the evaporator pinch point temperature difference DTpp; in this case, the fluid mass flow rate must be limited. The stack temperature (Fig. 16) is an indication of heat recovery from the GT exhaust gases. In a non-superheater layout, a lower stack temperature is shown for higher values of maximum oil temperature and expander inlet pressure. On the contrary, the trend is reversed when there is a superheater (when cyclopentane is used). The stack temperature depends on the working conditions of the economizer, which, in turn, depends on the conditions of the recuperator. The recovered heat in the economizer is shown in Fig. 17, where the exchanged heat is quite low when cyclopentane is used and so the high stack temperature observed is justified. At low maximum pressure values the economizer is unnecessary and, actually, the heat recovery in the recuperator is very high, as shown in Fig. 18. Fig. 19 points out the increase of heat exchanged in the evaporator because of the greater fluid mass flow rate and the increase of the subcooling temperature difference for benzene. Changing the maximum oil temperature, the oil curve becomes sharper because

500 8000

450

Pex,in =35. bar 400. | 360. | Toil max

400

HotGas Oil Fluid Fluid (REC)

6000

T [°C]

QREC [kW]

350

4000

300 250 200

2000

150

360 - 380 - 400 <= T Toluene (No SH) Benzene (No SH) CycloPentane (SH) CycloHexane (No SH)

0 20

25

30

35

40

45

Pex,in [bar] Fig. 18. Heat exchanged in recuperator versus inlet expander pressure of fluid varying maximum oil temperature and fluid.

100 50 0

5000

10000

15000

20000

Q [kW] Fig. 20. Temperature of fluid versus exchanged heat in heat exchangers for cyclopentane (with superheater) for two different maximum oil temperatures.

C. Carcasci et al. / Energy 65 (2014) 91e100

superheated vapor region enabling a regeneration to improve thermal efficiency; in this study, toluene is highly influenced by the maximum oil temperature, but it proves to be the best organic fluid among those selected for temperatures higher than 380  C. On the contrary, cyclopentane is the worst fluid in terms of performance. In the cases of benzene, cyclohexane and cyclopentane the maximum power is generated at the maximum allowed pressure and their performance is not greatly influenced by the oil temperature. It is, in fact, the oil working temperature which best indicates the fluid to be selected: cyclohexane is the best fluid for low oil temperature (below 363  C), benzene is the best choice when the oil temperature is approximately between 363  C and 378  C and toluene is the best for higher oil temperatures.

3800

3600

Wmax [kW]

3400

3200

Toluene Benzene CycloPentane CycloHexane

3000

2800 360

6. Conclusion 365

370

375

380

385

390

395

400

Toil max [°C] Fig. 21. Maximum net power versus maximum oil temperature for different organic fluids.

its mass flow rate decreases. Furthermore, the stack temperature (last point in the hot gas line) is higher when the maximum oil temperature decreases because the oil curve is not as sharp and so the exhaust oil temperature decreases. Likewise, Fig. 20 shows the case of cyclopentane: there is an excessive preheating in the recuperator and thus, the heat exchanged in the economizer is lower; the stack temperature and the oil temperature at the HRB inlet are higher; the subcooling temperature difference is constant with the imposed value (DTsub ¼ 15  C). For each maximum oil temperature a maximum power corresponding to an optimum pressure can be evaluated (Figs. 21 and 22). The optimum pressure is the pressure value relative to the maximum power output, considering the critical pressure Pcr as an upper limit for cyclopentane (the superheater is present) and Ps,max when the superheater is not present, as explained in Section 5.1. These figures permit to determine the optimum power plant conditions for each working fluid when the oil properties are known. Nishith et al. [39] consider that dry fluids are the most preferred working medium for the ORC system which utilizes low-grade heat sources, because the exhaust expander state point lies in the

46 44 42 40 38

Pex,in,opt[bar]

36 34 32 Toluene Benzene CycloPentane CycloHexane

30 28 26 24 22 20 18 360

99

365

370

375

380

385

390

395

400

Toil,max [°C] Fig. 22. Optimized pressure versus maximum oil temperature for different organic fluids.

The present paper has investigated a potential organic Rankine cycle. In Section 5.1, a thermodynamic analysis has been conducted by selecting the best configuration for several working fluids (toluene, benzene, cyclohexane and cyclopentane). The use of a superheater has been evaluated to transfer the working fluid into the superheated zone. Some organic fluids (toluene, benzene and cyclohexane) perform best without the superheater, while cyclopentane performs best with the superheater, although its performance output is the worst. Benzene, cyclohexane and cyclopentane present the optimal working pressure at the maximum value permitted. In Section 5.2, an analysis has been carried out by varying the maximum oil temperature and selecting the best configuration for each fluid. Generally, by increasing the oil temperature the net power increases (particularly, using toluene) until the heat recovery oil capacity limit is reached. Which organic fluid performs best depends on the oil working temperature: cyclohexane is the best fluid for low oil temperature, benzene is the best choice for medium oil temperature and toluene should be used for high oil temperature.

References [1] Cong CE, Velautham S, Darus AM. Sustainable power: solar thermal driven organic Rankine cycle. In: Proc Int Conf recent Adv in MechMater Eng. Paper No. 91, Kuala Lumpur, Malaysia May 2005. pp. 424e9. [2] Hung TC, Shai TY, Wang SK. A review of organic Rankine cycles (ORCs) for the recovery of low grade waste heat. Energy 1997;22(7):661e7. [3] Carcasci C, Facchini B, Harvey S. Modular approach to analysis of chemically recuperated gas turbine cycles. Energy Convers Manage 1998;39(16e18): 1693e703. [4] Carcasci C, Facchini B, Harvey S. Design issues and performance of a chemically recuperated aeroderivative gas turbine. Proc Inst Mech Eng, Part A 1998;212:315e29. [5] Badr O, Ocallaghan PW, Probert SD, et al. Rankine-cycle systems for harnessing power from low-grade energy-sources. Appl Energy 1990;36(4):263e 92. [6] Gu W, Weng Y, Wang Y, Zheng B. Theoretical and experimental investigation of an organic Rankine cycle for a waste heat recovery system. Proc Inst Mech Eng, Part A 2009;223:523e33. [7] Dai YP, Wang JF, Gao L, et al. Parametric optimization and comparative study of organic Rankine cycle (ORC) for low grade waste heat recovery. Energy Convers Manage 2009;50(3):576e82. [8] Roy JP, Mishra MK, Misra A. Parametric optimization and performance analysis of a waste heat recovery system using organic Rankine cycle. Energy 2010;35:5049e62. [9] Barbieri ES, Morini M, Pinelli M. Development of a model for the simulation of organic Rankine cycles based on group contribution techniques. Proc ASME Turbo Expo 2011;3:1011e9. [10] De Pascale A, Ferrari C, Melino F, Morini M, Pinelli M. Integration between a thermophotovoltaic generator and an organic Rankine cycle. Appl Energy 2012;97:695e703. [11] Gnutek Z, Bryszewska-Mazurek A. The thermodynamic analysis of multicycle ORC engine. Energy 2001;26(12):1075e82.

100

C. Carcasci et al. / Energy 65 (2014) 91e100

[12] Chen Y, Lundqvist P, Johansson A, Platell P, et al. A comparative study of the carbon dioxide transcritical power cycle compared with an organic Rankine cycle with R123 as working fluid in waste heat recovery. Appl Therm Eng 2006;26(17e18):2142e7. [13] Gang P, Jing L, Jie J. Analysis of low temperature solar thermal electric generation using regenerative organic Rankine cycle. Appl Therm Eng 2010;30: 998e1004. [14] Quoilin S, Orosz M, Hemond H, Lemort V. Performance and design optimization of a low-cost solar organic Rankine cycle for remote power generation. Sol Energy 2011;85. [15] Kosmadakis G, Manolakos D, Papadakis G. Simulation and economic analysis of a CPV/thermal system coupled with an organic Rankine cycle for increased power generation. Sol Energy 2011;85:308e24. [16] Heberle F, Brüggemann D. Exergy based fluid selection for a geothermal organic Rankine cycle for combined heat and power generation. Appl Therm Eng 2010;30:1326e32. [17] Astolfi M, Xodo L, Romano MC, Macchi E. Technical and economical analysis of a solaregeothermal hybrid plant based on an organic Rankine cycle. Geothermics 2011;40:58e68. [18] Sanchez D, Muñoz JM, Monje B, Chacartegui R, Sanchez T. Preliminary analysis of compound systems based on high temperature fuel cell, gas turbine and organic Rankine cycle. J Power Sources 2011;196:4355e63. [19] Chacartegui R, Sánchez D, Muñoz JM, Sánchez T. Alternative ORC bottoming cycles for combined cycle power plants. Appl Energy 2009;86:2162e70. [20] Muñoz JM, Sánchez D, Chacartegui R, Sánchez T. Part-load analysis of gas turbine & ORC combined cycles. Appl Therm Eng 2012;36:63e72. [21] Lai NA, Wendland M, Fischer J. Working fluids for high-temperature organic Rankine cycles. Energy 2011;36:199e211. [22] Saleh B, Koglbauer G, Wendland M, Fischer J. Working fluids for lowtemperature organic Rankine cycles. Energy 2007;32:1210e21. [23] Vankeirsbilck I, Vanslambrouck B, Gusev S, De Paepe M. Organic Rankine cycle as efficient alternative to steam cycle for small scale power generation. In: 8th international conference on heat transfer, fluid mechanics and thermodynamics, proceedings. Ghent, Belgium: Ghent University, Department of Flow, Heat and Combustion Mechanics; 2011. pp. 785e92. [24] Drescher U, Bruggemann D. Fluid selection for the organic Rankine cycle (ORC) in biomass power and heat plants. Appl Therm Eng 2007;27:223e8.

[25] Desai NB, Bandyopadhyay S. Process integration of organic Rankine cycle. Energy 2009;34:1674e86. [26] Quoilin S, Lemort V, Lebrun J. Experimental study and modeling of an organic Rankine cycle using scroll expander. Appl Energy 2010;87:1260e8. [27] Victor RA, Kim JK, Smith R. Composition optimisation of working fluids for organic Rankine cycles and Kalina cycles. Energy 2013;55:114e26. [28] Carcasci C, Ferraro R. Thermodynamic optimization and offdesign performance analysis of a toluene based Rankine cycle for waste heat recovery from medium size gas turbines. In: ASME 2012 gas turbine India conference, December 1, 2012, Mumbai, Maharashtra, India, GTIndia2012e9645. [29] Tchanche BF, Papadakis G, Lambrinos G, Frangoudakis A. Fluid selection for a lowtemperature solar organic Rankine cycle. Appl Therm Eng 2009;29:2468e76. [30] He C, Liu C, Gao H, Xie H, Li Y, Wu S, et al. The optimal evaporation temperature and working fluids for subcritical organic Rankine cycle. Energy 2012;38:136e43. [31] Bianchi M, De Pascale A. Bottoming cycles for energy generation: parametric investigation of available and innovative solutions for the exploitation of low and medium temperature heat sources. Appl Energy 2011;88:1500e9. [32] Dow chemical company product, “Line guide”. URL, http://www.dowtherm. com; May 2007. [33] Nuovo Pignone GE Oil & Gas, 2011 GE10 e 1 gas turbine brochure. URL, http:// www.ge.com/oil&gas; 2005. [34] Del Turco P, Asti A, Del Greco AS, Bacci A, Landi G, Seghi G. The ORegenTM waste heat recovery cycle: reducing the CO2 footprint by means of overall cycle efficiency improvement. Proc ASME Turbo Expo 2011. GT2011-45051. [35] Kang SH. Design and Experimental study of ORC (organic Rankine cycle) and radial turbine using R245fa working fluid. Energy May 2012;41(1):514e24. [36] Lukowicz H, Kochaniewicz A. Analysis of the use of waste heat obtained from coal-fired units in organic Rankine cycles and for brown coal drying. Energy September 2012;45(1):203e12. [37] Quoilin S, Aumann R, Grill A, Schuster A, Lemort V, Spliethoff H. Dynamic modeling and optimal control strategy of waste heat recovery Organic Rankine Cycles. Manuscript, Appl Energy June 2011;88(6):2183e90. [38] Sun J, Li W. Operation optimization of an organic Rankine cycle (ORC) heat recovery power plant. Appl Therm Eng August 2011;31(11e12):2032e41. [39] Nishith B, Desai S, Bandyopadhyay. Process integration of organic Rankine Cycle. Department of Energy Science and Engineering, Indian Institute of Technology Bombay, Powai, Mumbai 400076, India.