- Email: [email protected]
Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle
Energy Conversion and Management xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman
Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle Samer Maalouf ⇑, Elias Boulawz Ksayer, Denis Clodic Ecole des Mines de Paris, CES, 5 rue Léon Blum, Palaiseau, France
a r t i c l e
i n f o
Article history: Available online xxxx Keywords: Direct contact condensation Indirect contact condensation Organic Rankine Cycle (ORC) Waste heat recovery Wet flue gases
a b s t r a c t Low-temperature flue gases (<120 °C) exiting industrial processes could be recovered for electricity generation and constitute an effective mean to reduce primary energy consumption and carbon dioxide emissions. In the wet flue gases, substantial heat can be recovered if water vapor contained in the gases is condensed. Technical options include indirect contact water vapor condensation recovery, where heat is transferred between the two fluids (typically flue gases and working fluid) using an intervening wall (typically fin-and-tube heat exchanger) and direct contact water vapor condensation recovery, which involves direct mixing between flue gases and cooling fluid (typically water) through a condensing unit. In this paper, the two recovery processes are investigated using ORC (Organic Rankine Cycle). While the indirect contact condensation is the most favorable heat recovery scheme concerning the net output power, the direct contact heat exchanger has received attention because there are no heat-transfer surfaces exposed to corrosion. In a direct contact water–vapor condensation, the inlet flue-gas wet-bulb temperature determines the operating temperature levels throughout the system and limits the circulating water temperature. The maximal net turbine power for the direct contact system is reached for a final water temperature nearby the entering wet bulb temperature of the flue gases. The temperature pinch is as low as 0.5 K, which is possible with a direct contact heat exchanger. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Nowadays, more and more attention has been paid to the utilization of low-temperature flue gases (<120 °C) released from several industry sectors (cement, steel, refineries), for its potential in reducing fossil fuel consumption and alleviating environmental problems. Organic Rankine Cycle (ORC) is proposed to recover low-grade energy and transform it into power. It is a proven technology, which allows the generation of electricity from low-temperature heat sources in a far more efficient way than conventional steam cycles [1,2]. Much industrial flue gases may contain significant amount of moisture in vapor form (wet flue gases) due to many reasons such as flashing, washing, cleaning, and drying. The water dew point temperature of these flue gases could range between 55 °C and 65 °C. Therefore, a considerable amount of heat is available in the form of latent heat of water vapor in these gases and cannot be recovered if flue gases are not cooled down to temperatures lower ⇑ Corresponding author. E-mail addresses: [email protected] (S. Maalouf), elias. [email protected] (E. Boulawz Ksayer), [email protected] (D. Clodic).
than the flue-gas dew point. The recovery of this large amount of heat improves the overall efficiency of the recovery system [3]. An important factor influencing latent heat recovery is the corrosion problem associated to the cooling when flue gases contain sulfuric oxides (SOx), nitric oxides (NOx), and hydrochloric acid (HCl). However, many recovery technologies are already well developed and technically proven. Options include indirect contact condensation recovery and direct contact condensation recovery [3]. In an indirect contact condenser, the heat is transferred between the two fluids (typically flue gases and working fluid) using an intervening wall (typically fin-and-tube heat exchanger). In this case, the heat-exchanger design requires using advanced materials such as ‘‘Teflon” or equivalent coating to withstand exposure to corrosion problems. In a direct contact condenser, heat is transferred between the two fluids (typically gas and water) without an intervening wall thus there are no heat-transfer surfaces exposed to corrosion, clogging, and fouling. The two fluids move in a counter-flow direction, with one of them dispersed as small particles in a vertical column. Many investigations were carried out about low-grade heat recovery using ORC. Comprehensive researches on appropriate working fluids for low-temperature applications have been investigated by many authors such as [4–6]. Others researchers have
http://dx.doi.org/10.1016/j.enconman.2015.09.047 0196-8904/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
2
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
Nomenclature y
condensing unit mass enthalpy (kJ/kg) mass enthalpy of flue gases (kJ/kgdg) mass enthalpy of dry flue gases (kJ/kgdg) mass enthalpy of water vapor (kJ/kgwv) heat exchanger irreversibility (kJ/kg) mass flow rate (kg/s) molar mass of dry gases (kg/kmol) molar mass of water vapor (kg/kmol) Organic Rankine Cycle total pressure (Pa) water vapor pressure (Pa) heat capacity (kJ/kgdg) mass entropy (kJ/kg K) sub-cooling (K) temperature (°C) humidity ratio (kgwv/kgdg) power (kJ/kgdg)
Greek symbols efficiency (%)
g
Subscripts cond condensation/condenser dg dry gases dp dew point evap evaporation fg flue gases i components is isentropic o ambient conditions r working fluid wb wet bulb wf working fluid wv water vapor
focused on the parametric optimization and performance analysis of the ORC like [7–10]. However, these studies deal with the indirect contact condensation recovery based on sensible heat extraction from flue gases with low moisture contents, or even though, with high moisture contents [11], but cooling the gases to a minimum safe temperature in order to prevent water vapor condensation and acid formation during gas flow. The originality of this study is to extend the ORC applications to low-grade gas heat sources with high moisture contents by pointing out the effect of water vapor condensation on cycle performance using the two condensing heat recovery processes (direct and indirect heat exchange). 2. Heat load availability In the wet flue gases, heat is available in both sensible and latent forms. The sensible heat is determined by the temperature of the flue gases and the combined heating capacities of its constituents. The latent heat is determined by the amount of water present in the flue gases in gas form [4]. The characteristics of the flue gases used in this study are tabulated in Table 1. The flue gases to be cooled are considered a mixture of water vapor (H2O), nitrogen (N2), oxygen (O2), and carbon dioxide (CO2). The CO2 and O2 compositions correspond to those at the ‘‘Raw Mill” exhaust in the cement plants. The water dew point is varied between 55 and 65 °C. The enthalpy of a mixture of gases is equal to the sum of the individual partial enthalpies of the components [12]. Therefore, the mass enthalpy of flue gases can be written as follows:
hg ðTÞ ¼ hdg ðTÞ þ w hwv ðTÞ ¼ yO2 hO2 ðTÞ þ yCO2 hCO2 ðTÞ þ yN2 hN2 ðTÞ þ w hwv ðTÞ
ð1Þ
yO2 þ yCO2 þ yN2 ¼ 1
ð2Þ
where subscript ‘‘g” denotes gas, ‘‘dg” denotes dry gases, ‘‘wv” denotes water vapor, ‘‘T” is the temperature of the mixture, ‘‘w” is the humidity ratio, ‘‘y” is the dry mass fraction, and ‘‘h” represents the mass enthalpy. The mass enthalpies of the different components (O2, CO2, N2 and H2O) are calculated at the corresponding partial pressures. The humidity ratio is calculated by:
w¼
Mwv Pw M dg Pt Pw
ð3Þ
where ‘‘Mwv” and ‘‘Mdg” are the molar masses of the water vapor and dry gases respectively, ‘‘Pw” is the partial pressure of water vapor, and ‘‘Pt” is the total pressure. The thermodynamic data of gases adopted in the present work are calculated using REFPROP 9.0 [13]. The ambient pressure and temperature at the specified dead reference state (Po and To) are considered to be atmospheric pressure and 20 °C. The characterization of the temperature and the available quantity of heat referred to the ambient temperature are presented in Fig. 1. The sudden breaks in slope indicate initial water dew points. Water vapor in the flue gases is in superheated state above the initial water dew points. The cooling first reaches the initial water dew points at which condensation begins. The partial pressure of water vapor decreases continually and accompanied by a reduction in water dew point. Sensible heat recovery occurs down to water dew points. Cooling below this level increases the energy recovery rate by recovery of the latent heat.
140
Table 1 Flue-gas characteristics. Composition Inlet temperature Water dew point temperatures Corresponding wet bulb temperatures
mass fraction
T dp = 45°C
120
Temperature (°C)
CU h hg hdg hwv HEX I m Mdg Mwv ORC Pt Pw Q s SC T w W
T dp = 55°C 100
T dp = 65°C
80 60 40 20
Molar basis Tinlet Tdp Twb
CO2 – 15.5%, O2 – 6.8% 120 °C 55/60/65 °C 59.9/63.7/67.8 °C
0 0
100
200
300
400
500
600
700
∆h (kJ.kgdg -1) Fig. 1. Heat load availability.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
3
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
Fig. 1 shows that a large amount of heat within the streams is in the form of latent heat and can only be recovered by condensing water vapor. The fraction of latent heat is of course dependent on the water dew point temperature. The analysis carried previously shows that the sensible energy could be recovered at relatively high temperatures compared to the latent energy. However, the amount of energy available per temperature difference is much larger when recovering latent heat below the initial water dew point temperature.
3. Condenser:
Q cond ¼ mr ðh4 h1 Þ
ð9Þ
Icond ¼ E4 E1
ð10Þ
4. Pump:
W pump ¼ mr ðh2 h1 Þ
ð11Þ
Ipump ¼ W pump þ E1 E2
ð12Þ
3. ORC with indirect contact heat recovery The lay-out diagram of a simple ORC is presented in Fig. 2a. The ORC system consists of an evaporator, a turbine, a condenser, and a pump. As shown in Fig. 2a, the pump supplies the working fluid to the evaporator, where the working fluid is heated and vaporized by the flue-gas heat stream (Point i). The generated high-pressure (Point 3) vapor is expanded by the turbine, which is directly coupled to an electrical generator producing power, and then, the low-pressure vapor is condensed in a condenser whereupon it is pumped back to the evaporator. The T–s diagram for the studied ORC system is shown in Fig. 2b. Since the focus of this study is the thermodynamic ORC optimization, detailed calculations of pressure losses and heat transfer in evaporator and condenser are not taken into account since they depend strongly on materials and configurations of the system components [5]. In order to calculate the irreversibility ‘‘I” for each component in the ORC, the ‘‘exergy” will be used and is defined as:
E ¼ mr ðh T o sÞ
The evaporator superheat is given by:
SHevap ¼ T 3 T 2b
By subtracting the power consumed by the pump from the power generated by the turbine, the net turbine power is:
W net ¼ W turbine W pump
ð14Þ
The heat recovery from the flue gases is given by:
Q fg ¼ mdg ðhi he Þ
ð15Þ
where hi and he are respectively the inlet and outlet enthalpies of dry flue gases expressed in kJ/kgdg. Since the heat released by the flue gases is recovered by the working fluid, the working fluid mass flow rate could be calculated by combining Eqs. (5) and (15) as shown:
mr ¼ Q fg =ðh3 h2 Þ
ð4Þ
where ‘‘To” refers to a reference temperature state (20 °C). The thermodynamic equations for the components in the ORC of Fig. 2 are:
ð13Þ
ð16Þ
The isentropic efficiency of the turbine can be expressed as:
gis;turbine ¼ ðh3 h4 Þ=ðh3 h4s Þ
ð17Þ
The isentropic efficiency of the pump can be expressed as:
gis;pump ¼ ðh2s h1 Þ=ðh2 h1 Þ
1. Evaporator:
Q evap ¼ mr ðh3 h2 Þ
ð5Þ
Ievap ¼ Ei þ E2 Ee E3
ð6Þ
2. Turbine:
W turbine ¼ mr ðh3 h4 Þ
ð7Þ
Iturbine ¼ E3 W turbine —E4
ð8Þ
ð18Þ
Based on the first law of thermodynamics, the ORC thermal efficiency is defined as the ratio of the net power output to the heat addition:
gcycle ¼ ðW turbine W pump Þ=Q evap
ð19Þ
The thermal efficiency cannot reflect the ability to convert energy from low-grade waste heat into usable power and might be misleading since the maximum power output is not achieved concurrently with the maximum thermal efficiency. Therefore,
e
i
3 Evaporator 2 Pump Turbine
Generator 4
1
Temperature [K]
i e
DT
2a
4
2 2s 1
3
Pevap 2b
4s 4a
Pcond
4b
Condenser
(a)
Entropy [kJ/(kg.K)] (b) Fig. 2. Simple ORC: (a) lay-out diagram and (b) (T–s) diagram.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
4
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
the exergy efficiency is considered in order to evaluate the performance for waste heat recovery. The exergy efficiency of the ORC system can be expressed as:
gexergy ¼ ðEi RI Ee Þ=Ei
ð20Þ
The operating conditions of the ORC are given in Table 2. The turbine efficiency is supposed to be the overall turbine efficiency that includes the isentropic and the mechanical efficiencies. Consideration of the pinch value (DT) between the flue gases and the working fluid in the evaporator is essential to define the maximal heat recovery from the energy source. As pointed out by Butcher and Reddy [12] and Srinivasan et al. [14], the system performance is sensitive to the pinch point and should be as low as possible to improve the overall energy performance. However, this will result in a larger heat-exchanger area to transfer the same amount of heat, hence an increase in size and cost. Following calculations have been done for a 3-K pinch at the evaporator. By referring to [15], this value is considered as aggressive boundary conditions. Therefore, a technical and economic optimization will be performed in developing sensitivity analysis on cost variations around the pinch while thermodynamic optimization of energy system requires only that the pinch is as small as possible. For the working fluid selection, HFC-245fa is chosen. As stated elsewhere [16,17], HFC-245fa is a promising candidate for ORC operating on low-temperature heat sources, and presents one of the best compromise between thermal efficiency, power output, and environmental performance. The properties of HFC-245fa are calculated using REFPROP 9.0 [13] developed by NIST. A computer program employing ‘‘Excel/VBA” was developed to simulate the thermodynamic performances of the ORC system. A parametric optimization of the evaporation temperature was performed to obtain the ORC maximal net power output. As reported by Liu et al. [6], superheating is detrimental to ORC efficiency for the working fluids with the non-negative slope of the saturation vapor curves (dT/ds > 0, dry fluid) such as HFC-245fa, and a nil evaporator superheat should produce the highest turbine power. Fig. 3 shows the evolution of the working fluid mass flow rate (for a 1 kg/s of dry flue gases) as a function of the evaporation temperature. As shown in Fig. 3, the mass flow rate decreases when increasing the evaporation temperature, this could be explained
Table 2 ORC input data. Parameters
Values
Condensing temperature Condenser sub-cooling Turbine & pump efficiencies
25 °C 2K 80%/85%
Tcond SCcond
gturbine/gpump
by referring to Eq. (16). In fact, on one hand, an increase in the evaporation temperature leads to an increase in the outlet flue gas temperature for a given pinch value in the evaporator, and then less heat is extracted from the flue gases, thus ‘‘Qfg” decreases. While, on the other hand, for a given condensation temperature, an increase in the evaporation temperature leads to an increase in the enthalpy difference in the evaporator at the working side ‘‘h3 h2”. Fig. 4 shows the evolution of the net power output for various water dew point temperatures under a wide range of evaporation temperature. The value ‘‘Tevap = Tdp DT” breaks each corresponding curve into two regions classified as follows: Sensible heat region (Tevap > Tdp DT): where mostly sensible heat is recovered. Latent heat region (Tevap < Tdp DT): where a great portion of latent heat is recovered. The value of the net turbine power is strongly affected by the water dew point temperature. It should be noted that the net power is mainly managed by the turbine power since the pump power is almost negligible. Fig. 4 shows that the net turbine power increases when increasing the water dew point temperature. In fact, an increase in the water dew point temperature leads to an increase in the quantity of water vapor condensation; and then more latent heat is released in the evaporator. Thus, by referring to Eq. (15) and as shown clearly in Fig. 3, the working fluid mass flow rate increases when increasing the water dew point temperature, resulting to an increase in the net turbine power. As shown in Fig. 4, the net power output increases first and then decreases. This can be explained by the fact that, by referring to Eq. (7), the turbine power results from the products of two terms evolving differently with the evaporation temperature: the working fluid mass flow rate which decreases when the evaporation temperature increases as already shown in Fig. 3, and the enthalpy difference ‘‘h3 h4” which increases when the evaporation temperature increases for a given condensation temperature. Finally, Fig. 4 shows that an optimum for the net turbine power can be identified in each region: sensible and latent heat optima. The corresponding evaporation temperature at the sensible heat optimum (73 °C) is almost independent on the water dew point temperature, whereas it depends strongly on it at the latent heat optimum. The value of the net turbine power at the latent heat optimum exceeds widely that at the sensible heat optimum at high water dew point temperatures (Tdp = 60–65 °C). Fig. 5 shows the evolution of evaporator capacity as a function of the evaporation temperature. A high evaporator capacity is expected below water dew point due to the large available quantity of latent heat. Fig. 5 shows that the evaporator capacity decreases when increasing the evaporation temperature. In fact, for a given pinch value in the evaporator, an increase in the
3
m wf (kg/s)
2 2 1 1
25
W net (kJ/kgdg )
T dp = 55°C T dp = 60°C T dp = 65°C
3
Tdp = 55°C Tdp = 60°C Tdp = 65°C
20 15 10 5
0 20
40
60
80
100
120
T evap (°C)
0 20
40
60
80
100
120
T evap (°C) Fig. 3. Evolution of the working fluid mass flow rate as a function of the evaporation temperature.
Fig. 4. Evolution of the net power as a function of the evaporation temperature.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
5
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
Q evap (kJ/kgdg )
600
process” (Fig. 8). In this case, HEX 1 will recover almost all sensible heat above the water dew point temperature so no condensation will occur on the heat-exchanger surface, and the condensing unit
Tdp = 55°C Tdp = 60°C Tdp = 65°C
500 400 300 200 100
e
0 20
40
60
80
100
120
Condensing unit
T evap (°C) Fig. 5. Evolution of the evaporator capacity as a function of the evaporation temperature. Water inlet to CU
i water pump
25
Efficiencies (%)
Ƞ cycle 20
Ƞ exergy
15 10
3 Evaporator 2
5 Pump
0 0
20
40
60
80
100
120
Turbine
Generator 4
T evap (°C) 1
Fig. 6. Evolution of cycle and exergy efficiencies as a function of the evaporation temperature.
Condenser
evaporator temperature leads to an increase in the outlet flue gas temperature and then less heat is extracted from the flue gases. Fig. 6 shows the evolutions of the cycle and exergy efficiencies as a function of the evaporation temperature. The cycle efficiency increases by increasing the evaporation temperature, which can be deduced by referring to Eq. (19) since the evaporator capacity decreases by increasing the evaporation temperature. Indeed, the exergy efficiency presents the same evolution shape as for the net power (refer to Fig. 4). The irreversibility in the evaporator could be reduced and optimum exergy efficiency could be achieved by varying the operating temperature.
Fig. 7. ORC with direct contact condensation.
e
Condensing unit
4. ORC with direct contact heat recovery Direct condensation heat recovery involves direct mixing between flue gases and cooling fluid. Since these systems do not involve a separating wall across which heat must be transferred, they avoid some of the challenges of large heat-transfer surfaces required for indirect contact units. An example system is shown in Fig. 7. As the flue gases enter the condensing unit (Point i), they are cooled by cold water introduced at the top of the unit. The heated water stream recovered at the bottom of the condensing unit is then sent to an evaporator to transfer heat to the organic working fluid of the power generation unit. Once the heat is extracted from the water, a treatment system will ensure that the water is cleaned before being re-used for another heat extraction cycle. Because the circulating water can only reach a temperature equivalent to adiabatic saturation temperature of flue gases, heat can only be transferred to the working fluid at a temperature lower than the entering wet bulb temperature. In order to reach a higher circulating water temperature, an indirect contact heat exchanger (HEX 1) could be added resulting into a ‘‘hybrid heat recovery
water pump Water inlet to CU HEX1 i
3 Evaporator
Turbine
Generator 4
Condenser
Fig. 8. Hybrid heat recovery process.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
6
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
will recover most of the latent heat below water dew point temperature. The ORC operating conditions are given in Table 3. The condenser, turbine and pump parameters are the same as those indicated in Table 2. In the direct contact condensation process, a minimum pinch of 0.5 K is maintained in the condensing unit between the entering wet bulb temperature and the leaving water temperature. Using the same procedure as for Eq. (16), the mass flow rates of the circulating water and the working fluid are calculated by applying the heat balance in the condensing unit and in the evaporator respectively. In Figs. 9 and 10, a parametric optimization of the water temperature at the evaporator inlet was performed with respect to the net turbine power and the evaporator capacity. The vertical lines represent the boundary limit between the direct contact process (zone I) and the hybrid heat-recovery process (zone II). The small circles denote the temperatures at which the water temperatures at the evaporator inlet are equal to the corresponding wet bulb temperatures for the different water dew point temperatures. The vertical lines cross the horizontal axis at evaporator inlet water temperatures equal to the wet bulb temperatures minus 0.5 K (minimum pinch in condensing unit). In zone I, the net turbine power increases slightly by increasing the inlet water temperature
Table 3 Additional ORC input data. Parameters
Values
DTHEX 1 DTmin CU – Pwater
Pinch in HEX 1 Minimum pinch in CU Tout flue gases from CU Tin water to CU Water pressure after water pump
25
T dp = 55°C T dp = 60°C T dp = 65°C
20
W net (kJ/kgdg )
3K 0.5 K 2K 0.2 MPa
10
Zone I
Zone II
0 40
50
60
70
80
90
100
110
120
T water at evaporator inlet (°C) Fig. 9. Net power as a function of the water temperature at the evaporator inlet.
500
T dp = 55°C T dp = 60°C T dp = 65°C
Q evap (kJ/kgdg )
400 300 200 100
Zone I
Zone II
0 40
50
60
70
80
T water at evaporator
90
100
110
Wnet
maximal
(kJ/(kgdg/s))
Indirect contact condensation Direct contact condensation Hybrid heat recovery mode
Tdp = 55 °C
Tdp = 60 °C
Tdp = 65 °C
9.21 (100%) 7.79 (84.6%) 8.73 (94.8%)
13.90 (100%) 11.66 (83.9%) 13.03 (93.8%)
21.12 (100%) 17.26 (81.7%) 19.25 (91.1%)
Table 5 Comparison of the outlet flue gas temperature between the analyzed heat recovery modes at the corresponding maximal net power. T outlet flue gases (°C)
Tdp = 55 °C
Tdp = 60 °C
Tdp = 65 °C
Indirect contact condensation Direct contact condensation Hybrid heat recovery mode
43.5 45.1 42.2
45.2 47.4 43.7
46.8 49.9 45.1
at the evaporator and reaches the maximal value at the boundary limit between the two zones, whereas, the evaporator capacity is almost constant. In zone II, the net turbine power decreases by increasing the inlet water temperature at the evaporator inlet, because less latent heat is extracted from the heat source when moving away from the wet bulb temperature. The maximal value of the net turbine power occurs for a water temperature at the evaporator inlet equal to the wet bulb temperature. A comparison of the maximal net turbine power between the three heat recovery modes is shown in Table 4. While the indirect contact condensation is the most favorable heat recovery scheme regarding the net power output, the hybrid heat-recovery mode leads to a potential increase in the net power output by about 10% compared to the direct contact condensation. Table 5 shows a comparison of the outlet flue gases temperature between the three heat recovery modes at the corresponding maximal net power. The hybrid heat recovery mode presents the lowest outlet flue gas temperature since the heat is recovered using both heat exchangers: an indirect contact heat exchanger (HEX1) and a direct contact heat exchanger (CU).
5. Conclusions
15
5
Table 4 Comparison between the analyzed heat recovery modes.
120
inlet (°C)
Fig. 10. Evaporator capacity as a function of the water temperature at the evaporator inlet.
This study details heat recovery from low-temperature waste heat (T < 120 °C) in industrial processes for generating electrical power using an ORC system. The flue gases are characterized by a water dew point temperature ranging from 55 to 65 °C. Based on the analysis presented in this investigation, the following conclusions are drawn: While the amount of latent heat in the gas stream may look attractive, its recovery is hindered by the low-temperature level at which the recovered latent heat is available. In the indirect contact condensation, depending on the evaporation temperature, two operating regions can be identified in the evaporator: ‘‘sensible heat region” where mostly sensible heat is recovered, and ‘‘latent heat region” where a great portion of latent heat is recovered. In the direct contact condensation, the inlet wet-bulb temperature determines the operating temperature levels throughout the system and limits the circulating water temperature. The maximal net turbine power of the direct contact system is reached for a final water temperature nearby the entering wet bulb temperature of the flue gases. The hybrid heat-recovery process leads to a potential increase in the net power output compared to the direct contact condenser, while avoiding surface corrosion problems faced in the indirect contact condenser.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
As aforementioned in the introduction, an important factor influencing latent heat recovery is the corrosion problem accompanying the gas cooling, which requires high quality materials or frequently replacing system components especially for the indirect heat recovery mode. Thus, a multi-objective optimization taking into account the economic aspect is necessary after the thermodynamic optimization process. Acknowledgment The research leading to these results has received funding from the European Community’s Seventh Framework Program (FP7/2007-2013) under Grant agreement no. 256790 (‘LOVE’). References [1] Liu B, Rivière P, Coquelet C, Gicquel R, David F. Investigation of a two-stage Rankine cycle for electric power plants. Appl Energy 2012;100:285–94. [2] Tamamoto T, Furuhata T, Arai N, Mori K. Design and testing of the organic Rankine cycle. Energy 2001;26:239–51. [3] BCS, Incorporated. Report on waste heat recovery: technology and opportunities in U.S. industry. U.S. Department of Energy; March 2008. [4] Dai Y, Wang J, Gao L. Parametric optimization and comparative study of organic Rankine cycle (ORC) for low grade waste heat recovery. Energy Convers Manage 2009;50:576–82. [5] Hung TC, Shai TY, Wang SK. A review of organic Rankine cycles (OECs) for the recovery of low-grade waste heat. Energy 1997;22(7):661–7. [6] Liu BT, Chien KH, Wang CC. Effect of working fluids on organic Rankine cycle for waste heat recovery. Energy 2004;29:1207–17.
7
[7] Roy JP, Mishra MK, Misra A. Performance analysis of an organic Rankine cycle with superheating under different heat source temperature conditions. Appl Energy 2011;88:2995–3004. [8] Roy JP, Mishra MK, Misra A. Parametric optimization and performance analysis of a regenerative organic Rankine cycle using low-grade waste heat for power generation. Int J Green Energy 2011;8:173–96. [9] Roy JP, Mishra MK, Misra A. Parametric optimization and performance analysis of a regenerative organic Rankine cycle using R-123 for waste heat recovery. Energy 2012;39:227–35. [10] Wei D, Lu X, Lu Z, Gu J. Performance analysis and optimization of organic Rankine cycle (ORC) for waste heat recovery. Energy Convers Manage 2007;48:1113–9. [11] Roy JP, Mishra MK, Misra A. Parametric optimization and performance analysis of a waste heat recovery system using organic Rankine cycle. Energy 2010;35:5049–62. [12] Butcher C, Reddy B. Second law analysis of a waste heat recovery based power generation system. Int J Heat Mass Transf 2009;50:2355–63. [13] Lemmon EW, Huber ML, McLinden MO. Reference fluid thermodynamic and transport properties. NIST standard reference database 23, version 9.0. [14] Srinivasan K, Mago P, Krishnan S. Analysis of exhaust waste heat recovery from a dual fuel low temperature combustion engine using an organic Rankine cycle. Energy 2010;35(6):2387–99. [15] Becquin G, Freund F. Comparative performance of advanced power cycles for low-temperature heat sources. In: ECOS 2012: Proceedings of the 25th international conference on efficiency, cost, optimization, simulation, and environmental impact of energy systems, June 26–29, Perugia, Italy; 2012. [16] Kosmadakis G, Manolakos D, Kyritsis S, Papadakis G. Comparative thermodynamic study of refrigerants to select the best for use in hightemperature stage of a two-stage organic Rankine cycle for RO desalination. Desalination 2009;243:74–94. [17] Zyhowski GJ, Brown AP, Achaichia A. HFC-245fa working fluid in organic Rankine cycle – a safe and economic way to generate electricity from waste heat. In: ECOS 2010: Proceedings of the 23th international conference on efficiency, cost, optimization, simulation, and environmental impact of energy systems, June 14–17, Lausanne, Switzerland; 2010.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
Contents lists available at ScienceDirect
Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman
Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle Samer Maalouf ⇑, Elias Boulawz Ksayer, Denis Clodic Ecole des Mines de Paris, CES, 5 rue Léon Blum, Palaiseau, France
a r t i c l e
i n f o
Article history: Available online xxxx Keywords: Direct contact condensation Indirect contact condensation Organic Rankine Cycle (ORC) Waste heat recovery Wet flue gases
a b s t r a c t Low-temperature flue gases (<120 °C) exiting industrial processes could be recovered for electricity generation and constitute an effective mean to reduce primary energy consumption and carbon dioxide emissions. In the wet flue gases, substantial heat can be recovered if water vapor contained in the gases is condensed. Technical options include indirect contact water vapor condensation recovery, where heat is transferred between the two fluids (typically flue gases and working fluid) using an intervening wall (typically fin-and-tube heat exchanger) and direct contact water vapor condensation recovery, which involves direct mixing between flue gases and cooling fluid (typically water) through a condensing unit. In this paper, the two recovery processes are investigated using ORC (Organic Rankine Cycle). While the indirect contact condensation is the most favorable heat recovery scheme concerning the net output power, the direct contact heat exchanger has received attention because there are no heat-transfer surfaces exposed to corrosion. In a direct contact water–vapor condensation, the inlet flue-gas wet-bulb temperature determines the operating temperature levels throughout the system and limits the circulating water temperature. The maximal net turbine power for the direct contact system is reached for a final water temperature nearby the entering wet bulb temperature of the flue gases. The temperature pinch is as low as 0.5 K, which is possible with a direct contact heat exchanger. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction Nowadays, more and more attention has been paid to the utilization of low-temperature flue gases (<120 °C) released from several industry sectors (cement, steel, refineries), for its potential in reducing fossil fuel consumption and alleviating environmental problems. Organic Rankine Cycle (ORC) is proposed to recover low-grade energy and transform it into power. It is a proven technology, which allows the generation of electricity from low-temperature heat sources in a far more efficient way than conventional steam cycles [1,2]. Much industrial flue gases may contain significant amount of moisture in vapor form (wet flue gases) due to many reasons such as flashing, washing, cleaning, and drying. The water dew point temperature of these flue gases could range between 55 °C and 65 °C. Therefore, a considerable amount of heat is available in the form of latent heat of water vapor in these gases and cannot be recovered if flue gases are not cooled down to temperatures lower ⇑ Corresponding author. E-mail addresses: [email protected] (S. Maalouf), elias. [email protected] (E. Boulawz Ksayer), [email protected] (D. Clodic).
than the flue-gas dew point. The recovery of this large amount of heat improves the overall efficiency of the recovery system [3]. An important factor influencing latent heat recovery is the corrosion problem associated to the cooling when flue gases contain sulfuric oxides (SOx), nitric oxides (NOx), and hydrochloric acid (HCl). However, many recovery technologies are already well developed and technically proven. Options include indirect contact condensation recovery and direct contact condensation recovery [3]. In an indirect contact condenser, the heat is transferred between the two fluids (typically flue gases and working fluid) using an intervening wall (typically fin-and-tube heat exchanger). In this case, the heat-exchanger design requires using advanced materials such as ‘‘Teflon” or equivalent coating to withstand exposure to corrosion problems. In a direct contact condenser, heat is transferred between the two fluids (typically gas and water) without an intervening wall thus there are no heat-transfer surfaces exposed to corrosion, clogging, and fouling. The two fluids move in a counter-flow direction, with one of them dispersed as small particles in a vertical column. Many investigations were carried out about low-grade heat recovery using ORC. Comprehensive researches on appropriate working fluids for low-temperature applications have been investigated by many authors such as [4–6]. Others researchers have
http://dx.doi.org/10.1016/j.enconman.2015.09.047 0196-8904/Ó 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
2
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
Nomenclature y
condensing unit mass enthalpy (kJ/kg) mass enthalpy of flue gases (kJ/kgdg) mass enthalpy of dry flue gases (kJ/kgdg) mass enthalpy of water vapor (kJ/kgwv) heat exchanger irreversibility (kJ/kg) mass flow rate (kg/s) molar mass of dry gases (kg/kmol) molar mass of water vapor (kg/kmol) Organic Rankine Cycle total pressure (Pa) water vapor pressure (Pa) heat capacity (kJ/kgdg) mass entropy (kJ/kg K) sub-cooling (K) temperature (°C) humidity ratio (kgwv/kgdg) power (kJ/kgdg)
Greek symbols efficiency (%)
g
Subscripts cond condensation/condenser dg dry gases dp dew point evap evaporation fg flue gases i components is isentropic o ambient conditions r working fluid wb wet bulb wf working fluid wv water vapor
focused on the parametric optimization and performance analysis of the ORC like [7–10]. However, these studies deal with the indirect contact condensation recovery based on sensible heat extraction from flue gases with low moisture contents, or even though, with high moisture contents [11], but cooling the gases to a minimum safe temperature in order to prevent water vapor condensation and acid formation during gas flow. The originality of this study is to extend the ORC applications to low-grade gas heat sources with high moisture contents by pointing out the effect of water vapor condensation on cycle performance using the two condensing heat recovery processes (direct and indirect heat exchange). 2. Heat load availability In the wet flue gases, heat is available in both sensible and latent forms. The sensible heat is determined by the temperature of the flue gases and the combined heating capacities of its constituents. The latent heat is determined by the amount of water present in the flue gases in gas form [4]. The characteristics of the flue gases used in this study are tabulated in Table 1. The flue gases to be cooled are considered a mixture of water vapor (H2O), nitrogen (N2), oxygen (O2), and carbon dioxide (CO2). The CO2 and O2 compositions correspond to those at the ‘‘Raw Mill” exhaust in the cement plants. The water dew point is varied between 55 and 65 °C. The enthalpy of a mixture of gases is equal to the sum of the individual partial enthalpies of the components [12]. Therefore, the mass enthalpy of flue gases can be written as follows:
hg ðTÞ ¼ hdg ðTÞ þ w hwv ðTÞ ¼ yO2 hO2 ðTÞ þ yCO2 hCO2 ðTÞ þ yN2 hN2 ðTÞ þ w hwv ðTÞ
ð1Þ
yO2 þ yCO2 þ yN2 ¼ 1
ð2Þ
where subscript ‘‘g” denotes gas, ‘‘dg” denotes dry gases, ‘‘wv” denotes water vapor, ‘‘T” is the temperature of the mixture, ‘‘w” is the humidity ratio, ‘‘y” is the dry mass fraction, and ‘‘h” represents the mass enthalpy. The mass enthalpies of the different components (O2, CO2, N2 and H2O) are calculated at the corresponding partial pressures. The humidity ratio is calculated by:
w¼
Mwv Pw M dg Pt Pw
ð3Þ
where ‘‘Mwv” and ‘‘Mdg” are the molar masses of the water vapor and dry gases respectively, ‘‘Pw” is the partial pressure of water vapor, and ‘‘Pt” is the total pressure. The thermodynamic data of gases adopted in the present work are calculated using REFPROP 9.0 [13]. The ambient pressure and temperature at the specified dead reference state (Po and To) are considered to be atmospheric pressure and 20 °C. The characterization of the temperature and the available quantity of heat referred to the ambient temperature are presented in Fig. 1. The sudden breaks in slope indicate initial water dew points. Water vapor in the flue gases is in superheated state above the initial water dew points. The cooling first reaches the initial water dew points at which condensation begins. The partial pressure of water vapor decreases continually and accompanied by a reduction in water dew point. Sensible heat recovery occurs down to water dew points. Cooling below this level increases the energy recovery rate by recovery of the latent heat.
140
Table 1 Flue-gas characteristics. Composition Inlet temperature Water dew point temperatures Corresponding wet bulb temperatures
mass fraction
T dp = 45°C
120
Temperature (°C)
CU h hg hdg hwv HEX I m Mdg Mwv ORC Pt Pw Q s SC T w W
T dp = 55°C 100
T dp = 65°C
80 60 40 20
Molar basis Tinlet Tdp Twb
CO2 – 15.5%, O2 – 6.8% 120 °C 55/60/65 °C 59.9/63.7/67.8 °C
0 0
100
200
300
400
500
600
700
∆h (kJ.kgdg -1) Fig. 1. Heat load availability.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
3
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
Fig. 1 shows that a large amount of heat within the streams is in the form of latent heat and can only be recovered by condensing water vapor. The fraction of latent heat is of course dependent on the water dew point temperature. The analysis carried previously shows that the sensible energy could be recovered at relatively high temperatures compared to the latent energy. However, the amount of energy available per temperature difference is much larger when recovering latent heat below the initial water dew point temperature.
3. Condenser:
Q cond ¼ mr ðh4 h1 Þ
ð9Þ
Icond ¼ E4 E1
ð10Þ
4. Pump:
W pump ¼ mr ðh2 h1 Þ
ð11Þ
Ipump ¼ W pump þ E1 E2
ð12Þ
3. ORC with indirect contact heat recovery The lay-out diagram of a simple ORC is presented in Fig. 2a. The ORC system consists of an evaporator, a turbine, a condenser, and a pump. As shown in Fig. 2a, the pump supplies the working fluid to the evaporator, where the working fluid is heated and vaporized by the flue-gas heat stream (Point i). The generated high-pressure (Point 3) vapor is expanded by the turbine, which is directly coupled to an electrical generator producing power, and then, the low-pressure vapor is condensed in a condenser whereupon it is pumped back to the evaporator. The T–s diagram for the studied ORC system is shown in Fig. 2b. Since the focus of this study is the thermodynamic ORC optimization, detailed calculations of pressure losses and heat transfer in evaporator and condenser are not taken into account since they depend strongly on materials and configurations of the system components [5]. In order to calculate the irreversibility ‘‘I” for each component in the ORC, the ‘‘exergy” will be used and is defined as:
E ¼ mr ðh T o sÞ
The evaporator superheat is given by:
SHevap ¼ T 3 T 2b
By subtracting the power consumed by the pump from the power generated by the turbine, the net turbine power is:
W net ¼ W turbine W pump
ð14Þ
The heat recovery from the flue gases is given by:
Q fg ¼ mdg ðhi he Þ
ð15Þ
where hi and he are respectively the inlet and outlet enthalpies of dry flue gases expressed in kJ/kgdg. Since the heat released by the flue gases is recovered by the working fluid, the working fluid mass flow rate could be calculated by combining Eqs. (5) and (15) as shown:
mr ¼ Q fg =ðh3 h2 Þ
ð4Þ
where ‘‘To” refers to a reference temperature state (20 °C). The thermodynamic equations for the components in the ORC of Fig. 2 are:
ð13Þ
ð16Þ
The isentropic efficiency of the turbine can be expressed as:
gis;turbine ¼ ðh3 h4 Þ=ðh3 h4s Þ
ð17Þ
The isentropic efficiency of the pump can be expressed as:
gis;pump ¼ ðh2s h1 Þ=ðh2 h1 Þ
1. Evaporator:
Q evap ¼ mr ðh3 h2 Þ
ð5Þ
Ievap ¼ Ei þ E2 Ee E3
ð6Þ
2. Turbine:
W turbine ¼ mr ðh3 h4 Þ
ð7Þ
Iturbine ¼ E3 W turbine —E4
ð8Þ
ð18Þ
Based on the first law of thermodynamics, the ORC thermal efficiency is defined as the ratio of the net power output to the heat addition:
gcycle ¼ ðW turbine W pump Þ=Q evap
ð19Þ
The thermal efficiency cannot reflect the ability to convert energy from low-grade waste heat into usable power and might be misleading since the maximum power output is not achieved concurrently with the maximum thermal efficiency. Therefore,
e
i
3 Evaporator 2 Pump Turbine
Generator 4
1
Temperature [K]
i e
DT
2a
4
2 2s 1
3
Pevap 2b
4s 4a
Pcond
4b
Condenser
(a)
Entropy [kJ/(kg.K)] (b) Fig. 2. Simple ORC: (a) lay-out diagram and (b) (T–s) diagram.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
4
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
the exergy efficiency is considered in order to evaluate the performance for waste heat recovery. The exergy efficiency of the ORC system can be expressed as:
gexergy ¼ ðEi RI Ee Þ=Ei
ð20Þ
The operating conditions of the ORC are given in Table 2. The turbine efficiency is supposed to be the overall turbine efficiency that includes the isentropic and the mechanical efficiencies. Consideration of the pinch value (DT) between the flue gases and the working fluid in the evaporator is essential to define the maximal heat recovery from the energy source. As pointed out by Butcher and Reddy [12] and Srinivasan et al. [14], the system performance is sensitive to the pinch point and should be as low as possible to improve the overall energy performance. However, this will result in a larger heat-exchanger area to transfer the same amount of heat, hence an increase in size and cost. Following calculations have been done for a 3-K pinch at the evaporator. By referring to [15], this value is considered as aggressive boundary conditions. Therefore, a technical and economic optimization will be performed in developing sensitivity analysis on cost variations around the pinch while thermodynamic optimization of energy system requires only that the pinch is as small as possible. For the working fluid selection, HFC-245fa is chosen. As stated elsewhere [16,17], HFC-245fa is a promising candidate for ORC operating on low-temperature heat sources, and presents one of the best compromise between thermal efficiency, power output, and environmental performance. The properties of HFC-245fa are calculated using REFPROP 9.0 [13] developed by NIST. A computer program employing ‘‘Excel/VBA” was developed to simulate the thermodynamic performances of the ORC system. A parametric optimization of the evaporation temperature was performed to obtain the ORC maximal net power output. As reported by Liu et al. [6], superheating is detrimental to ORC efficiency for the working fluids with the non-negative slope of the saturation vapor curves (dT/ds > 0, dry fluid) such as HFC-245fa, and a nil evaporator superheat should produce the highest turbine power. Fig. 3 shows the evolution of the working fluid mass flow rate (for a 1 kg/s of dry flue gases) as a function of the evaporation temperature. As shown in Fig. 3, the mass flow rate decreases when increasing the evaporation temperature, this could be explained
Table 2 ORC input data. Parameters
Values
Condensing temperature Condenser sub-cooling Turbine & pump efficiencies
25 °C 2K 80%/85%
Tcond SCcond
gturbine/gpump
by referring to Eq. (16). In fact, on one hand, an increase in the evaporation temperature leads to an increase in the outlet flue gas temperature for a given pinch value in the evaporator, and then less heat is extracted from the flue gases, thus ‘‘Qfg” decreases. While, on the other hand, for a given condensation temperature, an increase in the evaporation temperature leads to an increase in the enthalpy difference in the evaporator at the working side ‘‘h3 h2”. Fig. 4 shows the evolution of the net power output for various water dew point temperatures under a wide range of evaporation temperature. The value ‘‘Tevap = Tdp DT” breaks each corresponding curve into two regions classified as follows: Sensible heat region (Tevap > Tdp DT): where mostly sensible heat is recovered. Latent heat region (Tevap < Tdp DT): where a great portion of latent heat is recovered. The value of the net turbine power is strongly affected by the water dew point temperature. It should be noted that the net power is mainly managed by the turbine power since the pump power is almost negligible. Fig. 4 shows that the net turbine power increases when increasing the water dew point temperature. In fact, an increase in the water dew point temperature leads to an increase in the quantity of water vapor condensation; and then more latent heat is released in the evaporator. Thus, by referring to Eq. (15) and as shown clearly in Fig. 3, the working fluid mass flow rate increases when increasing the water dew point temperature, resulting to an increase in the net turbine power. As shown in Fig. 4, the net power output increases first and then decreases. This can be explained by the fact that, by referring to Eq. (7), the turbine power results from the products of two terms evolving differently with the evaporation temperature: the working fluid mass flow rate which decreases when the evaporation temperature increases as already shown in Fig. 3, and the enthalpy difference ‘‘h3 h4” which increases when the evaporation temperature increases for a given condensation temperature. Finally, Fig. 4 shows that an optimum for the net turbine power can be identified in each region: sensible and latent heat optima. The corresponding evaporation temperature at the sensible heat optimum (73 °C) is almost independent on the water dew point temperature, whereas it depends strongly on it at the latent heat optimum. The value of the net turbine power at the latent heat optimum exceeds widely that at the sensible heat optimum at high water dew point temperatures (Tdp = 60–65 °C). Fig. 5 shows the evolution of evaporator capacity as a function of the evaporation temperature. A high evaporator capacity is expected below water dew point due to the large available quantity of latent heat. Fig. 5 shows that the evaporator capacity decreases when increasing the evaporation temperature. In fact, for a given pinch value in the evaporator, an increase in the
3
m wf (kg/s)
2 2 1 1
25
W net (kJ/kgdg )
T dp = 55°C T dp = 60°C T dp = 65°C
3
Tdp = 55°C Tdp = 60°C Tdp = 65°C
20 15 10 5
0 20
40
60
80
100
120
T evap (°C)
0 20
40
60
80
100
120
T evap (°C) Fig. 3. Evolution of the working fluid mass flow rate as a function of the evaporation temperature.
Fig. 4. Evolution of the net power as a function of the evaporation temperature.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
5
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
Q evap (kJ/kgdg )
600
process” (Fig. 8). In this case, HEX 1 will recover almost all sensible heat above the water dew point temperature so no condensation will occur on the heat-exchanger surface, and the condensing unit
Tdp = 55°C Tdp = 60°C Tdp = 65°C
500 400 300 200 100
e
0 20
40
60
80
100
120
Condensing unit
T evap (°C) Fig. 5. Evolution of the evaporator capacity as a function of the evaporation temperature. Water inlet to CU
i water pump
25
Efficiencies (%)
Ƞ cycle 20
Ƞ exergy
15 10
3 Evaporator 2
5 Pump
0 0
20
40
60
80
100
120
Turbine
Generator 4
T evap (°C) 1
Fig. 6. Evolution of cycle and exergy efficiencies as a function of the evaporation temperature.
Condenser
evaporator temperature leads to an increase in the outlet flue gas temperature and then less heat is extracted from the flue gases. Fig. 6 shows the evolutions of the cycle and exergy efficiencies as a function of the evaporation temperature. The cycle efficiency increases by increasing the evaporation temperature, which can be deduced by referring to Eq. (19) since the evaporator capacity decreases by increasing the evaporation temperature. Indeed, the exergy efficiency presents the same evolution shape as for the net power (refer to Fig. 4). The irreversibility in the evaporator could be reduced and optimum exergy efficiency could be achieved by varying the operating temperature.
Fig. 7. ORC with direct contact condensation.
e
Condensing unit
4. ORC with direct contact heat recovery Direct condensation heat recovery involves direct mixing between flue gases and cooling fluid. Since these systems do not involve a separating wall across which heat must be transferred, they avoid some of the challenges of large heat-transfer surfaces required for indirect contact units. An example system is shown in Fig. 7. As the flue gases enter the condensing unit (Point i), they are cooled by cold water introduced at the top of the unit. The heated water stream recovered at the bottom of the condensing unit is then sent to an evaporator to transfer heat to the organic working fluid of the power generation unit. Once the heat is extracted from the water, a treatment system will ensure that the water is cleaned before being re-used for another heat extraction cycle. Because the circulating water can only reach a temperature equivalent to adiabatic saturation temperature of flue gases, heat can only be transferred to the working fluid at a temperature lower than the entering wet bulb temperature. In order to reach a higher circulating water temperature, an indirect contact heat exchanger (HEX 1) could be added resulting into a ‘‘hybrid heat recovery
water pump Water inlet to CU HEX1 i
3 Evaporator
Turbine
Generator 4
Condenser
Fig. 8. Hybrid heat recovery process.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
6
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
will recover most of the latent heat below water dew point temperature. The ORC operating conditions are given in Table 3. The condenser, turbine and pump parameters are the same as those indicated in Table 2. In the direct contact condensation process, a minimum pinch of 0.5 K is maintained in the condensing unit between the entering wet bulb temperature and the leaving water temperature. Using the same procedure as for Eq. (16), the mass flow rates of the circulating water and the working fluid are calculated by applying the heat balance in the condensing unit and in the evaporator respectively. In Figs. 9 and 10, a parametric optimization of the water temperature at the evaporator inlet was performed with respect to the net turbine power and the evaporator capacity. The vertical lines represent the boundary limit between the direct contact process (zone I) and the hybrid heat-recovery process (zone II). The small circles denote the temperatures at which the water temperatures at the evaporator inlet are equal to the corresponding wet bulb temperatures for the different water dew point temperatures. The vertical lines cross the horizontal axis at evaporator inlet water temperatures equal to the wet bulb temperatures minus 0.5 K (minimum pinch in condensing unit). In zone I, the net turbine power increases slightly by increasing the inlet water temperature
Table 3 Additional ORC input data. Parameters
Values
DTHEX 1 DTmin CU – Pwater
Pinch in HEX 1 Minimum pinch in CU Tout flue gases from CU Tin water to CU Water pressure after water pump
25
T dp = 55°C T dp = 60°C T dp = 65°C
20
W net (kJ/kgdg )
3K 0.5 K 2K 0.2 MPa
10
Zone I
Zone II
0 40
50
60
70
80
90
100
110
120
T water at evaporator inlet (°C) Fig. 9. Net power as a function of the water temperature at the evaporator inlet.
500
T dp = 55°C T dp = 60°C T dp = 65°C
Q evap (kJ/kgdg )
400 300 200 100
Zone I
Zone II
0 40
50
60
70
80
T water at evaporator
90
100
110
Wnet
maximal
(kJ/(kgdg/s))
Indirect contact condensation Direct contact condensation Hybrid heat recovery mode
Tdp = 55 °C
Tdp = 60 °C
Tdp = 65 °C
9.21 (100%) 7.79 (84.6%) 8.73 (94.8%)
13.90 (100%) 11.66 (83.9%) 13.03 (93.8%)
21.12 (100%) 17.26 (81.7%) 19.25 (91.1%)
Table 5 Comparison of the outlet flue gas temperature between the analyzed heat recovery modes at the corresponding maximal net power. T outlet flue gases (°C)
Tdp = 55 °C
Tdp = 60 °C
Tdp = 65 °C
Indirect contact condensation Direct contact condensation Hybrid heat recovery mode
43.5 45.1 42.2
45.2 47.4 43.7
46.8 49.9 45.1
at the evaporator and reaches the maximal value at the boundary limit between the two zones, whereas, the evaporator capacity is almost constant. In zone II, the net turbine power decreases by increasing the inlet water temperature at the evaporator inlet, because less latent heat is extracted from the heat source when moving away from the wet bulb temperature. The maximal value of the net turbine power occurs for a water temperature at the evaporator inlet equal to the wet bulb temperature. A comparison of the maximal net turbine power between the three heat recovery modes is shown in Table 4. While the indirect contact condensation is the most favorable heat recovery scheme regarding the net power output, the hybrid heat-recovery mode leads to a potential increase in the net power output by about 10% compared to the direct contact condensation. Table 5 shows a comparison of the outlet flue gases temperature between the three heat recovery modes at the corresponding maximal net power. The hybrid heat recovery mode presents the lowest outlet flue gas temperature since the heat is recovered using both heat exchangers: an indirect contact heat exchanger (HEX1) and a direct contact heat exchanger (CU).
5. Conclusions
15
5
Table 4 Comparison between the analyzed heat recovery modes.
120
inlet (°C)
Fig. 10. Evaporator capacity as a function of the water temperature at the evaporator inlet.
This study details heat recovery from low-temperature waste heat (T < 120 °C) in industrial processes for generating electrical power using an ORC system. The flue gases are characterized by a water dew point temperature ranging from 55 to 65 °C. Based on the analysis presented in this investigation, the following conclusions are drawn: While the amount of latent heat in the gas stream may look attractive, its recovery is hindered by the low-temperature level at which the recovered latent heat is available. In the indirect contact condensation, depending on the evaporation temperature, two operating regions can be identified in the evaporator: ‘‘sensible heat region” where mostly sensible heat is recovered, and ‘‘latent heat region” where a great portion of latent heat is recovered. In the direct contact condensation, the inlet wet-bulb temperature determines the operating temperature levels throughout the system and limits the circulating water temperature. The maximal net turbine power of the direct contact system is reached for a final water temperature nearby the entering wet bulb temperature of the flue gases. The hybrid heat-recovery process leads to a potential increase in the net power output compared to the direct contact condenser, while avoiding surface corrosion problems faced in the indirect contact condenser.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047
S. Maalouf et al. / Energy Conversion and Management xxx (2015) xxx–xxx
As aforementioned in the introduction, an important factor influencing latent heat recovery is the corrosion problem accompanying the gas cooling, which requires high quality materials or frequently replacing system components especially for the indirect heat recovery mode. Thus, a multi-objective optimization taking into account the economic aspect is necessary after the thermodynamic optimization process. Acknowledgment The research leading to these results has received funding from the European Community’s Seventh Framework Program (FP7/2007-2013) under Grant agreement no. 256790 (‘LOVE’). References [1] Liu B, Rivière P, Coquelet C, Gicquel R, David F. Investigation of a two-stage Rankine cycle for electric power plants. Appl Energy 2012;100:285–94. [2] Tamamoto T, Furuhata T, Arai N, Mori K. Design and testing of the organic Rankine cycle. Energy 2001;26:239–51. [3] BCS, Incorporated. Report on waste heat recovery: technology and opportunities in U.S. industry. U.S. Department of Energy; March 2008. [4] Dai Y, Wang J, Gao L. Parametric optimization and comparative study of organic Rankine cycle (ORC) for low grade waste heat recovery. Energy Convers Manage 2009;50:576–82. [5] Hung TC, Shai TY, Wang SK. A review of organic Rankine cycles (OECs) for the recovery of low-grade waste heat. Energy 1997;22(7):661–7. [6] Liu BT, Chien KH, Wang CC. Effect of working fluids on organic Rankine cycle for waste heat recovery. Energy 2004;29:1207–17.
7
[7] Roy JP, Mishra MK, Misra A. Performance analysis of an organic Rankine cycle with superheating under different heat source temperature conditions. Appl Energy 2011;88:2995–3004. [8] Roy JP, Mishra MK, Misra A. Parametric optimization and performance analysis of a regenerative organic Rankine cycle using low-grade waste heat for power generation. Int J Green Energy 2011;8:173–96. [9] Roy JP, Mishra MK, Misra A. Parametric optimization and performance analysis of a regenerative organic Rankine cycle using R-123 for waste heat recovery. Energy 2012;39:227–35. [10] Wei D, Lu X, Lu Z, Gu J. Performance analysis and optimization of organic Rankine cycle (ORC) for waste heat recovery. Energy Convers Manage 2007;48:1113–9. [11] Roy JP, Mishra MK, Misra A. Parametric optimization and performance analysis of a waste heat recovery system using organic Rankine cycle. Energy 2010;35:5049–62. [12] Butcher C, Reddy B. Second law analysis of a waste heat recovery based power generation system. Int J Heat Mass Transf 2009;50:2355–63. [13] Lemmon EW, Huber ML, McLinden MO. Reference fluid thermodynamic and transport properties. NIST standard reference database 23, version 9.0. [14] Srinivasan K, Mago P, Krishnan S. Analysis of exhaust waste heat recovery from a dual fuel low temperature combustion engine using an organic Rankine cycle. Energy 2010;35(6):2387–99. [15] Becquin G, Freund F. Comparative performance of advanced power cycles for low-temperature heat sources. In: ECOS 2012: Proceedings of the 25th international conference on efficiency, cost, optimization, simulation, and environmental impact of energy systems, June 26–29, Perugia, Italy; 2012. [16] Kosmadakis G, Manolakos D, Kyritsis S, Papadakis G. Comparative thermodynamic study of refrigerants to select the best for use in hightemperature stage of a two-stage organic Rankine cycle for RO desalination. Desalination 2009;243:74–94. [17] Zyhowski GJ, Brown AP, Achaichia A. HFC-245fa working fluid in organic Rankine cycle – a safe and economic way to generate electricity from waste heat. In: ECOS 2010: Proceedings of the 23th international conference on efficiency, cost, optimization, simulation, and environmental impact of energy systems, June 14–17, Lausanne, Switzerland; 2010.
Please cite this article in press as: Maalouf S et al. Investigation of direct contact condensation for wet flue-gas waste heat recovery using Organic Rankine Cycle. Energy Convers Manage (2015), http://dx.doi.org/10.1016/j.enconman.2015.09.047