Exergy Analysis and Heat Integration of a Coal-Based Oxy

---

Article pubs.acs.org/EF

Exergy Analysis and Heat Integration of a Coal-Based Oxycombustion Power Plant Chao Fu* and Truls Gundersen Department of Energy and Process Engineering, Norwegian University of Science and Technology, Kolbjorn Hejes vei 1A, NO-7491 Trondheim, Norway S Supporting Information *

ABSTRACT: Oxy-combustion is a competitive technology to enable the capture of CO2 from coal-based power plants. The main challenge to implement this technology is the large energy penalty and investment cost related to CO2 capture. A comprehensive exergy analysis has been performed for an entire coal-based oxy-combustion power plant. The exergy flows and irreversibilities are presented. The thermal efficiency penalty related to CO2 capture is about 9.4 percentage points (on the basis of the higher heating value) and is mainly caused by the air separation unit and the CO2 compression and purification unit. The theoretical minimum is 3.4 percentage points when the two units are assumed to be reversible, where the air separation unit contributes 1.4 percentage points and the CO2 compression and purification unit is responsible for 2.0 percentage points. The compression processes are causing the largest exergy losses related to CO2 capture. The case studies show that the integration of the compression heat with the steam cycle can increase the thermal efficiency by up to 0.72 percentage points.

1. INTRODUCTION Carbon capture and storage (CCS) is a promising way to mitigate CO2 buildup in the atmosphere.1 The key idea of CCS is to capture CO2 from large point sources, such as power plants and some industrial processes, and inject it into geological formations, such as depleted oil and gas fields or saline formations.2 There are three main approaches to capturing CO2 from power plants: post-combustion, pre-combustion and oxycombustion. Figure 1 shows their main differences.3 A combustion reaction is composed primarily of four elements: the fuel, the oxidant, the flue gas, and the reaction heat (Q). The normal case in existing power plants is that coal, natural gas, or biomass is combusted with air as the oxidant. The flue gas then mainly contains CO2, N2, and H2O. CO2 can be separated directly from the flue gas by absorption, adsorption, membrane, or anti-sublimation. This approach is referred to as “post-combustion”. The main separation process in this case is the separation between CO2 and N2. Another way to capture CO2 is to convert the fuel into syngas (H2 and CO) by gasification or reforming. CO is further converted into CO2 by a water-gas shift process. CO2 can then be separated prior to the combustion of H2. This is called “pre-combustion”, and the main separation process is the separation between CO2 and H2. The third method is called “oxy-combustion”. The key idea is to use O2 or metal oxides (MeO) instead of air to react with the fuel. The advantage is that the flue gas will be composed mainly of CO2 and H2O. CO2 can be captured simply by condensing H2O and then purified by chilling. The separation between O2 and N2 is considered as the main separation process in this case. Thus, to capture CO2, measures are taken to deal with the flue gas, the fuel, or the oxidant in the three capture options. Oxy-combustion is a competitive option for CO2 capture, especially for coal-fired power plants, because the reduction in power efficiency and the increment of investments related to © 2013 American Chemical Society

CO2 capture are less than (or at least comparable to) postcombustion and pre-combustion according to the available literature.4,5 In addition, the emissions of SOx can be reduced because of the condensation of sulfates in the ducts and the absorption of sulfur in the ash.6 The emissions of NOx can also be significantly reduced for two reasons: (1) the thermal formation of NOx is suppressed because of the lack of N2 in the burner,7 and (2) NOx in the recycled flue gas (RFG) is converted into N2 because of the reburning effect.8 SOx and NOx can also be converted into acids, which are then condensed from the CO2 stream during the compression of CO2 when oxy-combustion is applied; thus, the flue gas desulfurization (FGD) unit and the selective catalytic reduction (SCR) system may not be required in some plants.4 Oxycombustion may also have an advantage when it comes to other environmental impacts, such as eutrophication, photochemical oxidation, and toxicity potentials.9 The concept of using the oxy-combustion technology in fossil-fuel-based power plants was proposed around 1980, and the primary purpose was to produce concentrated CO2 for enhanced oil recovery.10−13 Because of increasing concern about the greenhouse problem, the oxy-combustion technology was further developed in the late 1980s and 1990s10,12,14−16 and has been established as one of the three most common technology pathways to capture CO2. Another incentive for developing the oxy-combustion technology is to reduce NOx emissions from coal-based power plants.8,17,18 Oxy-combustion processes have been widely studied recently. The technology has reached maturity as an option for capturing CO2 from coalbased power plants.19 The main challenge is to reduce the energy penalty and investment cost related to CO2 capture. It is Received: May 9, 2013 Revised: September 25, 2013 Published: September 26, 2013 7138

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149

Energy & Fuels

Article

Figure 1. Classification of carbon capture processes.

Figure 2. Flowsheet of an oxy-combustion pulverized coal power plant: AU1, main air compressor; AU2, pre-purification unit; AU3, main heat exchanger; AU4, air distillation system; AU5, tail N2 turbine; AU6, waste N2; CU1, boiler area (including the combustion process, the steam generation and reheat process, and the air preheating process); CU2, HP steam turbines; CU3, remaining part of the steam system (including the IP and LP steam turbines, the condenser, the feedwater heaters, the deaerator, the boiler feedwater pump, the condensate pump, and the steam sealing system); CU4, electricity generator; CU5, flue gas processing system (including the ESP, the FGD unit, the fans, and the recycling of flue gas); RU1, first stage of the CO2 compression process; RU2, CO2 purification process; RU3, second stage of the CO2 compression process; RU4, tail gas turbine; and RU5, vented inerts.

commonly recognized that the energy penalty is mainly caused by two units: the air separation unit (ASU) and the CO2 compression and purification unit (CPU). Some studies have investigated the thermodynamic performance of such complex processes.20−22 However, very few detailed thermodynamic

analyses on the entire plant (including the ASU, the CPU, the combustion process, and the steam cycle) and investigations on heat integration opportunities can be found in the literature. The energy penalty related to CO2 capture via oxy-combustion is expected to be significantly reduced by process integration 7139

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149

Energy & Fuels

Article

composition of the flue gas is CO2, 70.1%; H2O, 15.0%; N2, 8.5%; O2, 3.5%; and Ar, 2.9%. The recycled stream (C3-1) must be reheated to increase the temperature by about 9 °C for this particular case to prevent entrained water droplets from passing through the air fans (CP3). The recycled flue gas (C3-3) is then mixed with O2 (C2-1) and preheated before entering the combustor. The rest of the flue gas first enters another direct contact aftercooler (R-DCA) to condense the water vapor and is then compressed to 32 bar by a five-stage compressor with water intercooling (R-P1). It is then dried in a molecular sieve twin-bed drier (R-S1) to avoid ice formation in the sub-ambient heat exchangers. The dried flue gas is cooled to −25 °C in a multi-stream heat exchanger (R-H1) and then separated into two streams in a flash drum (R-S2). The liquid stream (R2-1) is expanded though a Joule−Thomson valve and heated in RH1. The vapor stream (R3-1) is further cooled to −54 °C in another multi-stream heat exchanger (R-H2) and separated in the second flash drum (R-S3). The vapor stream (R5-1), mainly containing inert gases, is heated in the two multi-stream heat exchangers (R-H1 and R-H2). These inert gases are further heated against hot flue gas in the boiler area and expanded in a gas turbine (R-P5) to recover power. The liquid stream (R4-1) from the second flash drum (R-S3) provides necessary refrigeration by expansion and exchanging heat with other streams. Stream R4-5 is compressed to the same pressure as stream R2-3 and cooled by cooling water. The two streams are mixed and compressed to 78 bar by a two-stage compressor with water intercooling (R-P3). Compressed CO2 (R6-2) is further cooled to 25 °C by seawater or chilled cooling water. At this temperature and pressure, CO2 is in the dense phase and pumped to 150 bar for transportation and saline formation storage. The operating parameters (temperature and pressure) in the CPU are optimized on the basis of on a techno-economic analysis.26 The purity of the captured CO2 is 96.3 mol %. The recovery rate of CO2 (kilograms of CO2 in the product per kilogram of CO2 in the flue gas) is close to 95%. A corresponding air-fired power plant without CO2 capture has also been studied as the base case to compare the performance with the oxy-combustion case. The flue gas is vented directly into the atmosphere without capturing CO2. However, a SCR system is installed prior to the air heater to reduce the NOx emissions. In the oxy-combustion case, the NOx emission limit can be met by modifying the combustion technology.23 For the base case, the ASU, the CPU, the recycled flue gas system (streams C3-1, N4-1, and N4-2), and the steam for the ASU and the CPU (NASUa, NASUb, NCPUa, and NCPUb) in Figure 2 are not required. The ambient air is represented by the two streams C2-2 and C3-2.

(including the integration of mass and heat). The scope of this paper is to identify opportunities for improving the plant performance of oxy-combustion processes by exergy analysis and heat integration. The main body of the present paper consists of two parts. In the first part, a comprehensive exergy analysis is presented for a coal-based oxy-combustion process with CO2 capture. The coal to power process is based on a National Energy Technology Laboratory (NETL) report,23 while the ASU and the CPU are based on other common cases from the literature.24−26 The conventional cryogenic double-column air separation process is selected to produce O2 because the only commercially available air separation technologies for high volume oxygen production are based on cryogenic distillation. Detailed exergy flow diagrams for the ASU, the CPU, and the power cycle are presented in the paper. The results of the exergy analysis indicate that the compression processes are responsible for the largest exergy losses related to CO2 capture. In the second part of the paper, it is described how waste heat from the compression processes can be integrated with the steam cycle by preheating the boiler feedwater. Such heat integration has been investigated by several case studies. This part is an extension of the work by Fu and Gundersen.27

2. PROCESS DESCRIPTION Figure 2 presents the flowsheet of a 570 MW (net) supercritical pulverized coal power plant with CO2 capture. The high pressure (HP) steam is heated to 242 bar/600 °C with single reheat of the intermediate pressure (IP) steam to 45 bar/620 °C. The condenser pressure is 0.069 bar. The condenser, the feedwater heaters, the deaerator, the boiler feedwater pump, and the condensate pump are simplified as the condensing and preheating system in block CU3 in Figure 2. The traditional cryogenic double-column air separation process is applied to produce O2 with a mole fraction around 95%. Ambient air (A0) is compressed to 5.6 bar by a three-stage compressor (A-P1). The compression heat is removed by water-cooled intercoolers. The compressed air is cooled in a direct contact aftercooler (A-DCA) and enters a front-end temperature swing adsorption (TSA) prepurification unit (A-PPU) to remove H2O, CO2, and other impurities. The dry compressed air (A1-2) is cooled to near dew point temperature in the main heat exchanger (A-H1). The air is separated into O2 and N2 in the double-distillation columns. The reboiler in the lower pressure column (A-LP) is integrated with the condenser in the higher pressure column (A-HP). The temperature difference of the condenser/reboiler is maintained at around 1.5 °C. The N2 stream (A4-1) from the top of the higher pressure column mixes with another N2 stream (A7-3), after heat recovery and expansion in a tail gas turbine (A-P2). The mixed N2 stream first enters the A-PPU to cool the process air and is then vented to the atmosphere. The O2 product with molar composition (O2, 95%; Ar, 3.2%; and N2, 1.8%) is split into two streams (C2-1 and C2-2). A minor part of O2 (C2-2) enters the FGD unit and is used as the oxidant, while the major part (C2-1) reacts with coal in the combustor. To ensure complete combustion, excess O2 (the relative difference between the actual O2 supply and the theoretical O2 consumption in the case of stoichiometric combustion) at the combustor inlet is around 10%. The O2 content at the combustor outlet is around 3.0 mol %.23 The combustion process takes place at 1.1 bar. The adiabatic combustion temperature is 2080 °C (the heat loss is neglected, and the coal is assumed to combust completely). The combustion heat is converted to power by the steam cycle. The particulate matter in the flue gas is removed in an electrostatic precipitator (ESP). The flue gas is induced to a wet FGD system (limestone slurry is used) by the induced fan (C-P1). The SO2 removal rate is 98%. To avoid too high of a temperature in the combustor, a major part of the flue gas (72%) is recycled to the combustor after desulfurization. The molar

3. EXERGY ANALYSIS The entire process is simulated with Aspen Plus, version 7.3. The Peng−Robinson (PR) property method is used for the ASU, the combustion process, and the CPU. The NBS/NRC steam tables are used in the steam cycle. Detailed computational specifications are listed in Table 1. 3.1. Methodology. Exergy analysis based on the first and second laws of thermodynamics can be used to indicate the thermodynamic efficiency of a process, including all quality losses of materials and energy. This work uses the reference environment model defined by Szargut et al.28 The reference state (marked as “0”) is T0 = 25 °C and p0 = 1.013 25 bar (i.e., 1 atm). The standard chemical exergy of pure substances is also referenced by Szargut et al.28 A detailed route to perform exergy analysis has been described by Hinderink et al.29 The total exergy of a stream, Ė tot (MW), at process conditions (T and p) is given as ̇ = Ech ̇ + Eph ̇ + Emix ̇ Etot

(1)

where Ė ch, Ė ph, and Ė mix are the chemical exergy, physical exergy, and mixing exergy, respectively. 7140

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149

Energy & Fuels

Article

Table 1. Computational Specifications for the Entire Power Plant

Table 2. Coal Characteristics

turbomachinery units isentropic efficiency of HP/IP/LP steam turbines steam turbine mechanical efficiency generator mechanical efficiency isentropic/mechanical efficiency of compressors isentropic/mechanical efficiency of fans isentropic/mechanical efficiency of the tail gas turbine pump efficiency (including motor driver) gas outlet temperature of compression intercoolers (°C) ASU and CPU minimum temperature difference in sub-ambient heat exchangers (°C) temperature difference of the condenser/reboiler exchanger (°C) pressure drop in the pre-purification unit (bar) pressure drop in sub-ambient heat exchangers (%) pressure drop in the HP column (bar) pressure drop in the LP column (bar) inlet/outlet temperatures of cooling water (°C) inlet/outlet temperatures of seawater (°C) minimum temperature difference in cooling water heat exchangers (°C) cooling water pressure (bar) steam cycle pressure loss in feedwater heaters (bar) condenser pressure (bar)

proximate analysis (%) moisture volatile matter ash fixed carbon ultimate analysis (%) carbon (C) hydrogen (H) nitrogen (N) sulfur (S) chlorine (Cl) ash moisture (H2O) oxygen (O) high heating value (HHV) (MJ/kg) low heating value (LHV) (MJ/kg) chemical exergy (MJ/kg)

0.9/0.9/0.88 0.996 0.985 0.82/0.97 0.88/0.98 0.9/0.999 0.736 35

1.5 1.5 0.15 ∼3 0.1 0.1 25/35 15/25 10 2

̇ = [Fḣ − Emix

0.34 0.069

∑ (Fei ̇ ch,0i)

The way to calculate the chemical exergy of coal, kg), is described by Szargut et al.28 0 φ = ech,coal /(LHV)0

e0ch,coal

(MJ/

̇ = Wmin

i

−1

product

̇ Etot

−

̇ feed ∑ Etot feed

(8)

Here, (MW) is the exergy of a product stream and Ė feed tot (MW) is the exergy of a feed stream. To perform an exergy analysis of the power plant, the following computational assumptions are made: (1) The exergy of ash is assumed to be zero. (2) The exergy of the limestone slurry is calculated as the sum of the exergy of its two main constituents: water (70 wt %) and calcium carbonate (CaCO3). The exergy of gypsum is calculated similarly as the sum of the exergy of water (10 wt %) and solid gypsum (CaSO4·2H2O). The chemical exergy of CaCO3 and CaSO4·2H2O is obtained from Szargut et al.28 (3) The reaction heat in the FGD unit is ignored. The chemical exergy of the impurities in wastewater in the FGD unit is also ignored. (4) The physical exergy of solids (coal, limestone, and gypsum) is ignored. (5) The ambient air is assumed to be composed of N2, O2, Ar, CO2, and H2O. Other noble gases, organic substances, and others are ignored.

∑ (F0,̇ ih0,i)] (5)

i

(6)

−1

Ė product tot

where c, h, o, and n are the mass fractions of carbon, hydrogen, oxygen, and nitrogen in the ultimate analysis of coal (dry basis), respectively. Table 2 lists the coal characteristics.23 The physical exergy of a stream Ė ph (MW) is equal to the amount of work arising when changing a stream, consisting of the unmixed components, reversibly from process conditions (T and p) to reference conditions (T0 and p0) and is calculated by

∑ (F0,̇ is0,i)]

∑ product

(4)

i

71.73 5.06 1.41 2.82 0.33 10.91 0.00 7.74 30.53 29.45

in out (7) I ̇ = Ė − Ė where Ė in (MW) and Ė out (MW) are the exergy (including exergy of material streams and energy streams, such as heat and work) entering and exiting the unit, respectively. On the basis of eq 7, if all of the operations in a unit are reversible, the irreversibility is equal to zero. For a separation process, the theoretical minimum work, Ẇ min (MW), is then equal to the difference in exergy between the products and the feeds (assuming that all of the operations are reversible) and can be calculated by30

(3)

̇ − T0[∑ (Fs i i) −

63.75 4.50 1.25 2.51 0.29 9.70 11.12 6.88 27.14 26.17 31.95

Here, h (MJ/mol) and s (MJ mol °C ) are the molar enthalpy and entropy of the mixture, respectively. The exergy loss (irreversibility) of a unit operation, I ̇ (MW), can be calculated by

φ = 1.0437 + 0.1896(h/c) + 0.2499(o/c) + 0.0428(n/c)

i

0.00 39.37 10.91 49.72

i −1

Here, (LHV)0 (MJ/kg) is the lower heating value of coal at the reference conditions (T0 and p0), and φ is the ratio of the chemical exergy to the lower heating value, calculated by

̇ = [∑ (Fh ̇ Eph i i) −

11.12 34.99 9.70 44.19

̇ ̇ ̇ ∑ (Fh i i)] − T0[Fs − ∑ (Fs i i)] i

(2)

i

dry

The mixing exergy, Ė mix (MW), is equal to the amount of work arising when a stream, consisting of the pure components, is mixed at process conditions (T and p) and is calculated as follows:

The chemical exergy of a material stream, Ė ch (MW), is calculated by multiplying the standard chemical exergy, e0ch,i (MJ/mol), for component i with the molar flow, Ḟi (mol/s), of component i and then making a summation over all components that are present in the stream. ̇ = Ech

as received

−1

Here, hi (MJ/mol) and si (MJ mol °C ) are the molar enthalpy and entropy for component i, respectively. 7141

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149

Energy & Fuels

Article

Figure 3. Exergy flows (MW) in the ASU.

Figure 4. Exergy flows (MW) in the combustion process and the steam cycle.

(6) The exergy of impurities (H2O and CO2) in the ASU is ignored. 3.2. Results of Exergy Analysis and Discussion. Figure 3 presents the exergy flows (illustrated by the dashed areas) in the ASU. The labels of the flows are referred to the streams in Figure 2. The exergy loss for a sub-process (illustrated by the box with a dash-dot line) is represented by the dotted area. The power input is 128 MW, and the power recovered from the tail N2 turbine is 10.2 MW. The theoretical minimum work for separation to obtain 95 mol % O2 is calculated to be 26.9 MW by eq 8 at the reference state. The specific power consumption for O2 production in the actual case with the same purity is 0.23 kWh/kg of O2, while the theoretical minimum work for separation is 0.053 kWh/kg of O2. The exergy losses are divided into six parts in Figure 3. The air compressor (intercoolers included) and the distillation process are responsible for the two largest exergy losses: 34.0 and 29.6 MW, respectively. The exergy losses in the compression process can be reduced by increasing the compressor efficiency. The waste heat from the compression process (compression heat) could be used in other units of the plant, as illustrated in

section 4. The irreversibilities in the distillation process are mainly caused by irreversible heat and mass transfer between the descending liquid and the ascending vapor in the distillation columns. Such irreversibilities can be mitigated by a more efficient design of the distillation system, e.g., using the dual condenser/reboiler technology. When dual reboilers are used, the irreversibilities in both the distillation column and the compressor can be reduced and, thus, the total energy consumption is reduced by around 10%.31 The exergy loss in the main heat exchanger is small (11.5 MW) because of low temperature differences. However, the irreversibilities can be further reduced by transferring heat between the hot streams and cold streams with smaller temperature differences (1 °C for example). This can be achieved without area penalties if the heat transfer coefficient of the heat exchanger is improved or by accepting an increase in the heat transfer area to reduce irreversibilities. It should be noticed that N2 vented to the atmosphere causes a certain exergy loss (12.6 MW). The potential for using waste N2 may be limited because the amount of N2 is very large. 7142

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149

Energy & Fuels

Article

Figure 5. Exergy flows (MW) in the CPU.

The exergy flows in the combustion process and the steam cycle are shown in Figure 4. The exergy input by the coal is 2211 MW, while the gross power output from the steam turbines is 792 MW. The boiler area has the largest exergy loss, 1158 MW. Such exergy losses are mainly caused by the combustion of coal and the heat transfer between the flue gas and the water/steam. The irreversibilities related to combustion are normally unavoidable but can be reduced if other oxidation processes, such as gasification or chemical looping combustion, are implemented.32 Minor improvements can be realized by reducing the O2 excess amount. The exergy losses caused by heat transfer are mainly dependent upon the operating parameters (temperature and pressure) of the steam. These parameters are fixed by material constraints and economic factors. Higher temperatures and pressures reduce the temperature differences of the heat transfer and, thus, the exergy losses. If other working fluids or combined cycles are applied, the exergy losses may also be reduced. The HP steam turbines and the rest of the steam cycle are responsible for 13 and 112 MW, respectively. Other exergy losses can also be significantly compared to the exergy losses in the ASU and the CPU. However, such exergy losses are less related to CO2 capture, and the potential for improvement may be small because a lot of efforts have been devoted to such studies in the past, e.g., using regenerative feedwater preheating, improving turbine efficiencies, and reducing condenser losses. Figure 5 shows the exergy flows in the CPU. The total power input is 73 MW with contribution from the two-stage compression processes (54 and 19 MW). The power recovered from the tail gas turbine is 9 MW. Thus, the net power consumption in this unit is 64 MW, and the specific work is 0.114 kWh/kg of CO2 captured. The theoretical minimum work in this unit is calculated to be 37 MW, and the minimum specific work is 0.067 kWh/kg of CO2 captured. Again, the main exergy losses are caused by the CO2 compressors (intercoolers included). The first stage and the second stage of the CO2 compression process are responsible for 17.3 and 7 MW, respectively. The development of CO2 compressors and the use of the waste heat are important ways to reduce the losses. The losses in other units are relatively small. The losses in the sub-ambient heat exchangers can be reduced by optimizing the temperature profiles for the hot streams and the cold streams. CO2 in the inert gases (N2, O2, Ar, and CO2) may be further recovered by adsorption or membrane separation; thus, the losses in chemical exergy can be reduced. Figure 6 illustrates the distribution of exergy losses for the entire oxy-combustion plant and subunits. The combustor and the steam cycle cause the major exergy loss (91.3%) for the entire plant. However, this loss is not directly caused by CO2

Figure 6. Distribution of exergy losses: (a) entire oxy-combustion plant, (b) ASU, (c) CPU, and (d) combustor and steam cycle.

capture. The boiler area is mainly responsible for the loss (85.4%). The exergy losses in the ASU and the CPU are small compared to the losses in the combustor and the steam cycle; however, the two units are mainly responsible for the energy penalty related to CO2 capture. The compression processes cause major exergy losses in the two units. The plant performance for the oxy-combustion case and the base case are compared in Table 3. Other auxiliaries in the table Table 3. Plant Performance Comparison basic gross power generation (MW) air fans or the RFG fan (MW) induced fan (MW) condensate pumps (MW) ASU main air compressor (MW) N2 turbine (MW) CPU CO2 compression (MW) tail gas turbine (MW) other auxiliaries (MW) total auxiliaries (MW) net power (MW) net plant efficiency (%) (HHV) coal feed (kg/s) HHV (MJ/kg) thermal input of the coal feed (MW) 7143

585 4.2 8.0 0.8

oxy-combustion 792 4.1 8.4 1.1 128 −10.2

18.7 31.7 553 39.8 51.32 27.14 1392

73 −9 25.3 220.7 571 30.4 69.23 27.14 1878

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149

Energy & Fuels

Article

Figure 7. Effect of compressor efficiency on plant performance.

Figure 8. Steam cycle.

include the coal handling and conveying, the limestone handling and reagent preparation, the pulverizers, the ash handling, the precipitators, the FGD pumps and agitators, the steam turbine auxiliaries, the circulating water pumps, the cooling tower fans, the transformer loss, and the miscellaneous balance, including the plant control systems and lighting. The load for these auxiliaries is estimated to be 3.2% of the gross power production based on the NETL report.23 From Table 3, the thermal efficiency (the net power output divided by the thermal input of the coal feed, on the basis of HHV) is 39.8% for the base case and 30.4% for the oxy-combustion case. The efficiency penalty related to CO2 capture is thus 9.4 percentage points. The efficiency penalty is mainly caused by two units: the ASU and the CPU. If the shaftwork consumed in the ASU and the CPU (117.8 and 64 MW, respectively) is added to the net power production for the oxy-combustion case (the energy consumption for the pre-purification parts, front-end cooling pumps, motors, and transformer losses, and any other losses in the two units is small and, thus, not included), the thermal efficiency is calculated to be 40.1% (close to the value for the base case, 39.8%). The efficiency penalty is thus 9.7 percentage points, where the ASU contributes 6.3 percentage points and the CPU contributes 3.4 percentage points. Notice that the

base case with 39.8% efficiency is with air as the oxidant, while 40.1% efficiency refers to the use of O2 as the oxidant and then includes the power consumed in the ASU and the CPU. Assuming that the theoretical minimum work (calculated by eq 8) consumed in the ASU and the CPU can be achieved (in this case, an ideal, unspecified separation process is assumed, and thus, no steam is extracted for the pre-purification parts of the two units), the theoretical efficiency penalty related to CO2 capture is calculated to be 3.4 percentage points, where the ideal ASU (the O2 supply pressure is atmospheric) and the ideal CPU contribute 1.4 and 2.0 percentage points, respectively. According to the results from the exergy analysis of the ASU and the CPU, the compressors (intercoolers included) are causing the largest exergy losses. Figure 7 shows the effect of compressor efficiency on the plant performance. The net thermal efficiency (HHV) can increase from 29.3 to 31.4% with the assumption that the isentropic efficiencies of all compression stages in the ASU and the CPU increase from 0.74 to 0.9. However, such improvements are usually beyond the control of plant designers. It is expected that exergy losses related to CO2 capture processes can be reduced by optimizing the CO2 recovery rate and process integration between the ASU and other parts of the plant, such as the CPU. 7144

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149

Energy & Fuels

Article

Figure 9. Heat loads in the FWHs: (a) FWH1−FWH4 and (b) FWH6−FWH8.

The boiler feedwater is heated to 39.1 °C in the gland seal condenser and then heated to 147.1 °C in four closed-type feedwater heaters (FWHs) in the low-pressure section (FWH1−FWH4). The dissolved gases in the boiler feedwater are removed in the deaerator. The boiler feedwater is pumped to around 290 bar (N9) and then heated to 291 °C in three FWHs in the high-pressure section (FWH6−FWH8). Steam is extracted at several levels for preheating the boiler feedwater. The steam condensate normally flows into a lower pressure FWH. Thus, for the lower pressure FWHs, the major portion of heat comes from the corresponding extracted steam, while another portion is contributed by the condensates from the higher pressure FWHs. When the compression heat is used for partially (or fully) preheating the boiler feedwater, the steam extraction can be reduced. It is then essential to know the heat contribution from each extracted steam level in the FWHs. The heat loads for each FWH are decomposed on the basis of the heat contributions from the extracted steam and are illustrated in Figure 9. In each FWH, the cold stream (represented by the bottom line) is the BFW. The hot streams (represented by the lines above the bottom line) are decomposed heat loads of the extracted steam at different

4. HEAT INTEGRATION The results from exergy analysis indicate that compression processes in the ASU and the CPU are causing considerable energy penalty related to CO2 capture. The compression work can be reduced by increasing the compressor efficiency and/or reducing the operating temperature. Water coolers are normally used to remove the compression heat, thus reducing the operating temperature. The compression heat could be used in the steam cycle by preheating the boiler feedwater (BFW); thus, less steam is extracted for the preheating process. The temperature of the compression heat could be lifted using (fully or partially) adiabatic compression; thus, the BFW can be preheated to a higher temperature, and the extraction of steam at higher pressures can be reduced. However, more compression work is consumed in the case of adiabatic compression. Thus, the compression schemes should be optimized for the purpose of heat integration. Such optimization is rather complicated; instead, this paper investigates the heat integration potential by several case studies. 4.1. Decomposing the Steam Cycle. A more detailed flowsheet of the supercritical steam cycle is shown in Figure 8. 7145

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149

Energy & Fuels

Article

→ 49.1 °C), and Q3 (101.1 → 49.1 °C). The heat capacity is assumed constant and the same for the three heat streams. The mean heat capacity flow rate is calculated to be 0.64 MW/°C. On the basis of the supply temperatures, the compression heat can be used to reduce the steam consumption in FWH1− FWH3. Assuming that no steam is extracted for FWH1− FWH3; i.e., the heat is supplied by the compression heat; the condensed water is from FWH4; and the steam is from the SSR (N28; the sensible heat is very small and, thus, not included in the GCC), the GCC is shown in Figure 10.

pressure levels. They are actually mixed in each FWH. External drain coolers (flash type)33 are used. The condensate from each FWH is flashed to a lower pressure FWH (the deaerator is a special FWH) and condensed again at the lower saturation temperature. Thus, the temperature of each extracted flow of steam does not decrease continuously. When the compression heat is integrated with the steam cycle, the heat is transferred from the compressed gases to the boiler feedwater. The following assumptions are made: (1) The minimum temperature difference for the gas/water heat exchangers is specified to be 10 °C, and for the steam/water heat exchangers, it is assumed to be the same as in the oxycombustion case without integration. Thus, the temperature of the boiler feedwater to FWH1 (N3) is maintained (39.1 °C), and the compressed gases can be cooled to around 49.1 °C by the boiler feedwater. (2) The pressure drop on the gas side of the gas/water heat exchangers is neglected. (3) The pressure levels of the extracted steam are maintained. (4) The compressor efficiency is assumed to be constant and the same for all of the compression stages. (5) The steam seal regulator (SSR) and the deaerator are assumed not to be influenced by the heat integration. 4.2. Integration Steps. The integration study is performed by the following steps: (1) Determine the supply temperature of the compression heat and, thus, the FWHs by which the compression heat can be integrated. (2) Draw the grand composite curve (GCC)34 for the FWHs that can use the compression heat, when assuming that no steam is extracted. The GCC can be used to determine the heat demand at various temperature levels. (3) Use the GCC to determine the demand of the extracted steam at different pressure levels (the pressure levels are maintained as in the oxy-combustion case without integration). The condensed water in each FWH flows into the next lower pressure FWH (the deaerator and the gland seal condenser can be regarded as special FWHs) and, thus, supplies a small portion of heat for the lower pressure FWH. This portion of heat is neglected when determining the demand of extracted steam and primarily used when designing the FWHs. Thus, the compressed gases will be cooled to temperatures higher than 49.1 °C, resulting in larger temperature driving forces in the gas/water heat exchangers. (4) Further cooling of the compressed gases to 35.1 °C by cooling water to reduce the compression work or cooling duty in other heat exchangers. 4.3. Case Studies. The heat integration is investigated by four case studies: in cases 1 and 2, only the ASU is integrated; in case 3, only the CPU is integrated; and in case 4, both the ASU and the CPU are integrated. When the CPU is integrated, the compression heat from the two compressors R-P1 and R-P3 (in Figure 2) are integrated (the compression heat from the compressor R-P2 is too small and, thus, not integrated). Detailed description of the four cases is as follows: case 1, the main air compressor (MAC) has three stages with intercooling and the compression ratio for each stage is equal (i.e., 1.77); case 2: the MAC has two stages with intercooling and the compression ratio for each stage is equal (i.e., 2.35); case 3: compressor R-P1 has five stages with equal pressure ratio and compressor R-P3 has two stages with equal pressure ratio; and case 4, the compression schemes in case 1 (three-stage compression in the ASU) and case 3 (multiple-stage compression in the CPU) are used. The following part of this section takes case 1 as an illustrative example. The three heat streams from the outlet of each compression stage are Q1 (88.9 → 49.1 °C), Q2 (101.1

Figure 10. GCC for FWH1−FWH3 in case 1.

On the basis of the GCC, the heat demands for the feedwater heaters are 31.1 MW (FWH3), 6.7 MW (FWH2), and 0 MW (FWH1). The corresponding mass flows of the extracted steam (N25, N26, and N27) can thus be determined on the basis of Figure 9 and are shown in Table 4. The new decomposed heat Table 4. Results of Heat Integration

highest Tgas (°C) work in the ASU (MW) work in the CPU (MW) gross power (MW) net power (MW) thermal efficiency (%) steam flow (kg/s) N15 N16 N18 N24 N25 N26 N27

without integration

case 1

case 2

case 3

case 4

101.1 118

101.1 118

139.9 123

98.7 118

101.1 118

64

64

64

64

64

792 571 30.44

799 579 30.82

805 579 30.82

797 577 30.71

806 585 31.16

61.00 47.57 25.52 33.28 16.48 15.67 15.48

61.00 47.57 25.52 33.28 13.15 2.87 0.00

61.00 47.57 25.52 26.08 6.48 7.54 5.31

61.00 47.57 25.52 33.28 14.37 6.83 3.51

61.00 47.57 25.52 33.28 10.99 0.00 0.00

loads in FWH1−FWH4 are illustrated in Figure 11 (note that the compression heat is not included). When the compression heat is added, a heat balance can be achieved in FWH1− FWH3. One possible thermal configuration of the boiler feedwater preheating process is shown in Figure 12. The compression heat is illustrated by the dotted lines. In this case, the compressed air is first cooled to 56.8 °C. The heat is used to preheat the boiler feedwater. The air is then cooled to 35.1 °C by cooling water (not shown in Figure 12). The boiler 7146

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149

Energy & Fuels

Article

Figure 11. Decomposed heat loads for FWH1−FWH4 in case 1.

Figure 12. Heat balance for FWH1−FWH3 in case 1.

feedwater in each FWH is first slightly heated in an external drain cooler (flash type) by the condensate in the FWH (including the condensate from higher pressure FWHs), then heated by the compressed gases, and finally heated by the extracted steam (if necessary). On the basis of Figure 12, the heat exchanger network can be directly obtained and is shown in Figure 13. The multi-stream gas/water heat exchangers can be easily separated into regular two-stream heat exchangers by splitting the boiler feedwater into several streams. 4.4. Results of the Heat Integration Study. Cases 2−4 are investigated following the same procedure as for case 1. The results, including the highest achievable temperature of the gas (Tgas), the work consumption in the ASU and the CPU, the gross power output from the steam turbines, the net power output, the thermal efficiency, and the mass flow of the extracted steam, are summarized in Table 4. The integration of the ASU with the steam cycle increases the thermal efficiency by 0.38 percentage points in both cases 1 and 2; however, the temperature differences (driving forces) in the gas/water heat exchangers are larger in case 2. When the CPU is integrated, the thermal efficiency increases by 0.27 percentage points (case 3). The thermal efficiency increases by 0.72 percentage points when both the ASU and the CPU are integrated with the steam cycle (case 4). The results presented in Table 4 are obtained from four fixed case studies. It should be pointed out, however, that the compression ratio for each compressor stage, the supply and target temperatures of the compressed gas, and the pressure levels of the extracted steam are actually degrees of freedom

that could be optimized with respect to the energy saving potential and the investment cost. Such optimization is rather complex and needs further investigation.

5. CONCLUSION This paper performs an exergy analysis on an oxy-combustion supercritical pulverized coal power plant with CO2 capture. The thermal efficiency (on the basis of the higher heating value) is calculated to be 39.8% for an air-fired power plant without CO2 capture and 30.4% for the oxy-combustion plant with CO2 capture. The efficiency penalty is mainly caused by the ASU and the CPU. The theoretical minimum specific work for producing O2 with a purity of 95 mol % from air is calculated to be 0.053 kWh/kg of O2, while the actual work for the cryogenic distillation process that has been evaluated is 0.230 kWh/kg of O2. For the CPU, the minimum work consumption and the actual work consumed are calculated to be 0.067 and 0.114 kWh/kg of CO2 captured, respectively. The theoretical minimum energy consumption for the two units takes 3.4 percentage points of the thermal input of the coal feed. However, the actual value is 9.7 percentage points, where the ASU contributes 6.3 percentage points and the CPU contributes 3.4 percentage points. The compression processes are responsible for the largest exergy losses related to CO2 capture. The net thermal efficiency increases from 29.3 to 31.4% if the isentropic efficiencies of all of the compression stages increase from 0.74 to 0.9. Another way to reduce the exergy losses is to use the compression heat for preheating the boiler feedwater. The case studies show that 7147

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149

Energy & Fuels

Article

Figure 13. Possible integration scheme in case 1.

Ė tot = total exergy of a stream (MW Ė feed tot = exergy of a feed stream (MW) Ė product = exergy of a product stream (MW) tot Ḟ = molar flow (mol/s) h = molar enthalpy (MJ/mol); mass fraction of hydrogen in coal H = enthalpy (MW) I ̇ = irreversibilities (MW) LHV = lower heating value (MJ/kg) ṁ = mass flow rate (kg/s) n = mass fraction of nitrogen in coal o = mass fraction of oxygen in coal p = pressure (bar) Q = heat (MW) s = molar entropy (MJ mol−1 °C−1) T = temperature (°C) Tgas = temperature of the compressed gas (°C) Ẇ = work (MW) Ẇ min = theoretical minimum work (MW)

such heat integration can increase the thermal efficiency by 0.72 percentage points. However, the heat integration problem is rather complex, and further optimizations are required.

■

ASSOCIATED CONTENT

* Supporting Information S

Typical parameters for the main streams of the oxy-combustion case as well as the base case (without capturing CO2). This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Telephone: +47-73592799. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

NOMENCLATURE c = mass fraction of carbon in coal e0ch = standard chemical exergy (MJ/mol) e0ch,coal = chemical exergy of coal (MJ/kg) Ė in = exergy entering a unit (MW) Ė out = exergy exiting a unit (MW) Ė ch = chemical exergy of a stream (MW) Ė mix = mixing exergy of a stream (MW) Ė ph = physical exergy of a stream (MW)

Greek Letters

φ = ratio of the chemical exergy to the lower heating value Sub- and Superscripts

0 = reference state i = component index Abbreviations

ASU = air separation unit 7148

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149

Energy & Fuels

Article

(14) Payne, R.; Chen, S. L.; Wolsky, A. M.; Richter, W. F. CO2 recovery via coal combustion in mixtures of oxygen and recycled flue gas. Combust. Sci. Technol. 1989, 67, 1−16. (15) Herzog, H.; Golomb, D.; Zemba, S. Feasibility, modeling and economics of sequestering power plant CO2 emissions in the deep ocean. Environ. Prog. 1991, 10 (1), 64−74. (16) Kimura, N.; Omata, K.; Kiga, T.; Takano, S.; Shikisima, S. The characteristics of pulverized coal combustion in O2/CO2 mixtures for CO2 recovery. Energy Convers. Manage. 1995, 36 (6−9), 805−808. (17) Tree, D. R.; Mackrory, A. J.; Fletcher, T. A Mechanistic Investigation of Nitrogen Evolution and Corrosion with Oxy-combustion; Brigham Young University: Provo, UT, 2009; Report DE-FG2605NT42530. (18) Nakayama, S.; Noguchi, Y.; Kiga, T.; Miyamae, S.; Maeda, U.; Kawai, M.; Tanaka, T.; Koyata, K.; Makino, H. Pulverized coal combustion in O2/CO2 mixtures on a power plant for CO2 recovery. Energy Convers. Manage. 1992, 33 (5−8), 379−386. (19) International Energy Agency Greenhouse Gas R&D Programme (IEAGHG). Greenhouse News 104; IEA GHG: Cheltenham, U.K., 2011. (20) Romeo, L. M.; Usón, S.; Valero, A.; Escosa, J. M. Exergy analysis as a tool for the integration of very complex energy systems: The case of carbonation/calcination CO2 systems in existing coal power plants. Int. J. Greenhouse Gas Control 2010, 4 (4), 647−654. (21) Xiong, J.; Zhao, H.; Zheng, C. Exergy analysis of a 600 MWe oxy-combustion pulverized-coal-fired power plant. Energy Fuels 2011, 25 (8), 3854−3864. (22) Harendra, S.; Oryshcyhn, D.; Gerdemann, S.; Ochs, T.; Clark, J. Modeling energy flow in an integrated pollutant removal (IPR) system with CO2 capture integrated with oxy-fuel combustion. Energy Fuels 2012, 26 (11), 6930−6937. (23) U.S. Department of Energy (DOE)/National Energy Technology Laboratory (NETL). Pulverized Coal Oxycombustion Power Plants; DOE/NETL: Washington, D.C., 2008; 2007/1291. (24) Hands, B. A. Cryogenic Engineering; Academic Press: London, U.K., 1986. (25) Pipitone, G.; Bolland, O. Power generation with CO2 capture: Technology for CO2 purification. Int. J. Greenhouse Gas Control 2009, 3 (5), 528−534. (26) Fu, C.; Gundersen, T. Techno-economic analysis of CO2 conditioning processes in a coal based oxy-combustion power plant. Int. J. Greenhouse Gas Control 2012, 9, 419−427. (27) Fu, C.; Gundersen, T. Integrating the compression heat in oxycombustion power plants with CO2 capture. Chem. Eng. Trans. 2012, 29, 781−786. (28) Szargut, J.; Morris, D. R.; Steward., F. R. Exergy Analysis of Thermal, Chemical and Metallurgical Processes; Hemisphere: New York, 1988. (29) Hinderink, A. P.; Kerkhof, F. P. J. M.; Lie, A. B. K.; Arons, J. D. S.; Kooi, H. J. V. D. Exergy analysis with a flowsheeting simulatorI. Theory; calculating exergies of material streams. Chem. Eng. Sci. 1996, 51 (20), 4693−4700. (30) Kotas, T. J. The Exergy Method of Thermal Plant Analysis; Krieger Publishing Company: Malabar, FL, 1995. (31) Fu, C.; Gundersen, T. Using exergy analysis to reduce power consumption in air separation units for oxy-combustion processes. Energy 2012, 44 (1), 60−68. (32) Chakravarthy, V. K.; Daw, C. S.; Pihl, J. A. Thermodynamic analysis of alternative approaches to chemical looping combustion. Energy Fuels 2011, 25 (2), 656−669. (33) Thermoflow, Inc. STEAM PRO 21.0. Help Document; Thermoflow, Inc.: Southborough, MA, 2012; www.thermoflow.com. (34) Smith, R. Chemical Process Design and Integration; John Wiley and Sons, Ltd.: Hoboken, NJ, 2005.

CCS = carbon capture and storage CPU = CO2 compression and purification unit CW = cooling water DCA = direct contact aftercooler ESP = electrostatic precipitator FGD = flue gas desulfurization FWH = feedwater heater GCC = grand composite curve HHV = higher heating value HP = higher pressure; high pressure IP = intermediate pressure LHV = lower heating value LP = lower pressure; low pressure MAC = main air compressor MHE = main heat exchanger NETL = National Energy Technology Laboratory PPU = prepurification unit PR = Peng−Robinson RFG = recycled flue gas SCR = selective catalytic reduction SSR = steam seal regulator TSA = temperature swing adsorption

■

REFERENCES

(1) Intergovernmental Panel on Climate Change (IPCC). Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change; Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K. B., Tignor, M., Miller, H. L., Eds.; Cambridge University Press: New York, 2007. (2) Figueroa, J. D.; Fout, T.; Plasynski, S.; McIlvried, H.; Srivastava, R. D. Advances in CO2 capture technologyThe U.S. Department of Energy’s Carbon Sequestration Program. Int. J. Greenhouse Gas Control 2008, 2 (1), 9−20. (3) Fu, C.; Gundersen, T. Carbon capture and storage in the power industry: Challenges and opportunities. Energy Procedia 2012, 16 (Part C), 1806−1812. (4) Davison, J. Performance and costs of power plants with capture and storage of CO2. Energy 2007, 32 (7), 1163−1176. (5) Kanniche, M.; Gros-Bonnivard, R.; Jaud, P.; Valle-Marcos, J.; Amann, J.-M.; Bouallou, C. Pre-combustion, post-combustion and oxycombustion in thermal power plant for CO2 capture. Appl. Therm. Eng. 2010, 30, 53−62. (6) Buhre, B. J. P.; Elliott, L. K.; Sheng, C. D.; Gupta, R. P.; Wall, T. F. Oxy-fuel combustion technology for coal-fired power generation. Prog. Energy Combust. Sci. 2005, 31 (4), 283−307. (7) Scheffknecht, G.; Al-Makhadmeh, L.; Schnell, U.; Maier, J. Oxyfuel coal combustionA review of the current state-of-the-art. Int. J. Greenhouse Gas Control 2011, 5S, S16−S35. (8) Normann, F.; Andersson, K.; Johnsson, F.; Leckner, B. Reburning in oxy-fuel combustion: A parametric study of the combustion chemistry. Ind. Eng. Chem. Res. 2010, 49 (19), 9088−9094. (9) Singh, B.; Strømman, A. H.; Hertwich, E. G. Comparative life cycle environmental assessment of CCS technologies. Int. J. Greenhouse Gas Control 2011, 5 (4), 911−921. (10) Steinberg, M. History of CO2 greenhouse gas mitigation technologies. Energy Convers. Manage. 1992, 33 (5−8), 311−315. (11) Horn, F. L.; Steinberg, M. Control of carbon dioxide emissions from a power plant (and use in enhanced oil recovery). Fuel 1982, 61 (5), 415−422. (12) Allam, R. J.; Spilsbury, C. G. A study of the extraction of CO2 from the flue gas of a 500 MW pulverised coal fired boiler. Energy Convers. Manage. 1992, 33 (5−8), 373−378. (13) Abraham, B. M.; Asbury, J. G.; Lynch, E. P.; Teotia, A. P. S. Coal-oxygen process provides CO2 for enhanced recovery. Oil Gas J. 1982, 68−75. 7149

dx.doi.org/10.1021/ef400867d | Energy Fuels 2013, 27, 7138−7149