Applied Thermal Engineering 61 (2013) 115e122
Contents lists available at SciVerse ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Coal based power plants using oxy-combustion for CO2 capture: Pressurized coal combustion to reduce capture penalty Rengarajan Soundararajan*, Truls Gundersen Department of Energy and Process Engineering, Norwegian University of Science and Technology, Kolbjoern Hejes vei. 1B, NO-7491 Trondheim, Norway
a r t i c l e i n f o
a b s t r a c t
Article history: Received 14 November 2012 Accepted 2 April 2013 Available online 17 April 2013
The goal of this paper is to design and study a new variation of an oxy-combustion coal based power plant with CO2 capture. This variation employs a pressurized coal combustor that burns coal in an oxygen rich environment. The concept is compared with an atmospheric pressure oxy-combustion power plant (baseline case). Such analyses would provide us with information regarding potential heat integration and improvement opportunities of oxy-combustion coal based power plants. Also, this study highlights the efficiency improvement potential of the oxy-combustion technology for coal based power plants. The power cycle presented in this paper is a supercritical cycle that has a gross electric power output of 774 MW for the baseline case and 792 MW for the pressurized case. The auxiliary power consumption is reduced from 224 MW in the baseline case to 214 MW in the pressurized case due to the absence of air leakage into the boiler. The recovery of latent heat from the flue gases is increased due to the elevated dew point in the pressurized case. This results in a net LHV and HHV efficiency improvement of 1.7 percentage points each over the baseline case. In both the cases, over 90% of the produced CO2 is captured and compressed to 110 bar after removal of volatiles and other pollutants such as SOx and NOx. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: CO2 capture Oxy-combustion Capture penalty Pressurized coal combustion Heat integration
1. Introduction Carbon Capture and Storage (CCS) is a technology that collects and concentrates the CO2 emitted from large point sources such as power plants, transports it to a suitable storage location and stores it away from the atmosphere for a sufficiently long time to avoid warming of the atmosphere. Global energy consumption is projected to rise in the future driven by population growth and other factors such as the rapid economic expansion of large developing nations such as China, India and Brazil. Fossil fuels are expected to be a major part of the future energy mix according to projections by reputed international organizations [1]. In this context, technology such as CCS becomes inevitable to avoid atmospheric greenhouse gas emissions and related climate consequences. CCS is expected to play a pivotal role in stabilizing the atmospheric greenhouse gas levels within acceptable limits. It has been estimated that the average contribution of CCS in total emission reduction would range from 15% to 54% for stabilization targets of 750 ppmv to 450 ppmv CO2 [2]. The deployment of CCS could help bring down the overall cost of mitigation of climate change in the longer run [3]. There are three main ways to implement the CCS when it comes to
* Corresponding author. Tel.: þ47 4510 6327. E-mail address:
[email protected] (R. Soundararajan). 1359-4311/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.applthermaleng.2013.04.010
the capture part of the CCS chain. One of the three ways to capture CO2 is referred to as the oxy-combustion method in which the fuel is burnt in an oxygen rich environment instead of air, thus resulting in a flue gas that contains mainly water vapour and CO2. The oxygen required for combustion comes from the atmosphere after the associated nitrogen is stripped off before the combustor. Part of the flue gas is recycled back into the combustor to maintain the combustion temperature at acceptable levels. This method is considered to have several advantages such as reduced environmental impact [4] and competitive cost of electricity [5]. Oxy-combustion requires an upstream production of oxygen for combustion, which is energy intensive and hence expensive. Downstream purification of the flue gas using a Compression and Purification Unit (CPU) is also required to remove the volatile components such as nitrogen, argon, oxygen, etc., after condensing the water, to achieve the required purity before compression and pipeline transport. A cryogenic Air Separation Unit (ASU) is generally used for large-scale production of oxygen. The ASU and the CPU are responsible for an efficiency penalty of around 10 percentage points [6]. There is a need to reduce this efficiency penalty in order to make oxy-combustion power plants attractive and ready for commercial deployment. Heat integration techniques are widely suggested in order to reduce the capture penalty. Adiabatic compression in the ASU compressors helps to improve the heat integration [7]. One of the ways to reduce the capture
116
R. Soundararajan, T. Gundersen / Applied Thermal Engineering 61 (2013) 115e122
Nomenclature ASU BFW C1eC4 CPU CRH DoE E-1, E-2 EGR ESP F1eF4 FGD HHV HP-FWH HRSG LHV LP-FWH NETL V-1, V-2
air separation unit boiler feedwater compressors in the CPU CO2 compression and purification unit cold reheat steam Department of Energy multi stream heat exchangers exhaust gas recycle electrostatic precipitator flash drums in the CPU flue gas desulphurization higher heating value (kJ/kg) high pressure-feedwater heaters heat recovery steam generator lower heating value (kJ/kg) low pressure-feedwater heaters National Energy Technology Laboratory throttle valves
penalty is to go for a pressurized coal combustion system, which has several advantages over the atmospheric combustion system. Similar systems have been studied earlier [8] and pressurized coal combustion shows a reduction in capture penalty [9]. Gazzino et al. [8], have analysed pressurized oxy-combustion of coal and have found out that both char combustion and convective heat transfer is improved when compared to atmospheric pressure oxycombustion. More latent heat is also recovered in the acid condenser, improving the overall power plant efficiency. Their cycle employs a dual reheat steam cycle. Hong et al. [9], study the influence of operating pressure which is one of the most important parameters affecting the performance of pressurized oxycombustion systems. The study varies the combustor operating pressure from 1.238 bars to 10 bars. Their study shows that the savings achieved in the recirculation fan compression work and enhanced heat recovery in the acid condensers make 10 bars the optimal operating pressure of such systems. However, neither of the above studies has a base case atmospheric oxy-combustion power plant that is close to the state of art. A state of the art coal based power plant would use a single reheat supercritical steam cycle and a boiler that has a negative gage pressure with radiant heat transfer. Such a base case configuration would help us locate and quantify the efficiency improvements brought by pressurization. In this study, a pressurized oxy-combustion coal based power plant with CO2 capture is analysed and compared with an atmospheric counterpart (base case). The base case resembles a modern pulverised coal based power plant fitted with oxy-combustion technology for CO2 capture. The pressurized case employs a boiler design similar to that of the Gazzino et al. [8], while sharing the steam cycle, ASU and CPU design with the base case. The ambient conditions, coal composition and the CO2 pipeline specifications are also kept common between two cases. The only difference between the two cases is the boiler heat transfer mode. The atmospheric system has a radiant heat transfer section whereas the pressurized system does not. An acid condenser or a plastic heat exchanger is used in both cases to recover the low grade latent heat from the flue gases before it enters the CPU for purification, and the recovered heat is used in the steam cycle for additional power production [10]. Both cases have similar flue gas purification processes that remove SOx and NOx and then compress the resulting CO2 to pipeline specifications. The methodology used in the study is
described in the following section followed by detailed process descriptions, simulation parameters, performance results and conclusions. 2. Methodology The simulation package Aspen PlusÒ is used to model the steam cycle, the atmospheric and pressurized boiler islands and the CO2 Purification and Compression Unit. Thermoflow STEAM PROÒ is used to assist the simulation process by providing the parameters such as pressure drops and other thermodynamic assumptions for the boiler and the steam cycle. IAPWS-95 physical properties method was used to estimate the steam/water parameters in the steam cycle. Peng Robinson cubic equation of state with the BostoneMathias alpha function was used to simulate the boiler island while the same equation of state with kij binary interaction parameters was used to model the CPU. The ASU was not modelled, however, the energy required in producing and compressing the oxygen was taken into account from similar studies [8]. Initially, a steam cycle is modelled in Aspen PlusÒ assisted by STEAM PROÒ which gives the gross power produced and the thermal energy requirements. The steam cycle also provides the feedwater conditions and live steam parameters. Using this information, an oxy-combustion boiler can be designed that suits the steam cycle. A CPU is then designed for the boiler that removes the volatiles and compresses the flue gas to final pipeline specifications. Based on the above method, the fuel flow is kept constant at 64 kg/s so as to form a common ground for comparison. Other simulation parameters that have an impact over the results such as the excess oxygen, combustor exit temperature, steam cycle and CPU operating parameters are also kept consistent between the two cases. Various performance parameters such as the gross power produced, auxiliary power requirements and the net plant efficiency are calculated after combining the heat and mass balance models from Aspen Plus. Finally, the improvements brought by the pressurization are discussed along with the potential heat integration opportunities for the future. Comparisons of cases are mainly based on the overall efficiency perspective. Changes in capital investment, operating expenses and environmental impact are not considered in this study and could be investigated in the future. Thermodynamic assumptions and simulation parameters, cycle description and performance results are provided in the subsequent sections. 3. Process description In order to maintain a consistency and to have a baseline for future studies, the simulation parameters are taken from published reports as explained below. Ambient conditions, fuel composition, steam cycle and cooling system parameters are obtained from an EBTF common framework document [11] The feedwater preheating system, auxiliary power consumption and some of the boiler parameters are obtained from STEAM PROÒ. Other parameters such as the boiler pressure and pressure drop for the pressurized case are taken from Hong et al. [9],. The CPU simulation parameters including the configuration, flash temperatures, pressures and the compressor efficiency values are provided by Pipitone and Bolland, 2009 [12], and the Equation of State is from Posch and Haider, 2012 [13]. The coal composition is given in Table 1 and Table 2 gives the ambient air composition. Selected simulation assumptions are given in Table 3. Fig. 1 shows the overall schematic of the power plant except the CO2 purification and compression part. The central element of the boiler island is the combustor, the pressure of which depends on the case considered. Coal is fed into the combustor while ash is removed from the combustor in a molten state. The
R. Soundararajan, T. Gundersen / Applied Thermal Engineering 61 (2013) 115e122 Table 1 Coal composition and heating value.
117
Table 3 Selected simulation parameters for the cycle.
Bituminous Douglas premium coal characteristics
Parameter
Value
Proximate analysis wt.%
Steam cycle
Baseline
Pressurized
Main steam pressure Main steam temperature Reheat temperature Condenser pressure Feedwater heaters Feedwater final temperature Deaerator pressure
280 600 610 0.048 7 315
280 600 610 0.048 6 315
bar C C bar
18
18
bar
18
15 20 1.0124 3
15 20 10 3
bar C bar % (dry)
N/A 10/5 3 3
1850 95 1.04 71
1550 95 1.06 85.3
C % No unit Mass %
3 1 N/A 8
30/33 54/31 110
30/33 54/31 110
Moisture Ash Volatiles Fixed carbon Total sulphur
HHV (MJ/kg) LHV (MJ/kg)
Ultimate analysis wt.% 8.000 14.150 22.950 54.900 0.520
Carbon Nitrogen Hydrogen Total sulphur Ash Chlorine Moisture Oxygen
66.520 1.560 3.780 0.520 14.150 0.010 8.000 5.460
26.230 25.170
Table 2 Ambient air composition.
Nitrogen CO2 H2O Argon Oxygen Gas constant (J/(kg K)) Molecular weight
Volume fraction dry 78.09 0.03 0.93 20.95 287.06 28.964
Stream no. Fig. 1
C
11 11 12 13 N/A 10
Boiler island
boiler island also has an ash removal and handling system (ESP) to remove fly ash, an induced draft fan to overcome the gas-side pressure drop, a Heat Recovery Steam Generator (HRSG) and an acid condenser. An air leakage stream is present only in the atmospheric case due to a negative gage pressure in the combustor/ HRSG. The ASU is shown in the figure as a single block supplying oxygen to the combustor. Part of the flue gas after the fan is passed through the acid condenser. Heat from the flue gas is recovered in the acid condenser and used to preheat the feedwater before the low pressure-feedwater heaters. The oxygen stream from the ASU is mixed with recycled flue gas and fed into the combustor. The oxygen stream is preheated in the baseline case and compressed in the pressurized case before the combustor. The hot flue gases from the combustor is then mixed with more recycled flue gas in order to adjust the temperature to be suitable for heat transfer and steam production in the HRSG. The flue gas exiting the HRSG then passes through the ESP for ash removal. Part of the flue gas forms the recycle feed (Exhaust Gas Recycle-EGR). The power system is a large supercritical steam cycle that resembles those in operation at large power plants today. Steam, after expanding in the steam turbines, is condensed by cooling water from the cooling towers. The site ambient temperature of 15 C and a relative humidity of 60% determine the condensation pressure and hence the efficiency of the steam cycle. The condensate is then typically preheated in feedwater heaters that are supplied with condensing steam from the turbines. The LPFWH and HP-FWH in Fig. 1 represent the low pressure and the high pressure-feedwater heating trains. Regenerative preheating of boiler feedwater is essential to achieve high steam cycle efficiency and hence a better overall plant efficiency [14]. In a typical large steam cycle, as many as eight feedwater heaters are used to preheat the boiler feedwater by extracting steam from various extraction points in the steam turbines [15]. Also, a deaerator is used to remove gases and other impurities from the boiler feedwater. When using the acid condenser, part of this heating is performed by stream no. 7 (Fig. 1) from the boiler exhaust and hence corresponding steam extraction can be used in the steam turbines to generate additional power.
Component
Units
Volume fraction at 60% R.H 77.30 0.03 1.01 0.92 20.74 288.16 28.854
Evaporator pressure drop Boiler minimum design pinch Boiler operating pressure Excess oxygen at combustor outlet Combustor exit temperature Oxygen purity Fan pressure ratio Recycle ratio CO2 Purification Unit Stage 1 temperature/pressure Stage 2 temperature/pressure Final product pressure
Fig. 3 C/bar C/bar bar
27 31 39
Finally, the flue gas which is now cooled down in the acid condenser is fed to the CPU for purification and compression to the pipeline specifications. BFW in Fig. 1 represents the feedwater at its final preheated temperature before it enters the boiler. CRH is the cold reheat stream taken from the turbines before it is reheated to 610 C. Two pumps are used in the steam cycle; the condensate forwarding pump immediately after the condenser and a boiler feedwater pump after the deaerator. The boiler feedwater pump is driven by a low pressure condensing turbine that is supplied with steam at 9.4 bars. 3.1. ASU and the boiler island The ASU forms the central element of the whole oxycombustion CO2 capture process. It is also responsible for most of the energy penalty associated with the capture. Oxygen can be produced in many different ways, but the most commonly considered way to produce large quantities of oxygen is cryogenic distillation, which is the assumed technology in this study. As the ASU is a large and complex system in itself, to limit the complexity of the study, it is not simulated in detail. Instead, only the power required to produce the oxygen is taken into account. The specific energy required to produce an oxygen stream of 95% purity at atmospheric pressure is around 0.23 kWh/kg [8]. Commercial figures vary widely and also the value used in this report does not reflect the most recent developments made in the field of air separation [16]. In the atmospheric case, the oxygen stream is preheated before the combustor while in the pressurized case; the temperature of the oxygen stream at the inlet of the combustor is controlled by the choice of compression ratio and number of stages of the oxygen compressor. Compression of oxygen consumes additional energy, however, this saves considerable amount of energy in the downstream units such as the CPU. There are some differences in the way the atmospheric boiler and the pressurized boiler are constructed. The atmospheric boiler resembles the state of the art pulverized coal boiler with water walls around the furnace acting as evaporator (radiant heat transfer) followed by a convective section that includes the superheaters
118
R. Soundararajan, T. Gundersen / Applied Thermal Engineering 61 (2013) 115e122
Fig. 1. Overall schematic of the oxy-combustion power plant.
and reheaters. The pressurized boiler on the other hand has two sections with a separate high pressure combustor and a high pressure HRSG. The convective heat transfer rate is much higher in a high pressure HRSG that almost all of the heat transfer takes place only convectively [8]. This results in the lower temperature of flue gases entering the HRSG, increased flue gas mass flow rates and the associated fan compression work required. This is one of the fundamental differences between the atmospheric and the pressurized cases. The combustion temperature in the pressurized combustor is maintained at 1550 C by circulating 85.3% by mass of the flue gases after the HRSG. The ash resulting from the coal combustion is removed in two stages. One is from the combustor as a molten liquid and the other ash removal is after the HRSG by using an Electrostatic Precipitator. In the atmospheric combustor, there is a considerable amount of air leakage into the combustor due to the operating pressure while this is not the case with pressurized combustion. Air leakage is an operational issue in coal based power plants. It depends on many factors such as the size of the power plant, age, load, etc. While it is possible to minimize and control the air leakage, it is not possible to eliminate it completely from an atmospheric pressure system [17]. It is also difficult to precisely characterize the exact amount of air leakage as it varies with many factors such as the construction of the boiler. Hence, the value is assumed from an atmospheric pressure oxy-combustion power plant of similar size from the NETL report [18]. It is assumed to be 11 kg/s to the combustor in the base case. As a lot of energy is spent in the ASU to separate oxygen from air to produce a relatively pure Oxygen stream; air leakage into the boiler is considered a huge loss from the energy stand point. Also, more energy will be required in the CPU to remove the impurities to achieve pipeline specifications. A fan is required to overcome the gas-side pressure drop in both cases. The boiler fan consumes a considerable amount of power in the pressurized case due to both increased mass flows and a different pressure ratio. One of the main design parameters of the combustor is the excess oxygen at the combustor exit. An excess oxygen value of 3% on a dry basis at the combustor exit is chosen from a report by DOE which affects the oxygen production power, combustion efficiency and also the power required for the downstream purification process. An excess oxygen value of up to 3% does not affect the transport and storage of the captured CO2 [18].
Preheated feedwater at a supercritical pressure of 325 bar (BFW) enters the boiler economizer where it is heated before it is converted into steam. The steam reaches a maximum temperature of 600 C which is considered state of the-art of a modern supercritical steam power plant. There is also a reheat stream that is taken from the turbine at 55 bars (CRH) and then re-enters the steam turbine at 610 C and a pressure of 50 bars. Pipe pressure losses in the main steam pipe, cold and hot reheat pipes and other steam extraction pipes are assumed and maintained consistent between the cases. The preheat level for the boiler feedwater is chosen to be 310 C which is in accordance with modern steam cycle standards. This helps achieve better efficiency in the steam cycle. Due to this high level of preheating, the flue gas exit temperature from the HRSG is rather high at around 335 C. This has an impact on both the recirculation gas flow and fan power consumption. Also, there is no flue gas desulphurization required before the gas is recirculated back into the combustor, and thereby saving some energy. As the flue gas leaving the HRSG has a fairly high temperature and a lot of water vapour, utilising both the sensible and latent energy available in the flue gas after the HRSG becomes imperative. Hence, heat exchangers and an acid condenser are used downstream before the emission control system and the CPU. Finally, the boiler blow down is not considered in this simulation. 3.2. Steam cycle The main steam inlet conditions are 280 bars and 600 C. All the steam turbine isentropic efficiencies are assumed constant to make calculations simple. The steam turbine isentropic efficiencies are assumed to be 92%, 94% and 88% for the HP, IP and the LP stages, respectively. Steam seal losses and exhaust end losses are neglected for simplicity. There are two main pumps present in the steam cycle, i.e. the condensate forwarding pump and the boiler feedwater pump. The former is electric motor driven while the latter is driven by a low-pressure turbine. All the pumps are assumed to have an isentropic efficiency of 70%. The energy consumed by the pumps forms a significant part of the auxiliary power consumption. The feedwater preheating train is designed in such a way that it splits the feedwater temperature rise equally into 35 C intervals and supplied with extraction steam from the steam turbines. The
R. Soundararajan, T. Gundersen / Applied Thermal Engineering 61 (2013) 115e122
119
Fig. 2. Flue gas purification.
steam extraction pressure is computed based on the terminal temperature difference in the feedwater heater, the available degree of superheat in the extraction steam and the extraction pipe pressure losses which are assumed to be constant. Aspen Plus calculates the mass flows of the extraction steams from the turbines subjected to the Drain Cooler Approach temperature which is set at 5 C. The heat recovered from the flue gas by the acid condenser is used to supply a part of the feedwater heating requirements. This not only helps reduce the steam extraction from the turbines, but also eliminates one or more of the feedwater heaters and the associated capital cost. 3.3. The emission control system The emission control system is shown in two parts in Figs. 2 and 3. It is noteworthy to mention that for coals with a sulphur content of less than 1%, there is no need to remove sulphur from the exhaust gas before recirculation [18]. Although the concentration of sulphur compounds is amplified in the boiler due to flue gas recirculation, it will be well under the boiler design conditions for the coal considered in this study (0.52% sulphur). The flue gas stream is cooled to 25 C, any condensation is removed and then the flue gas is compressed to 15 bars. The flue gas is cooled again and water is added to remove sulphur as sulphuric acid (Stream 24) in the water wash column in Fig. 2. The conversion of SO2 to H2SO4 at 15 bars is catalysed by the presence of NO2. NO2 is converted into NO in the process. Then the flue gas is again compressed to 33 bars and cooled again. A second water wash column is required to convert the remaining NO2 and the NO produced in the first water wash column to HNO3. Traditional wet flue gas desulfurization is not required as it is easier to remove SOx and NOx together under high pressure in a water wash column [19]. The water wash
columns are designed to add a few seconds of holdup to the conversion process by use of a contacting column with pumped around liquid condensate [19]. Any moisture present is removed by using adsorption to avoid ice formation in the downstream purification where it will be cooled below the dew point. Flue gas stripped completely of SOx and moisture enters the double flash purification unit shown in Fig. 3. The gas enters a cold box (E-1) where it is cooled to 30 C (Stream 27) and separated in a flash drum (F3). Then again, the resulting vapour (Stream 28) is cooled to 54 C and separated in another flash drum (F4). Cooling for the above process is provided by evaporating part of the liquid streams (Stream 29 and 32) after throttling them to a lower pressure. While the impure stream rich in volatiles is vented to the atmosphere as the only emissions from the power plant, the CO2 rich stream is compressed, cooled and then pumped to the final pipeline pressure of 110 bars. 4. Performance and results Cycles for both the baseline case and the pressurized case are identical except for the operating pressure of the boiler island and the number of feedwater preheaters. In the pressurized case, as more latent heat can be extracted from the flue gas, only two low pressure-feedwater heaters are required. An oxygen compressor is also required to maintain the combustor pressure. The mass flow of the recycle feed changes between the two cases and so does the pressure drop on the gas side of the steam generator. The CPU remains unchanged between the two cases considered. The performance summary of both cases is presented in Table 4. It is evident that the pressurized cycle has some clear advantages over the atmospheric case. Total auxiliaries are reduced by 10 MW and the gross power has been increased by 18 MW resulting
Fig. 3. Compression and purification unit (CPU).
120
R. Soundararajan, T. Gundersen / Applied Thermal Engineering 61 (2013) 115e122
Table 4 Performance summary. Item
Baseline
Pressurized
Units
Fuel energy input, LHV Condenser duty ST shaft power ST-Gen losses Gross electric power Steam cycle aux. ASU power req. Boiler island aux. CPU power req. Total auxiliaries Net electric power Net plant efficiency, LHV Net plant heat rate, LHV Net plant efficiency, HHV CO2 emissions
1610.9 860.1 789.3 14.9 774.3 6.8 119.0 15.4 82.5 223.7 550.6 34.2 10,532 32.8 51
1610.9 899.7 807.5 15.3 792.2 6.8 148.4 24.9 33.7 213.8 578.4 35.9 10,026 34.5 21
MWth MWth MW MW MWe MWe MWe MWe MWe MW MW % kJ/kWh % g/kWh
in a net power increase of 28 MW. Table 5 provides a summary of major streams in the cycle for both cases. The air leakage stream is absent in the pressurized case as the boiler is operating at a pressure higher than the atmospheric pressure. In addition to the improvement in the efficiency of the power plant, the CO2 recovery rate is also increased. The air emission of the pressurized case is now 21 g/kWh compared to the 51 g/kWh of the atmospheric case. This represents a 2.8 percentage point increase in the CO2 recovery factor to 97.8%. Savings in terms of auxiliary power consumption is achieved in a number of ways in the pressurized case. Various factors that have a major impact on the overall efficiency are discussed in detail in the following sub-sections. 4.1. Oxygen compression effects Although more work is required to compress the oxygen rich stream before the combustor, savings achieved in the CPU
Table 5 Stream data, stream numbers from figures (a e atmospheric, b e pressurized). Stream no.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 22 23 24 27 28 29 33 39
Temperature C
Pressure bars
Mass flow kg/s
a
b
a
b
a
b
15.00 340.24 1876.89 N/A 335.00 340.19 340.19 340.19 57.38 310.00 600.02 610.83 32.15 88.12 32.16 32.44 83.12 172.00 355.58 292.9 25.00 29.18 30.00 30.00 30.00 15.00 33.49
132.25 342.97 1531.90 342.97 335.00 342.99 342.99 342.99 57.40 310.00 600.02 610.83 32.15 102.60 32.10 32.38 97.62 172.00 355.58 30.8 25.00 39.24 30.00 30.00 30.00 15.00 33.58
1.60 1.01 1.01 N/A 1.01 1.04 1.04 1.04 1.04 325.52 280.00 50.00 0.05 1.21 0.05 22.00 22.00 22.00 62.16 15 15 15 33.00 32.00 32.00 31.00 110.00
10.50 10.00 10.00 10.00 10.00 10.00 10.00 10.00 10.00 325.93 280.00 50.00 0.05 3.61 0.05 22.00 22.00 22.00 62.16 15 15 15 33.00 32.00 32.00 31.00 110.00
138.62 498.23 702.79 N/A 702.79 702.79 204.56 498.23 204.56 592.00 592.00 497.52 347.75 64.80 450.06 450.06 450.06 450.06 497.52 179.95 177.03 7.92 176.90 45.69 131.22 25.12 151.78
141.30 741.34 937.59 399.18 1336.77 1336.77 196.23 1140.53 196.23 602.80 602.80 506.61 364.47 55.45 458.17 458.17 458.17 458.17 506.61 169.70 168.77 5.93 168.65 21.00 147.65 11.80 156.85
compression work more than compensates for it. ASU power requirement is increased by 29.4 MW, while the CPU energy requirement is reduced by 48.8 MW resulting in a net savings of 19.4 MW. The mass flow of the oxygen stream is always smaller than the flue gas stream due to the addition of carbon from the fuel coal. The flue gas is mainly a combination of CO2 and H2O. The latter can easily be separated without much power consumption by means of condensation. At the same time, the heat recovered can be used to generate additional power. The CO2 present in the flue gas, however, has to be compressed by the CPU compressors. Also, the flue gas contains nitrogen, oxygen and argon both due to the purity level of the oxygen stream and air leakage in case of the atmospheric boiler. This also consumes some energy in the CPU. Pressurization helps compress only the oxygen stream and not the carbon from the fuel. While injecting the solid fuel into the combustor consumes some energy, it will be much less compared to gas compression of similar mass flows and pressure ratios. Thus, the power required to feed the coal into the combustor is not considered in the calculations. Besides that, pressurization eliminates air leakage into the boiler and hence further reduces the mass flow of the gases to be processed by the CPU. In an air fired coal based power plant, the air leakage presents challenges such as reduced efficiency and operability of the power plant [17]. But in an oxy-combustion coal based power plant, the air leakage is responsible for increased auxiliary consumption at the CPU. Atmospheric air leaking into the boiler goes against the objectives of the oxy-firing. 4.2. Fan power requirement The recycle ratio in the pressurized case is different from that of the baseline case because the combustor is considered adiabatic and no heat transfer to the water/steam takes place in the high pressure combustor. All the heat transfer takes place in a steam generator located after the combustor and hence the flue gas needs to be cooled before the steam generator to avoid hot corrosion. This leads to more flue gas being circulated for temperature control. Also, due to the change in pressure drop, the fan power requirement in the pressurized case is increased by 9.5 MW. The changes in auxilliary power consumption between the cases are shown in Fig. 4. 4.3. Enhanced heat recovery Addition of heat to the steam cycle has some effects on the efficiency of the steam cycle. For instance, heat added through the boiler area (HRSG) of the steam cycle is better converted to work in the steam turbines. Heat added in the feedwater preheating is not efficiently converted into work due to the lower quality of the heat. Every MW of additional heat in the HRSG generates 0.5 MW additional power in the steam turbines, whereas the same amount
Fig. 4. Auxiliary power consumption.
R. Soundararajan, T. Gundersen / Applied Thermal Engineering 61 (2013) 115e122
250 200
Atmospheric case Pressurized case
150 100 50 0 Heat recovery, MW
Steam Extraction, kg/s
Fig. 5. Acid condenser heat recovery and steam extraction.
Percentage points
Power in MW
Acid condenser increased heat recovery
0.2
Gross power increase due to fan compression
0.3 4.7 0.6
Aux. net savings due to pressure ratio & mass flow
0.6
can be recovered in the pressurized case. This helps generate additional power in the turbines by reducing the amount of steam that is extracted for feedwater preheating purpose. A pressurized boiler must have thicker walls and other features such as pressure seals that are gas tight in order to avoid hot gases leaking into the power plant premises. Such a system would be much expensive than today’s power plant. At the same time, as the pressure becomes high, the heat transfer is improved and size of the unit is reduced, saving some capital cost. An economic analysis is required to find out the cost optimal operating pressure. In this study, a net reduction in auxiliary power consumption is achieved despite an increase in fan power consumption. Besides, half of the additional auxiliary power consumed by the oxygen compressors and the recirculation fan is recovered back in the steam cycle. As the gross power output goes up and the auxiliary power consumption goes down, the net efficiency is increased. Efficiency improvement achieved is in the order of 1.7 percentage points. In addition, the CO2 recovery factor is improved by 2.8 percentage points to 97.8%. Acknowledgements
3.5
Gross power incrase due to oxygen compression
121
9.7
9.9
Fig. 6. Contribution of various factors to the overall efficiency improvement.
of heat added through the acid condenser generates only 0.12 MW of power. Due to a higher pressure of the flue gases, the dew point of the water vapour is raised and hence more of the latent heat available in the flue gas slip stream can now be recovered in the acid condenser. This takes the feedwater temperature prior to the low pressure-feedwater heater to 98.3 compared to the 83.7 degrees of the baseline case. In other words, as much as 30 MW more heat is added to the steam cycle in the pressurized case. This results in an increase in the gross power produced by about 3.5 MW. As a result, only six feedwater heaters, including the deaerator, are required in the steam cycle. This saves both capital cost as well as some extraction steam from the turbines. It is noteworthy to mention that almost half the power spent to compress the oxygen stream and the hot gases in the HRSG (fan) is now recovered back in the steam cycle, thereby driving up the gross power by 18 MW. Fig. 5 shows the heat recovery in the acid condenser and the corresponding steam extraction from the turbines for feedwater preheating. Fig. 6 shows the contribution of various factors leading to the overall improvement in the power plant efficiency of the pressurized case.
5. Conclusion The simulation results show that a pressurized oxy-combustion power cycle is more efficient than its atmospheric counterpart. By compressing a smaller amount of gas (oxygen) before the combustor, considerable savings can be achieved in the compression work of the exhaust gases after the combustor. The savings in auxiliary power consumption can mainly be attributed to the elimination of air leakage into the boiler. In addition, more of the latent heat available in the flue gas which is rich in water content
This publication has been produced with support from the BIGCCS Centre, performed under the Norwegian research program, Centres for Environment-friendly Energy Research (FME). The authors acknowledge the following partners for their contributions: Aker Solutions, ConocoPhillips, Det Norske Veritas, Gassco, Hydro, Shell, Statoil, TOTAL, GDF SUEZ and the Research Council of Norway (193816/S60). References [1] World Energy Outlook, International Energy Agency, Paris, France, 2011. [2] T. Morita, N. Nakicenovic, J. Robinson, Overview of mitigation scenarios for global climate stabilization based on new IPCC emissions scenarios, Environmental Economics and Policy Studies 3 (2000) 65e88. [3] J.A. Edmonds, P. Freund, J.J. Dooley, The role of carbon management technologies in addressing atmospheric stabilization of greenhouse gases, in: Proceedings of the Fifth International Conference on Greenhouse Gas Control Technologies, Cairns, Australia, 2000, pp. 46e51. [4] Z. Nie, A. Korre, S. Durucan, Life cycle modelling and comparative assessment of the environmental impacts of oxy-fuel and post-combustion CO2 capture, transport and injection processes, Energy Procedia 4 (2011) 2510e2517. [5] M. Kanniche, R. Gros-Bonnivard, P. Jaud, J. Valle-Marcos, J.-M. Amann, C. Bouallou, Pre-combustion, post-combustion and oxy-combustion in thermal power plant for CO2 capture, Applied Thermal Engineering 30 (2010) 53e62. [6] C. Fu, T. Gundersen, Heat integration of an oxy-combustion process for coal fired power plants with CO2 capture by pinch analysis, Chemical Engineering Transactions 21 (2010) 181e187. [7] C. Fu, T. Gundersen, Integrating the compression heat in Oxy-combustion power plants with CO2 capture, Chemical Engineering Transactions 29 (2012) 781e786. [8] M. Gazzino, G. Benelli, Pressurised oxy-coal combustion rankine-cycle for future zero emission power plants: process design and energy analysis, in: Proceedings of Energy Sustainability, ES2008-54268, Jacksonville, Florida, Usa, vol. 2, 2008, pp. 269e278. [9] J. Hong, R. Field, M. Gazzino, A.F. Ghoniem, Operating pressure dependence of the pressurized oxy-fuel combustion power cycle, Energy 35 (2010) 5391e 5399. [10] S. Espatolero, C. Cortes, L.M. Romeo, Optimization of boiler cold-end and integration with the steam cycle in supercritical units, Applied Energy 87 (2010) 1651e1660. [11] F. Franco, O. Bolland, N. Booth, E. Macchi, G. Manzolini, R. Naqvi, A. Pfeffer, S. Rezvani, M.A. Zara, Common framework definition document EBTF, in: F. Franco (Ed.), Politecnico di Milano/Alstom UK, 2009. vol. FP7-energy.2007.5.1e04. [12] G. Pipitone, O. Bolland, Power generation with CO2 capture: technology for CO2 purification, International Journal of Greenhouse Gas Control 3 (2009) 528e534. [13] S. Posch, M. Haider, Optimization of CO2 compression and purification units (CO2CPU) for CCS power plants, Fuel 101 (2012) 254e263. [14] M.A. Habib, S.M. Zubair, Second law based thermodynamic analysis of regenerative-reheat rankine-cycle power plants, Energy 17 (1992) 295e301. [15] F. Cziesla, H. Kremer, U. Much, J.-E. Riemschneider, R. Quinkertz, Advanced 800þ MW steam power plants and future CCS options, in: COAL-GEN Europe 2009, Katowice, Poland, 2009.
122
R. Soundararajan, T. Gundersen / Applied Thermal Engineering 61 (2013) 115e122
[16] C. Fu, T. Gundersen, Power reduction in air separation units for oxycombustion processes based on exergy analysis, Computer Aided Chemical Engineering 29 (2011) 1794e1798. [17] M.S. Bhatt, Effect of air ingress on the energy performance of coal fired thermal power plants, Energy Conversion and Management 48 (2007) 2150e2160.
[18] Pulverized coal oxycombustion power plants, in: Bituminous Coal to Electricity, vol. DOE/NETL-2007/1291, U.S Department of Energy, USA, 2008. [19] V. White, L. Torrente-Murciano, D. Sturgeon, D. Chadwick, Purification of oxyfuel-derived CO2, International Journal of Greenhouse Gas Control 4 (2010) 137e142.