Article pubs.acs.org/EF
Sulfur as a Fuel Source in a Combined Power Cycle Equipped with a Dry Flue Gas Desulfurization System Yasmine Hajar,† Kim McAuley,† and Frank Zeman*,‡ †
Department of Chemical Engineering, Queen’s University, Kingston, Ontario K7L 3N6, Canada Department of Chemistry and Chemical Engineering, Royal Military College, Kingston, Ontario K7K 7B4, Canada
‡
ABSTRACT: This paper investigates the novel use of elemental sulfur as a fuel in a combined cycle power plant, wherein sulfur is oxidized in the combustion chamber to produce work from a gas turbine. Produced sulfur dioxide is then reacted with calcite to produce anhydrite (CaSO4) in a flue gas desulfurization unit. The desulfurization reaction is exothermic, producing heat that can be converted to electricity by passing the exhaust gas through a heat recovery steam generator. The cases studied in this work are a 410 MW natural gas combined cycle, used as a reference case, a sulfur combined cycle (SCC) consisting of a typical industrial gas turbine with heat recovery followed by desulfurization, and a third case of a SCC followed by desulfurization before the heat recovery, thereby producing power from the exothermic desulfurization reaction. Energy calculations of the last case show promising results for converting sulfur to anhydrite through a combined cycle that produces 615 MW of electrical power with CO2 emission of 0.584 kg/kWh, a value lower than the representative value of a new, supercritical coal power plant.
1. INTRODUCTION Recently, sulfur has been largely produced as a byproduct from fossil fuel industries, resulting in a drastic reduction in sulfur mining.1 This increase in production is driven by environmental concerns regarding sulfur dioxide emissions to the atmosphere, leading to the removal of sulfur from petroleum and subsequent storage in solid form. Fort McMurray, Alberta, is an example of this type of sulfur production, where more than 9 million tonnes (Mt) of sulfur are stored in its elemental solid form.2 This is a multi-step process, wherein sulfuric compounds in fuel undergo a first step of hydrogenation, followed by the Claus process to convert it to an elemental form.3,4 This type of storage is by definition temporary because elemental sulfur is far from its thermodynamic ground state and will eventually form other compounds.5 Unlike oil refineries, where sulfur is removed in a precombustion process, coal power plants combust sulfur then mitigate SO2 emissions in a post-combustion process called flue gas desulfurization (FGD).6 To prevent its emission into the atmosphere, the FGD system scrubs the SO2-laden flue gas with a limestone sorbent as follows:
In this study, elemental sulfur is converted to anhydrite in a two-stage process; the first stage is based on the sulfur oxidation reaction shown as follows: S(s) + O2 (g) → SO2 (g)
(2)
In the second stage, SO2 is reacted with a source of calcite, e.g. limestone, to produce anhydrite in a dry FGD system as per reaction 1; the overall reaction would then be CaCO3(s) + S(s) + 3/2O2 (g) → CaSO4 (s) + CO2 (g) ΔH ° = −620.4 kJ/mol
ΔH ° = −324.2 kJ/mol
(1)
The use of limestone as a sorbent, instead of hydrated lime [Ca(OH)2], has been of industrial interest as a result of the lower cost of the limestone and its availability as a raw material. A dry FGD, where limestone is burned with SO2 in a boiler, is considered for this work. More information about various types of FGDs and their applications is available in the literature.7−14 The FGD reaction produces anhydrite (CaSO4) that, if hydrated, is called “synthetic” gypsum to differentiate it from mined gypsum.15 Gypsum is used in the production of Portland cement, plasters, and wallboard. It is also used as a soil conditioner for agriculture fertilization.16 © XXXX American Chemical Society
(3)
The two sub-reactions, reactions 1 and 2, are both exothermic, suggesting that energy recovery is feasible and a thermodynamic study is warranted to evaluate the amount of energy that can be recovered. These reactions present an alternative to stockpiling elemental sulfur that is becoming an environmental burden.17 The objective of this work is not only to convert sulfur to the benign form anhydrite but also to produce electrical power. Sulfuric acid plants already produce electrical power from the thermal energy released during H2SO4 production.18 This analysis considers electricity production from a gas turbine combined cycle that will be referred to as a sulfur combined cycle (SCC). The fact that sulfur combustion produces gaseous SO2 with no other solid biproducts suggested electrical power production in a gas turbine. The SCC is based on the natural gas combined cycle (NGCC), in which the fuel is switched from natural gas (NG) to sulfur. NGCC is used as a basis for the analysis because it is currently the state of the art in large-scale electricity production.19 In typical NGCC plants, the
CaCO3(s) + SO2 (g) + 1/2O2 (g) → CaSO4 (s) + CO2 (g)
ΔH ° = −296.8 kJ/mol
Received: June 14, 2016 Revised: August 26, 2016
A
DOI: 10.1021/acs.energyfuels.6b01454 Energy Fuels XXXX, XXX, XXX−XXX
Article
Energy & Fuels
Figure 1. NGCC (reference case). The black line is the gas stream, and the blue line is the water stream.
Figure 2. SCC with no heat recovery from FGD in the HRSG (case 1). The black line is the gas stream; the blue line is the water stream; and the red line is the heat stream.
Figure 3. SCC with heat recovery from FGD in the HRSG (cases 2a and 2b). The black line is the gas stream, and the blue line is the water stream.
using concentrated sulfuric acid21 or a conventional desiccant.22 Similar to the reference case, a HRSG system is used for extra heat recovery. SO2 removal from flue gas through the FGD system can occur before or after the HRSG unit. In case 1, shown in Figure 2, the HRSG is followed by the FGD system; in cases 2a and 2b, the FGD system is moved upstream of the HRSG, as shown in Figure 3. The objective in cases 2a and 2b is the recovery of the heat released in the FGD (from reaction
turbine is equipped with a heat recovery steam generator (HRSG) of a “three pressure level + reheat” type, as shown in Figure 1, that results in an increase in the total efficiency of the system, defined here as total electrical work produced over total available thermal energy.20 The reference case is compared to three different SCC cases (cases 1, 2a, and 2b). In all three cases, sulfur is oxidized to SO2 (reaction 2) using compressed “bone dry” air. The water vapors in the air could be removed B
DOI: 10.1021/acs.energyfuels.6b01454 Energy Fuels XXXX, XXX, XXX−XXX
Article
Energy & Fuels Table 1. Main Assumptions Used for Modeling the Processes number
assumptions
1
air: dry molar fraction (%), O2, 20.95; N2, 78.08; CO2, 0.04; and Ar, 0.93; for the reference case, the air has a relative humidity of 60%; for case 1 and cases 2a and 2b, the air is assumed to be dry NG: molar fraction (%), CH4, 89.0; C2H6, 7.0; C3H8, 1.0; C4H10, 0.1; N2, 0.9; and CO2, 2.0; the standard enthalpy of combustion for the NG is 837.6 kJ/mol (LHV) sulfur: specific heats of liquid and solid sulfur are provided from Table 9 in ref 5; ΔH of fusion is equal to 1.61 kJ/mol, at a melting point of 119.6 °C turbine and compressors: the work performed by the turbine is positive; a total of 19.4% of the air exiting the compressor is used to cool the inlet to the gas turbine combustion and energy: adiabatic combustion is assumed in the combustion chamber and FGD; kinetic and potential energy changes are negligible in the evaluation of energy flow; ideal heat capacities are used in the evaluation of enthalpies where the effect of pressure is neglected; enthalpies of formation for species are used from ref 33; constants in the evaluation of specific heat are used from ref 34; complete combustion is assumed in the combustion chamber and the FGD unit HRSG/steam turbines: efficiency of heat transfer from flue gas to water in the HRSG is 35.6% found from ref 23 in the reference case and is used in tuning cases 1 and 2 to find steam turbine water work, where the temperature of water/steam in the different streams matches the reference case by changing the mass flow; mass fractions of steam entering HP, IP, and LP turbines are 70, 84, and 10%, respectively; enthalpy values for water/steam were obtained from the steam table in ref 34
2 3 4 5
6
Table 2. Equations Used for Modeling the NGCC and SCC Processes number
equation name
equation
ΔḢk = 1
∑ ΔḢi(Tk) = ∑ ni̇ ∫ i
enthalpy of the gas mixture in the kth stream
where
Cp,igi dT
= Ai + Bi T + CiT 2 + DiT −2
R
standard enthalpy change as a result of combustion (LHV)
° ΔḢcomb,unit =
3
total standard enthalpy released by reactions
−ΔḢ T° = −
4
enthalpy change between inlet and outlet streams for a unit
ΔḢ unit =
2
298.15
i
Cp,igi
Tk
∑ ni̇ ,unit(viΔHf,°i) i ° ΔḢcomb,unit
∑ unit = CC,FGD
∑ ΔḢk ,out − ∑ ΔḢk ,in k ,out
k ,in
° ΔḢcomb,unit
5
energy balance on the combustion chamber and FGD
ΔḢ unit +
6
power of compressor
̇ Wcomp = −ΔḢcomp
7
power of turbine
̇ = −ΔḢ turb Wturb
8
power of steam turbines
̇ Wsteam =−
9
net work from the compressor and turbine
̇ = Wturb ̇ + Wcomp ̇ Wnet
10
total work by the combined cycle
̇ + Wcomp ̇ ̇ ẆT = Wturb + Wsteam
11
electrical efficiency
ηth =
12
fraction of enthalpy from flue gas converted to work in the HRSG
13
sulfur enthalpy in stream 4
=0
ΔḢ unit
∑ unit = HP,IP,LP
ẆT −ΔḢ T°
ηHRSG = {ṁ water [x HP(H8 − H7) + x IP(H11 − H9) + x LP(H11 − H10)]} ̇ − ΔḢ 6) /(ΔH14 ΔH4̇ = nṠ
∫T
T25 ° C
119.6 ° C
Cp,S(s) dT + nṠ hm,S(119.6 °C) + nṡ
∫T
T119.6 ° C
Cp,S(l) dT
160 ° C
single NG turbine and one HRSG turbine, reducing the power output approximately in half. The reference case NGCC plant is also similar to a single gas turbine combined cycle used by Amrollahi et al.24 Assumptions related to the case study simulations are shown in Table 1, and the thermodynamic equations used are provided in Table 2. Stream mass flow rates, temperatures, pressures, and compositions for the reference case are presented in Table 3. As shown in Figure 1, air from stream 1 is pressurized in the compressor before entering the combustion chamber at a temperature of 417 °C. NG in stream 4 enters the combustion chamber after preheating to 160 °C by a heat exchanger as per Manzolini et al. In the adiabatic combustion chamber, the exothermic combustion reaction of NG increases the temperature to a combustor outlet temper-
1) in the HRSG. The turbine outlet temperature (TOT) was decreased in case 2b to increase energy recovery because the FGD provides sufficient heat to maintain HRSG inlet conditions, albeit with a reduction in bleed air (streams 7 and 12). Thermodynamic calculations based on this process configuration can be used to justify future research into energy recovery and anhydrite production from sulfur.
2. RESULTS AND DISCUSSION 2.1. NGCC Plant (Reference Case). The NGCC plant used in the analysis is based on the NGCC case study plant considered by Manzolini et al.23 Their reference plant involves two NG turbines and a single HRSG turbine that produces 829.9 MW of electrical power. The current analysis considers a C
DOI: 10.1021/acs.energyfuels.6b01454 Energy Fuels XXXX, XXX, XXX−XXX
Article
Energy & Fuels
Table 3. Mass Flow Rate, Pressure, Temperature, and Composition of Main Streams in NG Reference Case Adapted from Ref 20 composition (mol %) stream 1 2 3 4 5 6 7 8 9 10 11 12 13 14
ṁ (kg/s)
T (°C)
650 15 523 417.5 15.3 10 15.3 160 539 1443 (COT) 665 608 76.9 559.5 76.9 337.7 92.5 561.0 21.9 299 109 32.2 6.58 230 110 32.2 665 86.8 total power output, Ẇ (MW) 410
P (bar)
Ar
N2
O2
CO2
H2O
1 0.92 77.31 20.74 0.04 0 18.16 0.92 77.31 20.74 0.04 0 70 see NG composition in Table 1 70 17.6 0.88 73.72 11.06 4.88 10.04 1.04 0.89 73.72 10.48 3.97 8.34 120.9 100 28 100 22.96 100 3.52 100 0.048 100 28 100 1 100 1.01 0.89 74.40 12.40 3.97 8.34 net electricity efficiency, η (%) CO2 emission (kg of CO2/kWh) 58.1 0.36
Table 4. Mass Flow Rate, Pressure, Temperature, and Composition of Main Streams of SCC with Heat Recovery for FGD from the HRSG (Case 1) composition (mol %) stream 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
ṁ (kg/s)
T (°C)
650 15 523 417.5 72.0 25 72.0 160 596 1328 (COT) 722 608 56.2 559.5 56.2 337.7 67.4 561 8.0 299 75.4 32.2 4.8 230 80.2 32.2 722 87 225 25 306 900 641 900 total power output, Ẇ (MW) 353
P (bar)
Ar
N2
O2
CO2
H2O
1 18.16 1.01 1 17.6 1.04 120.9 28 22.96 3.52 0.048 28 1 1.01 1 1 1
0.93
78.08
20.95
0.04
0
SO2 0
elemental sulfur (100%) 0.93 0.93
78.08 78.08
0.93
78.08
8.41 10.92
0.04 0.04
0 0 100 100 100 100 100 100 100 0
12.54 10.03
10.92 0.04 10.03 calcium carbonate, CaCO3 (100%) calcium sulfate, CaSO4 (100%) 0.98 82.20 6.22 10.60 0 0 net electricity efficiency, η (%) CO2 emission (kg of CO2/kWh) 25.3 1.013
ature (COT) of 1443 °C. A fraction (19.5%) of the compressed air bypasses the combustion chamber via stream 2′ to lower the turbine inlet temperature (TIT) to a value of 1270 °C. The gas mixture expands in the turbine, producing work that is converted to electricity. To increase the efficiency of the cycle, stream 6 is fed to a three-pressure level HRSG where the flue gas heats the water stream in a counter-current fashion. All of the liquid water that enters the HRSG (stream 13) is pumped and evaporated to produce low-pressure (LP) steam at 3.5 bar and 299 °C. Then, 10% of this steam is fed to the LP turbine via stream 10. A stream containing 70% of the inlet water is pressurized to a high pressure (HP) of 121 bar. It absorbs a large fraction of the heat from the flue gas, producing stream 7 at 560 °C. Stream 8 exits the HP turbine at 338 °C and a pressure of 28 bar. Stream 8 is reheated in the HRSG, and 91% of it is combined with the
intermediate-pressure (IP) stream (containing 20% of the inlet water) to form stream 9 at 561 °C and 23 bar. Stream 9 enters the IP turbine and produces work as it expands to give stream 11, which exits the turbine at 32 °C. Stream 14 is 9% of stream 8 used to preheat NG. Solving the equations in Table 2 gives a predicted total power output of 410 MW. This power production is slightly less than half of Manzolini’s value (415 MW)23 and higher than the value reported by Amrollahi et al. (384.4 MW) for a similar single gas turbine plant.24 The overall thermal efficiency of 58.1% based on the lower heating value (LHV) for the plant in Figure 1 is similar to the value reported by Manzolini et al. (58.3%) and higher than that of Amrollahi et al. (56.4%). These results suggest that the NGCC plant model is a reasonable reference case for calculations involving the SCC plant models described below as cases 1, 2a, and 2b. The cases considered are D
DOI: 10.1021/acs.energyfuels.6b01454 Energy Fuels XXXX, XXX, XXX−XXX
Article
Energy & Fuels
Table 5. Mass Flow Rate, Pressure, Temperature, and Composition of Main Streams of SCC with Heat Recovery from FGD in the HRSG (Case 2a) composition (mol %) stream 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
ṁ (kg/s)
T (°C)
650 15 523 516 72.0 25 72.0 160 595 1328 (COT) 722 608 297 15 225 15 306 900 1255 900 1442 608 197 15 158 559.5 158 337.7 190 138 22.6 299 213 32.2 13.6 230 226 32.2 1442 87 total power output, Ẇ (MW) 537
P (bar)
Ar
N2
1 18.16 1.01 1 17.6 1.1 1 1 1 1.06 1.04 1 120.9 28 22.96 3.52 0.048 28 1 1
0.93 0.93
78.08 78.08
0.93 0.93 0.93
78.08 78.08 78.08
0.96 0.96 0.93
CO2
H2O 0 0
8.41 0.04 10.92 0.04 20.95 0.04 calcium carbonate, CaCO3 (100%) calcium sulfate, CaSO4 (100%) 80.96 10.64 7.43 80.64 11.79 6.61 78.08 20.95 0.04
0.96 80.64 net electricity efficiency, η (%) 38.4
all based on the same inlet air mass flow rate (stream 1) for the reference case shown in Table 3. 2.2. Sulfur Combined Cycle with no Heat Recovery from FGD (Case 1). Consider the SCC system shown in Figure 2 (case 1) where the difference from the reference case is that sulfur is the fuel rather than NG and that a FGD system is added to desulfurize the flue gas after the HRSG. Sulfur is first heated to 160 °C and then is injected into the combustion chamber as a liquid. The enthalpy of sulfur in stream 4 can be calculated as per Table 2. To match a COT of 1443 °C and match the reference case value, the mass flow of sulfur required for combustion was 72 kg/s, 5 times greater than that of NG required to heat the 650 kg/s of compressed air. The gas mixture expands in the turbine and drops in temperature to a TOT in stream 6 to match the reference case. The gas turbine produces 253 MW of work (see eq 9 in Table 2), a value slightly smaller than that in the reference case, owing to the difference in the flue gas composition. All of the relevant mass flows are presented in Table 4. The electrical efficiency of the HRSG from the reference case (ηHRSG = 35.6%; see eq 12 in Table 2) was used to calculate the flow rate of feedwater ṁ w required to match temperatures, pressures, and flow rate fractions (xHP, xIP, and xLP) within the HRSG. As a result, the steam turbines produced similar power Ẇ HRSG as in the reference case (135 MW compared to 138 MW). The HRSG output gas mixture, at a temperature of 87 °C set to the reference case value, enters the FGD as stream 14. A feed of calcite is added in stream 15 to react with SO2 through reaction 1, producing CaSO4(s) in stream 16. The operational temperature for this reaction is taken as 900 °C.25 This is a conservative estimate that ensures calcination of the limestone. The heat released from reaction 1 is insufficient to reach 900 °C, and an additional 96 MWth is required. This heat, shown as Q̇ in Figure 2, can be supplied from the steam in the
O2
20.95 0.04 20.95 0.04 elemental sulfur (100%)
11.79
0 0 0
SO2 0 0
12.54 10.03 0
0 0 0 0 0 0 100 100 100 100 100 100 100 6.61 0 0 CO2 emission (kg of CO2/kWh) 0.670
HRSG, which would lower the power produced from 135 to 101 MW. The total (electrical) power output is then 353 MW, with a total electrical efficiency of 25.7%. CO2 emitted from this process is 1.013 kg/kWh, a value higher than the reference value for a supercritical pulverized coal plant (0.762 kg/ kWh).26 Note that, in this case, there is no heat recovery from the FGD. 2.3. SCC with FGD Heat Recovery and TOT Matched to the Reference Case (Case 2a). To increase the heat recovery and reduce the CO2 emissions, the FGD could be located upstream of the HRSG, as shown in Figure 3. The flue gas coming out of the gas turbine in stream 6 is at 608 °C; therefore, if the flue gas is fed directly to the FGD, less energy will be required to heat the materials to the reaction temperature (900 °C). This proposed change makes the process more efficient in that all thermal energy needed is provided by reaction 1. Energy balance calculations show that the hot gas in stream 6, when reacted with calcite from stream 8, results in FGD temperatures that exceed the design value of 900 °C. Therefore, excess ambient air is added through stream 7 to regulate the temperature at the design value. The flue gas in stream 10, which exits the FGD, is also mixed with ambient air (stream 12) to lower the gas temperature entering the HRSG and match the reference case value (608 °C). The resulting mass flow rate of the flue gas in stream 11 is significantly higher than in the reference case, almost an order of magnitude. This results in a proportional increase of the mass flow of water required in the HRSG to match temperature conditions of the reference case. As shown in Table 5, the power output of the HRSG increases to 285 MW and the total power output is now 537 MW. The resulting CO2 emissions decrease to 0.670 kg/kWh. 2.4. SCC with Heat Recovery and TOT Reduced from the Reference Case (Case 2b). One possible method for improving the performance of the SCC system of case 2a is to E
DOI: 10.1021/acs.energyfuels.6b01454 Energy Fuels XXXX, XXX, XXX−XXX
Article
Energy & Fuels
Table 6. Mass Flow Rate, Pressure, Temperature, and Composition of Main Streams of SCC with Heat Recovery from FGD in the HRSG and a Lower TOT (Case 2b) composition (mol %) stream 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
ṁ (kg/s)
T (°C)
650 15 523 516 72.0 25 72.0 160 595 1328 (COT) 722 457 178 15 225 15 306 900 1136 900 1260 608 124 15 138.6 559.5 138.6 337.7 166.3 138 19.8 299 186.1 32.2 11.9 230 198 32.2 1260 87 total power output, Ẇ (MW) 615
P (bar)
Ar
N2
1 18.16 1.01 1 17.6 1.1 1 1 1 1.06 1.04 1 120.9 28 22.96 3.52 0.048 28 1 1
0.93 0.93
78.08 78.08
0.93 0.93 0.93
78.08 78.08 78.08
O2
CO2
H2O
20.95 0.04 20.95 0.04 elemental sulfur (100%)
8.50 0.04 10.92 0.04 20.95 0.04 calcium carbonate, CaCO3 (100%) calcium sulfate, CaSO4 (100%) 81.41 9.04 8.58 81.22 9.73 8.08 78.08 20.95 0.04
0.97 0.97 0.93
0.97 81.22 net electricity efficiency, η (%) 44.0
SO2
0 0
0 0
0 0 0
12.45 10.03 0
0 0 0 0 0 0 100 100 100 100 100 100 100 8.08 0 0 CO2 emission (kg of CO2/kWh) 0.584
9.73
reduce the energy available for recovery. For the model above, both effects result in linear reductions in power output and increases in CO2 emissions, with results presented in Table 7.
increase the output of the gas turbine by lowering the TOT to a value typical for a simple Brayton cycle.27 Abam et al. reported a TOT value of 457 °C, which is considerably lower than the 608 °C value specified in the NGCC reference case. Changing the TOT to 457 °C in the energy balance calculations leads to an increased gas turbine power of 366 MW. The lower gas temperature in stream 6 reduces the cooling airflow rate required in stream 7 to obtain a FGD temperature of 900 °C (i.e., 31.5 kg/s compared to 117 kg/s in case 2a). If the gas mixture in stream 10 is mixed with 374 kg/s of additional ambient air, the reference case temperature of 608 °C is obtained for stream 11, which has a total mass flow rate of 672 kg/s. To match the HRSG temperatures, a total inlet flow rate of water in stream 19 is ṁ w = 198 kg/s, which is lower than 226 kg/s of inlet water in case 2a, as shown in Table 6. The power output from the HRSG is lower than that in case 2a, but the total power output from the SCC system increases to 615 MW. The thermal efficiency increases to 44.0%, and the CO2 emission is 0.584 kg/kWh, an improved value lower than the representative value for coal power (0.762 kg/kWh).26 2.5. Sensitivity Analysis for SO2 Condensation. The model presents the case where the efficiency of reaction 1 is 100% and there is no risk of SO2 condensing as sulfuric acid in the HRSG. Vapor−liquid equilibrium between SO2 and H2O is complex and a function of trace gases.28 One method of reducing this risk is to apply a desiccant system to all inlet air to reduce relative humidity.22 On the basis of the enthalpy of vaporization at 15 °C, such a system would consume 33 MW and increase CO2 emissions to 0.615 kg/kWh. More conventional alternatives would be to increase the limestone feed to the FGD reactor and/or increase the outlet temperature (T20) from the HRSG. Adding limestone to the FGD reactor would reduce the heat released in the FGD, through additional endothermic load, and, therefore, the energy available to the HRSG. An increase of the outlet temperature would similarly
Table 7. Sensitivity Analysis for SO2 Condensation rate of change Ẇ totala
CO2 emissionb
Ẇ totalc
CO2 emissiond
0−27
−0.945
0.00696
589
0.774
87−207
−0.462
0.000482
559
0.642
range excess lime (mol %) T20 (°C)
limit values
a
In units of MW/mol % and MW/°C. bIn units of (kg/kWh)/mol % and (kg/kWh)/°C. cIn units of MW. dIn units of kg/kWh.
Excess lime has the added complication of increasing the CO2 partial pressure and, therefore, calcination temperature in the FGD. Combining the two methods (25% excess lime and T20 = 207 °C) reduces the power output to 541 MW with 0.828 kg of CO2/kWh. Adding a desiccant system would further reduce power to 514 MW and emissions would increase to 0.871 kg of CO2/kWh. While these are significant reductions, it is important to note that no energy recovery is considered from the solids (anhydrite), leaving the FGD at 900 °C. The available thermal energy here is 268 MWth, which would produce 95 MW of electricity if injected into the HRSG. In addition, exhaust gases could be pass through these solids to reduce CO2 emissions through carbonation of unreacted lime. 2.6. Fugitive Emission Considerations. There are several sources of fugitive emissions that may affect the overall climate benefit and economics of this proposed technology. An exhaustive life cycle analysis (LCA) is not warranted at this stage; however, some obvious sources can be considered independently to gauge their effects. A first estimate of fugitive SO2 emissions can be derived from LCA work on the NGCC F
DOI: 10.1021/acs.energyfuels.6b01454 Energy Fuels XXXX, XXX, XXX−XXX
Article
Energy & Fuels
the 251 MWth required to maintain the FGD at 900 °C may be available as waste heat from another source. Also, heat recovery from hot FGD anhydrite could have been a potential source of extra energy (268 MWth). Cases 2a and 2b have markedly improved power outputs because no external heat source for the FGD is required as a result of the high temperature of the exit stream from the gas turbine, directly entering the FGD. The exit temperature from the gas turbine is lower in case 2b compared to case 2a, resulting in two effects: an increase in the gas turbine power output and a decrease in the HRSG power output, with the net result being a higher overall power output. The HRSG steam temperatures are matched to the reference case in the three SCC cases considered. The HRSG power output in case 1 is lower than the reference case (101 MW compared to 138 MW); however, as a result of the increased mass flow rates in case 2a and, to a lesser extent, in case 2b, the power output from the HRSG increases to 285 and 249 MW, respectively, increasing the total power outputs. Another criterion considered in this study is the CO2 emissions per kilowatt hour. In comparison to the NGCC value of 0.36 kg/kWh, the SCC system with no FGD heat recovery (case 1) produced a total of 1.013 kg/kWh, whereas the system with FGD heat recovery produced 0.670 kg/kWh when the turbine outlet temperature was matched to the reference case and an improved value of 0.584 kg/kWh when the turbine outlet temperature was reduced to 457 °C. These values are below the representative value of 0.762 kg of CO2/ kWh for new supercritical coal plants. Overall, this study shows that the case 2b SCC system with FGD heat recovery shows promise as an electrical power plant with lower CO2 emissions than a coal gasification combined cycle plant. This study compares three possible systems for electricity production from sulfur using a NGCC system as a reference case. In all three cases, the sulfur combustion unit and gas compressor/turbine system are combined with a FGD unit and a HRSG system. The first case considered has relatively low efficiency because it does not consider energy recovery from the exothermic SO2 conversion to anhydrite in the FGD unit. Cases 2a and 2b obtain improved performance by placing the FGD unit before the HRSG, so that additional energy recovery can occur. Case 2b is more efficient than case 2a because more energy is recovered by the gas turbine and less cooling air is used in the FGD to achieve the specified operating temperatures. The CO2 emissions from case 2b are the lowest among the SCC cases considered. Although the CO2 emissions are higher than from a comparable NGCC plant, they are lower than a typical modern coal gasification combined cycle plant. More importantly, long-term management of sulfur piles is eliminated, and in the case of Alberta, with annual sulfur production approaching 3 Mt, ∼800 MW of power can be generated. This analysis is a step toward a longer term goal of converting waste sulfur to benign anhydrite and electrical energy with low CO2 emissions. Appropriate process design and optimization will be required to determine whether increased efficiency of power production and reduced CO2 emissions can be obtained in comparison to the values presented here. Further research is also required to obtain improved knowledge about any impediments (e.g., construction materials, reaction kinetics, and achievable SO2 emissions) to practical operation of this type of SCC system. Further criteria, such as capital cost, capacity factor, and the need for
system, which provides a value of 2.82 mmol of CH4/kWh from electricity generation.29 This value can be converted to 0.18 g of SO2/kWh or 0.04% of fuel feed. This would increase the cost of electricity by a negligible amount (∼7.2 × 10−6 $/kWh) using the latest United States Environmental Protection Agency (U.S. EPA) market prices ($40/t of SO2).30 There is no climate risk associated with SO2 emissions as a result of their rapid return to the surface environment as acid rain. The electricity generation system proposed here differs from conventional systems in that a chemical reagent is hauled to site and solid product must be shipped off site. The limestone (CaCO3) imported will also require size reduction, a parasitic energy load. Using the industry standard Bond equation,31 this would consume 5.96 kWh/t of CaCO3 from boulders to 500 μm. Given that there are 1.437 kg of CaCO3 needed/kWh, this would reduce the power output by 0.009 kWh/kWh or ∼1%. CaCO3 and CaSO4 would produce fugitive CO2 emissions during haulage. These have been estimated at 1.335 kg of CO2/ km loaded (40 tonnes) and 0.859 kg of CO2/km empty.32 For discussion purposes, we cite the source and sink at 100 km from the power plant. Therefore, a round trip would generate 219.4 kg of CO2 and bring 40 000 kg of CaCO3 to the site (5.485 × 10−3 kg of CO2/kg of CaCO3). Consuming 1.437 kg of CaCO3/kWh would emit 0.008 kg of CO2/kWh from CaCO3. Anhydrite being heavier, 1.36 times, would add 0.011 kg of CO2/kWh for a total of 0.019 kg of CO2/kWh or a 3.3% increase. This suggests that optimization would be helpful but not critical to the siting of a SCC system. Additional fugitive emissions may result from post-processing of CaSO4.
3. CONCLUSION The performances of the NGCC reference case, the SCC with no FGD heat recovery (case 1), and the SCC with FGD heat recovery (cases 2a and 2b) are compared in Table 8. The Table 8. Results for the Different Cases ṁ fuel (kg/s) TOT (°C) Ẇ gas turbine (MW) Ẇ HRSG (MW) Ẇ total (MW) ηgas turbine (LHV base) (%) ηtotal (LHV base) (%) CO2 emission (kg/kWh)
reference case
case 1
case 2a
case 2b
15.3 608 272 138 410 38.5 58.1 0.360
72.0 608 252 101 354 37.8 25.3 1.013
72.0 608 252 285 537 37.8 38.4 0.670
72.0 457 366 249 615 54.7 44.0 0.584
NGCC reference case shows a power generation efficiency of 58.1%, comparable to the reference NGCC plant. The predicted overall efficiency for the SCC with no FGD heat recovery, using the assumptions in Table 1, is 20.5%. For the SCC with FGD heat recovery cases, the power generation efficiencies are of higher values (i.e., 37.7% in case 2a and 43.6% in case 2b). All of the calculated SCC efficiencies are lower than that of the NGCC reference case, owing to the heat required to operate the FGD in case 1 and the restriction on the FGD temperature limit of 900 °C, to maintain compatibility with the reference case HRSG. In comparison to the total power output of the reference case, 410 MW, case 1 results in a lower value power output of 354 MW. Cases 2a and 2b produce higher power outputs of 537 and 615 MW, respectively. The low power output computed for case 1 is a worst-case value because G
DOI: 10.1021/acs.energyfuels.6b01454 Energy Fuels XXXX, XXX, XXX−XXX
Article
Energy & Fuels new technology development, can be considered but are not presented in the current work.
■
AUTHOR INFORMATION
■
Corresponding Author
*Telephone: +001-613-541-6000. E-mail:
[email protected]. Notes
REFERENCES
(1) United States Geological Survey (USGS). Mineral Commodity Summaries 2014-Sulfur; USGS: Reston, VA, 2014. (2) Natural Resources Canada. Canadian Mineral Yearbook (CMY)Sulphur; Natural Resources Canada: Ottawa, Ontario, Canada, 2009. (3) Song, C. Catal. Today 2003, 86, 211−263. (4) Elsner, M. P.; Menge, M.; Müller, C.; Agar, D. W. Catal. Today 2003, 79−80, 487−494. (5) Meyer, B. Chem. Rev. 1976, 76 (3), 367−388. (6) Srivastava, R. K.; Jozewicz, W. J. Air Waste Manage. Assoc. 2001, 51 (12), 1676−1688. (7) Lee, K.; Jung, J.; Keel, S.; Yun, J.; Min, T.; Kim, S. Sci. Total Environ. 2012, 429, 266−271. (8) Fuertes, A. B.; Velasco, G.; Fuente, E.; Alvarez, T. Fuel Process. Technol. 1994, 38 (3), 181−192. (9) Katolicky, J.; Jicha, M. Chem. Eng. Technol. 2013, 36 (1), 156− 166. (10) Dam-Johansen, K.; Østergaard, K. Chem. Eng. Sci. 1991, 46 (3), 827−837. (11) Wang, C.; Jia, L.; Tan, Y.; Anthony, E. Fuel 2010, 89 (9), 2628− 2632. (12) Chen, J.; Yao, H.; Zhang, L. Fuel 2012, 102, 386−395. (13) Laursen, K.; Duo, W.; Grace, J. R.; Lim, J. Fuel 2000, 79 (2), 153−163. (14) Siagi, Z. O.; Mbarawa, M.; Mohamed, A. R.; Lee, K. T.; Dahlan, I. Fuel 2007, 86 (17−18), 2660−2666. (15) United States Geological Survey (USGS). Gypsum; USGS: Reston, VA, 2014. (16) Olson, D. W. United States Geological Survey (USGS): Reston, VA, 2001; pp 35.31−35.35. (17) United States Geological Survey (USGS). USGS: Reston, VA, 2014; p 156. (18) Rappold, T. A.; Lackner, K. S. Energy 2010, 35 (3), 1368−1380. (19) Pequot Publishing, Ltd. Gas Turbine World Handbook; Pequot Publishing, Ltd.: Fairfield, CT, 2009; Vol. 27. (20) Horkeby, K. Simulation of Heat Recovery Steam Generator in a Combined Cycle Power Plant. Thesis, Technical University Linkoping, Linkoping, Sweden, 2012. (21) Wilson, R. E. J. Ind. Eng. Chem. 1921, 13 (4), 326−331. (22) Madhiyanon, T.; Adirekrut, S.; Sathitruangsak, P.; Soponronnarit, S. Chem. Eng. Process. 2007, 46 (4), 282−290. (23) Manzolini, G.; Macchi, E.; Binotti, M.; Gazzani, M. Int. J. Greenhouse Gas Control 2011, 5 (2), 214−225. (24) Amrollahi, Z.; Ystad, P. A. M.; Ertesvåg, I. S.; Bolland, O. Int. J. Greenhouse Gas Control 2012, 8, 1−11. (25) García-Labiano, F.; Rufas, A.; de Diego, L. F.; Obras-Loscertales, M. d. l.; Gayán, P.; Abad, A.; Adánez, J. Fuel 2011, 90 (10), 3100− 3108. (26) Metz, B.; Davidson, O.; de Coninck, H.; Loos, M.; Meyer, L. Carbon Dioxide Capture and Storage; Cambridge University Press: Cambridge, U.K., 2005. (27) Abam, D. P. S.; Moses, N. N. GJRE-A 2011, 11 (1), 37−44. (28) Beutier, D.; Renon, H. Ind. Eng. Chem. Process Des. Dev. 1978, 17 (3), 220−230. (29) Spath, P. L.; Mann, M. K. Life Cycle Assessment of a Natural Gas Combined-Cycle Power Generation System; National Renewable Energy Laboratory (NREL): Golden, CO, 2000; p 56, NREL/TP-570-27715. (30) United States Environmental Protection Agency (U.S. EPA). Market Activity Figures; U.S. EPA: Washington, D.C., 2015, https:// www3.epa.gov/airmarkets/progress/reports/market_activity.html (accessed Aug 26, 2016). (31) Bond, F. C. Crushing and Grinding Calculations; Allis-Chalmers Manufacturing Company: Milwaukee, WI, 1961.
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors acknowledge the financial support of the Natural Sciences and Engineering Research Council through Grant 424223-2011.
■
th = thermal turb = turbine u = operating unit w = water
NOMENCLATURE
Abbreviations
COT = combustor outlet temperature FGD = flue gas desulfurization HP = high pressure HRSG = heat recovery steam generator IP = intermediate pressure LHV = lower heating value LP = low pressure NG = natural gas NGCC = natural gas combined cycle PC = pulverized coal SCC = sulfur combined cycle TIT = turbine inlet temperature TOT = turbine outlet temperature Symbols
Cp = specific heat (J mol−1 K−1) h = enthalpy ΔH° = standard enthalpy change (J/mol) ΔḢ ° = standard enthalpy change rate (W) ṁ = mass flow (kg/s) Mt = million tonne η = efficiency ṅ = molar flow (mol/s) P = pressure (bar) Q̇ = heat input rate (W) R = ideal gas constant (J mol−1 K−1) T = temperature (°C) v = coefficient of reaction Ẇ = work rate or power (W) x = fraction
Subscripts
CC = combustion chamber comb = combustion comp = compressor f = formation HP = high pressure HRSG = heat recovery steam generator i = species ig = ideal gas IP = intermediate pressure k = stream number l = liquid LP = low pressure m = melting S = sulfur s = solid steam = steam in the HRSG H
DOI: 10.1021/acs.energyfuels.6b01454 Energy Fuels XXXX, XXX, XXX−XXX
Article
Energy & Fuels (32) Eloneva, S.; Puheloinen, E. M.; Kanerva, J.; Ekroos, A.; Zevenhoven, R.; Fogelholm, C. J. J. Cleaner Prod. 2010, 18 (18), 1833−1839. (33) Smith, J. M.; Van Ness, H. C. Introduction to Chemical Engineering Thermodynamics, 4th ed.; McGraw-Hill: New York, 1987. (34) Lide, D. R. Standard thermodynamic properties of chemical substances. In Handbook of Chemistry and Physics, 85th ed.; CRC Press: Boca Raton, FL, 2015.
I
DOI: 10.1021/acs.energyfuels.6b01454 Energy Fuels XXXX, XXX, XXX−XXX
