Energy recovery from effluents of supercritical water oxidation reactors

1 Energy Recovery from Effluents of Supercritical Water Oxidation 1 Reactors 2 Yoana García-Rodríguez, Fidel A. Mato*, Alexandra Martín, M. Dolores Bermejo and 3 M. José Cocero 4 5 High Pressure Processes Group, Department of Chemical Engineering and 6 Environmental Technology. EII Sede Mergelina. University of Valladolid. 47011 7 Valladolid, SPAIN 8 * Corresponding author, TEL: +34-983423169, e-mail: [email protected] 9 2 Abstract 10 Supercritical Water Oxidation (SCWO) reactors can process waste effluents achieving high 11 conversions, but the required extreme pressure and temperature operational conditions entail 12 high-energy operational expenditure. SCWO has the potential to be considered a clean energy 13 generation process, as the process effluent is a high temperature, high pressure stream with a high 14 enthalpy content that can be converted to heat and shaft work. This ensures the self-sustained 15 reaction and can generate excess shaft power to drive both the high-pressure pump and the air 16 compressor. On the contrary, an efficient heat and power recovery from SCWO reactors outlet 17 streams using conventional procedures presents several problems. First, Rankine cycles impose 18 indirect heat transfer to the working fluid and are unable to recover the pressure energy and 19 second, direct expansion of the effluents entails costly development of specific, efficient 20 expansion equipment. 21 In this work, we investigate the options for energy recovery of SCWO reactors coupled with 22 commercial gas turbines (GT). SCWO outlet streams are mainly composed of water, nitrogen and 23 carbon dioxide. These operating values nearly resemble the well-known and already-implemented 24 GT steam injection procedures. The temperature of the flue gases (approx. 500 ºC) and the direct 25 shaft work usage offers adequate energy integration possibilities for both feed preheating and 26 compression. The wide range of commercially available GT sizes enables process scaling. 27 Keywords: SCWO, shaftwork, energy recovery, gas turbine (GT), steam injection, 28 simulation. 29 3 1. Introduction 30 Supercritical Water Oxidation (SCWO) is an intensive energy process to eliminate 31 organic wastes. For many years the process has been developing technical solutions to 32 achieve results for corrosion and plugins problems [1, 2]. Although its industrial 33 development progresses slowly, in 2013 two industrial plants for chemical weapons and 34 sludge treatment were under construction [3]. 35 One of the SCWO challenges is the energy recovery to get shaft work and heat in 36 order to get net energy [4]. Existing literature on SCWO process focusing on clean energy 37 production has been reviewed. Most of the practical development is based on recovering 38 the heat released by waste oxidation and generating steam. Many theoretical works point 39 that the process would be much more efficient if the compression energy could be 40 recovered as work. The efficient thermal and pressure energy recovery will open the 41 opportunity to use SCWO as an efficient and clean energy production processes from 42 wastes or biomass [5]. 43 Depending on the SCWO process different alternatives can be applied for heat 44 recovery. Conventional tubular reactors are thin tubes, with evident plugging problems 45 from solid precipitation. In practice, industrial plants work with two reactors, one under 46 operation and the other undertaking the cleaning of deposited solids. Even isolated tubular 47 reactor loss energy by the long surface area, and furthermore cleaning is a highly energy 48 and time consuming step. These reactors can operate with air or oxygen, both alternatives 49 work properly. Oxygen is the most usual oxidant to reduce the energy consumption of the 50 air compressor. The oxidation by oxygen requires lower reactor volume and less work to 51 compress the liquid oxygen than the gas air, but the oxygen cost is the limit issue. The 52 election depends on the economic balance. For operation below ignition temperature, 53 4 reaction time is about several minutes and the reactor volume is minimized by the use of 54 oxygen. Air is more conventional oxidant but requires higher reactor volume associated 55 to nitrogen. To implement the use of air as oxidant the reactor volume could be minimized 56 by the use of faster kinetic and by recovering the energy associated to the compression if 57 the work from effluent depressurization could be retrieved by a turbine. 58 The reactor effluent energy can be recovered by a Closed Rankine Cycle through 59 indirect heat transfer to a working fluid but the process is still highly energy demanding 60 [6]. 61 For operation at temperatures above the ignition, supercritical water oxidation with 62 hydrothermal flame as internal heat source allows to use air or oxygen and the faster 63 kinetics minimizes the reactor volume. The operation under hydrothermal flames allows 64 total oxidation of the waste within milliseconds residence times, which opens the 65 possibility of developing small combustors to produce high-pressure gas/vapor streams. 66 The application of hydrothermal flames opens a wide field for the production of energy 67 from wastes [7]. The cooled wall reactor developed at University of Valladolid is the only 68 reactor prototype currently in operation with hydrothermal flame as internal heat source 69 that produces a reduced liquid effluent with dissolved solids and a high-pressure and high70 temperature effluent at 600-650 ºC and 23 MPa, that is able to produce work and thermal 71 energy in a more efficient way that the below ignition tubular reactors effluent [8]. 72 Even when the option of direct expansion of the effluent is, by far, the most 73 energetically efficient, it will be not applicable in the short term. This is mainly due to the 74 fact that the composition of the effluent (50-80% mole of water, carbon dioxide and 75 nitrogen if air is used as oxidant) makes it not suitable for expansion in a conventional 76 turbine. This composition makes the effluent one of intermediate characteristics between 77 5 the pure water used in steam turbine and the flue gases, products of combustion used in 78 gas turbines. The starting conditions of this mixture, around 600 ºC and 23 MPa, 79 determine the near-isentropic path needed for an efficient expansion and route it down 80 this path to an early condensation in terms of a full harnessing of the mixture enthalpy 81 content; depending on course on the specific composition of the mixture. Thus, technical 82 issues concerning the expansion of two-phase streams prevent the effective 83 implementation of direct expansion in the short term. Furthermore, the detailed design of 84 a dedicated, effective turbine would be costly and would take a long time to be carried 85 out. Moreover, the design of such a turbine would be highly dependent on the mass flow 86 rate of the effluent stream, not allowing for wide variation without loss of efficiency. 87 Therefore, a commercial gas turbine is proposed, where the reactor outlet stream is 88 injected in or after the combustor. Before the injection, this stream is mixed with the 89 combustion gases, this method allows the energy recovery using a conventional 90 equipment (expander turbine section) because this doesn’t change in excess the 91 expanding flue gases stream properties. 92 2. Material and methods 93 2.1. Pilot Plant description 94 The simplified PFD (Process Flow Diagram) of the cooled wall reactor facility placed 95 at Universidad de Valladolid is shown in Figure 1. The plant can be used to oxidize 96 various compounds with air as oxidant in an aqueous environment. The maximum 97 operating pressure is 30 MPa at temperatures between 400°C and 700°C with a maximum 98 treatment capacity of 25 kg/h of feed. 99 The main equipment of this pilot plant is the reactor. This device has three inlet lines 100 and two outlet lines [8]: the feed line, entering at the bottom of the reactor vessel and 101 6 proceeding down-up inside of a tubular injector to the top of the reactor, consist of a 102 pumpable mixture of water and fuel which is pressurized and preheated electrically; air 103 line is introduced at the bottom of the reactor after compression, heating and mixing with 104 the feed; and the third inlet line consists of an auxiliary downward flow of water at the 105 top of the reactor intended to protect the reactor wall from high temperature. The liquid 106 products line leaves the reactor from the bottom and is mainly composed of water and 107 salts; and the vapor line flows from the top of the reactor and is mainly composed of water 108 vapor, nitrogen and carbon dioxide with composition depending on the nature of the fuel 109 waste. The outlet lines are cooled and depressurized. 110 The reaction chamber consists of a vertical tube. It is surrounded and contained in a 111 pressure vessel. Between the pressure vessel and the reaction chamber the down flow of 112 cooling water keeping the temperature of the pressure standing wall under 400°C. The 113 feed is premixed with air and enters the reaction chamber through a tubular injection lance 114 [9]. Usually the hydrothermal flame is produced above the lance, at the top of the reaction 115 chamber, where the maximum temperature is detected [9]. To preheat the reactor at the 116 start up of the process there are two electrical heaters. The room temperature cooling 117 water enters at the top end of the reactor flowing down between the walls of the reaction 118 chamber and pressure vessel. At the bottom end it forms a pool of liquid water where it 119 mixes with the reaction products and can solve salts to avoid large salt deposits inside the 120 reaction chamber. 121 Data from this facility are used as the base of this work [8]. 122 123 124 125 7 2.2. Energy Integration 126 As stated above, the use of supercritical water as reaction media requires extreme 127 pressure and temperature operational conditions entailing high-energy operational 128 expenditure. Liquid water can be compressed using a pump with affordable energy costs. 129 The use of supercritical fluids makes necessary to supply heat of high quality (≈ 400ºC). 130 Because of this, it is necessary to study reasonable solutions which are able to solve this 131 part of the process with a viable efficiency. One solution could be the integration of 132 supercritical processes with energy production in cogeneration or Combined Heat and 133 Power (CHP) cycles. Cogeneration is defined as the simultaneous production of various 134 forms of energy –being the most frequent heat and shaft work, i.e., power– from one 135 power source. The implementation of CHP processes is often joined to the use of gas 136 turbines (GT). Nowadays, the most extended fuel used in gas turbines is natural gas. This 137 kind of internal combustion turbines own several advantages over steam turbines and 138 diesel engines, such as, higher yields, better flexibility and higher efficiency [10]. 139 Besides, it is a compact engine, with lower manpower operating needs and ready 140 availability [11]. Also, the gas turbine engine is further recognized for its better 141 environmental performance manifested in curbing of air pollution and reducing the 142 greenhouse effect [12]. For all these advantages it is proved that over the last two decades, 143 GT has seen tremendous development and market expansion. Gas turbines representing 144 only twenty percent of the power generation market twenty years ago, they now claim 145 approximately forty percent of new capacity additions [13]. 146 The SCWO process produces a high pressure reactor outlet stream, being these mainly 147 composed of water, nitrogen and carbon dioxide and can be thermally integrated if there 148 is a necessity of heat in other parts of the process. If there are no other heat requirements, 149 8 it is possible to use the excess heat to implement a steam injection in the gas turbine, 150 which will improve the efficiency of the global process. This mechanism links the process 151 of SCWO with the cogeneration process. Steam injection is a technique which can 152 increase the ability of a plant to generate extra power without burning extra fuel and 153 requiring moderate capital investment. Furthermore a decrease in NOx emissions from 154 the gas turbine is produced and also the electric generation efficiency of the simple and 155 regenerative cycles is improved [14]. Steam Injected Gas Turbines (STIG) systems 156 operate as an enhancement to the Brayton cycle. High quality steam is used to increase 157 the power output and improve operating efficiency of the basic Brayton cycle. The 158 definite place at which this steam is injected differs according to the design of the 159 particular gas turbine; however mainly, high pressure steam is injected into the high160 pressure sections of the gas turbine via the combustor fuel nozzles [11]. In its most basic 161 form, steam injection works by increasing the global mass flow rate through the gas 162 turbine without increasing the mass of air to be compressed. This increase in the expanded 163 mass flow generates an increase in the rotational torque and power output. Steam injection 164 technology offers a clear improvement over the Brayton cycle while providing a fully 165 flexible operating cycle [15]. 166 One of the key parameters that must be considered for the design of a SCWO system 167 for energy production is the choice of the oxidant. From the reaction point of view, using 168 air or oxygen shows no influence on the conversion of the feed oxidized [16]. Air is the 169 cheapest material, but it contains a large amount of nitrogen that has to be pressurized, 170 and that acts as a diluent that reduces the temperature of effluents and, therefore, its 171 thermal quality. On the other hand, cryogenic liquid oxygen carries no diluents, and air 172 compressors could be replaced by low consumption cryogenic pumps. Furthermore, pure 173 9 oxygen does not need to be preheated up to feed injection temperature. However, the cost 174 and energy consumption of producing pure oxygen could affect the viability of the 175 process. An intermediate option is the use of oxygen-enriched air [4]. 176 2.3. Analyzed schemes and methods 177 In this research, different possibilities for energy recovery from the upper stream of 178 the SCWO cooled wall reactor are explored. This stream is gaseous and mainly composed 179 of water, nitrogen and carbon dioxide. Energetic efficiencies are studied and compared 180 using a simulation software. Also, the mass and energy balances are calculated for the 181 proposed schemes. 182 For carrying out these studies, Aspen Plus V8.0 software is used. This software can 183 be used for a wide variety of simulation chemical engineering tasks, from parallel process 184 monitoring to operation modes exploration to grass root design. The approach adopted in 185 this work is to develop an Aspen simulation flow-sheet that validates against experimental 186 runs of the pilot plant and then apply this flowsheet to explore different process setups 187 for the recovery of energy from the top reactor effluent. In order to model the 188 thermodynamic behavior of the mixtures the Peng-Robinson thermo package with 189 Boston-Mathias (PRBM) modifications was used. 190 The initial values used in this simulation are experimental data which were obtained 191 from the pilot plan referred above. 192 The feed consists of solutions of lactose in water (mass fraction: 87% H2O and 13% 193 C12H22O11) at room conditions (20ºC and 1 bar) with a mass flow rate of 13.5 kg/h. The 194 mass flow rate of cooling water necessary is 5.6 kg/h at 20ºC and 1 bar. 195 Into the reactor the next reaction happens: 196 𝐶12𝐻22𝑂11 +12𝑂2 → 12𝐶𝑂2+11𝐻2𝑂 197 16 Different final valve expansion pressures can have a significant influence in shaft 340 work recovery, but this is difficult to assess due to strong dependencies of the maximum 341 allowable value of this pressure on the specific equipment (GT) and injection details. 342 Case 0 is the most basic configuration, there isn’t gas injection, and being for this 343 reason the net work produced the lowest. With this configuration, heat integration is 344 achieved to just preheat inlet stream. Energy integration is improved with gas injection in 345 case 1. 346 In case 2, 3 and 4 the high pressure is used to increase energy production using an 347 ejector. The simulation software employed doesn't include an ejector or jet-steam unit, 348 and for this reason a simplified configuration was used. 349 If intermediate pressure is high, the net work is higher, therefore, case 4 is better than 350 case 2 and case 3. These cases are improved whit the case 5. All the outlet pressure reactor 351 is used for the ejector. 352 And finally, the efficiencies obtained (Table 7) in every cases are over 25 % and going 353 to up 34.6 % in case 5. 354 Acknowledgements 355 Y.G.R. & A.M. thanks to MS3 for PhD financial support. M.D.B. thanks 356 MINECO for RyC fellowship (RYC-2013-13976) & MINECO project CTQ2013357 44143-R for financial support. 358 17 References 359 360 [1] G. Brunner, Hydrothermal and supercritical water processes, in, Elsevier, Hamburg, 2014, pp. 361 2-666. 362 [2] M.D. Bermejo, M.J. Cocero, Supercritical water oxidation: A technical review, AIChE J., 52 363 (2006) 3933-3951. 364 [3] P.A. Marrone, Supercritical water oxidation - Current status of full-scale commercial activity 365 for waste destruction, J. Supercritical Fluids, 79 (2013) 283-288. 366 [4] J.P.S. Queiroz, M.D. Bermejo, F. Mato, M.J. Cocero, Supercritical water oxidation with 367 hydrothermal flame as internal heat source: Efficient and clean energy production from waste, 368 The Journal of Supercritical Fluids, 96 (2015) 103-113. 369 [5] M.D. Bermejo, Á. Martín, J. Queiroz, P. Cabeza, F. Mato, M.J. Cocero, Supercritical Water 370 Oxidation (SCWO) of Solid, Liquid and Gaseous Fuels for Energy Generation, in: Z. Fang, C. 371 Xu (Eds.) Near-critical and Supercritical Water and Their Applications for Biorefineries, Springer 372 Netherlands, 2014, pp. 401-426. 373 [6] E.D. Lavric, H. Weyten, J. De Ruyck, V. Pleşu, V. Lavric, Delocalized organic pollutant 374 destruction through a self-sustaining supercritical water oxidation process, Energy Convers. 375 Manage., 46 (2005) 1345-1364. 376 [7] J.P.S. Queiroz, On the development of computational tools for the modeling and simulation 377 of SCWO process intensified by hydrothermal flames., in, University of Valladolid, 2014. 378 [8] P. Cabeza, Studies in the development of supercritical water oxidation vessel reactors with 379 hydrothermal flame as an internal heat source., in, University of Valladolid, 2012. 380 [9] P. Cabeza, J.P.S. Queiroz, S. Arca, C. Jiménez, A. Gutiérrez, M.D. Bermejo, M.J. Cocero, 381 Sludge destruction by means of a hydrothermal flame. Optimization of ammonia destruction 382 conditions, Chemical Engineering Journal, 232 (2013) 1-9. 383 [10] M.E. McKay, A. Rabl, A case study on cogeneration, Energy, 10 (1985) 707-720. 384 18 [11] L. Langston, Market drivers for electric power gas turbines: reasons for the revolution., in: 385 Global gas turbine news, 1996. 386 [12] Y.S.H. Najjar, M. Akyurt, O.M. Al-Rabghi, T. Alp, Cogeneration with gas turbine engines, 387 Heat Recovery Systems and CHP, 13 (1993) 471-480. 388 [13] T.G. Koivu, New technique for steam injection (STIG) using once through steam generation 389 (GTI/OTSG) Heat recovery to improve operational flexibility and cost performance., Industrial 390 application of gas turbines committe. , Paper No: 07-IAGT-02.02 (2007). 391 [14] K. Nishida, T. Takagi, S. Kinoshita, Regenerative steam-injection gas-turbine systems, 392 Applied Energy, 81 (2005) 231-246. 393 [15] D.A. Cantero, L. Vaquerizo, F. Mato, M.D. Bermejo, M.J. Cocero, Energetic approach of 394 biomass hydrolysis in supercritical water, Bioresource Technology, 179 (2015) 136-143. 395 [16] B.D. Phenix, J.L. DiNaro, J.W. Tester, J.B. Howard, K.A. Smith, The effects of mixing and 396 oxidant choice on laboratory-scale measurements of supercritical water oxidation kinetics, 397 Industrial and Engineering Chemistry Research, 41 (2002) 624-631. 398 399 400 401 402 403 404 405 406 407 408 409 19 Tables 410 Table 1 411 Temperature (ºC) Pressure (bar) Mass flow (kg/h) Partial molar flow (kmol/h) Feed 20 1 13.5 0.005 C12H22O11 0.652 H2O Air Reactor 20 1 10 0.073 O2 0.274 N2 Reactor Inlet (Feed and Air Reactor) 400 230 23.5 0.073 O2 0.274 N2 0.005 C12H22O11 0.652 H2O Cooling Water 35.4 230 5.6 0.311 H2O Lower Reactor Outlet 700 230 9.469 0.526 H2O Upper Reactor Outlet 700 230 19.631 0.011 O2 0.274 N2 0.062 CO2 0.494 H2O Air Turbine 20 1 80.9 0.589 O2 2.215 N2 Natural Gas 20 15.6 1.349 0.003 CO2 0.076 CH4 Gas Turbine Flue Gases 583.8 1 82.249 0.438 O2 2.215 N2 0.079 CO2 0.151 H2O Cooled Gas Turbine Gases 192 1 82.249 0.438 O2 2.215 N2 0.079 CO2 0.151 H2O 412 413 20 Table 2 414 Temperature (ºC) Pressure (bar) Mass flow (kg/h) Partial molar flow (kmol/h) Injected Stream 676.1 15.6 19.631 0.011 O2 0.274 N2 0.062 CO2 0.494 H2O Gas Turbine Flue Gases 539.9 1 101.889 0.449 O2 2.489 N2 0.140 CO2 0.645 H2O Cooled Gas Turbine Flue Gases 235.7 1 101.889 0.449 O2 2.489 N2 0.140 CO2 0.645 H2O 415 416 21 Table 3 417 Temperature (ºC) Pressure (bar) Mass flow (kg/h) Partial molar flow (kmol/h) Ejector Inlet (Valve Outlet) 680.7 50 19.631 0.011 O2 0.274 N2 0.062 CO2 0.494 H2O Air Compressor Turbine 20 1 77.552 0.565 O2 2.123 N2 Air Ejector 20 1 3.358 0.024 O2 0.092 N2 Ejector Outlet 611.1 15.6 22.989 0.036 O2 0.366 N2 0.062 CO2 0.494 H2O Gas Turbine Flue Gases 523.3 1 101.889 0.449 O2 2.489 N2 0.140 CO2 0.645 H2O Cooled Gas Turbine Flue Gases 227.5 1 101.889 0.449 O2 2.489 N2 0.140 CO2 0.645 H2O 418 419 22 Table 4 420 Temperature (ºC) Pressure (bar) Mass flow (kg/h) Partial molar flow (kmol/h) Ejector Inlet (Valve Outlet) 686.8 100 19.631 0.011 O2 0.274 N2 0.062 CO2 0.494 H2O Air Compressor Turbine 20 1 75.929 0.553 O2 2.079 N2 Air Ejector 20 1 4.981 0.036 O2 0.136 N2 Ejector Outlet 584.1 15.6 24.612 0.048 O2 0.410 N2 0.062 CO2 0.494 H2O Gas Turbine Flue Gases 519.6 1 101.889 0.449 O2 2.489 N2 0.140 CO2 0.645 H2O Cooled Gas Turbine Flue Gases 223.5 1 101.889 0.449 O2 2.489 N2 0.140 CO2 0.645 H2O 421 422 23 Table 5 423 Temperature (ºC) Pressure (bar) Mass flow (kg/h) Partial molar flow (kmol/h) Ejector Inlet (Valve Outlet) 692.3 150 19.631 0.011 O2 0.274 N2 0.062 CO2 0.494 H2O Air Compressor Turbine 20 1 75.089 0.547 O2 2.056 N2 Air Ejector 20 1 5.821 0.042 O2 0.159 N2 Ejector Outlet 571 15.6 25.451 0.054 O2 0.433 N2 0.062 CO2 0.494 H2O Gas Turbine Flue Gases 517.6 1 101.889 0.449 O2 2.489 N2 0.140 CO2 0.645 H2O Cooled Gas Turbine Flue Gases 221.5 1 101.889 0.449 O2 2.489 N2 0.140 CO2 0.645 H2O 424 425 426 427 428 429 430 431 432 433 434 435 436 24 Table 6 437 Temperature (ºC) Pressure (bar) Mass flow (kg/h) Partial molar flow (kmol/h) Air Compressor Turbine 20 1 74.290 0.541 O2 2.034 N2 Air Ejector 20 1 6.620 0.048 O2 0.181 N2 Ejector Outlet 559.1 15.6 26.251 0.059 O2 0.455 N2 0.062 CO2 0.494 H2O Gas Turbine Flue Gases 515.8 1 101.889 0.449 O2 2.489 N2 0.140 CO2 0.645 H2O Cooled Gas Turbine Flue Gases 219.5 1 101.889 0.449 O2 2.489 N2 0.140 CO2 0.645 H2O 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 25 Table 7 453 Case 0 Case 1 Case 2 Case 3 Case 4 Case 5 Ejector Inlet Pressure (bar) - - 50 100 150 230 Air Ejector (mass fraction) (%) - - 4.150 6.156 7.194 8.183 Outlet combustor temperature (ºC) 1041.4 950.1 939.3 934.1 931.4 928.8 Gas Turbine Flue Gases Temperature (ºC) 583.8 530.9 523.3 519.6 517.6 515.8 Cooled Gas Turbine Flue Gases Temperature (ºC) 192 235.7 227.5 223.5 221.1 219.5 Energy consumption by compressor-turbine (kW) 9.874 9.873 9.463 9.265 9.163 9.065 Energy production by turbine (kW) 12.674 15.333 15.189 15.120 15.084 15.050 Energy consumption by feed pump (kW) 0.330 0.330 0.330 0.330 0.330 0.330 Energy consumption by cooling water pump (kW) 0.142 0.142 0.142 0.142 0.142 0.142 Energy consumption by air compressor 2.328 2.328 2.328 2.328 2.328 2.328 Net work (kW) 0 2.660 2.926 3.055 3.121 3.185 Net work from turbine (kW) 2.800 5.460 5.726 5.855 5.921 5.985 Improvement percentage with respect to case 0 (%) 95 104.5 109.107 111.464 113.75 Improvement percentage with respect to previous case (%) 95 4.872 2.253 1.127 1.081 Efficiency (%) 0 28.934 31.828 33.231 33.949 34.645 454