The benefits of flue gas recirculation in waste incineration

Waste Management 27 (2007) 106–116 www.elsevier.com/locate/wasman The benefits of flue gas recirculation in waste incineration Giuseppe Liuzzo, Nicola Verdone *, Marco Bravi Dipartimento di Ingegneria Chimica dei Materiali, delle Materie Prime e Metallurgia – Università degli Studi di Roma ‘‘La Sapienza’’, via Eudossiana, 18, 00184 Roma, Italy Accepted 6 January 2006 Available online 3 March 2006 Abstract Flue gas recirculation in the incinerator combustion chamber is an operative technique that offers substantial benefits in managing waste incineration. The advantages that can be obtained are both economic and environmental and are determined by the low flow rate of fumes actually emitted if compared to the flue gas released when recirculation is not conducted. Simulations of two incineration processes, with and without flue gas recirculation, have been carried out by using a commercial flowsheeting simulator. The results of the simulations demonstrate that, from an economic point of view, the proposed technique permits a greater level of energy recovery (up to +3%) and, at the same time, lower investment costs as far as the equipment and machinery constituting the air pollution control section of the plant are concerned. At equal treatment system efficiencies, the environmental benefits stem from the decrease in the emission of atmospheric pollutants. Throughout the paper reference is made to the EC legislation in the field of environmental protection, thus ensuring the general validity in the EU of the foundations laid and conclusions drawn henceforth. A numerical example concerning mercury emission quantifies the reported considerations and illustrates that flue gas recirculation reduces emission of this pollutant by 50%. Ó 2006 Elsevier Ltd. All rights reserved. 1. Introduction Incineration processes featuring energy recovery and aimed at minimizing the emitted flue gas flow rate for a given load of waste material have well recognized benefits from both the economic and the environmental points of view (Urban, 1979, 1982; Liuzzo et al., 1992, 1995). The economic benefit, in qualitative terms, derives essentially from the lower flow rate of the flue gas and is twofold: (1) the consequent larger attainable energy recovery, and (2) the lower capital costs associated with the smaller installed capacity of the downstream processing equipment. Environmental benefits stem from the decrease in atmospheric pollution resulting from the smaller flow rate of flue gas. The most efficient process scheme is that featuring flue gas recirculation (FGR) in the combustion chamber. Usually, flue gas from the economizer outlet is used to partly replace secondary combustion air, thus reducing furnace * Corresponding author. Tel.: +39 6 44585819; fax: +39 6 4827453. E-mail address: [email protected] (N. Verdone). 0956-053X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.wasman.2006.01.002 temperature and oxygen concentration simultaneously. By contrast, in the analyzed configuration the flow rate of the re-circulated flue gas flow rate is used to control furnace temperature, while the air flow rate is used to control the oxygen content of the fumes leaving the combustion chamber to be as close to 6% as possible, i.e., the minimum allowed for existing plants to ensure completion of the combustion reactions, and determines the corresponding minimum flue gas flow rate. A reference process diagram of an incineration plant equipped with FGR is shown in Fig. 1. Although not frequently implemented in the thermal treatment of wastes, FGR can be applied to both new and existing incineration plants and is one of the most promising techniques to improve their performances (Brem, 2003). The application of FGR in new plants has allowed a reduction of the total amount of incineration air and flue gas in the range of 10–15%. It has also entailed a proportional reduction of the flue gas treatment system load and of the relevant emissions, while boosting the thermal efficiency of the plant by approximately 2–3%. In retrofit applications, however, FGR can be very expensive: new ducts, fans, dampers and control equipment may be G. Liuzzo et al. / Waste Management 27 (2007) 106–116 107 Nomenclature EM ES FO FT FU FGR FGT LHV MSW OM measured emission concentration (mg N m3 dry basis) calculated emission concentration corrected to 11% oxygen level (mg N m3 dry basis) correction factor for oxygen content, dimensionless correction factor for temperature, dimensionless correction factor for moisture content, dimensionless flue gas recirculation flue gas treatment lower heating value (kJ kg1) municipal solid waste measured oxygen concentration (% v/v dry basis) required, while the increased gas flow may cause operational and maintenance problems (cleaning and even corrosion); furthermore, the effective reduction of oxygen may result in high levels of CO and other PICs. An important environmental benefit of FGR is the reduction of nitrogen oxides of thermal origin entrained by the lowered flame temperature (Tillman et al., 1989) and oxygen and nitrogen partial pressures in the fumes. According to Bowman (1992), the kinetics of formation of thermal NOx, whose major constituent is NO, can be modeled combining the Zel’dovich mechanism equations and correcting for the presence of excess air: d½y NOx  ¼ 1:45  1017 T 1=2 dt   1=2 69460 P 1=2  exp ½y N2 ½y O2  T RT ð1Þ P PDF PIC R RDF t T U V gb gc ge pressure (atm) plastic derived fuel products of incomplete combustion gas constant refuse derived fuel time (s) temperature (°C) moisture content (% v/v) gas flow rate (m3 h1) boiler efficiency (%) net overall efficiency (%) net cycle efficiency (%) subscripts nr without FGR r with FGR where T (K) is the flame temperature, t the time, R the gas constant, P the system pressure, and y NOx , y N2 and y O2 are mole fractions. The equation indicates that NOx formation is an exponential function of temperature and a square root function of oxygen concentration and shows that, by manipulating temperature and oxygen concentration, FGR represents an effective, dilution-based combustion control technique reducing the formation of thermal NOx. It is solely for the purpose of reducing nitrogen oxides emission that FGR was originally tried out in Japan (Hirayama, 1975), where the recirculation of 20–25% of the effluent fumes from the thermal recovery section led to a decrease of approximately 25% in the production of nitrogen oxides. It was also shown that an increase in the ratio of recirculation provided no further benefits in nitrogen oxides reduction. Recently, the potential of FGR in reducing NOx formation was confirmed in a field applica- Fig. 1. Incineration plant with flue gas flow rate minimized by FGR. 108 G. Liuzzo et al. / Waste Management 27 (2007) 106–116 tion of waste thermal treatment (NOx reduced by 30–37%), allowing the plant to meet the emissions standards (Burnley and Aplin, 1999), while a theoretical study based on computational fluid dynamics (CFD) techniques, which referred to a coal-fired stoker furnace, predicts a 27% decrease in furnace exit NOx (Valentine et al., 2003). Due to the high complexity of modeling thermal decomposition and combustion processes in solid biomass fuel beds, only a few references can be found in the literature expressly referring to CFD models of waste incinerators equipped with FGR. However, a CFD analysis of air staging and flue gas recirculation in biomass grate furnaces (Scharler et al., 2000) has demonstrated that an optimized FGR can control both CO emissions (decreasing concentrations to 40%) and temperature peaks, thus reducing the thermal NOx formation rate. Finally, industrial suppliers claim a NOx reduction in the range 50–80% (Agrawal and Wood, 2002). Although the importance of thermal NOx in nitrogen oxides emission from the incineration plants is limited if compared to the case of fossil fuel-fired plants, FGR helps in meeting the environmental regulation standards and will reduce the consumption of ammonia or urea if an SCR or SNCR system is fitted. Furthermore, since FGR is mostly targeted at flue gas minimization, the recirculation ratio will be pushed beyond the limits of effectiveness for NOx reduction (Lacquaniti et al., 1996). Decreasing the oxygen content of the fumes to treatment makes the compliance with emission limits easier; this latter is a useful benefit which assumes a great importance in view of the far more restrictive provisions laid down by Directive 2000/76/EC (EC, 2000), compliance with which is required both for existing facilities and in the planning of new ones. With the current waste lower heating value (LHV), the temperature of the fumes in the combustion chamber could become excessive for the strength of the refractory materials employed. To avoid this effect, two possible alternatives are: (1) adopting an adiabatic combustion chamber and using as much excess air as necessary to maintain the desired temperature in the combustion chamber, and (2) withdrawing heat from the combustion chamber by providing water pipes in the chamber and producing steam. The first alternative involves a greater output of fumes, and hence larger capital costs for processing of the fumes. At the same time, the utilization of the waste thermal energy is impaired by the larger flow rate of the emitted fumes and, hence, the greater amount of energy lost with the fumes themselves, which leave the thermal recovery section at 200–250 °C. The system based on FGR finds its ideal application in the very case of an adiabatic combustion chamber (a configuration regarded as advisable by many designers) whose temperature is controlled with very large amounts of excess air. However, even in systems equipped with a radiant section in the combustion chamber, the recirculation of fumes improves controllability and helps in maximizing energy recovery in the presence of thermal power fluctuations deriving from fluctuations of the processed waste LHV. Even though these systems process waste with a higher LHV, the heat subtracted in the radiant section permits the minimization of the air flow rate while ensuring the completion of combustion and meeting both the limit on the minimum free oxygen content of the fumes leaving the combustion section and the maximum temperature allowance therein (Liuzzo et al., 1992, 1995). However, given that the heat removal rate exhibits a limited variation over the entire operating condition range, controlling the combustion chamber temperature by manipulating the air flow rate may lead to under- or over-design value of the combustion chamber temperature in low- or high-load conditions, respectively; both of these problems (the first of which appears to be quite serious) could be avoided by adopting FGR in the combustion chamber and regulating it as shown in Fig. 1. With this configuration, the surface of the radiant section should be sized in such a way that the design temperature is obtained in the combustion chamber in conditions of minimum load and with practically no recirculated fumes; in this way, when the thermal load increases, the temperature in the combustion chamber can be controlled by increasing the flow rate of the re-circulated fumes and regulating air inlet so as to maintain the free oxygen content of the fumes leaving the combustion chamber at 6% v/v, simultaneously also minimizing the flow rate of fumes to be processed. The major aim of this paper is to demonstrate and quantify the significant operative environmental benefit enjoyed by facilities adopting process schemes including FGR in ensuring the compliance with the emission limits. As FGR introduced a further accessory benefit in terms of efficiency, this latter benefit is also discussed. 2. Operative benefits and major drawbacks of FGR The operative environmental benefit mentioned in Section 1 has recently gained a great deal of importance in Italy with the issue of a new environmental regulation which incorporates the relevant UE Directives on the incineration of non-hazardous waste (EC, 2000) and sets the emission limits for ‘‘new facilities’’ (and hence the requisites for the revamping of existing ones). While these limits are generally compatible with the currently available flue gas processing technologies, they could pose substantial problems for at least some pollutants, especially in the revamping of existing facilities. One such case is mercury when MSW and hospital waste are treated together, for which compliance with the currently enforced limit (0.05 mg/N m3) is a problem; for this case and the many more that may soon fall into this class, this paper aims at pointing out the benefit offered by the mentioned process schemes and, in particular, that involving the recirculation of fumes in the combustion chamber. To understand the benefit, it is necessary to consider that a treatment system for the fumes discharges the 109 G. Liuzzo et al. / Waste Management 27 (2007) 106–116 processed gas with a well determined pollutant(s) concentration at the outlet of the fumes treatment system and this determines the pollutant removal efficiency (and not vice versa). The pollutant concentration emitted into the environment depends upon the effectiveness of the treatment system, but is completely independent of the macro-composition of the fumes with respect to the main constituents (CO2, N2, H2O and O2) and of the initial concentration of the pollutant. For the purposes of a comparison with the limits set by the law, the compliance monitoring for waste incinerators requires that the pollutant concentration measured in the combustion gas be corrected to 11% oxygen by applying the FO factor defined by: ES ¼ 21  11  EM ¼ F O  EM 21  OM ð2Þ where EM is the pollutant concentration expressed on a dry volume basis, ES is the concentration corrected to 11% O2, OM is the percent dry oxygen concentration in the actual conditions, and FO is the correction factor defined by Eq. (2). The correction factor FO increases as OM decreases (see Fig. 2), thus discouraging any attempt to comply with the limits on emissions by improperly diluting the fumes with air. It is easy to verify that the application of the FO correction factor represents a benefit for facilities with a smaller specific (i.e., relative to the processed waste) output of flue gas; indeed, a smaller output of flue gas can only be achieved with a lower specific utilization of total combustion air and is therefore associated with a lower oxygen content in the gas emissions. The lower the oxygen content of the fumes, the lower the correction factor FO, making a larger allowance for the effective pollutant concentration in the fumes EM. In addition to the operative benefit previously discussed, FGR produces other environmental and energy-related advantages. During the processes occurring on the grate, the combustible waste materials are transformed into a gaseous form consisting in a mixture of many volatile components. In order to assure a complete oxidation of these compounds, secondary air is injected into the furnace through multiple high velocity nozzles, obtaining at the mean time an improvement in the mixing and homogeneity of the flue gas. However, the use of more secondary air than necessary results in higher flue gas quantities, reducing the energy efficiency of the plant and leading to larger flue gas treatment units and, therefore, to higher costs. Furthermore, injection of air with normal oxygen content can produce a local increase of the temperature at the nozzle heads at values above 1400 °C, responsible for higher NOx production. By replacing part of the secondary air with re-circulated flue gases, the flue gas volume is reduced downstream of the extraction point and at the point of emission. The reductions in the fresh oxygen supply (from air) to the furnace may help to reduce NOx emission. Thus, a reduction in reagent consumption for NOx control is expected. However, FGR introduces some drawbacks that should be properly corrected in order to ensure safe operation of the plant. In general, the recirculation extraction point is downstream the flue gas treatment system (FGT) to limit corrosion and other operational problems caused by raw flue gas. This choice, however, involves some energy losses and the FGT system must be designed for a larger flow. On the contrary, if the flue gases are re-circulated from upstream of the FGT system, then the size of the related pieces of equipment can be reduced, although the system should be operated in order to treat more polluted flue 6 Correction factor FO 5 4 3 2 1 0 0 2 4 6 8 10 12 14 16 Oxygen content [%vol.] Fig. 2. Oxygen correction factor vs. oxygen content in flue gas. 18 20 110 G. Liuzzo et al. / Waste Management 27 (2007) 106–116 gases because of the increased concentration. The risk of erosion, corrosion and fouling is higher; corrosion in the recirculation ducting actually has been reported. But, it is also reported that this drawback can be overcome by the elimination of joints and by using effective insulation of ducting to prevent cold spots where condensation of the flue gas and corrosion can rapidly occur. Corrosion may also occur in the boiler due to lower oxygen levels in the flue gas. Operators should be attentive, otherwise corrosion can be very rapid. In such cases the expected operational savings are quickly turned into higher repair costs and loss of plant availability. Corrosion risk is reduced if the hotter parts of the boiler are covered by special claddings. However, when this cladding is installed, the O2 excess concentration at the boiler exit can be reduced even without FGR; thus some benefits are accordingly reduced. Finally, depending on furnace design details, the effective reduction of oxygen content induced by high rates of FGR may result in elevated CO (and other PIC) levels. Care must, therefore, be taken to ensure that replacement rates are optimized. The FGR technique involves additional investments for new plants and significant cost for retrofitting existing plants; however, the attainable benefits may reduce and even compensate the supported costs. 3. Quantification of the operative benefit of FGR In order to quantitatively illustrate the major FGR benefits, a simulation was carried out – for different operating conditions and different waste characteristics – of two energy-recovery incinerators, both characterized by the presence of an adiabatic combustion chamber, a dry processing section for acidic gases and a bag filter. Only one of them (shown in Fig. 1) used a system whereby the fumes from the boiler were re-circulated in the combustion chamber under the control of a regulation loop. Two control loops take care of the temperature in combustion chamber and of the free oxygen content in the gas produced by the combustion process (at the 6% v/v on wet basis concentration set point, according to the current legislation). The process scheme for the system not including the FGR and defined as standard scheme is shown in Fig. 3. In this latter case, just the temperature in the combustion chamber is controlled. In both process schemes, power is generated by a power cycle with a two-bleed regenerative feed water heating. The gas cleaning section reduces the hydrogen chloride and sulphur dioxide content in the flue gas emitted to the values required by the EC regulation and modeling was limited to the extent required for pressure drop and pumping power estimation. Finally, the flue gas temperature at the boiler exit was set at 200 °C, in both cases, in accordance with the actual trend to increase the energy efficiency. All of the simulations were performed by using the PRO/II-Provision v. 6.01 flowsheeting simulator (SimsciEsscor, 2004). PRO/II is a comprehensive computer simulation system for process engineers in the chemical, petroleum, natural gas and power industries. It combines the data resources of a large chemical component library and extensive thermodynamic property prediction methods, with the most advanced and flexible unit operations techniques, providing the process engineer with the computational facilities to perform all mass and energy balance calculations needed to model most steady-state processes. For all of the considered cases, the simulations were carried out assuming an identical value of thermal input (22 MW) with respect to three types of waste: municipal solid waste (MSW), refuse derived fuel (RDF) and plastic derived fuel (PDF) (Table 1) and varying the input waste flow rate accordingly. Furthermore, the simulations were repeated, adopting four different temperature values for the combustion chamber: 850 °C, the minimum value for the temperature in the combustion section of incinerators handling non-hazardous waste, and 1100 °C, set for incinerators handling hazardous waste for halogenated organic Fig. 3. Standard incineration plant. G. Liuzzo et al. / Waste Management 27 (2007) 106–116 Table 1 Characteristics of waste used in simulations Waste MSW RDF PDF Proximate analysis (%wt) Moisture Ash Fixed carbon and volatile matter 25.2 24.4 50.4 19.6 8.4 74.7 2.0 2.0 96.0 Ultimate analysis dry ash free (%wt) Carbon Hydrogen Oxygen Nitrogen Chlorine Sulfur 50.8 6.8 40.3 1.0 0.8 0.3 56.0 7.9 33.4 1.3 1.0 0.4 74.1 9.2 16.7 – – – Lower heating value (kJ/kg) 9570 16,700 31,750 substances content >1%, which are the values established by EU Regulation (EC, 2000). Two other temperatures, 950 and 1200 °C, were selected in order to refine the trend plots in the simulation results. The characteristic operating parameters adopted in the simulations for the equipment and machinery present in the systems are shown in Table 2, and the main results of the simulations are given in Table 3 for each type of waste considered. For the system not equipped with FGR, a downward trend can be noted for the flow rate of the fumes emitted (Table 3). Conversely, for the system equipped with FGR, the flue gas flow rate has a constant value throughout the temperature range corresponding to the excess air needed to obtain a 6% v/v concentration of oxygen in the fumes at the boiler exit. Furthermore, in the system without the recirculation of fumes, for each set of combustion temperature values, the simulations show an increase in the volume of fumes released in the atmosphere going from the lowest LHV waste (MSW) to the highest LHV (PDF), an effect explained by the lower specific enthalpy content associated with a lower water content of the waste itself. In controlling temperature, water acts as a thermal driving force both through its latent heat of vaporization and Table 2 Characteristic parameters of equipment functioning in simulated cycles Adiabatic efficiency of steam turbine (%) Losses at discharge of steam turbine (kPa) Efficiency of pumps and blowers (%) Organic efficiency of reducer (%) Organic efficiency of alternator (%) Additional consumption condenser/duty condenser (%) Thermal loss from incinerator (MW) DP boiler water side (kPa) DP boiler fumes side (kPa) DP system for treatment of fumes (kPa) DP bag filter (kPa) Discharge head of supply pump (kPa) DP drum control valve (kPa) DP valve regulating entry of steam into turbine (kPa) Steam pressure at condenser (kPa) Temperature of superheated steam (°C) 84 5.6 70 97 98 1.0 0.93 304.0 5.0 1.5 2.5 4762 203 203 7.8 360 111 through the greater specific heat of steam with respect to that of the other gases making up the fumes. In the process scheme with fumes recirculation, the opposite is true: the flow rate of the released fumes decreases passing from the lowest LHV waste to the highest LHV waste as a result of the greater re-circulated volume required to control oxygen concentration and combustion temperature. As the waste LHV decreases, the fumes specific enthalpy also decreases following a corresponding drop in their moisture content. The net overall efficiency gc of the conversion of the waste energy content into electric energy can be written: gc ¼ gb  ge ð3Þ where gb is the boiler efficiency, i.e., the fraction of thermal energy in the fumes that is captured by the boiler, calculated as the ratio of the enthalpy difference of the produced steam and the boiler feedwater to the plant thermal input, and ge is the net cycle efficiency of the conversion of thermal energy into electric energy, defined as the ratio of the net electric power production considering the plant electric loading to the power gained by the water into the boiler. Efficiency results reported in Table 3 show that for systems equipped with FGR, the overall net efficiency vs. temperature in combustion chamber values assume a slight upward trend, while, for systems without fumes recirculation, a more marked upward trend is evident. Simulations also show the translation of the above trends to higher levels as the waste LHV increases, due to the lower water content of the waste and hence the smaller loss of enthalpy with the discharged fumes. The highlighted behaviour can be explained considering apart the contribution of the boiler and cycle efficiencies as previously defined to the overall net efficiency. FGR heavily affects the boiler efficiency, as can be noted from the results reported in Fig. 4, as the great increase in boiler efficiency induced by the hot gas re-circulated in the combustion chamber with respect to the standard cases is evident, especially at the lower temperatures. As far as the temperature in the combustion chamber is increased, FGR decreases resulting in a lower and lower boiler efficiency. On the contrary, in the standard cases, an expected marked upward trend is observed as temperature increases. The influence of the cycle efficiency is less marked and always assumes the same trend (Fig. 5), due to the decrease in pumping energy requirement for flue gas handling at higher combustion temperatures. A deeper insight of the reasons why FGR increases the efficiency in power production can be derived considering the heat recovery temperature profiles reported in Fig. 6 as an example for the case of MSW processing at the combustion chamber temperature of 850 °C. In the process scheme featuring FGR, both the flue gas flow rate and water vapor content increase with respect to the values characteristic of the corresponding process scheme not equipped with FGR. The specific heat of water vapour is significantly greater than the specific heats of the other flue 112 G. Liuzzo et al. / Waste Management 27 (2007) 106–116 Table 3 Simulation results Temperature (°C) 850 Waste – MSW Air (kg/h) FGR (kg/h) Flue gas at the stack (kg/h) Steam (kg/h) Power generated (kWs)a Blower consumption (kWe)b Pump consumption (kWe) Aux. cond. consum. (kWe) Net power (kWe) Net overall efficiency (%) H2O in flue gas (stack) (%v raw) O2 in flue gas (stack) (%v raw) 74,809 – 82,679 22,674 5231 481 47 118 4586 20.8 11.9 11.9 40,564 39,294 48,159 25,119 5795 490 52 130 5123 23.2 19.4 5.7 64,675 – 72,356 23,314 5379 420 48 121 4790 21.7 13.2 10.7 40,567 26,761 48,163 25,032 5775 423 52 130 5171 23.5 19.4 5.7 53,073 – 60,538 24,047 5548 350 50 125 5023 22.8 15.2 8.8 40,563 13,376 48,159 24,934 5753 350 52 129 5222 23.7 19.4 5.7 47,016 – 54,368 24,430 5637 314 51 127 5146 23.3 16.5 7.5 40,536 6771 48,159 24,887 5741 314 52 129 5247 23.8 19.4 5.7 Waste – RDF Air (kg/h) FGR (kg/h) Flue gas at the stack (kg/h) Steam (kg/h) Power generated (kWs) Blower consumption (kWe) Pump consumption (kWe) Aux. cond. consum. (kWe) Net power (kWe) Net overall efficiency (%) H2O in flue gas (stack) (%v raw) O2 in flue gas (stack) (%v raw) 78,842 – 84,784 22,899 5283 494 47 119 4623 21.0 9.3 12.8 38,872 46,721 44,420 25,757 5943 507 53 134 5249 23.8 16.2 5.7 68,625 – 74,380 23,546 5432 433 49 122 4829 21.9 10.2 11.7 38,874 33,655 44,408 25,664 5921 437 53 133 5298 24.0 16.2 5.7 56,940 – 62,480 24,285 5603 363 50 126 5064 23.0 11.6 10.1 38,872 19,692 44,426 25,564 5898 362 53 133 5351 24.3 16.2 5.7 50,843 – 56,270 24,670 5692 326 51 128 5187 23.5 12.5 9.0 38,872 12,801 44,420 25,517 5887 325 53 132 5377 24.4 16.2 5.7 Waste – PDF Air (kg/h) FGR (kg/h) Flue gas at the stack (kg/h) Steam (kg/h) Power generated (kWs) Blower consumption (kWe) Pump consumption (kWe) Aux. cond. consum. (kWe) Net power (kWe) Net overall efficiency (%) H2O in flue gas (stack) (%v raw) O2 in flue gas (stack) (%v raw) 83,882 – 88,723 23,396 5398 519 48 121 4709 21.4 7.9 13.5 37,408 55,540 40,998 26,720 6165 530 55 139 5441 24.7 12.2 5.7 73,455 – 78,013 24,055 5550 455 50 125 4920 22.3 8.4 12.6 37,407 41,704 40,987 26,623 6142 457 55 138 5493 24.9 12.2 5.7 61,546 – 65,781 24,807 5723 383 51 129 5161 23.4 9.2 11.2 37,412 26,921 40,991 26,520 6119 379 55 138 5548 25.2 12.2 5.7 55,337 – 59,403 25,200 5814 345 52 131 5286 24.0 9.7 10.2 37,410 19,622 40,990 26,470 6107 340 55 137 5575 25.3 12.2 5.7 a b 950 1100 1200 kWs, mechanical power. kWe, electric power. gas components, so that the average specific heat of the fumes increases. These two conditions increase the heat transferred from the fumes per degree of cooling, which is proportional to the flue gas flow rate times the average specific heat. In the examined case, stepping from no FGR to FGR raises the water vapor content from 9.0% to 15.6% v/v, and the flue gas flow rate and the average specific heat increase by 6.2% and 4.3%, respectively. These combined effects increase the heat transferred from the flue gas per degree of cooling to the water side in the heat recovery section by 10.8%. In the MSW thermal treatment, the increase in power production net efficiency obtained by the facilities equipped with FGR, with respect to the conventional ones, is in the range 11.7–2%, for the minimum (850 °C) and maximum (1200 °C) values of the temperature set in the combustion chamber, respectively. For RDF and PDF processing, the increase in power production efficiency lies in the range 11.3–1.6% and 15.5–5.5%, respectively, for the same extreme values of temperature (Table 3). These results were obtained with a power production cycle of a relatively low complexity, the level of power production (5 MWe) not justifying the adoption of more sophisticated and efficient power production schemes. As far as the oxygen content of the fumes at the stack (dry basis) and the consequent correction factor FO defined by Eq. (2) are concerned, these quantities assume in any case values in a very narrow range, around 7% v/v and 0.7, respectively, for the system equipped with FGR (Fig. 7). For the system without FGR, the downward trends (albeit at different levels depending on the LHV of the waste) are linked to the decrease in excess air, to the increase in temperature in the combustion section and to the rise in excess air accompanying the increase in LHV. 113 G. Liuzzo et al. / Waste Management 27 (2007) 106–116 94 92 90 ηb [%] 88 86 84 82 MSW 80 RDF PDF 78 FGR no FGR 76 800 900 1000 Temperature [˚C] 1100 1200 Fig. 4. Boiler efficiency vs. temperature in combustion chamber. 27.8 27.6 ηe [%] 27.4 27.2 27.0 26.8 MSW RDF PDF 26.6 FGR no FGR 26.4 800 900 1000 1100 1200 Temperature [˚C] Fig. 5. Net cycle efficiency vs. temperature in combustion chamber. The correction factor for the system with FGR, FO,r, measures the extent to which the recirculation of fumes eases compliance with the emission limits for all pollutants. FGR-equipped systems tolerate higher actual pollutant concentrations than systems without recirculation in compliance with the limits (or with the same deviation from them) by a figure equal to the ratio between the correction factor computed in the case where FGR is not operated, FO,nr, to the correction factor calculated where FGR is present, FO,r. This ratio scales up as temperature in the combustion chamber decreases, the other parameter being equal (Fig. 8). It is possible to verify that the ratio of the volume of fumes for the system without recirculation (Vnr) to the volume for the system with recirculation (Vr) is identical to the analogous ratio between the FO correc- tion factors shown in Fig. 8. It follows that the following equality is always verified:     Vr F O;r ð4Þ ¼ V nr T F O;nr T For an assigned thermal input, the incineration facilities that produce a lower volume of fumes, and hence a lower content of free oxygen in the fumes, are ‘‘rewarded’’ with a proportionally lower FO, thus allowing them to produce fumes with a greater effective concentration of all of the atmospheric pollutants compared to facilities not equipped with FGR while still meeting the same established limits. In actual practice, however, there is absolutely no reason to assume that this will always happen. Indeed, it is very likely that FGR will bring about a reduction in the emis- 114 G. Liuzzo et al. / Waste Management 27 (2007) 106–116 900 MSW - 850 ˚C 800 flue gas economizer 700 Temperature [˚C] vaporizer superheater 600 FGR 500 no FGR 400 300 200 100 0 0 2 4 6 8 10 12 Heat exchanged [MW] 14 16 20 18 Fig. 6. Heat recovery temperature profiles in the sample case. 4 14 MSW 3 RDF 10 PDF 2.5 FGR 8 no FGR 2 6 1.5 4 1 2 0.5 0 800 Correction factor FO Oxygen content in flue gas [%vol. dry] 3.5 12 0 900 1000 1100 1200 Temperature [˚C] Fig. 7. Oxygen concentration in flue gas at the stack and correction factor vs. temperature in combustion chamber. sions of all pollutants at significantly lower levels than those characterizing systems without recirculation. For a given thermal power, the adoption of operating conditions or processing systems making it possible to lower the volume of fumes and hence the correction factor FO – the effective concentration of all the atmospheric pollutants being independent of the volume of fumes, treatment systems being equal – will certainly make compliance with the limits on emissions set by law easier. At the same time, and to the same extent, it will also permit a substantial reduction of emissions. Facilities will thus benefit to the extent that they produce a lower flow rate of fumes and hence enjoy a proportionally lower correction factor. It can also be argued that, with the introduction of the FO factor defined by Eq. (2), the technical legislation on emission limits assumed as reference (EC, 2000), while apparently setting limits on the concentrations (mg N m3) of the pollutants specified therein only, actually sets very precise limits on the mass emissions (mg h1) of these pollutants and, for a specific type of waste, also on the emission factors of the facility in relation to the unit mass of waste treated ðmg kg1 waste Þ. In other words, for an incinerator designed to treat a specific type of waste, the FO factor imposes, for each pollutant, a limit on the emission factor equal to that prescribed for a facility operated in such a way that the fumes had a free oxygen content equal to the reference value (11% vol.). Although no constraint is placed on the actual free oxygen content of the fumes emitted, a ‘‘moving’’ – with the oxygen content – ‘‘boundary’’ is implicitly established for each pollutant present in the fumes, which decreases in such a way to keep the maximum value of the product of the actual value of the volumetric 115 G. Liuzzo et al. / Waste Management 27 (2007) 106–116 2.4 MSW 2.2 RDF FO,nr /FO,r 2 PDF 1.8 1.6 1.4 1.2 1 800 900 1000 1100 1200 Temperature [˚C] Fig. 8. Correction factor ratio vs. temperature in combustion chamber. flow rates and the concentrations of the different pollutants constant, and hence the maximum value of their effective emissions. As an example of the exposed considerations, let us consider mercury, one of the pollutants that pose substantial problems in regard to compliance with the set limits. Consider the case of two facilities processing RDF with the temperature in the combustion chamber set at 850 °C, operated in such a way as to produce fumes characterized by equal values of flue temperature (160 °C, correction factor for temperature FT = (273 + 160)/273 = 1.59) but different values of free oxygen and moisture content (Table 3). Owing to the recirculation of fumes in the combustion chamber and the control loop introduced to set the O2 concentration at the boiler outlet at 6% v/v, one facility can be assumed to produce fumes with [O2] equal to 6.9% v/v dry at the stack (FO = 0.71) and moisture content of 16.2% v/v (correction factor for moisture, FU = 1/(1  U) = 1.19). Given the excess air required to keep the temperature in the combustion chamber at 850 °C, it can be assumed that the other, which is not equipped for recirculation, produces fumes with [O2] equal to 14.1% v/v dry at the stack (FO = 1.45) and moisture content of 9.3% v/v (FU = 1.10). In complying with the limit (0.05 mg N m3) set on mercury, the facility using recirculation is allowed to produce fumes which have an effective post-treatment mercury concentration equal to: ½Hglim 0:05 ¼ 0:04 mg N m3 ¼ F U  F T  F O 1:19  1:59  0:71 while the facility without FGR is forced to stay within a much tighter limit: ½Hglim 0:05 ¼ 0:02 mg N m3 ¼ F U  F T  F O 1:10  1:59  1:45 This benefit depends almost exclusively on the value of the ratio between the correction factors (FO,nr/FO,r = 1.45/ 0.71 = 2.05), which remains quantitatively unchanged also for facilities that have been revamped to comply with the emission limits set by the legislation. This benefit can be enjoyed for all of the regulated atmospheric pollutants by facilities that adopt the recirculation of fumes in the combustion chamber and thus succeed in minimizing the free oxygen content of the fumes emitted together with their flow rate. 4. Conclusions The recirculation of flue gas (FGR) in the combustion chamber of waste incinerators has been proven to be a promising technique to obtain both operative and environmental benefits by minimizing the flow rate of fumes emitted to the environment with lower oxygen content. The simulations carried out on a commercial process simulator have quantified the effective decrease in the flue gas emitted, when FGR was operated and the free oxygen content in fumes at the boiler exit was set at 6% v/v. Compared to conventional plants, the reduction in flue gas flow rate at the stack is in the range 11–54% (Table 3), depending on the characteristics of waste processed and the temperature value set in the combustion chamber, and the efficiency in net power production increases in the range 1–15% (Table 3) under the same conditions and for a production of 5 MWe. Waste incinerators equipped with FGR, from the environmental point of view, will be characterized by lower emission factors, the efficiency of the flue gas treatment units and the effective concentration of pollutants in the released flue gas being equal. Furthermore, from the point of view of facility management, easier compliance 116 G. Liuzzo et al. / Waste Management 27 (2007) 106–116 with the limits on emissions set by the regulation will be obtained. The extent of the gained benefit is given by the value of the correction factor applied to the concentrations of pollutants measured in the flue gas at the stack [FO = (21– 11)/(21–OM)], which, in the case of facilities employing FGR, is almost halved. As the correction factor FO diminishes, the actual concentration of pollutants in the fumes effluent from the treatment section is allowed to reach proportionally higher values. References Agrawal, R.K., Wood, S.C., 2002. Innovative solutions for cost-effective NOx control. Pollution Engineering 34 (6), 18–25. Bowman, C.T., 1992. Control of combustion-generated nitrogen oxide emissions: technology driven by regulation. In: 24th Symposium (International) on Combustion, Proceedings, pp. 859–878. Brem, G., 2003. Advanced grate furnaces for future waste incineration. ISWA Beacon Conference Waste to Energy: State of the Art and Latest News. Malmö (S). Available from: <www.iswa.ch/Info/Documents/WastetoEnergy03/Brem.pdf>. Burnley, S.J., Aplin, P., 1999. Controlling oxides of nitrogen emissions from a municipal-waste incinerator by flue-gas recirculation. Journal of the Institute of Energy 72 (493), 165–169. EC, 2000. Directive 2000/76/EC of the European Parliament and of the Council of 4th December 2000 on the incineration of waste. Official Journal of the European Communities. L. 332/91. Available <http://europa.en.int/comm/environment/wasteinc/newdir/ from: 2000-76_en.pdf>. Hirayama, N., 1975. Control of NOx emissions from municipal refuse incinerators. In: Convers. Refuse Energy, 1st Int. Conf. Tech. Exhibit., Montreaux (CH), pp. 197–201. Lacquaniti, L., Liuzzo, G., Palitto, M., 1996. Incinerator with low output of fumes for combustible waste. Italian Patent IT1258402. Liuzzo, G., Bemporad, E., Lacquaniti, L., Palitto, M., 1992. I prodotti di combustione negli impianti termici e negli inceneritori di rifiuti (Combustion products in thermal power plants and waste incinerators). Ingegneria Sanitaria-Ambientale 39 (5), 21–44. Liuzzo, G., Verdone, N., Bemporad, E., Lacquaniti, L., Palitto, M., 1995. Rendimenti di conversione energetica mediante il trattamento dei fumi ad alta temperatura negli inceneritori di rifiuti (Energy conversion efficiencies through high temperature flue gas cleaning in waste incinerators). GEA 8 (2), 44–52. Scharler, R., Obernberger, I., Längle, G., Heinzle, J., 2000. CFD analysis of air staging and flue gas recirculation in biomass grate furnace. In: 1st World Conference on Biomass for Energy and Industry, Sevilla (E), vol. 2, pp. 1935–1939. Simsci-Esscor. 2004. Invensys Systems, Inc. Available from: <http:// www.simsciesscorr.com>. Tillman, D.A., Rossi, A.J., Vick, K.M., 1989. Incineration of Municipal and Hazardous Solid Wastes. Academic Press, Inc. Urban, A.I., 1979. Experiments pertaining to the recirculation of flue gas in waste incinerator system. In: International Recycling Congress, Berlin (D), pp. 405–410. Urban, A.I., 1982. Optimization of combustion by recirculation of flue gas. In: Recycl. Int.: Recovery Energy Mater. Residues Waste, Int. Recycl. Congr., Berlin (D), pp. 259–264. Valentine, J.R., Davis, K.A., Denison, M.K., Cron, D., Morrow, R., Giaier, T., 2003. A computational model study of NOx reduction strategies for a coal-fired stoker furnace. In: Int. Tech. Conf. on Coal Utilization and Fuel Systems, 28th, vol. 1, pp. 177–186. Available from: <http:// www.reaction-eng.com/downloads/denison_clear- water_2003.pdf>.