Denitrification techniques for biomass combustion

Denitrification techniques for biomass combustion

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect Renewable and Sustainable Energy Reviews journ...

1MB Sizes 0 Downloads 84 Views

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews journal homepage: www.elsevier.com/locate/rser

Denitrification techniques for biomass combustion ⁎

Milica Mladenović , Milijana Paprika, Ana Marinković University of Belgrade, Vinča Institute of Nuclear Sciences, Laboratory of Thermal Engineering and Energy, Mihaila Petrovića Alasa 12-14, 11351 Belgrade, Serbia

A R T I C L E I N F O

A B S T R A C T

Keywords: Biomass combustion NOx emission DeNOx techniques

In order to achieve the main Applicable combustion control systems in grate-fired bogoals of sustainable development through the harmonization of rising energy needs with environmental protection, modern society promotes the use of biomass as a renewable energy source. Biomass, like any taother fuel, emits certain pollutants from combustion, nitrogen oxides (NOx) being one of them. Control of NOx emission, originated in biomass combustion, is becoming a very significant technical challenge due to the imposition of increasingly strict emission limits. The experimental research and industrial experiences (that are not always easily available) were analyzed in order to make an overview of proven and prospective technical solutions, as well as directions for practical applications for reducing NOx emissions originating from biomass combustion. The denitrification techniques according to the broadest classification (pre-combustion, combustion control and post-combustion) have been analyzed. As the NOx emission is more strongly influenced by the nitrogen content of biomass fuels (especially of those with significant nitrogen content) rather than the operating conditions, the emphasis is placed on the post-combustion (secondary denitrification) measures and the most successful among them - selective catalytic (SCR) and non-catalytic reduction (SNCR). The SCR catalysts, as well as commonly used amine-based reagents (in both SCR and SNCR), are analyzed in accordance with various parameters (activity temperature domain, the location of installation and structural configuration). The special challenges for SCR and SNCR application were considered, and a comparative overview of advantages and disadvantages are given, in accordance with several different criteria. In addition, the applicability of deNOx techniques from the aspect of individual biomass combustion technology is given. Guidelines for the selection of denitrification measures are created, depending on the biomass combusted, combustion technology used, and the installation capacity.

1. Introduction

level ozone and acid rain, visibility impairment, causing damage to human and animal health, to natural ecosystems and crops [1,2]. Another severe negative effect is the significant contribution to particulate matter (PM) in the ambient air, from aerosols formed by nitrogen (NO, NO2) and sulfur (SO2, SO3) oxides [3]. Over 90% of nitrogen oxides emitted due to the combustion process makes NO, while the rest is attributed to NO2. In the atmosphere NO is converted to NO2, so the environmental protection regulations treat all nitrogen oxides as NO2. Due to its greenhouse effect and its long-term stability, nitrous oxide (N2O) is also important, but its emitted quantity is significantly less than of the previous two. In addition, emission level of N2O is not a subject of regulations; therefore N2O is not primarily considered herein. NOx is formed both from atmospheric nitrogen - N2, and from nitrogen contained in the combusting fuel. The mechanisms responsible for the conversion of the atmospheric nitrogen are [4]:

The use of biomass as a fuel is related to the very beginnings of human civilization. Nevertheless, it still remains one of the largest renewable energy resources in the world. One major difference occurring nowadays is the usage of highly efficient combustion systems burning traditional and new biomass resources, with strictly controlled pollutant emission. Biomass conversion technologies are generally considered environmentally friendly (as biomass is CO2 neutral and its pollutant emissions are less pronounced in comparison to coal) and economically sound. In that frame, it is essential to understand emissions of individual hazardous compounds, their environmental and health effects in order to reduce and minimize it. The emission of nitrogen oxides (NOx) is one of the most important challenges in the field. Multiple negative effects of nitrogen oxides are reflected primarily in the formation of photochemical smog, ground



• thermal

(Zeldovich) mechanism - high temperature (> 1300 °C)

Corresponding author. E-mail addresses: [email protected] (M. Mladenović), [email protected] (M. Paprika), [email protected] (A. Marinković).

http://dx.doi.org/10.1016/j.rser.2017.10.054 Received 10 October 2016; Received in revised form 7 September 2017; Accepted 26 October 2017 1364-0321/ © 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: Mladenovic, M.R., Renewable and Sustainable Energy Reviews (2017), http://dx.doi.org/10.1016/j.rser.2017.10.054

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

2. DeNOx techniques

oxidation of atmospheric nitrogen by oxygen in combustion air and mechanism - combustion of atmospheric N2 and hydrocarbons in the rich mixture conditions /very low air-fuel ratios. The mechanism responsible for the conversion of the nitrogen contained in fuel is: oxidation of fuel-bound nitrogen which forms fuel - NOx also called feed NOx.

• prompt •

The selection of biomass combustion technology is determined mainly by the characteristics of the fuel, its accessibility and annual availability, existing environmental legislation, the costs, and performance of the equipment available, as well as the energy and capacity needed (heat, electricity). For any biomass combustion application, emission reduction, besides efficiency improvement, is a major goal. Accordingly, controlling NOx emissions is becoming increasingly demanding technical challenge as tightening emissions regulations are being imposed. The NOx control technologies can broadly be classified into:

Comprehensive analysis of the biomass combustion investigations [4–17] have derived a few facts related to NOx emissions, crucial for improving technologies of biomass combustion and for the proper and adequate choice of deNOx technique:

• The amount of thermal and prompt NO •







1. pre-combustion which involves the use of low nitrogen biomass or biomass blends with reduced the total N content, 2. combustion control or primary measures - modifying the design and operating features of the combustion unit, and 3. post-combustion techniques (end-of-pipe treatment) or secondary measures - flue gas treatment (FGT) after the combustion process.

x is negligible due to relatively low combustion temperatures conditioned with low melting temperature of biomass [4–8,17]. During biomass combustion, the oxidation of fuel-bound nitrogen is the dominant mechanism of forming NOx [1] and an emission-related problem could be expected at fuel-N concentrations above 0.6 wt% (d.b) [5]. This especially refers to straw, cereals, grasses, grains and fruit residues. The formation of NOx from fuel-bound N takes place predominantly in the gas phase oxidation of the nitrogenous species released with the volatiles (66–75%) and less through the heterogeneously catalyzed oxidation of the nitrogen retained in the char (< 25%) [5]. N released with the volatiles from the biomass fuels generally ends up as NH3 (which is attributed to the main N functional groups present in biomass - amino groups), rather than as HCN which is the most important precursor for N2O formation [9,11]. So quite low N2O concentration has been detected during biomass combustion [7–10]. Both NH3 and HCN can be oxidized to NO during subsequent combustion: 4NH3 + 5O2 → 4NO + 6H2O. At the same time, the two precursors (especially NH3) can also serve as reducing agents for NO reduction, as follows: 2NH3 + NO + NO2→2N2 + 3H2O. A catalytic effect of char and ash on NOx formation and reduction has been recorded [12]: char provides a catalytic surface for the gas phase NO reduction by CO and, similarly, the ash, especially the presence of CaO, MgO, and Fe2O3, can catalyze the reduction of NO and NO2. NOx reduction catalytic effect of biomass char is less pronounced compared to coal char because the biomass chars have higher oxidation reactivity resulting from the high levels of oxygen content, the presence of alkali and alkaline-earth elements (catalytically active) in the char matrix, and larger char surface area in biomass fuels [13–16,18]. It is difficult to reduce CO and NOx simultaneously - decreasing one may result in an increase of the other [4,19–21].

The pre-combustion measures refer to some affordable solutions as an informed choice of biomass (e.g. knowledge on fertilizer treatment, the length of storage and harvest time because natural senescence decreases N content, as the N is remobilized to the roots or rhizomes) and/or pretreatments with a target of minimizing heterocyclic-N –compounds.1 The chemical composition of a fuel has a direct influence on NOx emissions therefore in selecting the type of biomass it is necessary to bear in mind the classification given in Table 1. Proper selection of the biomass is illustrated in the following example - burning the kernel of various non-food cereal crops, as third class biomass fuel (see Table 1), leads to the high nitrous oxide (NOx) emissions. The answer to this issue is to burn oats because it has a lower N content in relation to other cereals. The lower the protein level, the lower the nitrogen content of the kernel and, consequently, the lower the resulting NOx emission. For that reason, oat grain is the most common second-rate cereal2 used for residential heating in Sweden [23]. The pre-combustion measures are also a modification of the fuel composition by usage of fuel additives [24], biomass fuel blending [25] and biomass co-combustion with fossil fuels [26,27]. Significant reduction potential was observed for blends with wood and herbaceous biomass [25]. The use of biomass/coal co-firing decreases both NOx and SOx emissions [28]. Thus coke (the quenched char from coal) as ultralow nitrogen fuel can be very successful at co-combustion of biomass rich in nitrogen because nitrogen in the volatile fraction of the coal is removed in making coke, while it is very rich in carbon (acting as a catalyst for the NOx reduction). Primary and secondary measures of NOx reduction are given in Fig. 1.

The literature from the field of the denitrification techniques in the industrial applications is not generally easily obtainable. In order to enable a better understanding of the DeNOx mechanisms and techniques, and help with decision making in praxis, an attempt has been made here to systematize experimental and practical experiences and to give a comprehensive overview and analysis. The goal was to point out their advantages and disadvantages, the availability, affordability, and maturity of application, and to enable easier selection of NOx control technology for a selected biomass fuel, either from among those already proven or from a growing number of new and promising. The focus was the post-combustion measures and the most successful among them selective catalytic (SCR) and non-catalytic reduction (SNCR). The special challenges for SCR and SNCR application were considered, and a comparative overview of advantages and disadvantages is given, in accordance with several different criteria. In addition, guidelines for the selection of denitrification measures are given, depending on the biomass combusted, combustion technology used, and the installation capacity.

3. Combustion control systems Combustion control systems are commonly applied with various possible options, each resulting in a more or less significant reduction of NOx-levels formed during combustion. These options often combine several measures that rely on any of following strategies: (a) reducing peak temperatures in the combustion zone: (b) reducing the gas residence time in the high-temperature zone, (c) reducing oxygen concentrations in the combustion zone and (d) improving mixing conditions. These can be achieved either through process modifications or by modifying operating conditions on existing furnaces without the use of additional reactor/flue gas pollution abatement equipment behind the 1 heterocyclic N compounds seem to decompose mostly through HCN, while amino acids and proteinic nitrogen appear to produce mostly ammonia NH3. 2 waste grain unsuitable for human or animal nutrition.

2

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

increased by using of additional compounds, such as ammonia, urea or metals, being introduced simultaneously with the main reburning fuel. This modification is known in the literature as advanced reburning [31]. An application of NH3, in this case, has a few drawbacks, such as ammonia slip3 - the build-up of ammonium salts on downstream heattransfer surfaces and the creation of ammonium chloride (which will be discussed later). Consequently, iron or low-alloy steel waste as low-cost

Table 1 Classification of biomass fuels [22]. Class

Type

Nitrogen content [w% d.a.f]

1 2 3

Woodlike Strawlike N-rich (agricultural & herbaceous)

< 0.3 0.3–1 >1

Fig. 1. Overview of deNOx techniques.

main combustion zone. Process modifications include the application of a low NOx burner, air/combustion staging, gas recirculation, reduced air preheating and combustion rates, water or steam injection, and low excess air or surplus oxygen present for combustion. The method of combustion control used depends on the type of the boiler and the applied biomass combustion technology. As biomass combustion usually takes place in relatively low-temperature conditions for inhibiting or alleviating the alkali-related issues, the first two strategies ((a) and (b)) are not of interest. Water or steam injection, as usage of different oxidant (with lower N content), are effective only in controlling thermal NOx and therefore are not of interest for this paper, too. Staging the introduction of combustion air is a measure of splitting the combustion air stream which creates a fuel-rich primary and fuel-lean secondary zone. The basic idea of air-staging is to reduce NOx formation by reducing oxygen availability in the flame and by lowering flame temperature peaks. In the air-staged combustion process, the fuel-rich zone reduces NOx formation. Namely, fuel-N is converted to intermediate volatile components such as HCN and NHi (i = 0, 1, 2, 3). These can be oxidized to NOx if oxygen is available, which is the case in conventional combustion. But if less oxygen is present in the reduction zone, intermediates can interact and form N2 in reactions such as NO +NH2 = N2+H2O. The complete combustion is achieved after the addition of air in the second (burnout) zone. Using air staging with fine control of the residence time and primary excess air ratio, an NOx reduction of even 50–80% [19,29] is recorded. Significant reductions in NOx emissions (20–56%) have been recorded in the grate (fixed bed) combustion of wood type biomass (pellets, logs …) and some biomass fuel mixtures [1], in small-scale combustion applications [20,30]. Due to the likely increase in flame length and flame instability, in general, this method requires very precise control of the combustion process. Fuel staging (reburning) known as a cost-effective primary method of NOx removal involves the injection of a proportion (10–20%) of fuel above the combustion zone, creating a fuel-rich secondary combustion zone where NOx formed from the primary combustion (fuel-lean) zone is reduced through decomposition [27]. The de-NOx efficiency can be

and readily available compounds have been used in fuel-rich combustion zones as an additive in advanced reburning. In that case, the predominant reactions are: NO+Fe→ FeO+1/2N2

(1)

CO+ FeO→ Fe+ CO2

(2)

CO+ NO→ CO2+1/2N2

(3)

The role of CO is to chemically reduce the oxides back to metallic iron in reaction (2). The method of advanced reburning is of particular interest to fluidized bed combustor [32], but is still under investigation and has no application in real-scale processes. In experimental studies, it has also been shown that hybrid NOx reduction measure combining fuel blending (coal and woody biomass) and air/fuel-staging [27,33,34], applied to a pulverized fuelled furnaces, can significantly reduce NOx, even more than 50% [27]. Flue gas recirculation (FGR) technique can significantly reduce, primarily, thermal NOx production by reducing flame temperatures and overall excess air. Thereby, the recirculated flue gas acts as an inert gas containing mainly CO2, H2O, with a low level of O2, N2, and CO, depending on combustion efficiency. The high concentrations of CO2 and particularly CO caused by FGR could play a significant role in NOx emissions as the reducing agents. This measure has high NOx reduction potential (more than 20%) for low nitrogen fuels. [35] It is necessary to bear in mind that the implementation of FGR reduce flue gas temperatures and so reduce boiler output [36]. Low NOx burner is a type of burner that is typically used in utility boilers to produce steam and electricity and therefore may be of interest for the coal and biomass co-combustion. Also, the fluidized bed combustion (FBC) conditionally falls into the

3 Tailpipe emissions of ammonia that occur when exhaust gas temperatures are too low for the reduction reaction to occur or when too much reductant is feed for the amount of NOx present.

3

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

involves flue gas irradiation with fast (300–800 keV) electrons, which interact with the main flue gas components (N2, O2, H2O, and CO2) and generate oxidants (OH∗, O∗, O3). These oxidants react with SO2 and NOx forming sulfuric and nitric acids, respectively, which in turn react with NH3, added to the flue gas prior to irradiation, thus forming ammonium nitrate and sulfate. Although associated with high capital and operating costs, the technique has been applied on a few industrial fossil-fired boilers with results of 80% deNOx and 70% SO2 removal. It also destroys volatile organic compounds [42]. The most widely used dry deNOx techniques are a selective catalytic reduction (SCR) and selective non-catalytic reduction (SNCR), therefore a special attention will be given to them.

“combustion control” category, because this technology enables the simultaneous removal of SO2 and NOx at relatively high efficiencies. The conditions as temperature and the residence time of particles in boilers are very favorable for achieving low emissions of the above-mentioned pollutants. The FBC deNOx will be further discussed later in this paper. 4. Flue gas treatment (FGT) The NOx emission appeared to be more strongly influenced by the nitrogen content of biomass fuels rather than the operating conditions [4] so combustion control techniques alone are often insufficient to comply with the stringent emission standards and/or cannot be applied in existing combustors. Additional reduction is required which is achieved by means of end-of-pipe flue gas treatment technologies. FGT is more effective in NOx reduction, especially for biomass with high nitrogen content, although at a higher cost. Therefore the operation range and overall economics of the facility should be considered before applying FGT. Water soluble gases can be removed effectively either by wet methods or dry absorption. In wet methods of FGT, the liquid is used as an agent to extract the contaminants or pollutants from flue gas and was originally designed as simultaneous SOx-NOx (sulfur-/nitrogen oxides) removal systems. Although simultaneous SO2-NOx removal may be a potential economic advantage, these methods, so far, are limited to small waste stream cleanup and are not suitable for NOx removal from large volumes of flue gases typical for biomass combustion. The reasons for this are:

4.1. Selective catalytic reduction (SCR) Selective catalytic reduction (SCR) is the most advanced and effective method for reducing NOx emissions and can do so by up to 60–90%. SCR entails the reaction of NOx with a reducing agent within a heterogeneous catalytic bed in the presence of O2. Ammonia or urea NH3 (NH3-SCR), hydrocarbons (HC-SCR) and hydrogen (H2-SCR) are used as reducing agents. Both latter technologies (HC-SCR & H2-SCR) have some drawbacks that prevent them from being commercialized for control NOx emission from stationary sources as well as NH3-SCR. Neither HC nor H2 has the ability to react selectively with NOx in flue gases, on the contrarystrongly prefers to react with O2. Apart of that, at temperatures above 500 °C, all of the hydrocarbons are consumed by combustion reactions. Although H2-SCR reduce NOx at low temperatures (T < 200 °C) on supported noble metals (Pt and Pd), this type of catalysis has very highcost issues and catalysts have questionable water and SO2 resistance [43,44], so more studies to the practical application are needed. Despite the fact that these systems cannot offer the performance of NH3-SCR systems, both technologies, especially HC-SCR, are promising deNOx methods in the automotive industry since they use unburned hydrocarbons present in the engine exhaust as reductants, so it does not require special reactants supply. In case SCR entails the reaction of NOx with NH3 (as a reducing agent, stored in the form of liquid anhydrous ammonia, aqueous ammonia, or urea), it occurs within a heterogeneous catalytic bed in the presence of O2 at temperatures in the range of 250–450 °C (for conventional NH3-SCR). Amine-based reagents, ammonia (NH3) and urea (CO (NH2)2), are the unique reducing agents in the SCR process as preferring to react with NOx rather than the far more plentiful oxygen – hence term “selective reduction”.

• nitric oxide (NO) is not soluble in water, although NO •

2 and NH3 are soluble, their concentrations in flue gas are very low compared to NO, therefore, costly reagents are needed for converting NO in the soluble state; the major drawback lies in the necessity of recycling of the collected contaminated liquid used for the flue gas cleaning, by either evaporation or biological treatment.

Several dry adsorption techniques are available for simultaneous control of NOx and sulfur oxides (SOx). This kind of treatment, which uses a chemical/sorbent to absorb NOx or an adsorber to hold it, could be applied in: the combustion chamber (1), the flue gas duct leading to the baghouse (2), or to the electrostatic precipitator (3). One type of treatment uses activated carbon with ammonia (NH3) injection to simultaneously reduce the NOx to nitrogen (N2) and oxidize the SO2 to sulfuric acid (H2SO4). If there is no sulfur in the fuel, the carbon acts as a catalyst for NOx reduction only. Another adsorption system uses a copper oxide (CuO, Cu2O) catalyst that adsorbs sulfur dioxide to form copper sulfate (CuSO4, Cu2SO4). Both copper oxide and copper sulfate are reasonably good catalysts for the selective reduction of NOx with NH3 [37,38]. A spray dryer system, which sprays slurry of powdered limestone and aqueous ammonia into the flue gas, could also be used. The limestone preferentially reacts with the sulfur while the ammonia with the NOx [17]. The byproducts formed by sorption are gypsum (calcium sulfate) usable in the construction industry, and ammonium nitrate that could be used to make either an explosive or a fertilizer. Non-selective catalytic reduction (NSCR) of NOx is associated with three-way catalysts of automotive vehicles and is not an option as a deNOx technique for biomass combustion. The pulse corona-induced plasma process has been shown to remove NOx from flue gases with deNOx efficiencies of 60% being reported [39]. The major issues associated with this technology are the energy consumption required to achieve adequate reduction and the formation of undesirable by-products [40]. The technology is still in the development stage. Electron beam flue gas treatment is a promising new technology that enables simultaneous abatement of SO2 and NOx, generating no waste except a useful by-product (agricultural fertilizer) [40,41]. The method

4.1.1. SCR reactions The selective catalytic reduction of NOx with NH3 (NH3-SCR) is the best developed and widely applied deNOx technique due to its high efficiency, good selectivity and cost effectiveness. The injected liquid ammonia quickly turns into gas, mixes with the NOx in gas and the mixture then reacts at the catalyst (Fig. 2). The predominant reactions are [45]: 4NO + 4NH3 + O2 → 4N2 + 6H2O

(4)

6NO2 + 8NH3 → 7N2 + 12H2O

(5)

NO + NO2 + 2NH3 = 2N2 + 3H2O

(6)

If urea is used as a reducing agent the corresponding reactions are [46]: 4NO + 2 CO (NH2)2+2H2O+O2 → 4N2 + 6H2O+2CO2

(7)

6NO2 + 4 CO (NH2)2+4H2O → 7N2 + 12H2O+4CO2

(8)

Urea is used in a form of white crystal granules, dissolved in water. Aqueous ammonia solution is preferable for SCR, while urea, generally, to the SNCR technology. 4

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

Fig. 2. Scheme of selective catalytic reduction.

with inert elements) and zeolites (crystalline, highly porous natural or synthetic aluminosilicates). These can be extruded into a honeycomb structure or coated/dispersed directly onto a metallic or ceramic support.

Reaction (4) is the dominate reaction for the overall stoichiometry of the NH3-SCR process. Since NO2 commonly accounts for only 5% of the NOx, reactions (4) and (5) play a minor role in the process. Although the aqueous ammonia is commonly used by SCR there is no wet residue or by-product collected: all the reactants leave the system as gases (a mixture of N2 and water vapor H2O - Fig. 2) without any environmental concern. Reactions of complete conversion to N2 and water vapor are, in general, attainable with the best catalysts depending on their selectivity as well as of the amount of NH3 added to the reaction, but in non-ideal cases, undesirable oxidation reactions can also take place: 4NH3 + 5O2 = 4NO + 6H2O

4.1.3. SCR system arrangement The most common and successfully commercialized industrial catalyst for NOx removal is vanadium pentoxide (V2O5), supported on titanium dioxide (TiO2) promoted by the addition of either tungsten trioxide (WO3) or molybdenum trioxide (MoO3). TiO2 is considered as the best support for SCR catalysts for two main reasons: it has high SCR activity and is resistant to SO2 poisoning, thus prolonging catalyst life. Titania is used in the form of anatase, because of their larger surface area of 50–120 m2/g in comparison to less than 10 m2/g for the rutile phase. A typical commercial vanadia catalyst consists of 1–5 wt% V2O5 and 10 wt% WO3 (alternatively 6 wt% MoO3) supported on high-surface-area TiO2 (anatase). The role of WO3 and MoO3 is to prevent the transformation of anatase to rutile and to block adsorption of SO3, thereby preventing sulfation of the support [47]. The ideal operating range titanium/vanadium catalysts are 260–425 °C, which falls under medium operating temperature, while the zeolites operate at a higher temperature range of 345–590 °C (see Table 2. and Fig. 3). In both cases specified temperature windows makes necessary that the SCR reactors have to be located upstream of the desulfurization and particulate control device in order to avoid expenses of flue gas reheating (Fig. 4, high dust case). This location brings inherent problems like an accelerated catalyst deactivation (through exposure to high concentrations of small particulates and SO2) and installation difficulties. Namely, the optimal gas temperature may be often in the center of an existing economizer or air heater, resulting in considerable facility structural modifications and additions, such as new flues/ducts and the necessity of upgrading the induced draft fans. Because of the possible extensive modifications required, a conventional SCR may not be cost effective to retrofit into smaller units. Mentioned problems could be avoided by locating the

(9)

4NH3 + 3O2 =2N2 + 6H2O

(10)

2NH3 + 2O2 =N2O + 3H2O

(11)

SO2 + 1/2O2 = SO3

(12)

Reactions (9)-(11) entail the consumption of ammonia in reversal process of the NOx removal and in the formation of N2O as a byproduct. These reactions occur on SCR catalysts in the absence of NO in the flue gas, but become negligible in the presence of NOx. In the case of sulfurcontaining fuels, SO2 produced during combustion oxidize to SO3 over the catalyst (12) which could further react with water present in flue gas and with unreacted NH3 leaving the catalyst bed (slip NH3) to form a highly reactive sulfuric acid (13) and ammonium sulfates (14 and 15): SO3 + H2O = H2SO4

(13)

NH3 + SO3 + H2O = NH4SO4

(14)

2NH3 + SO3 + H2O = (NH4)2SO4

(15)

In spite of the fact that slip NH3 and SO3 are generally present both at ppm levels, this still could cause severe problems such as corrosion, catalyst deactivation due to ammonium salts deposition on the catalyst (especially if the temperature is less than 300 °C) and consequently pressure drop problems. Accordingly, the SCR catalysts should be highly selective, particularly with respect to SO2 oxidation/poisoning.

Table 2 SCR classification as a function of catalytic activity temperature domain.

4.1.2. SCR catalysts The selection of catalyst is crucial to the operation and performance of the SCR system. The catalyst may comprise a single component, multi-component or active phase with the support structure, which gives thermal or structural stability or increases surface area. Most of SCR catalysts are composed of active metals or ceramics with a highly porous structure. The reduction reaction occurs within the catalyst's pores in so-called activated sites. After the reduction reaction occurs, the site reactivates via rehydration or oxidation. Over time, the catalyst activity decreases, requiring replacement. Depending on the process parameters, primarily operating temperatures, efficiency of NOx removal, and catalyst regeneration, various catalysts have been used for NH3-SCR including noble metals, metal oxides (titanium, vanadium, iron), activated carbon (pulverized high-carbon coal or coke, mixed

Temperature catalyst

5

High Zeolite/zeolite-based materials on ceramic substrate

Medium V/Ti/W on honeycomb or plate type shapes

Low Pt- or Mn-based, porous extrudates in bed reactor

345–590 °C very high NOx conversion

260–425 °C most broadly used

very low NH3 slip

mature technology

NH3 destruction sulfur tolerant above 425 °C

sulfur tolerant

150–300 °C narrow temperature window temperature window shifts not sulfur tolerant

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

dioxin/furan removal.

• Active carbon fiber (ACF) is a superior adsorbent due to its large

• Fig. 3. Three major families of SCR catalyst [48].





Fig. 4. SCR location: B - boiler, AH - air preheater, ESP -electrostatic precipitator/another particulate collector, SCR- selective catalytic reduction, FGD - flue gas desulfurization [48].

SCR unit at the very end of the flue gas pollution abatement units (Fig. 4, tail end case) where the flue gas is relatively clean which implies the application either of: (a) low-temperature SCR catalysts or (b) an additional heat source to heat the commercial vanadium catalyst (classic TiO2/V2O5/WO3), which can be achieved: ● by reheating of the flue gas (by using either oil- or natural gasfired duct burners) or ● by economizer bypass resulting in less energy transfer to the feed water, which significantly increases the SCR operational costs.

number of well distributed micro pores on the high surface area (1000–2000 m2 g−1) which enables metal-oxides to be highly dispersed during the preparation process of the catalyst. Mn and Ce had the best effects as the active component in ACF [50], but the inadequate SO2 tolerance is still a problem for their industrial application. However, this can be enhanced by loading V2O5 with extraordinary inherent resistibility to SO2, on carbon materials [51]. But, despite the advantages of absorption and despite all modifications, the stability of carbon containing a catalyst, ACP, and ACF, in O2 is low. Al2O3 as the carrier has high-temperature resistance, withstands wear and tear and has a surface area below 200 m2/g. The tested catalyst of MnOx/CeO2–ZrO2–Al2O3 (MnOx/CZA), with 10% manganese, prepared by incipient wetness impregnation method, has shown a reduction of 90% NO to N2, in the temperature range of 143–300 °C, using NH3 as a reducing agent [52]. Alumina supported silver catalysts prepared by coprecipitation, impregnation and single step sol–gel methods was studied for the selective NOx reduction by C3H6 in the presence of oxygen. The catalyst prepared by sol–gel method had the highest activation for selective reduction of NOx, with almost 100% [53]. Over the last few decades, there have been increasing interests in the development of LTC containing transition metal oxides, such as V2O5, Fe2O3, CuO, and MnOx. TiO2 supported Mn-, Cu-, and Crcatalyst is highly active at low temperature (80–250 °C). Among all low-temperature SCR (LTC), Mn-series catalysts have superior lowtemperature SCR activity [49,54]. In accordance to that, up to now various Mn-based catalysts were developed: Mn/TiO2 [55] Mn–Fe/ TiO2, Mn–Ce/TiO2 [56], Mn–Ni/TiO2, Mn/Al2O3, Mn/SiO2, MnOx/ AC, MnOx/CeO2–ZrO2–Al2O3, Mn–Ce oxide, Mn–W oxide [49], Mn–Fe oxide, Mn–Zr oxide, Mn–Cu oxide, Mn–Fe spinel, and mesoporous MnO2–Fe2O3–CeO2–TiO2 [49,51]. Although V2O5/TiO2 -based catalysts are highly commercialized, there are continuous efforts in the improvement of their low-temperature activity under abundant SO2 conditions. It was found that addition of Ce and Mn can significantly promote their low-temperature activity and the catalytic performance could be further promoted by loading V2O5 on a TiO2–SiO2–MoO3 support using a co-precipitation method. In this case, the SO2 resistance is improved by increasing SiO2 content in the support [51].

However, whether they are the Mn-, V2O5- or noble metal-based LTC, generally, they suffer from SO2, some of them even from H2O poisoning, which still has not been satisfactorily solved (Table 2). After all, since most of LTC was only researched in laboratory conditions, more studies of real-scale application are needed. Besides high dust and tail end SCR system locations, there is also a low dust SCR system arrangement (Fig. 4) which bypasses the drawbacks of the high dust configuration. A low-dust SCR system increases catalyst life by reducing concentrations of particulates and catalyst poisons in the SCR reactor due to the fact that ash removed by the ESP typically contains harmful constituents (sodium, potassium, lead, arsenic, etc.). Furthermore, fewer particles mean lower catalyst volume which reduces the costs for SCR. Also, a temperature drop of the flue gas, after it flows through the ESP/other dust precipitators, often do not decrease to the point where reheating is required. Just because the low dust SCR operates without supplementary heating, the installation of high-temperature ESP/dust precipitators, in this arrangement is necessary, which significantly increases the investment in the case of retrofitting power units. Moreover, the air heater is more subjected to the deposition of ammonium salts (Fig. 4). When selecting SCR location it is necessary to bear in mind all pros and cons of all three possible cases and carefully choose the most economical option. Despite that tail end systems improve the most

The application of low-temperature catalysts solves problems of SCR displacement at tail end case as they can work well around 250 °C or even lower [49]. For that purpose, initially, Pt-based catalysts were used. Due to the narrow temperature window (180–250 °C) and a fact that above 220 °C significant amounts of N2O (11) are produced, Ptbased catalysts are not commonly used today. Up to now, low-temperature catalysts (LTC) have been studied without carriers or using: ACP (activated carbon pellet), ACF (activated carbon fiber), GAC (granular active carbons), Al2O3, TiO2, zeolites, and SAPO (silica aluminophosphate) as the carriers.

• Mn–Ce mixed-oxide catalyst had the best performance among LTC without carriers. • Activated carbon, in the form of sintered pellets of pulverized coal mixed with inert elements (ACP), is used for integrated NOx and

6

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

Fig. 5. SCR catalyst support geometries: a) extruded ceramic monolith; b) composite catalyst (either on a metallic or ceramic monolith) c) zeolite pellets and d) plate type, e) metal wire gauzes, f) ceramic foam.

type catalysts for high flue dust applications are preferable for biomass combustion (see Table 3).

significantly the lifetime of the catalyst, current operating experience and the availability of improved catalysts for the high dust SCR system makes the tail end SCR system the least attractive of the three options listed (Fig. 4). It would be a more cost effective option in the future with development and successful application of new low temperatures catalysts.

4.1.5. Fixed and fluidized bed arrangements of SCR reactor A high variety of catalyst supports (pellets beds, foams, metal wire gauzes, honeycomb monoliths and pleated metal plate with different channels shapes-Fig. 5) are commonly constructed in the form of a fixed-bed/packed bed reactor, which provides high surface area to volume ratio. Due to its structural integrity and ease of scale-up, the catalytic monolith and pleated metal plate structures have been the most widely used. Catalyst elements placed in a frame form a catalyst module, which protect the elements from damage and keep them in place. This design enables easy handling during cleaning or replacement. The modules stack together in multiple layers creates a fixed reactor bed in a manner that provides total required catalyst volume [57,58]. Pellets have greater surface area than honeycomb or plate type catalysts but are more susceptible to plugging so their use, in fixed bed arrangements, is limited to very clean fuels such as natural gas, so inapplicable to the biomass combustion system, except when biogas is using. Besides pellet for biogas engine applications packed beds of wire gauze and foam filled reactors are also applicable [59,60] (Fig. 5). Apart from the packed structure in fixed bed arrangements, pellet catalysts in fluidized beds are also available. In this approach, flue gas, containing NOx and NH3, fluidizes a bed containing granulated or small catalyst particles [61] (Fig. 6). SCR fluidized bed reactor requires mechanically strong catalysts to minimize attrition losses and show advantages over fixed beds including a better gas solid contact, less chance for dust clogging which allows treating of flue gases with high dust loadings but with the pressure drop higher than that through conventional packed beds.

4.1.4. SCR catalyst support geometries As well as material, its configuration determines the properties of a catalyst. The conventional configuration of SCR catalysts is the catalytic monolith structure, with the catalysts preformed into two shapes, honeycomb-type (extruded and self-supported blocs in the form of squares or honeycombs with catalyst material extruded and dispersed inside them) and plate-type, which have a metal support covered with the deposited active substance (Fig. 5). The catalyst material application involves also a procedure of wash coating a layer of catalyst onto a monolith matrix consisting of cordierite (a silicate mineral) or corrugated ceramic paper. The plate type catalysts have a higher resistance to erosion and deposition of dust particles than honeycomb ones and are commonly used for high dust installations (see Table 3). Pellets (especially of activated carbon) are also used as the catalyst geometry. Zeolite catalysts are produced in granular form/pellets and also as honeycombs. Because of zeolite wider operating temperature range (see Table 2) and that they do not contain toxic metals such as vanadium, zeolites based SCR received much attention recently. As the reserves of natural zeolite are very limited, and the amounts of catalysts demanded by the SCR-NOx technologies are very high and difficult to be accomplished, the focus has been moved to synthetic zeolite production for the catalysts preparation, making them cost prohibitive. Considering the generally lower heating value of biomass fuel compared with conventional fossil fuels, biomass combustion generates a greater flow i.e. the higher volume of flue gas. Consequently, the plate 7

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

Table 3 Characteristics of plate type and honeycomb catalysts.

Honeycomb

Plate

pitch • Variable dust deposition • Low erosion resistance • High poisoning resistance • High drop • Low-pressure clearance and height variable • Plate-to-plate flue dust applications • High • Biomass combustion applications

specific surface area • High activity in low-dust flue gases (low volume) → Low-dust flue gas applications • High length and number of cells variable, ~ 6 to 300 (cpsi = cells/in .) • Honeycomb • High-grade fuels combustion 2

4.1.7. Brief overview of the special challenges for SCR application The lifetime of the catalyst dominates the cost of an SCR. Deactivation of catalysts could be chemical, thermal and mechanical, caused by mechanisms of poisoning by vapors of volatile metals and SO2, thermal degradation/thermal sintering, fouling and plugging by dust, vapor-solid, and solid-solid reactions, and attrition/crushing resulting in catalyst bed erosion. The SCR process can be also problematic due to high risks over-oxidation of NH3 to N2O (11) and NO (9), the oxidation of SO2 to SO3 (12) and the potential occurrence of an ammonia slip-stream in the exhaust. One of the major challenges for SCR application in biomass combustion system are alkali metals present in fairly large quantities in biomass-fuels, which can seriously deactivate catalyst either by physically masking or by chemically poising. Catalyst regeneration can be achieved by washing the catalyst with water (and/ or sulfuric acid), after taking the catalyst out of the system or by in-situ washing/blowing system [63]. When catalysts contaminated with biomass ashes are washed by water it is necessary to be particularly cautious. As the alkali metal salts deposited on the catalyst are water soluble, plentiful water has to be used during the washing process to minimize the alkali metal concentration in the washing solution, avoiding possible additional catalyst deactivation. Apart of the aforementioned drawbacks, there are some concerns associated with anhydrous ammonia storage and disposal of the spent catalyst. Therefore, the capital costs associated with SCR systems are high.

Fig. 6. Schematics of deNOx fluidized bed reactor.

4.1.6. Basic design of typical SCR deNOx reactor The NH3-SCR system utilizes gas phase ammonia (NH3), in a way that the ammonia, either in anhydrous or aqueous form, has to be vaporized and subsequently diluted with air before injection through spray grid/system of nozzles in front of the catalyst (Fig. 7) [58]. A solution of urea can be used as well. Mixing achieved by a system of nozzles can be further improved by a static mixer, placed in the flue gas duct. As anhydrous ammonia is a gas at a normal atmospheric temperature, and therefore must be transported and stored under pressure, which often requires special permits, aqueous ammonia is more commonly employed, although it requires more storage capacity than anhydrous ammonia. Thus necessary equipment, except those listed, includes: SCR reactors, ammonia storage tank, air compressor, vaporizers and mixing chamber, piping and pumping equipment, steam supply piping for the SCR reactor sootpiping blowers, air ductwork between air blowers, flue gas ductwork consisting of insulated duct, static mixers, turning vanes, and expansion joints, ash handling system, induced draft fans, and associated control instrumentation/PLC (see Fig. 7). In case the boiler is operating at a lower load, an economizer bypass has to be used in order to increase the flue gas temperature to the optimum range for SCR systems which used commercial catalysts (metal oxides based). For a given NOx removal efficiency, higher NOx levels at the SCR inlet require more catalyst volume. Hence, SCR is generally more cost effective for sources which emit less NOx, since the required catalyst volume is minimal.

4.2. Selective non-catalytic reduction (SNCR) Selective non-catalytic reduction (SNCR) is a process of reduction of NOx to N2 in the presence of oxygen by reaction with amine-based reagents, either ammonia (NH3) or urea (CO(NH2)2) at the temperature window of 800–1100 °C, the higher temperature being needed for urea (Fig. 8). Due to perform at high temperatures, this process does not need a catalyst to initiate the reactions. Although the reagent can react with a number of flue gas components, the NOx reduction reaction is favored over other chemical reaction processes for a specific temperature range and in the presence of oxygen; therefore, it is also considered a selective chemical process. SNCR systems can reduce NOx emission by 30–70% [64], but that is highly variable for different applications. Peak NOx removal is occurring at 870 for ammonia i.e. at 1000 °C if urea is used [65] and when the reaction temperature increases over 1000 °C, the NOx removal rate decreases due to thermal decomposition of ammonia/urea. On the other hand, at lower temperatures the NOx and the ammonia/urea do not react and undesirable ammonia slip-stream may increase (ammonia can react with other combustion species, such as sulfur trioxide (SO3), to form ammonium salts). In case NH3 is the reagent, the reaction scheme is the same as with SCR (4−6) and if urea is the reagent used the reaction are (7) and (8). 8

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

Fig. 7. SCR process flow diagram [62].

injection at temperatures too low for effective reaction with NOx or from reagent uneven distribution. The reagent injection system must be able to place the reagent where it is most effective within the boiler because NOx distribution varies within the cross section. Thus proper SNCR design implies that, more ammonia must find its way to the center of the furnace where more NOx is formed and less near the walls, as they are cooler, otherwise NO in the center meets insufficient ammonia for reduction and excess NH3 near the walls slips through affecting on downstream equipment by forming ammonia salts. Hence, distribution of the reagent can be especially difficult in larger boilers because of the long injection distance required to cover the relatively large cross-section of the boiler. Apart from that, the temperature of flue gas in the boiler vary also by height and may be caused by changes in the boiler load or calorific value of the fuel (which is particularly

The reagent ammonia or urea can be injected directly into the combustion chamber. Urea-based SNCR has advantages over ammoniabased systems because urea is non-toxic and as less volatile, thus it can be stored and handled more safely. Further, urea solution droplets injected into the boiler can penetrate further, enhancing the mixing with the flue gas, which is particularly important for the large units. However, urea is more expensive than ammonia and urea based reduction generates more N2O than ammonia-based systems [66]. 4.2.1. Special challenges for SNCR application A critical issue for SNCR systems is finding a proper injection location in accordance with the reduction of ammonia slip and the appropriate temperature window for all operating conditions and boiler loads. Ammonia slip from SNCR systems could occur either from

Fig. 8. Scheme of selective non-catalytic reduction.

9

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

operation but with the increase of deposits on the heating surfaces the temperature window moves upwards. In order to overcome uneven temperature distribution and imbalances resulting from the boiler load, it is necessary to predict additional injection points and levels to accommodate operations at low loads (see Fig. 9). Another solution is the addition of enhancers such as H2, CO, H2O2, C2H6, light alkanes and alcohols to the reagent to lower the temperature range at which the NOx reduction reaction occurs. The use of enhancers reduces the need for additional injection locations [68]. The temperature window for ammonia is 850–1000 °C, with the optimum temperature being 870 °C and, and when using urea, the temperature window is wider (800–1100 °C), with an optimal temperature of 1000 °C. So another possible solution, although not in a common practice, is shifting of the reactants: to shift the system to ammonia at low-load operation while urea is used at fullload (as ammonia corresponds to lower optimum temperature). 4.2.2. Basic design of typical SNCR deNOx reactor A typical SNCR system involves reagent storage and handling systems (similar to those for SCR systems), multi-level reagent-injection equipment, and associated control instrumentation. Because of the higher stoichiometric ratios required at equivalent efficiency, both NH3 and urea SNCR processes require larger quantities of reagent than SCR systems to achieve similar NOx reductions [45]. The technology is attractive due to its relative simplicity, the catalyst-free system (and hence free of associated problems), ease of installation on existing plants, applicability to all types of stationary-fired equipment, lower capital cost, the fact that it is largely unaffected by fly ash and usability with other NOx emission control technologies.

Fig. 9. Example of three injection levels with the possibility of switching to a single injection ports/ individual lances [67].

evident at the biomass use) or be result of the increasing deposits thickness on the heating surfaces or be result of the burner configuration, etc. The location of the optimum temperature region for SNCR application shifts within the lower part of the boiler during low-load

4.3. Main differences between SCR and SNCR The main differences between the two most commonly used post combustion deNOx systems are given in Table 4.

Table 4 Comparison of SCR and SNCR. Operating conditions

SCR

SNCR

Temperature Tolerance of temperature fluctuations

250–450 °C (classic TiO2/V2O5/WO3) ± 90 °C For low-load operations, an economizer bypass can be used to increase the flue gas temperature

Reagent injection Amount of reagent

Occurs downstream of the combustion unit

NH3/NOx ratio [36,69,70] Residence time within temperature range/ space requirement Reaction zone

0.8–1 Fractions of seconds/minimum space

800–1100 °C Higher Cost effective for seasonal or variable load applications Additional injection points/ addition of enhancers are required to accommodate operations at low loads Inside the combustion unit 3–4 times as much reagent is required to achieve NOx reduction similar to those of SCR 1.5–2.5 Seconds/large space

Reaction zone geometry NH3-slip Cost

investment operating

Design Good candidate

deNOx efficiency Environmental impact

Use of a catalyst Reaction chamber/bed being separate from the combustion unit No limits. Pitch variable

Higher High due to the increased energy consumption for: the blower overcoming the pressure drop, reheating the flue gases and regenerating/replacing the catalysts Standard design applicable. Without major modification of boiler design. More cost effective for sources which emit less NOx Can be used to remove both NOx and dioxins/furans (PCDD/ PCDF) from the flue gases [46]

• • •

Higher [71] production of catalysts higher energy consumption disposal of the spent catalysts

• • •

10

/ Reaction chamber = combustion unit Limited by spray distance and mixing zone At least 2 times higher than SCR High NH3 slip at low emission Higher for reductants

High sophisticated flow design + control mandatory More cost effective at higher levels of uncontrolled NOx better suited for applications with high levels of particulate matter in the flue gas stream than SCR Lower NOx reductions than SCR

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

catalyzes the conversion of NH3 to NOx whereas decompose N2O. Put simply, the more lime (for SOx control, and for HCl and HF capture as well) is added into the FB, the more NOx and less N2O is formed [44,49]. This remark is worth mentioning in the case of the co-combustion of coal and biomass since biomass itself has negligible sulfur content (generally less than 0.2% on a dry matter basis, with a few exceptions as e.g. for canola and beans S-content, are 0.5% 0.7% d.b., respectively [74]).

When a high NOx removal is required and installation space is limited, combined tandem application of NOx control techniques- hybrid SNCR-SCR (first SNCR, then SCR in the duct) could be the optimal choice. A hybrid SNCR-SCR system is typically designed in a manner that intentionally generates ammonia slip from SNCR is used as the reagent feed to the SCR catalyst, which provides additional NOx removal [72]. A hybrid SNCR-SCR system can be also installed in configurations that including:

• SNCR with catalyzed air preheater (AH with catalytically active heat transfer elements), • SNCR with a combination of in duct SCR and catalyzed air preheater [48] and • SNCR with in-duct SCR and then adsorption before the ESP which is

Applicable combustion control systems in FBC

• The NO

x formation can be further limited by minimizing the excessair ratio. Once again, it is necessary to bear in mind that lower O2/ fuel ratio is preferred for NOx and N2O reduction but it increases SO2 emission for the conventional bed material [75] (important fact in the case of biomass and coal co-combustion)

referred to as

“polishing”.

• Usual combustion control measure applied in modern FB boilers is

In this way, an overall NOx reduction up to 90% can be achieved [5]. The catalyst volume required in a hybrid system can be reduced from that of an SCR-only application, which reduces the space and investments needed, the amount of the reagent is smaller than when the systems are used separately, so that the hybrid system could represent a fine balance between capital and operating costs for a specific NOx reduction level. However, one of the major limitations of the application of this systems is inadequate mixing of the flue gas downstream of the SNCR zone in order to redistribute NOx and ammonia [72]. Furthermore, experience with these systems is limited. Although FGT/post combustion control techniques may have advantages over many other combustion control options for given operating conditions and environmental restrictions, cost studies should be conducted in order to determine the best NOx control application prior to choosing a particular technology solution.



CFB: Air staging is not easily feasible due to high solid densities in freeboard and near the furnace walls and can be applied only to high loads.

• It is necessary to bear in mind that air staging decreases the sulfur capture efficiency which is one of the benefits of CFB technology. • Fuel reburning is a cost-effective method but has no application data in real-scale biomass FB installations.

5. Applicability of deNOx techniques to individual biomass combustion technology

Applicable secondary measures of NOx reduction in FBC:

• Both selective and non-selective catalytic reduction processes (SCR

5.1. Fluidized bed combustion (FBC) systems



Fluidized Bed (FB) boilers could efficiently combust a broad variety of biomass or other low-grade fuels that are difficult or impractical to burn in conventional boilers, from small CHP plants to large utility power plants. Although NOx emissions are noticeably lower than in conventional pulverized fuel combustion, the FBC boilers may also generate a significant proportion of the NOx emission, especially when N-rich fuel is used.

• The low combustion temperature of FBC (800–950 °C) is advanta• • •



the air staging, with combustion air supplied to the furnace on two or three different levels. BFB: Below 50% load, the air staging is limited; the optimum load of the BFB boiler for this combustion control technique is between 70% and 100%.

geous for the suppression of NOx emissions as the formation of thermal NOx is avoided. All NOx emissions are caused by fuel-bound nitrogen. For BFB (bubbling fluidized bed) boilers the correlation of the NOx emissions and nitrogen content of the biomass fuel is obvious. For instance, some agro fuels, especially cereals, have high nitrogen content [4] emitting higher NOx than wood, which has much lower N- content. For CFB (circulating fluidized bed) boilers NOx emissions do not depend on the fuel nitrogen content as much as in the BFB boiler. At CFB heterogeneous reaction of NOx reduction on residual char surfaces are more intensive due to the significant amounts of residual char in CFB hot loop circulation. Thus, the NOx emission at biomass combustion which has high N but low volatile/high char content can be very low in CFB [73]. FBC has been developed as an in-situ SO2 capture combustion technology which means the capture of SO2 by limestone feed directly in FB without additional scrubbing system. As a cautionary note, by this desulfurization improvement, any unreacted lime

• •

or SNCR), as well as their combination - hybrid SNCR-SCR, have been applied to biomass-fired FBC boilers. SNCR has been used widely as a well-established technology for biomass combustion. In particular, Sweden and Finland stand out in applying these systems at the biomass combustion in FB. For the proper SNCR process, the location of ammonia/urea injection and the suitable radial distribution of injected reactant are of crucial importance, as mentioned earlier. If, at low loads, ammonia/urea injection is too close to the bottom of FB combustor that can lead to NOx increases which are attributed to the reduced residence time of injected reactant and to the high O2 concentration zone that lead to oxidation of NH3/CO (NH2)2. Application of SNCR is problematic in fuels with high S-content, because of unreacted lime from desulphurization process in bad, catalyze the oxidation of ammonia to NOx. As has already been said, this fact is important in the case of co-combustion of biomass and coal. In large FB utility power (PP) and CHP plants prevails use of SNCR systems (Table 5)

5.2. Grate firing system >Grate firing of biomass fuel is the combustion system, most frequently used in small power facilities. In size of over 5 MW it is particularly popular in Sweden, Denmark Netherlands, Poland etc. where some biomass and waste fuels are fired (e.g. wood pellets, bark, straw, plywood and chipboard waste, as well as municipal waste). Today, biomass grate-fired boilers and stokers are commonly used in combined heat and power (CHP) plants, in the range from 4 to 300MWe (many in the range of 20–50MWe). 11

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

Table 5 Overview of some references for FBC systems and applied deNOx technique. Plant

[MWfuel]

Boiler type

Main fuel

deNOx

Kaukaan Voima (FI) [76]

CFB/CHP

Biomass and peat

SNCR

Kymin Voima (FI) [76] Porin Prosessivoima (FI) [76] Rauman Voima (FI) [76] Stora Enso Veitsiluoto (FI) [76] Tornion Voima (FI) [76] Plant Sicet (Ospitale di Cadore) (IT) [77]

510 125 MWe/ 385 MWth 294 206 120 280 143 63

BFB/CHP CFB/CHP BFB/CHP BFB/CHP CFB/CHP BFB/CHP

Biomass, peat, and sludge Biomass, peat and REF Biomass, peat, REF, and sludge Biomass, peat, and sludge Peat, biomass, and CO gas Biomass: wood chips, bark, and forest residues

Połaniec Power Plant, (PL) [78] Konin (PL) [78]

447 154 MWth /55 MWe

CFB/PP CFB/CHP

Czechnica K2, (PL) [78] Söderenergi AB, Igelsta, (SE) [79]

76 MWth /25 MWe 240 MWth/ 73 MWe

BFB/CHP CFB/CHP

Hedesbyn, Skelleftea Kraft, (SE) [80]

25

BFB/CHP

100% biomass/ agricultural and woodlike biomass 100% biomass/ rapeseed residue, straw briquettes, energy willow, cherry stones, oat husk 100% Wet Biomass co-fire mixtures of biomass (mainly wood residues) with max 25%en of waste (REF pellets) peat 20%, wood pellets 80%

staged -air system SNCR SNCR – SNCR multiple air staging, SNCR SNCR+SCR SNCR

• Fuel NO

Applicable combustion control systems in grate-fired boilers:

• The most economical combustion modification to reduce NOx is air

• •



Except in Denmark, combustion facilities of these type have been constructed in Schkoelen, Germany (3.15 MWth) Duernkurt, Austria (2.18 MWth), Serbia (1.5 MWth) [4] and etc. Applicable control systems in cigar-burner boilers:

x is the major source of NOx from biomass combustion in grate-fired boilers.



staged -air system SNCR

• Nitrogen oxide emissions are reduced by combustion control sys-

staging. Applicable secondary measures of NOx reduction in grate-fired boilers: An attractive method for this system -selective catalytic reduction (SCR), could be less favorable for biomass with high potassium and other alkali content, because of the accelerated deactivation of the catalysts caused mainly by alkali salts present in the submicron ash. So the high-dust SCR is not a viable solution. The spraying of an aqueous solution of sulfate, (NH4)2SO4, into the hot flue gases upstream of the superheaters, originally proposed to reduce the deposition/corrosion rates for super heater tubes, can drastically reduce the NOx emissions from these boilers [15]. The SNCR deNOx system, similar to FBC technology, is the most preferred in large utility power (PP) and CHP plants (Table 6).



tems (staged air supply system) and/or use of SNCR, and/or use of SCR. For this type of combustion, Denmark has used a special technique based on injecting ozone into flue gas where O3 react with NOx forming N2O5 which is easily removed later by water scrubber [85].

5.4. Pulverized firing Pulverized combustion is rarely used for biomass as single fuel, for the following reasons: – due to its fibrous structure biomass is difficult to mill and it requires a high consumption of extra energy, – biomass is a highly reactive fuel and, as such, may cause problems in the drying system, – not infrequent changes in fuel quality require the use of supporting fossil fuels, – independent use of biomass for pulverized firing causes corrosion and fouling problems.

5.3. Cigar burner combustion system Cigar firing technology originally developed in Denmark [83] has been designed exclusively for the combustion of straw bales and is deemed suitable for the combustion of whole-crop bales without any reserves [84]. Here, there is no need for shredding the bales, as they go directly into the furnace burning like cigarette - "cigar burner" system. Table 6 Overview of some of the grate-fired references with deNOx. Plant

[MWfuel] Power/heat output

Boiler type Main fuel

deNOx

Tilbury Green Power (UK) [81] Cofely - BCN (FR) [80] Cofely - BESVSG (FR) [81] Novopan (DK) [81] DONG Energy, Avedøre 2 Power Station (DK) [82] Helius CoRDe Ltd (UK) [81] Zignago Power (IT) [81] FunderMax - Neudörfl (AT) [81]

125 55 50 27 105

40MWe 38MWt/12 MWe 25MWt/ 17MWe 18MWt/4 MWe 35 MWe

PP CHP CHP CHP CHP

Waste wood Clean wood, forestry wood Clean wood Waste wood, sander dust Straw

SNCR/urea SNCR/urea SNCR/urea SNCR/urea low-NOx burners & SCR catalyst

34 49 50

8MWt/8 MWe CHP 4.5MWdh/18 MWe CHP-DH 8MWto/10 MWe CHP

SNCR/urea SNCR/urea SNCR/urea

Boehringer Ingelheim (DE) [81] Kronoply (DE) [81] Sierra Power, California, USA [81]

70 65 9

– 10MWt/20MWe –

Clean wood & distillery by-product (draff) Clean wood and agro-waste Clean wood, demolition wood (A1A2) & dust Demolition wood (A1-A4) Demolition wood (A1-A4) and dust Wood and demolition wood

Type:

CHP CHP

PP = power plant CHP = combined heat and power / cogeneration CHP-DH = combined heat and power with district heating

12

Heat:

SNCR/urea SNCR/urea SCR located downstream the cyclone dh = district heating to = thermal oil t = thermal (= steam)

6. Summary and conclusion remarks

13

low-NOx burners & SCR catalyst existing SNCR was retrofitted with a high-dust SCR

high-dust SCR tail-end SCR PC/PP PC/PP

PC/PP

Power/heat output

2 × 350 MWe

2 × 250 MWe

400 MWth/125 MWe

– 350 MWth

800 MWth

DONG Energy, Studstrup Power Station, (DK) [81,87] Vattenfall Nordic, Amager Power Station Unit 1, (DK) [81,87] E.ON Langerlo Power Station, (BE) [81,87]

DONG Energy, Avedøre 2 Power Station, (DK) [81,82] Vattenfall Nordic, Uppsala, (SE) [81,88]

Table 7 Overview of some DeNOx SCR systems applied in PC units.

Small-scale combustion units require special concern, as they need simple and affordable solutions. In this case, preference is given to primary measures, first of all to air- and fuel-staging, if woody biomass and the grate (fixed bed) combustion is used. In the case of the use of biomass with a high content of bound nitrogen, these measures are used combined with the fuel blending. Flue gas recirculation (FGR), as a primary measure has a medium and large-scale combustion application and low NOx burners, has application exclusively in utility pulverized firing boilers for the coal and biomass co-combustion. Primary denitrification measures are economically sound, however, in the case of N-rich biomass combustion, especially for installations of more than 50 MW, preference is given to secondary measures with a tolerance of economic costs (investment and operational). In general, SCR and SNCR are the most promising deNOx option for biomass fired medium and large-scale combustion units. SCR-only application is cost effective for biomass fuels that emit less NOx since the required catalyst volume is smaller. For the SCR systems, usage of the NH3-SCR with the commercial catalysts, based on the metal oxides (V/Ti/W), plate-type structure, at a low-dust location, in fixed-

[MWfuel]

• the maximum expected emissions at the boiler outlet, • the maximum allowable NOx emissions for the specific site, • the maximum expected flue gas temperature, • space availability (especially for retrofits), • requirements for energy recovery, etc.

Plant

Boiler type

Main fuel

This paper gives a detailed overview of measures for reducing NOx emissions as well as applicability deNOx techniques to individual combustion technology. The effects of different parameters on NOx emissions cannot always be generalized because all boilers are unique, but it is certain that the physical and chemical properties of fueled biomass and its N content are the most influential. Analogously, there is no unique formula or a unique technology that can be suitable for all biomass fuelled installations. Besides NOx removal effectiveness and cost considerations, other parameters play important role in selection, such as:

PC/PP PC/CHP



x reduction systems are the same as with fluidized bed combustion. Use of staged air supply system and/or use of SNCR are recommended. An overview of some DeNOx SCR systems applied in pulverized combustion units is given in Table 7.

86% coal & 3% olive residue, 4% dried sewage sludge and 7% wood dust coal & pulverized wood pellets peat and up to 30% wood dust

• NO

coal & up to10%en co-firing of straw coal, pelletized straw, and wood pellets

deNOx

However, a few plants still grind wood pellets and use pulverized firing as a way to make low-carbon power. It is done at a small number of power stations in Europe including Hasselby in Sweden, Ironbridge Power Station in the UK, and Tampere in Finland. This combustion technology is particularly attractive for co-combustion of biomass and coal in large power systems. Biomass co-firing has been successfully demonstrated in over 230 installations worldwide (mostly in North Europe and the United States, with a capacity from 50 to 700 MWe) [86] with a wide range of biomass feedstock (wood and herbaceous biomass, crop residues, and energy crops) and boiler types (mainly pulverized firing, but also in fluidized bed and grate firing system). Usually, up to 20% of the biomass in a pulverized firing system is burned with the main fuel. In a power plant in Denmark the most common biomass in co-combustion are pulverized wood and to a lesser extent straw and wood from energy crops. Biomass briquettes, up to 10% (mass based), are co-fired in a lignite-fired power plant of 280 MWe in Germany, without significant problems. Up to 7% (mass based) pressed olive stones (wood from pressed olive stones) was co-combusted in a power plant in Greece. Applicable control systems in pulverized fuel boilers:

high-dust SCR

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

2008;22:1004–11. [9] Abelha P, Gulyurtlu I, Cabrita I. Release Of nitrogen precursors from coal and biomass residues in a bubbling fluidized bed. Energy Fuels 2008;22:363–71. [10] Tarelho LAC, Neves DSF, Matos MAA. Forest biomass waste combustion in a pilotscale bubbling fluidised bed combustor. Biomass Bioenergy 2011;35:1511–23. [11] Obernberger I. The present state and future development of industrial biomass combustion for heat and power generation. Conference on Thermal and Environmental Issues in Energy Systems Sorrento, Italy; 2010. [12] Werther J, Saenger M, Hartge EU, Ogada T, Siagi Z. Combustion of agricultural residues. Progr Energy Combust Sci 2000;26:1–27. [13] Sidney P. Huey KAD, Robert H. Hurt, Mary J. Wornat. Comparison of biomass and coal char reactivities. Biomass Fuels Symposium. Chicago, IL (USA); 1995. p. 694–98. [14] Timmer KJ. Carbon conversion during bubbling fluidized bed gasification of biomass. Iowa State University; 2008. [15] Yin C, Rosendahl LA, Kær SK. Grate-firing of biomass for heat and power production. Progr Energy Combust Sci 2008;34:725–54. [16] Di Blasi C, Buonanno F, Branca C. Reactivities of some biomass chars in air. Carbon 1999;37:1227–38. [17] US Environmental Protection Agency. Nitrogen Oxides (NOx) Why and How they are Controlled; 1999. 〈http://www.epa.gov/ttncatc1/dir1/fnoxdoc.pdf〉. [18] Huey SP, Davis KA, Hurt RH. Comparison of biomass and coal char reactivities. Conference: Biomass fuels symposium, Chicago, IL (United States), 20-26 Aug 1995; Other Information: PBD: [1995]; 1995. p. 5. [19] Khodaei H, Guzzomi F, Patiño D, Rashidian B, Yeoh GH. Air staging strategies in biomass combustion-gaseous and particulate emission reduction potentials. Fuel Process Technol 2017;157:29–41. [20] Eskilsson D, Rönnbäck M, Samuelsson J, Tullin C. Optimisation of efficiency and emissions in pellet burners. Biomass Bioenergy 2004;27:541–6. [21] Duan F, Zhang J-P, Chyang C-S, Wang Y-J, Tso J. Combustion of crushed and pelletized peanut shells in a pilot-scale fluidized-bed combustor with flue gas recirculation. Fuel Process Technol 2014;128:28–35. [22] Hustad J, Skreiberg Ø, Sønju O. Biomass combustion research and utilisation in IEA countries. Biomass Bioenergy 1995;9:235–55. [23] Bäfver LS, Rönnbäck M, Leckner B, Claesson F, Tullin C. Particle emission from combustion of oat grain and its potential reduction by addition of limestone or kaolin. Fuel Process Technol 2009;90:353–9. [24] Sommersacher P, Brunner T, Obernberger I, Kienzl N, Kanzian W. Application of novel and advanced fuel characterization tools for the combustion related characterization of different wood/kaolin and straw/kaolin mixtures. Energy Fuels 2013;27:5192–206. [25] Zeng T, Weller N, Pollex A, Lenz V. Blended biomass pellets as fuel for small scale combustion appliances: influence on gaseous and total particulate matter emissions and applicability of fuel indices. Fuel 2016;184:689–700. [26] Saikaew T, Supudommak P, Mekasut L, Piumsomboon P, Kuchonthara P. Emission of NOx and N2O from co-combustion of coal and biomasses in CFB combustor. Int J Greenh Gas Control 2012;10:26–32. [27] Hodžić N, Kazagić A, Smajević I. Influence of multiple air staging and reburning on NOx emissions during co-firing of low rank brown coal with woody biomass and natural gas. Appl Energy 2016;168:38–47. [28] Baxter L. Biomass-coal co-combustion: opportunity for affordable renewable energy. Fuel 2005;84:1295–302. [29] Nussbaumer T. Combustion and co-combustion of biomass: fundamentals, technologies, and primary measures for emission reduction. Energy Fuels 2003;17:1510–21. [30] Carroll JP, Finnan JM, Biedermann F, Brunner T, Obernberger I. Air staging to reduce emissions from energy crop combustion in small scale applications. Fuel 2015;155:37–43. [31] Lasek JA. Investigations of the reduction of NO to N2 by reaction with Fe under fuel-rich and oxidative atmosphere. Heat Mass Transf 2014;50:933–43. [32] Fennell PS, Dennis JS, Hayhurst AN. Latter stages of the reduction of NO to N2 on particles of Fe while simultaneously oxidizing Fe to its oxides. Energy Fuels 2011;25:1510–20. [33] Smajevic I, Kazagic A, Music M, Becic K, Hasanbegovic I, Sokolovic S, et al. Cofiring Bosnian coals with woody biomass: experimental studies on a laboratoryscale furnace and 110 MWe power unit. Therm Sci 2012;16:789–804. [34] Sung Y, Lee S, Kim C, Jun D, Moon C, Choi G, et al. Synergistic effect of co-firing woody biomass with coal on NOx reduction and burnout during air-staged combustion. Exp Therm Fluid Sci 2016;71:114–25. [35] Skalska K, Miller JS, Ledakowicz S. Trends in NOx abatement: a review. Sci Total Environ 2010;408:3976–89. [36] Mahmoudi S, Baeyens J, Seville JPK. NOx formation and selective non-catalytic reduction (SNCR) in a fluidized bed combustor of biomass. Biomass Bioenergy 2010;34:1393–409. [37] Cooper CD, Alley FC. Air pollution control: a design approach. 3rd ed Waveland Press Inc; 2002. [38] World Bank Group, United Nations Industrial Development Organization. Pollution prevention and abatement handbook, 1998: toward cleaner production; 1999. 〈http://documents.worldbank.org/curated/en/758631468314701365/Pollutionprevention-and-abatement-handbook-1998-toward-cleaner-production〉. [39] Lin H, Gao X, Luo Z, Cen K, Huang Z. Removal of NOx with radical injection caused by corona discharge. Fuel 2004;83:1349–55. [40] Hu X, Zhang J-J, Mukhnahallipatna S, Hamann J, Biggs MJ, Agarwal P. Transformations and destruction of nitrogen oxides—NO, NO2 and N2O—in a pulsed corona discharge reactor. Fuel 2003;82:1675–84. [41] Basfar AA, Fageeha OI, Kunnummal N, Al-Ghamdi S, Chmielewski AG, Licki J, et al.

bed arrangements is recommended. In the case that the tail-end application is selected option, a low-temperature catalyst from the Mn-series is preferred. The contaminated catalyst should be rinsed with large amounts of water. SNCR urea-based systems are recommended with the high-capacity units. In order to overcome the uneven temperature distribution, caused by the boiler load, more injection points are necessary at the different levels. Instead of that, usage of enhancers is possible. SNCR systems have lower cost in comparison to SCR. Generally, they reduce the NOx emission to the levels just below limit required by the current legislation. Values much below the permitted NOx limits lead to higher operating costs due to higher reductant consumption, which should be avoided. This technology is capable to meet the future stringent restrictions on emissions of nitrogen oxides but with the use of costlier and more sophisticated diagnostic temperature and finer control reactants injection systems. Currently, SCR installations have a slight advantage in achieving low NOx emission values in comparison to SNCR systems. In the view of the anticipated imposing the lower NOx limits, SCR systems will be able to meet them without any significant changes, but with a tolerance of certain economic (high O & M costs), energy (high consumption) and environmental disadvantages (issues with worn catalyst disposal). Forcing the use of biomass as a renewable energy source is the current global trends. However, it should be noted that the massive and uncontrolled use of this fuel, especially in urban areas, could cause negative effects, including those in terms of NOx emissions. Therefore, it is necessary to restrict the use of biomass for heating in individual households in urban areas and make efforts to use this fuel only in plants of medium and large capacity, because the application of the denitrification system in them is economically viable. The beneficial effect of district heating on the local air quality should be pointed out, too. A larger and professionally-maintained boiler facilities with flue gas cleaning and with high stacks replace a great number of individual heating installations with low stacks and often poorly controlled combustion and pollutant emission. Acknowledgement The authors wish to thank the Serbian Ministry of Education, Science and Technological Development for financing the projects: “Improvement of the industrial fluidized bed facility, in scope of technology for energy efficient and environmentally feasible combustion of various waste materials in fluidized bed”, (Project TR33042), and “Development and improvement of technologies for efficient use of energy of several forms of agricultural and forest biomass in an environmentally friendly manner, with the possibility of cogeneration”, (Project III42011). Thanks are also due to Dr. Dragoljub Dakić for inspiration and support during the research. References [1] Houshfar E, Skreiberg Ø, Todorović D, Skreiberg A, Løvås T, Jovović A, et al. NOx emission reduction by staged combustion in grate combustion of biomass fuels and fuel mixtures. Fuel 2012;98:29–40. [2] UNEP Division of Early Warning and Assessment. UNEP Year Book 2014 emerging issues update. 2014. 〈http://staging.unep.org/yearbook/2014/〉. [3] European Environment Agency. Air quality in Europe — 2014 report. 2014. 〈https://publications.europa.eu/en/publication-detail/-/publication/297909e61ab1-4800-b362-96107744c95a/language-en〉. [4] Mladenović M, Dakić D, Nemoda S, Paprika M, Komatina M, Repić B, et al. The combustion of biomass - the impact of its types and combustion technologies on the emission of nitrogen oxide. 70. Hemijska industrija; 2016. p. 11. [5] Obernberger I, Brunner T, Bärnthaler G. Chemical properties of solid biofuels—significance and impact. Biomass Bioenergy 2006;30:973–82. [6] Khan A, De Jong W, Jansens P, Spliethoff H. Biomass combustion in fluidized bed boilers: potential problems and remedies. Fuel Process Technol 2009;90:21–50. [7] Leckner B, Karlsson M. Gaseous emissions from circulating fluidized bed combustion of wood. Biomass Bioenergy 1993;4:379–89. [8] Chyang C-S, Qian F-P, Lin Y-C, Yang S-H. NO and N2O emission characteristics from a pilot scale vortexing fluidized bed combustor firing different fuels. Energy Fuels

14

Renewable and Sustainable Energy Reviews xxx (xxxx) xxx–xxx

M. Mladenović et al.

[42]

[43]

[44] [45]

[46]

[47] [48] [49]

[50]

[51] [52]

[53]

[54]

[55]

[56]

[57] [58]

[59]

[60]

[61] [62]

[63]

[64] [65]

Electron beam flue gas treatment (EBFGT) technology for simultaneous removal of SO2 and NOx from combustion of liquid fuels. Fuel 2008;87:1446–52. Licki J, Chmielewski AG, Iller E, Zimek Z, Mazurek J, Sobolewski L. Electron-beam flue-gas treatment for multicomponent air-pollution control. Appl Energy 2003;75:145–54. Cheng X. Modeling and simulation of a novel internal circulating fluidized bed reactor for selective catalytic reduction of nitrogen oxides [Thesis]. Vancouver, Canada: University of British Columbia; 2013. Liu Z, Li J, Woo SI. Recent advances in the selective catalytic reduction of NOx by hydrogen in the presence of oxygen. Energy Environ Sci 2012;5:8799–814. Mahmoudi S, Baeyens J, Seville JP. NOx formation and selective non-catalytic reduction (SNCR) in a fluidized bed combustor of biomass. Biomass Bioenergy 2010;34:1393–409. Tzimas E, Peteves S. NOx and dioxin emissions from waste incineration plants. European Commission Energy technology observatory, Institute for Energy 2002;EUR 20114 EN. Argyle M, Bartholomew C. Heterogeneous catalyst deactivation and regeneration: a review. Catalysts 2015;5:145. Heck RM. Catalytic abatement of nitrogen oxides–stationary applications. Catal Today 1999;53:519–23. Nam KB, Kwon DW, Hong SC. DRIFT study on promotion effects of tungstenmodified Mn/Ce/Ti catalysts for the SCR reaction at low-temperature. Appl Catal A: Gen 2017;542:55–62. Fu M, Li C, Lu P, Qu L, Zhang M, Zhou Y, et al. A review on selective catalytic reduction of NOx by supported catalysts at 100-300 [degree]C-catalysts, mechanism, kinetics. Catal Sci Technol 2014;4:14–25. Shan W, Song H. Catalysts for the selective catalytic reduction of NOx with NH3 at low temperature. Catal Sci Technol 2015;5:4280–8. Zhang Q, Qiu C, Xu H, Lin T, Lin Z, Gong M, et al. Low-temperature selective catalytic reduction of NO with NH3 over monolith catalyst of MnOx/ CeO2–ZrO2–Al2O3. Catal Today 2011;175:171–6. Seker E, Cavataio J, Gulari E, Lorpongpaiboon P, Osuwan S. Nitric oxide reduction by propene over silver/alumina and silver–gold/alumina catalysts: effect of preparation methods. Appl Catal A: Gen 1999;183:121–34. Ettireddy PR, Ettireddy N, Mamedov S, Boolchand P, Smirniotis PG. Surface characterization studies of TiO2 supported manganese oxide catalysts for low temperature SCR of NO with NH3. Appl Catal B: Environ 2007;76:123–34. Marbán G, Valdés-Solı́s T, Fuertes AB. Mechanism of low-temperature selective catalytic reduction of NO with NH3 over carbon-supported Mn3O4. J Catal 2004;226:138–55. Jin R, Liu Y, Wang Y, Cen W, Wu Z, Wang H, et al. The role of cerium in the improved SO2 tolerance for NO reduction with NH3 over Mn-Ce/TiO2 catalyst at low temperature. Appl Catal B: Environ 2014;148:582–8. MurCal. Catalysts & Modules. 〈http://www.murcal.com/johnson_matthey/johnson_ matthey_catalysts.html〉; 2017 [accessed 8 August 2017]. McGill AirClean LLC. New Brochure Details McGill AirClean’s DeNOx SCR Reactors. 〈http://www.mcgillairclean.com/textDocs/news/deNOx_news.htm〉; 2017 [accessed 8 August 2017]. Jodłowski PJ, Kryca J, Rogulska A, Gil B, Iwaniszyn M, Łojewska J, et al. Advantages of a wire gauze structured reactor with a zeolite (Cu-USY) catalyst for NH3-SCR of NOx. Chem Eng J 2013;214:319–26. Patcas FC, Garrido GI, Kraushaar-Czarnetzki B. CO oxidation over structured carriers: a comparison of ceramic foams, honeycombs and beads. Chem Eng Sci 2007;62:3984–90. Chou TS. FCC process with fines tolerant SCR reactor. United States patent US 5, 413,699; 1995. Mussatti DC, Srivastava R, Hemmer PM, Strait R. EPA air pollution control cost manual. Air quality strategies and standards division of the office of air quality planning and standards. Research Triangle Park, NC: US Environmental Protection Agency; 2002. p. 27711. European Commission. Integrated Pollution Prevention and Control - Reference Document on Best Available Techniques for Large Combustion Plants; 2006. 〈http://eippcb.jrc.ec.europa.eu/reference/BREF/lcp_bref_0706.pdf〉. Bajpai P. Green chemistry and sustainability in pulp and paper industry. Springer International Publishing; 2015. Sorrels JL, Randall DD, Fry CR, Schaffner KS. Selective noncatalytic reduction. US

Environmental Protection Agency; 2016. [66] US Environmental Protection Agency. Air pollution control technology fact sheet. 2003〈https://www3.epa.gov/ttncatc1/dir1/fsncr.pdf〉. [67] von der Heide B. Advanced SNCR technology for power plants. Las Vegas, USA: POWER-GEN International; 2011. [68] Miller BG. Coal energy systems. Academic Press; 2004. [69] Girard J, Snow R, Cavataio G, Lambert C. The Influence of Ammonia to NOX Ratio on SCR Performance (No. 2007-01-1581): SAE Technical Paper; 2007. [70] Radojevic M. Reduction of nitrogen oxides in flue gases. Environ Pollut 1998;102:685–9. [71] Van Caneghem J, De Greef J, Block C, Vandecasteele C. NOx reduction in waste incinerators by selective catalytic reduction (SCR) instead of selective non catalytic reduction (SNCR) compared from a life cycle perspective: a case study. J Clean Prod 2016;112:4452–60. [72] Nguyen TDB, Lim Y-I, Eom W-H, Kim S-J, Yoo K-S. Experiment and CFD simulation of hybrid SNCR–SCR using urea solution in a pilot-scale reactor. Comput Chem Eng 2010;34:1580–9. [73] Li J, Yang H, Wu Y, Lv J, Yue G. Effects of the updated national emission regulation in China on circulating fluidized bed boilers and the solutions to meet them. Environ Sci Technol 2013;47:6681–7. [74] Clarke S, Preto F. Biomass burn characteristics. Northern Ontario: Ministry of Agriculture, Food and Rural Affairs; 2011. [75] Shimizu T, Tachiyama Y, Fujita D, Kumazawa K, Wakayama O, Ishizu K, et al. Effect of sulfur dioxide removal by limestone on nitrogen oxide (NOx) and nitrous oxide emissions from a circulating fluidized bed combustor. Energy Fuels 1992;6:753–7. [76] Finnish Energy Industries Federation and Finnish Forest Industries Federation. NOx emission study – theory and experiences of selected fluidized bed boilers; 2011. [77] Modern Power Systems. Italy’s first biomass fired fluidised bed power plant. 〈http://www.modernpowersystems.com/features/featureitaly-s-first-biomassfired-fluidised-bed-power-plant/〉; 2001. [Accessed 8 August 2017]. [78] Natunen M, Jäntti T, Goral D, Nuortimo K, Oy FWE. First operating experiences of 55 MWe konin and 205 MWe połaniec CFB boilers firing 100% biomass. Europe: POWER-GEN; 2013. [79] Nevalainen T, Jäntti T, Nuortimo K. Advanced CFB technology for large scale biomass firing power plants. Bioenergy For 2012:29. [80] IEA Bioenergy Task 32. Database of Biomass Cofiring initiatives; 2009. 〈http:// www.ieabcc.nl/database/cofiring.html〉. [81] Aalborg Energie Technik a/s. Welcome to a Greener World. 〈http://www.aetbiomass.com/en-GB/Home/References/Biomasas-Fired-Plants.aspx〉; 2017. [Accessed 8 August 2017]. [82] DONGenergy The Avedøreværket CHP plant. 2017. 〈https://stateofgreen.com/ files/download/5194〉. [83] Marutzky R, Seeger K. Energie aus Holz und anderer Biomasse. Grundlagen, Technik, Entsorgung, Recht DRW-Verlag Weinbrenner, Leinfelden-Echterdingen; 1999. [84] Mladenović R, Dakić D, Erić A, Mladenović M, Paprika M, Repić B. The boiler concept for combustion of large soya straw bales. Energy 2009;34:715–23. [85] Jensen TK, Jørgensen L, Ørtenblad M, Stamate E, Simonsen P, Tobiassen L, et al. NOx reduction obtained by low-temperature plasma generated ozone. International Gas Union research conference2008. p. 2541-52. [86] Lempp P. Biomass Co-firing in Coal Power Plants. Energy Technology System Analysis Programme and International Renewable Energy Agency 2013;Technology-Policy Brief E21. [87] Jensen-Holm H, Castellino F, White TN. SCR DeNOx catalyst considerations when using biomass in power generation Kgs. Lyngby, DK: Haldor Topsøe A/S; 2012. [Technical report from the internet 〈http://www.topsoe.com/sites/default/files/ scr_denox_catalyst_considerations_when_using_biomass_in_power_generation_2012. ashx__0.pdf]〉. [88] Vattenfall Heat Uppsala, Safety, Health and Environment 2014. 2015. 〈https:// www.vattenfall.se/globalassets/fjarrvarme/miljoredovisningar/miljoredovisninguppsala-engelsk-text.pdf?_t_id=1B2M2Y8AsgTpgAmY7PhCfg%3d%3d & _t_ q=Vattenfall+Heat+Uppsala & _t_tags=language%3asv%2csiteid%3a885ae1f981f5-42fb-b768-e0d96523dc00 & _t_ip=147.91.1.44%3a59869 & _t_hit. id=Vattenfall_CustWeb_UI_Models_Media_GenericMedia/_efb375e9-efb0-440686cd-e21e5449ab8a & _t_hit.pos=2〉.

15