Alkali metals association in biomass and their impact on ash melting behaviour

Alkali metals association in biomass and their impact on ash melting behaviour

Fuel 261 (2020) 116421 Contents lists available at ScienceDirect Fuel journal homepage: www.elsevier.com/locate/fuel Full Length Article Alkali me...

5MB Sizes 0 Downloads 26 Views

Fuel 261 (2020) 116421

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Alkali metals association in biomass and their impact on ash melting behaviour

T

Agata Mlonka-Mędralaa, , Aneta Magdziarza, Marcin Gajeka, Katarzyna Nowińskab, Wojciech Nowaka ⁎

a b

AGH University of Science and Technology, Mickiewicza 30 Av., 30-059 Krakow, Poland Silesian University of Technology, Akademicka 2 St., 44-100 Gliwice, Poland

ARTICLE INFO

ABSTRACT

Keywords: Biomass Ash Thermal analysis Chemical leaching Ash fusion behaviour Alkali metals

Selected agricultural and energy crop biomass ashes represented by two mixed cereal straws, corn straw, Miscanthus × Gigantus and Salix Viminalis were chosen for ash behaviour investigation and prediction of operating problems during biomass combustion. The presence of aggressive species in the ash generate operational problems of heat exchanging surfaces in power boilers connected with slagging and fouling processes, limiting the use of biomass direct combustion for energy production. In this work, thermal behaviour characteristics, transformation properties of the inorganic components and ash fusion temperatures of biomass ashes were investigated using thermal analysis (STA), X-ray diffraction (XRD) and high-temperature microscope. The special attention was focused on the evaluation of potassium compounds presented in ashes. Potassium was detected as KCl, K2SO4, K2CO3 and K3PO4. It was noted, that presence and concentration of alkali metals, silicon and calcium compounds has the major impact on fusion temperatures of studied ashes. Leaching process of ash elements using water, ammonium acetate and hydrochloric acid solutions was performed to determine the association of alkali metals in the raw material. Based on the results, a mineral matter composition recalculation model was proposed to predict alkali compounds concentration in the fuel. The model might be also used to determine the risk of eutectics formation, which have the strongest influence on ash melting behaviour.

1. Introduction

combustion and co-combustion technologies [3,4], as well as gasification and pyrolysis installations attempt to use biomass [5,6]. Unfortunately, during the thermal processes of biomass conversion the ash related problems frequently came up. The biomass ash contains wide range of inorganic elements which form complex compounds in gaseous, liquid and solid phases during thermal conversion of the fuel. The mineral matter transformation is crucial phenomenon in biomass combustion. The presence of aggressive species in the ash generate operational problems of heat exchanging surfaces in power boilers connected with slagging and fouling processes [7]. Although the amount of ash generated during biomass combustion in comparison to coal is significantly lower, the biomass ash generates more damages and needs more attention these days then coal ash. The nature of biomass is characterised by huge heterogeneity and depends on biomass origin and processing, it is also difficult to systematise and predict to prevent the unwanted effects in combustion system. The generation of deposit layers on heat exchanger surfaces depends on many factors mainly, chemical composition of the fuel, combustion temperature and residence time of the fuel samples.

The necessity of environmental protection, decreasing of fossil fuel resources and EU regulations promote renewable fuels use in energy and heat production. The European Union has an ambitious goal for renewable energy production to achieve at least 27% of final energy consumption leading to 40% reduction of greenhouse gas emissions [1]. Biomass is one of the major renewable energy resources and energy production from biomass is expected to be further increased. Based on EUCO27 scenario, it is expected that by 2030, the share of biomass will be around 50% of overall renewable energy production. Biomass is mainly supplied from forest and agriculture industries and urban waste resources. The origin of biomass is directly associated with its advantages like, renewable form enables to reduce amount of wastes with neutral CO2 emission. The application of biomass as environmental friendly fuel for energy production still faces new challenges which require sustainable solutions and has to be managed [2]. Thermal utilization technologies of different kind of biomass go into a good direction. Biomass is successfully combusted via direct ⁎

Corresponding author. E-mail address: [email protected] (A. Mlonka-Mędrala).

https://doi.org/10.1016/j.fuel.2019.116421 Received 2 September 2019; Received in revised form 4 October 2019; Accepted 11 October 2019 Available online 26 October 2019 0016-2361/ © 2019 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/).

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

The biomass ash mainly consists of calcium, silicon, potassium and phosphorus as well as noticeable amounts of chlorine. The alkali metals like K and Na together with chlorine are the most important elements in biomass ash affecting unwanted processes. During fuel combustion reactive compounds evaporate and alkali metals might form aerosols (KOH, KCl, K2SO4, NaCl, Na2SO4) in the gas phase. When the exhaust gases temperature decreases, aerosols can condensate on fly ash particles and form deposits - small agglomerates of ash particles leading to preliminary slag layer. Additionally, aerosols of alkali metals can react with each other or other ash components like SiO2 and Fe2O3 to form eutectic mixtures [8]. Si, K and Ca are the main elements responsible for agglomeration process, based on the knowledge of SiO2–CaO–K2O system the fusibility tendencies can be predicted [9]. Alkali metals can exist also in solid phase in the form of silicates (e.g. K2Si2O5, Na2SiO3) and aluminosilicates (e.g. KAlSi3O8, KAlSiO4, NaSi3AlO8) [10]. Beside of some disadvantages associated with ash related issues in power boilers, biomass ash is characterised by appropriate chemical composition for its further application. The amount of ash generated during combustion depends on the origin of the fuel and it is in the range of c.a. 1% (woody biomass) up to more than 10% (agriculture biomass) [11]. Silica-rich ash can be used as a substrate for both calcium silicate hydrate-based and sodium aluminosilicate-based binders and in consequence can be successfully applied in ceramic and cement sectors [12]. There are many studies concerning the formation and transformation of the mineral phases during thermal processes depending on the process parameters [11,13,14]. Many methods used in previous studies including: simultaneous thermal analysis [10,15,16], XRD [16,17] and chemical fractionation [18] were incorporated in this study to find most appropriate combination of laboratory scale methods used to predict potential ash related operating problems associated with direct biomass combustion. Such combination of procedures is essential for prediction of the potential problems before introduction of the new, unknown fuel in the existing unit. This paper has to offer a multifaceted analysis of five biomass ashes using combination of chemical and instrumental methods. The leaching process was investigated to predict the behaviour of ash elements during combustion process. Based on the results, a modified recalculation model was proposed to predict alkali compounds concentration in the fuel. The model can be used to predict the risk of low-melting eutectic mixtures formation in the ash phase. For deeper studies the ash fusion temperatures and its thermal behaviour were in detail investigated using thermal analysis and high-temperature microscope. Additionally, achieved results can help in developing the biomass ash utilization methods and its appropriate management.

SELECTIVE CONSECUTIVE LEACHING CHEMICAL FRACTIONATION PROCEDURE

2. Material and methods 2.1. Materials In the study more than 15 kinds of biomass (woody, energy crops and samples of agricultural origin) were collected from Polish market. The mineral matter composition was determined for all studied materials. The main aim of this study was to determine the influence of alkali metals on ash melting behaviour. Therefore, only five kinds of biomass, characterised by the highest concentration of potassium and varying concentration of chlorine and sulphur were chosen for further analysis. Three samples of herbaceous biomass of agricultural origin (two mixed cereal straw and one corn straw samples) and two energy crops were selected and denoted as MCS1 – mixed cereal straw 1, MCS2 – mixed cereal straw 2 (fuel sample collected from Polaniec Power Plant), CS – corn straw, MxG – Miscanthus × Giganteus and SV – Salix Viminalis. 2.2. Methods 2.2.1. Mineral matter composition Inductively Coupled Plasma Atomic Emission Spectroscopy (ICPAES) was used for detection of inorganic elements in raw biomass and ash samples using ICP-AES-JY 2000 apparatus. The concentration of the following elements was determined: Na, Mg, Al, Si, P, K, Ca, Ti and Fe. The method is very useful for low content elements concentration measurements in the liquid samples, but it can be also used for elements present in higher concentrations in the sample. Solid samples prior analysis are mineralised in acid mixtures. In the study, raw biomass and ash samples were dissolved using the acid digestion method with a mixture of nitric (HNO3) and hydrofluoric (HF) acids. The ash analysis including carbon, hydrogen, nitrogen and sulphur concentrations was carried out using an Elemental Analyser Truespec CHN and S Leco (CHNS628). The method is dedicated to raw biomass samples analysis, but with some restrictions it might be used in case of ashes as well. The analysed sample is heated up in pure oxygen at 950 °C, for C, H, N and 1350 °C for S determination. The chlorine content was measured using chemical analysis (titration method) based on PN-EN 196-2:2013-11 standard. 2.2.2. Chemical fractionation The fundamental assumption of the chemical fractionation method is that a given element will be dissolved in water or acidic solvent according to its association in the fuel matrix. First, the procedure was developed and applied for determination of coal inorganic behaviour [19,20], later on it was adapted for biomass studies as well [21–23]. The overall concept of the method was shown in Fig. 1 and the procedure applied in this study was determined in Fig. 2. A consecutive

Water leachable compounds

Reactive

1 M NH4Ac leachable compounds

Mostly reactive

1 M HCl leachable compounds

Mostly nonreactive

Insoluble residue

Nonreactive

Fig. 1. Chemical fractionation general concept [24]. 2

Released to the gas phase, actively reacts with other combustion products

Remain stable in the solid ash phase

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

Fig. 2. Chemical fractionation general procedure.

leaching by water (H2O), 1 M ammonium acetate (NH4Ac), and 1 M hydrochloric acid (HCl) was introduced. Concentrations of selected elements in the initial raw samples and leachates after the leaching procedure were determined using ICP-AES method. Water leachable compounds are expected to be present in the fuel as alkali metal chlorides, sulphates and carbonates, on the other hand, ionic compounds leachable in ammonium acetate solution will be associated with organic fuel matrix. Both, water and ammonium acetate leachable compounds are assumed to be reactive compounds in the fuel. Most probably reactive compounds will be evaporated during combustion process and released to the gas phase in the form of aerosols, participating in the second order reactions with other ash and flue gas components [24]. Compounds dissolved in acid solution and insoluble in all previous solvents will remain stable in solid phase during combustion [25,26]. In general, insoluble residue is composed mostly of oxides, silicates and carbonates etc. [27]. In the chemical fractionation method, secondary reactions affecting reactiveness of mineral matter components are not included. Most important one is high-temperature alkali metals adsorption on silica. According to selective leaching results, silicon is a non-reactive ash component, but as its concentration in the fuel is high it will affect the reactive components like potassium according to reaction no. (1), limiting the release and condensation of the metal during combustion process. It was proved by thermodynamic equilibrium calculations, that an increase of ratio between non-reactive and reactive mineral matter components will decrease the alkali aerosols formation by 50 % [24]. However, when high concentrations of KCl are present in the fuel, formation of low melting eutectics with silicon oxide is a risk and bad agglomeration might be observed in fluidized-bed boilers combusting highly contaminated biomass. 2KCl (g ) + Al2 O3 (s ) + 6SiO2 (s ) + H2 O (g )

straw is hard, but not brittle material and proper sample preparation generates many problems. Additionally, biomass samples differs much between each other in terms of bulk density and water adsorptivity. Therefore, final leaching procedure (sample pre-treatment, additional filtration step, proper separation and leachates amounts) should be adapted according to a given fuel properties. A detailed procedure of the biomass particle size reduction introduced for fuels examined in this study was presented in [30]. The chemical fractionation method is not a standard procedure and in the literature different methodologies might be found, most frequently found differences concern leaching times, number of repeats, solvent types, amounts of the leaching agents and process temperatures. In this study sequential leaching method was used and the detailed procedure was shown in Fig. 3. As the evaporation of solvent during the last step of the procedure was observed, the amount of leaching agent was increased up to 7 ml/1 g of fuel. The initial concentrations of the analysed elements were measured in the raw fuel sample and in 3 collected leachates using inductively coupled plasma atomic emission spectroscopy. In biomass studies, the experimental results show that for both woody and herbaceous biomass: sodium, potassium, chlorine, phosphorus and in some cases sulphur major fraction is water soluble [21,31–33]. 2.2.3. Free alkali index To determine the risk of highly corrosive alkali metals hydroxides formation during the combustion process an index called “free alkali index” was introduced in [34,35]. The indices is determined based on the assumption that alkali metals soluble in acetic acid solution will be released to the gas phase during combustion and will form hydroxides, chlorides and sulphates. In the calculations, formation of carbonates is assumed to be a secondary reaction and it is not included in the model. When thermodynamic equilibrium conditions are achieved, at high CO2 concentration in the flue gas, hydroxides are stable above 700 °C [35]. In the wall boundary layer, depending upon the temperature, alkali metals will be present mostly in the form of chlorides, sulphates, hydroxides and carbonates. The free alkali index is very useful in case of fuels rich in alkali metals, but with smaller content of chlorine and sulphur:

K2 O ·Al2 O3·6SiO2 (s ) + 2HCl (g )

(1) Other process leading to ash deposits melting behaviour modification is alkali chlorides reaction with sulphur (2). It might be achieved by injection of sulphur-containing additives (elementary sulphur, ammonium sulphate) to the combustion chamber or co-combustion with high-sulphur fuels.

2KCl (g ) + SO2 (g ) + H2 O (g ) +

1 O2 (g ) 2

K2 SO4 (l, s ) + 2HCl (g )

(2)

AI =

The kinetics of the reaction (2) is rather slow and in the reducing conditions, temporarly present in the combustion chamber, alkali chlorides are the stable form [28]. The reliability of the chemical fractionation results mostly depends on the particle size distribution in the sample [29]. Biomass, especially

(Nasol + Ksol (2S + Cl)) LHV

where: AI – free alkali index, mol/MJ, Nasol – acetic acid soluble sodium fraction, mol/g, 3

(3)

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

Fig. 3. Chemical fractionation detailed procedure used in this study, according to [21].

LHV – lower heating value, MJ/g.

Table 1 Biomass mineral matter composition (wt%). Sample

MCS1

MCS2

CS

MxG

SV

Aa Na2O MgO Al2O3 SiO2 P2O5 K2O CaO TiO2 Fe2O3 CLT CS HLT HS S Cl LHV, MJ/kg AI, mol/MJ

10.0 0.92 2.75 4.00 53.70 2.57 17.47 10.37 0.18 1.79 1.64 1.68 0.28 0.34 0.87 1.75 17.28 0.0003

10.6 0.24 2.09 0.98 65.04 3.21 14.09 16.23 0.08 0.56 1.46 2.1 0.26 0.31 0.74 1.33 15.65 0.0018

13.3 0.18 3.62 2.97 51.13 4.01 23.85 8.38 0.28 1.54 1.57 2.87 0.29 0.31 0.73 1.51 14.59 0.0078

5.0 0.22 1.69 1.04 64.18 7.19 15.90 13.14 0.08 0.80 1.49 1.85 0.27 0.35 0.47 0.85 14.20 0.0098

1.4 0.19 4.64 0.34 2.91 13.50 22.04 39.60 0.02 0.51 2.83 6.41 0.68 0.27 1.01 0.33 15.82 0.0136

In case of biomass, the amount of reactive alkali metals is almost equal to the total concentration of alkali metals in the sample [34]. In the calculations, the total amount of sodium and potassium soluble in water and 1 M ammonium acetate was assumed as Nasol and Ksol, respectively. 2.2.4. Simultaneous thermal analysis (STA) Simultaneous thermal analysis (STA) can be used to determine changes of mass (Thermogravimetry - TG) and caloric reactions (Differential Scanning Calorimetry DSC) of a sample in the wide temperature range and at the same time. The application of this method allows to carry out the tests under perfectly identical conditions like atmosphere, gas flow rate, vapour pressure, heating rate, thermal contact to the sample crucible and sensor, and radiation effect [36]. Taking into account the problems associated with a precise biomass ash melting temperature determination, the STA method is the most adequate one and was proposed by many researchers [15,37]. During heating process of biomass ash sample, two types of phase transition processes: evaporation and melting processes take place. The melting process of pure substance takes place when mass of sample does not change (any changes on TG are observed) and the peak on DSC curve appears. The pick temperature determines the melting temperature of a

Ksol – acetic acid soluble potassium fraction, mol/g, S – sulphur content, mol/g, Cl – chlorine content, mol/g,

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% INSOLUBLE RESIDUE 1M HCl 1M HCl 1M CH3COONH4 1M CH3COONH4 H2O H2O

Mg

Al

Si

P

11%

98% 100% 98%

K

Ca

Ti

Fe

Na

9%

87%

99%

97%

82%

6%

1%

0%

1%

3%

4%

0%

0%

7%

31%

0%

0%

1%

11%

3%

0%

1%

4%

52%

1%

0%

1%

76%

5%

1%

2%

7%

Fig. 4. Chemical fractionation results for MCS1 sample, in wt%. 4

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% INSOLUBLE RESIDUE 1M 1MHCl HCl 1M 1MCH3COONH4 CH3COONH4 H2O H2O

Mg

Al

Si

P

K

Ca

Ti

Fe

Na

51%

95% 100% 100% 38%

52%

99%

98%

91%

4%

2%

0%

0%

15%

0%

1%

2%

10%

1%

0%

0%

7%

12%

0%

0%

1%

34%

2%

0%

0%

55%

22%

1%

1%

5%

Ti

Fe

1%

Fig. 5. Chemical fractionation results for MCS2 sample, in wt%.

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

INSOLUBLE RESIDUE 1M1M HCl HCl 1M1M CH3COONH4 CH3COONH4 H2O H2O

Mg

Al

Si

P

K

Ca

Na

9%

99% 100% 99%

15%

97% 100% 98%

97%

16%

0%

0%

0%

2%

1%

0%

1%

1%

23%

0%

0%

0%

8%

1%

0%

0%

1%

52%

0%

0%

0%

75%

1%

0%

1%

1%

Fig. 6. Chemical fractionation results for CS sample, in wt%

given compound and the area under DSC pick gives the melting process enthalpy value [38]. The method might be also used to determine melting behaviour of eutectic mixtures present in the ash phase [39]. For biomass ash (the mixture of compounds characterized by different melting points) many peaks can appear on DSC curve [15]. During evaporation process, a mass reduction is observed as well. Generally, STA method is a very useful technique for describing the thermal processes of a wide range of materials, the results provide a rapid information about the studied sample e.g. starting and end temperatures of a given process, maximum reactivity temperature etc. As the biomass ash is characterised by a very complex chemical composition and some process can go simultaneously, the determination of ash melting temperature by means of STA method can be used only in comparative analysis. The secondary reactions in gas and solid phases taking place during combustion of biomass (high-chlorine content fuel) or co-combustion with coal (high-sulphur content fuel) strongly influence ash melting behaviour and cannot be included in STA [40]. Beside the fact, the method makes some problems in interpretation of biomass ash samples, it can be indirectly helpful in the prediction of slagging and fouling of biomass ash on heat transfer surfaces in boilers.

The special methodology of biomass ash preparation is required for STA studies. Low-temperature ash samples were prepared, by gradual heating of the fuel 2 °C/min up to 550 °C and isothermally heated through next 24 h [36]. The STA conditions were as follows: the biomass ash samples were heated in alumina crucibles (Al2O3) from an ambient temperature up to 1200 °C (for all cases, except SV sample for which the analysis was performed up to 1500 °C) at a constant rate of 10 °C/min and at a 50 ml/min flow of nitrogen (inert atmosphere). The reference substances (pure salts) analysis was conducted under the same conditions as examined ashes. The masses of samples were 5 mg and 20 mg for biomass ash and reference samples, respectively. As an output, the TG and DSC curves were obtained. TG curve presents the mass changes of studied samples in contrast to the initial mass under increasing temperature, whereas DSC curves reflect the thermal effects (endothermic and exothermic). Additionally, the DTG curve was created. DTG is the mathematical conversion, a first derivative of the TG. DTG curve was calculates as dm/dt = f(t), (where: m – mass of a sample, t – time). DTG curve has allowed, with a better precision to analyse the thermal behaviour of studied ashes.

5

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

INSOLUBLE RESIDUE 1M1M HClHCl 1M1M CH3COONH4 CH3COONH4 H2O H2O

Mg

Al

Si

P

K

Ca

Ti

Fe

Na

46%

93% 100% 99%

26%

95%

98%

99%

98%

6%

3%

0%

0%

3%

2%

1%

0%

2%

16%

1%

0%

0%

8%

1%

0%

0%

1%

32%

3%

0%

0%

63%

2%

1%

0%

0%

Ti

Fe

Na

Fig. 7. Chemical fractionation results for MxG sample, in wt. %.

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% INSOLUBLE RESIDUE 1M 1MHCl HCl 1M 1MCH3COONH4 CH3COONH4 H2O H2O

Mg

Al

Si

P

14% 100% 100% 99%

K

Ca

22%

38% 100% 100% 96%

13%

0%

0%

0%

2%

26%

0%

0%

2%

17%

0%

0%

0%

10%

13%

0%

0%

1%

56%

0%

0%

0%

65%

23%

0%

0%

1%

Fig. 8. Chemical fractionation results for SV sample, in wt%.

Fig. 9. Modified recalculation model for determination of alkali metal compounds concentration in the fuel.

2.2.5. Ash melting behaviour To complement STA results additional tests using high-temperature microscope were introduced. Ash fusion tendencies were determined and visualized using High-Temperature Microscope Misura HSM 3 M. The use of camera allowed to record the sequence of the characteristic

shape changes and the temperature throughout the experiment. The samples were prepared in the shape of cylinders (ø = 2 mm, h = 3 mm) and heated up to 1400 °C with 10 °C/min heating rate in ambient atmosphere. The heating process of the ash cylinder was observed and monitored via video camera to determine behaviour of ash samples in a 6

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

The phase analysis of the studied ashes was examined by X-ray Diffraction (XRD) using X’Pert PRO PANalytica with Cu Kα radiation in Bragg-Brentano geometry. The scans for all of the samples were collected from 10 to 80° (2θ) with a scanning speed of 0.008°/min and 4 h scan time. For phase identification, a data base PDF-4 product of ICDD was used.

1.4% 1.2% 1.0% 0.8% 0.6%

3. Results and discussion

0.4%

3.1. Mineral matter composition

0.2% 0.0%

K

Na

Cl

S

MCS1

1.160%

0.036%

0.420%

0.136%

MCS2

1.285%

0.047%

0.324%

0.134%

CS

1.180%

0.096%

0.308%

0.080%

MxG

1.178% 1.278%

0.089% 0.050%

0.091% 0.019%

0.040%

SV

The mineral matter composition results show great differences between analysed samples. All fuels were characterized by moderate concentration of sodium, below 1 % of Na2O and diverse potassium content, varying from 14 up to 24 % of K2O. Such numbers have influenced content of other mineral matter components, e.g. main ash components like: Ca and Si. Aluminium oxide (Al2O3) concentration was low, and does not exceed 4.0 %, in most cases it was even lower than 1 %. It might be concluded that analysed ashes will be characterised by low content of aluminosilicates and low ash melting temperatures should be expected. Phosphorus content in the analysed samples were between 2.5 up to 13.5 %, the highest values were noted for energy crops. Sulphur content was similar and it was almost in all cases lower than 1 %, the highest concentration of S in SV ash might be associated with low ash content in the fuel. Chlorine content in the ash did not exceed 2 %, but the values were the highest for straw samples and few times smaller in case of energy crops. In the study additional comparison of C and H was performed for low-temperature ash, denoted as LT and standard ash, denoted as S (higher initial sample mass, achieved according to a Polish standard: PN-EN ISO 18122:2016-01) to determine the influence of low-temperature incineration on the organic matter concentration in the ash. Straw samples MCS1, MCS2 and CS were characterised by the highest ash content (above 10 %). In case of MCS2 and CS samples lowtemperature incineration decreased the total carbon content by 30 % (mostly organic carbon). MxG sample, with a similar to straw samples ash composition, has almost two times smaller ash content and in both ash samples (low-temperature and according to a standard) carbon content was similar. In case of samples characterised by high ash content (straw samples), it was very hard to achieve representative ash samples for further analyses when standard incineration procedure was

0.084%

Fig. 10. Measured potassium, sodium, chlorine and sulphur concentrations in the dry raw fuel samples, in wt%.

wide temperature range. 2.2.6. Phase analysis X-Ray Diffraction (XRD) is a non-destructive, advanced instrumental method for phase analysis based on monochromatic X-ray application to obtain information about the structure of crystalline materials. The technique allows to identify and characterize the crystal compounds based on their diffraction pattern. The intensities of the diffracted waves depend on the kind and arrangement of atoms in the crystal structure. In general, the method is dedicated for investigation of crystalline materials (e.g. minerals, alloys, building materials, ceramics, catalysts, zeolites), but it is also frequently used for ash analysis as well. Unfortunately, the biomass ash, especially low-temperature ash mostly consists of non-crystalline compounds (amorphous phase) which are difficult for detection by means of XRD method [41]. It is huge disadvantage of this method as the amorphous compounds presented in biomass ash have the strongest impact on ash deposits formation [42,43].

Fig. 11. Calculated potassium salts, hydroxide and oxide concentration in a dry fuel sample, (wt%). 7

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

Fig. 12. Calculated sodium salts, hydroxide and oxide concentration in a dry fuel sample, (wt%).

Fig. 13. TG and DSC evaluation profiles for KCl under nitrogen atmosphere.

used. During fast heat up of large amount of sample with a high ash content, fuel material will be partially subjected to a pyrolysis process and char will be formed. Small, black inclusions formed in the ash will not decompose at temperature as low as 550 °C and chars will be the main source of unburned organic carbon in the ash. However, in case of MCS1 ash samples the organic carbon concentration in both samples was similar and the ash content was one of the highest noted, most probably due to the initial raw biomass sample size in case of standard incineration procedure. The sample SV was characterised by the smallest ash content and in both cases obtained ash samples were bright white, the difference in carbon concentration, in this case will be associated with thermal decomposition of low-temperature carbonates, not with organic carbon incineration. Therefore, it is recommended to use low-temperature ash procedure in case of fuels characterised by high ash content, only. Otherwise, due

to long heat up, low-temperature carbonates will be decomposed and sample properties might change. However, even low amounts of organic carbon will affect STA ash analysis, both in case of TG and DSC signals. Organic carbon combustion in ash will show mass reduction on TG and as thermal decomposition and melting are both endothermic processes and combustion exothermic one, the DSC signal will be also disturbed. Additional analysis of hydrogen content in the ash showed high hygroscopicity of the material and most probably also small amounts of hydroxides in the ash samples. Free alkali index derived to predict the risk of aggressive hydroxides formation is especially recommended to be used in biomass fuels predictive studies. When, AI index is positive, hydroxides or carbonates (depending upon the temperature and CO2 concentration) will definitely appear in the flue gas. The formation of alkali metal hydroxides is associated with high concentration of alkali metals and low 8

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

Fig. 14. TG and DSC evaluation profiles for CaCO3 under nitrogen atmosphere.

Fig. 15. TG and DSC evaluation profiles for K2SO4 under nitrogen atmosphere.

concentration of chlorine and sulphur in the fuel sample. Based on chemical composition of biomass ashes and calculated AI index it was noted, that the lowest risk of KOH and NaOH formation was determined for agricultural samples and the highest for energy crops, especially SV sample. It might be concluded, that combustion of agricultural biomass will be associated with ash slagging and fouling processes, but energy crops combustion will be subjected to different problems connected with carbonates and hydroxides formation, very sensitive to combustion parameters. All results concerning biomass mineral matter composition, together with AI index were collected in Table 1.

and ammonium acetate solution and will be released to the gas phase and subjected to further reactions with other flue gas and ash components. Magnesium and calcium will be also reactive components of the ash, mostly released as carbonates (MgCO3 and CaCO3) decomposing during fuel combustion. Based on the obtained results, sodium will behave in a different manner than potassium and most probably it will remain stable in the ash phase. Other mineral matter components like: Si, Ti, Fe, P and Al will be also non-reactive components during biomass fuel combustion. Obtained results are similar to those published in [29], it might be concluded that potassium, magnesium and calcium in agricultural biomass and energy crops will be released to the flue gas during combustion and its compounds will have the major impact on the formation of ash deposits on heat exchanging surfaces in power boilers. Chemical leaching is especially useful method when some new, unknown fuels are planned to be introduced and essential knowledge about its properties is necessary to predict risk of operating problems

3.2. Chemical fractionation Chemical fractionation procedure results in Figs. 4–8, showed huge similarities between all fuel samples. In case of both fuel groups: straw and energy crops, major fraction of potassium was leached out by water 9

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

Fig. 16. TG and DSC evaluation profiles for K2CO3 under nitrogen atmosphere.

Fig. 17. TG, DTG and DSC evaluation profiles for low-temperature cereal mixed straw 1 (MCS1) under nitrogen atmosphere.

like: slagging, fouling or corrosion. Additionally, most valuable results are achieved after water and ammonium acetate leaching, therefore in case of biomass, acid leaching might be excluded from the procedure.

Potassium can occur in gas phase as KOH, KCl, K2SO4, K2CO3, and can form eutectics like KCl-K2CO3, KCl-K2SO4 [45–47]. Moreover, in the studies [48,49] K2Ca(SO4)2 and K3Na(SO4)2 were found as a significant phases affecting ash slagging process. In this study, chemical fractionation results were used to determine the concentration of alkali metal compounds: chlorides, sulphates, hydroxides, carbonates and oxides in the fuel sample. The main assumption is that water and ammonium acetate soluble alkali metals will be released as salts during combustion process. Additionally, due to low stability of carbonates in the gas phase, the free alkali metals, assumed to be reactive compounds, but not in form of sulphates or chlorides will be present in the form of carbonates or hydroxides, depending upon the combustion process conditions. In Fig. 9: Cld are Sd denote chlorine and sulphur concentration in dry fuel samples, in wt%, respectively; Ksol and Nasol are potassium and sodium concentration soluble in water and 1 M NH4Ac, in wt%, respectively and Kash and Naash are potassium and sodium concentration

3.3. Alkali metal salts in fuel – prediction model A so-called quasi-chemical equilibrium approach derived by researchers from Åbo Akademi University [44] was further developed in this paper and the main concept was presented in Fig. 9. In the original model, the oxide-based bulk ash chemical composition was used to determine the concentrations of potassium and sodium chlorides, sulphates and carbonates. The new model proposed in this study link this idea with the general concept of AI index. It is assumed that only reactive part of potassium and sodium in the fuel will take part in alkali metal salts and hydroxides formation. The remaining part (soluble in HCl and in the solid residue) will be present in the form of non-reactive oxides. 10

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

Fig. 18. TG, DTG and DSC evaluation profiles for low-temperature cereal mixed straw 2 (MCS2) under nitrogen atmosphere.

Fig. 19. TG, DTG and DSC evaluation profiles for low-temperature corn straw (CS) under nitrogen atmosphere.

soluble in 1 M HCl and retained in solid residue, in wt%, respectively. In the calculations, total Cl and S concentrations in dry fuel samples were taken into consideration, chlorine might be partially released in the form of HCl or Cl2 and S as SO2, but in secondary reactions in combustion zone additional sulphates and chlorides will be formed. Therefore, the highest possible concentration of sulphates and chlorides was assumed in the calculations. The initial concentrations of alkali metals, chlorine and sulphur in the raw fuel samples were shown in Fig. 10. All samples were characterised by high and similar concentrations of potassium, around 1.2% and very low sodium contents, below 0.1%. Biomass of agricultural origin (straw) has the highest content of chlorine, above 0.3% and moderate sulphur content, around 0.1%. The content of chlorine and sulphur in energy crop samples was low, in both cases Cl and S concentration did not exceed 0.1%. The results of calculations based on the proposed model were shown for potassium and sodium compounds separately, in Figs. 11 and 12,

respectively. Major fraction of potassium in all cases was water and ammonium acetate soluble. Therefore, it might be assumed that potassium will form mostly reactive compounds in the flue gas and deposits, mainly in the form of: KOH or K2CO3, and next KCl and K2SO4 in case of fuels with low Cl and S concentrations and inversly for fuels rich in Cl and S. Second alkali metal, sodium was mainly insoluble and it will remain non-reactive in the oxide form (Na2O) in the ash phase. Additionally, by means of the proposed recalculation model, it is possible to determine tendencies of mineral matter to form eutectics. The composition in the eutectic point of the most preferable eutectic in biomass ash: KCl - K2SO4 is K2SO4 = 45.9 wt% and KCl = 54.1 wt%. In case of agricultural biomass samples the ratios KCl/KCl + K2SO4 were: 54.6 wt%, 48.5 wt% and 61.1 wt% for MCS1, MCS2 and CS samples respectively. It means that, in case of samples MCS1 and CS the presence of eutectics in the ash phase might be expected. However, the great impact of other ash components will also affect the overall ash 11

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

Fig. 20. TG, DTG and DSC evaluation profiles for low-temperature Miscanthus × Giganteus (MxG) under nitrogen atmosphere.

Fig. 21. TG, DTG and DSC evaluation profiles for low-temperature Salix Viminalis (SV) under nitrogen atmosphere.

melting behavior.

processes of major biomass ash components. Based on DSC curves (endothermic peak) the melting temperatures of pure salts were determined. Experimentally determined melting points frequently reflects the literature data. The evaporation and decomposition of the salts proceed in wide temperature range and the analysis of pure salts is essential in further analysis of biomass ashes. According to literature data, KCl melts at 772/776 °C and it was also proved in experimental results (Fig. 13). Then, KCl evaporates and the final evaporation was observed c.a. 1100 °C and the following reaction proceed:

3.4. Simultaneous thermal analysis (STA) The simultaneous thermal analysis was used to investigate the thermal behaviour of the studied ashes. Additionally, the thermal analysis was carried out for pure potassium salts as a reference. The potassium salts were chosen as potassium was the major alkali metal in the analysed biomass ashes and it will have the highest impact on slagging, fouling and corrosion processes in power boilers. In the real process, potassium compounds (as well as other alkali metals) can evaporate, decompose and then condensate during the combustion process. Additional tests were conducted for two other main ash components: CaCO3 and SiO2. The decomposition and melting temperatures are known based on literature, but more detailed knowledge about pure salts behaviour is needed for biomass ashes STA results proper interpretation. Figs. 13–16 present melting, evaporation and decomposition

KCl (s , l)

KCl (g )

(4)

Decomposition of calcium carbonate runs in wide temperature range of 550 up to 825 °C. To detect the carbonates (crystalline phases using XRD) in studied biomass ashes, the ashes were obtained at 550 °C to avoid the decomposition of carbonates according to reaction: 12

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

Table 2 Characteristic parameters obtained from TG and DTG profiles of the studied ashes during the heating process under nitrogen atmosphere. Temperature range, °C

Solid residue

MCS1

T, °C Δm, %

25–200 −1.67 H2O

200–500 −1.34 MgCO3

500–700 −1.21 CaCO3

MCS2

T, °C Δm, %

25–250 −2.86 H2O

250–515 −1.64 MgCO3

515–670 −0.81 CaCO3

CS

T, °C Δm, %

25–215 −4.59 H2O

215–400 −1.93 MgCO3

MxG

T, °C Δm, %

25–260 −2.68 H2O

SV

T, °C Δm, %

25–300 −3.20 H2O

CaCO3 (s )

700–1050 −5.52 CaCO3 + KCl + K2CO3

1050–1200 −0.87 K2CO3 + K2SO4

1200 91.06

670–850 −3.79 CaCO3 + KCl

850–1050 −6.02 KCl + K2CO3

1050–1200 −7.76 K2CO3 + K2SO4

1200 79.98

400–500 −0.78 XOH/XHCO3/XHPO4

500–650 −1.21 CaCO3

650–1015 −4.12 CaCO3 + KCl + K2CO3

1015–1200 −0.81 K2CO3 + K2SO4

1200 91.15

260–500 −0.91 MgCO3

500–670 −0.83 CaCO3

670–850 −1.50 CaCO3 + KCl

850–1050 −1.31 KCl + K2CO3

1050–1200 −1.67 K2CO3 + K2SO4

1200 93.78

300–470 −6.65 MgCO3

470–770 −6.56 CaCO3

770–915 −2.80 K2CO3

1200–1500 −53.66 K3PO4

1500 30.33

(5)

CaO (s ) + CO2 (g )

highest mass loss of 4 % was observed for CS ash (the most noticeable DTG peak was observed), and then c.a. 1.5-3 % for other ones. The second mass pick on DTG curve was between c.a. 200 °C and 515 °C for all samples and it was associated with thermal decomposition of MgCO3, which decomposes in temperature range from 170 to 320 °C [52]. A significant mass loss of 10% was noted for SV ash with maximum temperature of DTG peak at 380 °C. Above 600 °C many effects were observed on TG, DTG and DSC curves. Table 2 and Figs. 17–21 shows the mass losses and temperature ranges of significant changes for all samples. To interpret the results of STA for biomass ashes, the results for reference substances were used. In the case of CaCO3, KCl and K2CO3 partial overlapping of peaks reflected to evaporation and thermal decomposition processes was observed. Regardless this fact, it was possible to identify the occurring processes (qualitative analysis) because DTG peaks correspond to the initial evaporation and thermal decomposition. The processes in the temperature range 700–1200 °C were associated with the evaporation of potassium salts, what was clearly observed for cereal mixed straw ashes, corn straw and Salix Viminalis ashes. It confirmed that potassium present in the biomass was mainly in the form of salts (chlorides, carbonates and sulphates). In the temperature range up to 1200 °C the most significant mass lost was observed for cereal mixed straw 2 ash (MCS2), but it partially come from the oxidation of organic matter present in the ash. MCS2 lowtemperature ash was characterised by several dark inclusions and most probably high unburned carbon content. To minimize oxidizing process thermal analysis was performed under nitrogen atmosphere. The SV ash contained significant amounts of volatile compounds mainly in the form of low-temperature salts and CaCO3, whereas these compounds were not observed for MxG. The experimental results for SV sample above 1200 °C showed high concentration of potassium phosphate in the sample. The test for this sample was performed at higher temperature, as the melting temperature of the SV sample was the lowest noted. The final step observed in all samples except SV, was thermal decomposition of K3PO4 connected with high-temperature reaction between potassium and silica. For all studied ashes the final product after STA analysis was partially vitrified slag, it means that analysis under higher temperature could cause final vitrification of the ash.

Fig. 15 presents the thermal behaviour of K2SO4. The small endothermic peak was observed in temperature range 350–550 °C. Probably, it could be attributed with small amount of contamination presented in salt. DSC peak at 585 °C had confirmed the phase transition of K2SO4 [15]. The melting point of K2SO4 had appeared at 1071 °C leading to its thermal decomposition (reaction no. (6)). The final decomposition did not take place up to 1200 °C.

K2 SO4 (s, l)

K2 O (s ) + SO2 (g ) +

1 O 2 (g ) 2

(6)

K2CO3 thermal decomposition took place according to the reaction (7) and had started at 850 °C. The DSC peak with maximum at 897 °C reflected meting point of K2CO3. At 1200 °C, the mass loss was 58%, but the complete thermal decomposition was not observed.

K2 CO3 (s )

K2 O (s ) + CO2 (g )

(7)

The thermal analysis of SiO2 had confirmed that SiO2 is thermally stable in studied temperature range and it does not decompose below 1200 °C. Not significant mass loss had appeared under 1050 °C. In general, the melting points defined based on STA experiments were very close to literature data: KCl − 772/776 °C, K2CO3 – 891 °C, K2SO4 – 1069 °C, KCl-K2CO3 − 632 °C, KCl-K2SO4 – 694 °C [35,50]. Some differences have appeared for potassium eutectics, most probably due to imperfect preparation procedure (e.g. problems with mixing and sample contamination), in the case of pure salts the results were nearly the same [51]. Taking into account the investigation concerning the influence of biomass ash chemical composition on slagging and fouling process, the presence of inorganic compounds in gas phase is crucial. The STA method allows to determine the mass changes during heating the biomass sample, but the most important is the temperature determination of occurring processes. The analysis of pure salts had provided the information about phase transition, melting point and other physical processes to help better understanding of the thermal processes going for studied biomass ashes. The biomass ash thermal analysis results are very difficult to interpret. Biomass ash consists of variety of elements leading to complex chemical reactions and transformations during the heating. DTG curves were created additionally as STA output. Figs. 17–21 present the TG, DTG and DSC curves for studied biomass ashes. Based on obtained STA results it was noted that thermal decomposition for all studied ashes took place in few stages and significant differences were observed. In all cases, the presence of moisture was confirmed, proving hygroscopic properties of biomass ashes. In the temperature range 60–200 °C, the water was released (Figs. 17–21). The

3.5. Phase analysis The crystalline compounds in ash samples were identified by XRD and the results are shown in Fig. 22 and summed up in Table 3. XRD analysis had confirmed the presence of potassium phases, such as KCl, K2SO4 and K2CO3 in all studied ashes besides SV ash, which contained mainly CaCO3 and Ca5(PO4)3(OH,F,Cl). The presence of SiO2 13

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

Fig. 22. Ash diffractograms (XRD) for: MCS1, MCS2, CS, MxG and SV samples (1 – KCl, 2 – SiO2, 3 – CaCO3, 4 - Ca5(PO4)3(OH,F,Cl), 5 – K2CO3, 6 – K2SO4).

amorphous forms, detected in major extent in the analysed biomass ashes (especially in case of: MCS1, MCS2 and MxG shown in Fig. 22), which were difficult to identified using XRD. Due to low temperature of incineration even a special long-term preparation procedure of ashes did not allow to obtain high-crystalline form of the ash. The most significant signal of KCl was observed for CS sample. The presence of high content of potassium compounds was reflected in lower ash fusion temperatures of MCS1, MCS2, CS and MxG ashes. The SV ash contained the highest amount of calcium and phosphorus and CaCO3 and Ca5(PO4)3(OH,F,Cl) were the major crystalline compounds detected in the ash. XRD results strongly depend on ash temperature preparation [17,54] and as the biomass ash incineration temperature is low, the XRD results are very hard to interpret. With an increase of temperature some phases will form (e.g. silicates, aluminosilicates) but other ones will decompose, like chlorides and carbonates.

Table 3 Crystalline compounds identified in studied ashes (based on XRD method). Sample

Mineral phase

MCS1 MCS2 CS MxG SV

SiO2, KCl, K2SO4, K2CO3 SiO2, KCl, K2SO4, K2CO3 SiO2, KCl, K2SO4, K2CO3 SiO2, KCl, K2SO4, K2CO3 CaCO3, Ca5(PO4)3(OH,F,Cl)

in MCS1, MCS2, CS and MxG had corresponded to high concentration of silicon in these ashes. The silicon compounds were expected as the major ash component, due to a significant amount of Si in ashes. It is known that in biomass ash different forms of aluminosilicates can occur [17,53]. Unfortunately, more complex silicates were not identified, and only quartz (SiO2) was the main detected crystalline phase. Most probably silicates or aluminosilicates were associated with an 14

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

Table 4 Ash fusion tendencies of MCS1, MCS2, CS, MxG and SV ashes (based on high-temperature microscope). Sample

200 °C

900 °C

1100 °C

Temperatures 1206 °C

MCS2

1202 °C

CS

1162 °C

MxG

1261 °C

SV

1400 °C

AREA [%]

MCS1

biomass ash obtained at higher temperature e.g. 800–900 °C some chemical reactions and transformation are observed and partial thermal decomposition and evaporation of mineral matter components will take place. Additionally, the presence of unburned coal can disturb the results. In this study, a high-temperature microscope was used to investigate low-temperature ash fusion behaviour. The characteristic temperatures and shape of ash samples prepared to determine ash fusion behaviour are presented in Table 4. The changes of the ash samples shape with temperature associated with melting and thermal decomposition of the samples are shown in Fig. 23. All studied ashes, besides SV ash had decreased its volume above 700 °C what was probably connected with the decomposition of CaCO3. For MCS2, CS and MxG ashes, the increase of volume at 850 °C reflecting melting and in consequence the evaporation of KCl was observed. In addition, the thermal decomposition of K2CO3 was confirmed for MCS1 and MCS2 samples. At 1050 °C MCS2 ash volume had increased once again resulting from thermal decomposition of K2SO4. CS ash had melt at the lowest temperature. Two stages were observed for MCS1 ash. The first stage was connected with melting and evaporation of KCl (up to 1050 °C) and the second - melting and evaporation of K2SO4. The MxG ash had increased its volume most significantly. The evaporation and melting processes of KCl and K2SO4 probably overlapped, what was confirmed based on STA results (two stages of mass loss were very near to each other). The SV ash contained mainly the carbonates, and some sulphates and phosphates that is why the sample had started to shrink under heating. A decrease of the sample at 1100 °C reflected K3PO4 decomposition. The determination of ash melting behaviour using high-temperature microscope connected with thermal analysis gives information about the behaviour of ash during heating

140 120 100 80 60 40 20 0 200

300

400

500

MCS1

600

700 800 900 1000 1100 1200 1300 1400 Temperature [°C] MCS2 CS MxG SV

Fig. 23. Ash samples behaviour at higher temperatures.

3.6. Ash melting behaviour Characteristic ash melting temperatures are well known parameters defining biomass ash melting behaviour [55,56]. However, despite of a detailed specification on determination of characteristic biomass ash melting temperatures, obtained results are highly uncertain and poorly repeatable [57]. In many cases, the ash melting process started in temperatures noticeably lower than laboratory determined initial deformation temperature (IDT) [38]. In the paper the visual interpretation was proposed to compare the samples ash melting behaviour. In general, melting behaviour of low-temperature ash obtained at 550 °C should reflect the ash properties with high precision. For 15

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al.

and is helpful in predicting of possible reactions and transformations in the ash phase. Unfortunately, the chemical and phase compositions of the ash are necessary to be known.

[7]

4. Conclusions

[8]

Selected agricultural and energy crop biomass ashes were chosen in this study for ash behaviour investigation and prediction of operating problems during biomass combustion. Biomass ashes were characterised by high potassium content and various chlorine and sulphur content. The ashes were investigated in details by means of several analytical methods: chemical fractionation procedure, thermal analysis (STA), X-ray diffraction (XRD) and high-temperature microscope. Potassium in the ashes was detected in the form of its salts: KCl, K2SO4, K2CO3 and K3PO4. The presence, concentration of alkali metals and possibility of low-melting eutectics formation has the major impact on melting temperatures of studied ashes. Leaching process of ash elements using water, ammonium acetate and hydrochloric acid solutions was performed to determine the association of alkali metals in the raw material. Based on the results, a mineral matter composition recalculation model was proposed to predict alkali compounds concentration in the fuel, revealing that the highest concentrations of KOH or K2CO3 might be expected in case of fuels with lower, in comparison to potassium, chlorine and sulphur concentrations. The proposed model might be used also to determine the risk of eutectics formation, the highest risk was determined for MCS1 and CS samples characterised also by the lowest melting temperature, what proves eutectics presence in the MCS1 and CS ash. The influence of the analysed biomass ashes on high-temperature corrosion of steel was further analysed in [58]. The results shows that the most suitable methods for prediction of ash related issues in biomass-firing units are chemical fractionation method, complemented with calculation method and verified by means of simultaneous thermal analysis and ash melting behaviour prediction using high-temperature microscope. The results of phase analysis are hard to interpret, due to low-temperature incineration of biomass.

[9] [10] [11] [12] [13] [14]

[15] [16] [17] [18]

[19] [20]

Declaration of Competing Interest

[21]

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

[22]

Acknowledgement

[23]

This work was supported by National Science Centre, Poland (grant no. 2016/21/N/ST8/03720).

[24]

References

[25]

[1] PricewaterhouseCoopers et al. Sustainable and optimal use of biomass for energy in the EU beyond 2020. VITO, Utrecht University, TU Vienna, INFRO, Rütter Soceco & PwC;2017:170. [2] Zeng T, Mlonka-Mędrala A, Lenz V, Nelles M. Evaluation of bottom ash slagging risk during combustion of herbaceous and woody biomass fuels in a small-scale boiler by principal component analysis. Biomass Convers Biorefinery 2019. https://doi.org/ 10.1007/s13399-019-00494-2. [3] Lee SH, Lee TH, Jeong SM, Lee JM. Economic analysis of a 600 mwe ultra supercritical circulating fluidized bed power plant based on coal tax and biomass cocombustion plans. Renew Energy 2019;138:121–7. https://doi.org/10.1016/j. renene.2019.01.074. [4] McIlveen-Wright DR, Huang Y, Rezvani S, Redpath D, Anderson M, Dave A, et al. A technical and economic analysis of three large scale biomass combustion plants in the UK. Appl Energy 2013;112:396–404. https://doi.org/10.1016/j.apenergy.2012. 12.051. [5] Pedrazzi S, Santunione G, Minarelli A, Allesina G. Energy and biochar co-production from municipal green waste gasification: a model applied to a landfill in the north of Italy. Energy Convers Manage 2019;187:274–82. https://doi.org/10.1016/j. enconman.2019.03.049. [6] Thallam Thattai A, Oldenbroek V, Schoenmakers L, Woudstra T, Aravind PV. Experimental model validation and thermodynamic assessment on high percentage (up to 70%) biomass co-gasification at the 253 MWe integrated gasification

[26] [27] [28] [29] [30] [31]

[32]

16

combined cycle power plant in Buggenum, The Netherlands. Appl Energy 2016;168:381–93. https://doi.org/10.1016/j.apenergy.2016.01.131. Kassman H, Pettersson J, Steenari BM, Åmand LE. Two strategies to reduce gaseous KCl and chlorine in deposits during biomass combustion – injection of ammonium sulphate and co-combustion with peat. Fuel Process Technol 2013;105:170–80. https://doi.org/10.1016/j.fuproc.2011.06.025. Lindberg D, Backman R, Chartrand P, Hupa M. Towards a comprehensive thermodynamic database for ash-forming elements in biomass and waste combustion – current situation and future developments. Fuel Process Technol 2013;105:129–41. https://doi.org/10.1016/j.fuproc.2011.08.008. Lindström E, Öhman M, Backman R, Boström D. Influence of sand contamination on slag formation during combustion of wood derived fuels. Energy Fuels 2008;22:2216–20. https://doi.org/10.1021/ef700772q. Du S, Yang H, Qian K, Wang X, Chen H. Fusion and transformation properties of the inorganic components in biomass ash. Fuel 2014;117:1281–7. https://doi.org/10. 1016/j.fuel.2013.07.085. Vassilev SV, Baxter D, Vassileva CG. An overview of the behaviour of biomass during combustion: part II. Ash fusion and ash formation mechanisms of biomass types. Fuel 2014;117:152–83. https://doi.org/10.1016/j.fuel.2013.09.024. Chaunsali P, Uvegi H, Traynor B, Olivetti E. Leaching characteristics of biomass ashbased binder in neutral and acidic media. Cem Concr Compos 2019;100:92–8. https://doi.org/10.1016/j.cemconcomp.2019.04.001. Wang Y, Wu H, Sárossy Z, Dong C, Glarborg P. Release and transformation of chlorine and potassium during pyrolysis of KCl doped biomass. Fuel 2017;197:422–32. https://doi.org/10.1016/j.fuel.2017.02.046. 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. https://doi.org/10.1016/ j.fuel.2016.07.047. Arvelakis S, Jensen PA, Dam-Johansen K. Simultaneous thermal analysis (STA) on ash from high-alkali biomass. Energy Fuels 2004;18:1066–76. https://doi.org/10. 1021/ef034065+. Magdziarz A, Gajek M, Nowak-Woźny D, Wilk M. Mineral phase transformation of biomass ashes e experimental and thermochemical calculations. Renew Energy 2018;128:446–59. https://doi.org/10.1016/j.renene.2017.05.057. Reinmöller M, Schreiner M, Guhl S, Neuroth M, Meyer B. Ash behavior of various fuels: the role of the intrinsic distribution of ash species. Fuel 2019;253:930–40. https://doi.org/10.1016/j.fuel.2019.05.036. Pettersson A, Zevenhoven M, Steenari BM, Åmand LE. Application of chemical fractionation methods for characterisation of biofuels, waste derived fuels and CFB co-combustion fly ashes. Fuel 2008;87:3183–93. https://doi.org/10.1016/j.fuel. 2008.05.030. Benson SA, Holm PL. Comparison of inorganics in three low-rank coals. Ind Eng Chem Prod Res Dev 1985;24:145–9. https://doi.org/10.1021/i300017a027. Van Dyk JC, Baxter LL, Van Heerden JHP, Coetzer RLJ. Chemical fractionation tests on South African coal sources to obtain species-specific information on ash fusion temperatures (AFT). Fuel 2005;84:1768–77. https://doi.org/10.1016/j.fuel.2005. 04.006. Werkelin J, Skrifvars BJ, Zevenhoven M, Holmbom B, Hupa M. Chemical forms of ash-forming elements in woody biomass fuels. Fuel 2010;89:481–93. https://doi. org/10.1016/j.fuel.2009.09.005. Baxter LL, Miles TR, Miles Jr TR, Jenkins BM, Milne T, Dayton D, et al. The behavior of inorganic material in biomass-fired power boilers: field and laboratory experiences. Fuel Process Technol 1998;54:32. https://doi.org/10.1016/S0378-3820(97) 00060-X. Zevenhoven-Onderwater, Blomquist MJ, Skrifvars B, Backman R, Hupa M. The prediction of behaviour of ashes from five different solid fuels in fluidised bed combustion. Fuel 2000;79:1353–61. https://doi.org/10.1016/S0016-2361(99) 00280-X. Nutalapati D, Gupta R, Moghtaderi B, Wall TF. Assessing slagging and fouling during biomass combustion: a thermodynamic approach allowing for alkali/ash reactions. Fuel Process Technol 2007;88:1044–52. https://doi.org/10.1016/j. fuproc.2007.06.022. Theis M, Skrifvars BJ, Zevenhoven M, Hupa M, Tran H. Fouling tendency of ash resulting from burning mixtures of biofuels. Part 3. Influence of probe surface temperature. Fuel 2006;85:2002–11. https://doi.org/10.1016/j.fuel.2005.12.011. Stam AF, Livingston WR, Cremers MFG, Brem G. Review of models and tools for slagging and fouling prediction for biomass co-combustion. Rev Artic IEA 2010:1–18. Laumb JD, Folkedahl BC, Zygarlicke Chris. Chapter 4, characteristics and behavior of inorganic constituents. In: Miller BG, Tillman David A, editors. Combust. Eng. Issues Solid Fuel Systems. Elsevier; 2008. Baxter L. Biomass-coal co-combustion: opportunity for affordable renewable energy. Fuel 2005;84:1295–302. https://doi.org/10.1016/j.fuel.2004.09.023. Miles TR, Baxter LL, Bryers RW, Jenkins BM, Oden LL. Alkali deposits found in biomass power plants. NREL Rep 1995;1:1–122. I. Mlonka-Mędrala A, Magdziarz A, Dziok T, Nowak W. Laboratory studies on the influence of biomass particle size on pyrolysis and combustion using TG GC/MS. Fuel 2019;252:635–45. https://doi.org/10.1016/j.fuel.2019.04.091. Piotrowska P, Zevenhoven M, Hupa M, Giuntoli J, De Jong W. Residues from the production of biofuels for transportation: characterization and ash sintering tendency. Fuel Process Technol 2013;105:37–45. https://doi.org/10.1016/j.fuproc. 2011.09.020. Glarborg P, Jensen PA, Dam-Johansen K, Illerup JB, Karlström O, Brink A, et al. editors. Scientific tools for fuel characterization for clean and efficient biomass combustion: SciToBiCom final report; 2013.

Fuel 261 (2020) 116421

A. Mlonka-Mędrala, et al. [33] Wang X, Bai S, Jin Q, Li S, Li Y, Li Y, et al. Soot formation during biomass pyrolysis: effects of temperature, water-leaching, and gas-phase residence time. J Anal Appl Pyrolysis 2018;134:484–94. https://doi.org/10.1016/j.jaap.2018.07.015. [34] Blomberg T. Which are the right test conditions for the simulation of high temperature alkali corrosion in biomass combustion? Mater Corros 2006;57:170–5. https://doi.org/10.1002/maco.200503905. [35] Blomberg T, Makkonen P, Hiltunen M. Role of alkali hydroxides in the fireside corrosion of heat transfer surfaces, a practical approach. Mater Sci Forum 2004;461–464:883–90. https://doi.org/10.4028/www.scientific.net/MSF.461-464. 883. [36] Frandsen FJ, van Lith SC, Korbee R, Yrjas P, Backman R, Obernberger I, et al. Quantification of the release of inorganic elements from biofuels. Fuel Process Technol 2007;88:1118–28. https://doi.org/10.1016/j.fuproc.2007.06.012. [37] Hansen LA, Frandsen FJ, Dam-Johansen K, Sørensen HS, Skrifvars B-J. Characterization of ashes and deposits from high-temperature coal−straw co-firing. Energy Fuels 1999;13:803–16. https://doi.org/10.1021/ef980203x. [38] Tortosa Masiá AA, Buhre BJP, Gupta RP, Wall TF. Characterising ash of biomass and waste. Fuel Process Technol 2007;88:1071–81. https://doi.org/10.1016/j.fuproc. 2007.06.011. [39] Nielsen HP, Frandsen FJ, Dam-Johansen K. Lab-scale investigations of high-temperature corrosion phenomena in straw-fired boilers. Energy Fuels 1999;13:1114–21. https://doi.org/10.1021/ef990001g. [40] Garcia-Maraver A, Mata-Sanchez J, Carpio M, Perez-Jimenez JA. Critical review of predictive coefficients for biomass ash deposition tendency. J Energy Inst 2017;90:214–28. https://doi.org/10.1016/j.joei.2016.02.002. [41] Vassilev SV, Baxter D, Andersen LK, Vassileva CG, Morgan TJ. An overview of the organic and inorganic phase composition of biomass. Fuel 2012;94:1–33. https:// doi.org/10.1016/j.fuel.2011.09.030. [42] Tiainen MS, Ryynänen JS, Rantala JT, Patrikainen HT, Laitinen RS. Determination of amorphous material in peat ash by X-ray diffraction. Impact Miner Impurities Solid Fuel Combust 2005:217–24. https://doi.org/10.1007/0-306-46920-0_15. [43] Ward CR, Gupta RP, Wall TF, Baxter L, editors. Mineral Characterization for Combustion impact of mineral impurities in solid fuel combustion, Springer; 2002. [44] Frandsen FJ. DTU Ash Course, January 2015, Exercise 01, PhD Course on Ash and Deposit Formation, Corrosion in Utility Boilers Technical University of Denmark, 19-23.01.2015. 2015. [45] Reinmöller M, Klinger M, Schreiner M, Gutte H. Relationship between ash fusion temperatures of ashes from hard coal , brown coal , and biomass and mineral phases under different atmospheres: A combined FactSage TM computational and network theoretical approach 2015;151:118–23. DOI:10.1016/j.fuel.2015.01.036. [46] Pettersson J, Folkeson N, Johansson LG, Svensson JE. The effects of KCl, K2SO4 and

[47] [48] [49] [50] [51] [52]

[53] [54] [55]

[56] [57] [58]

17

K2CO3 on the high temperature corrosion of a 304-type austenitic stainless steel. Oxid Met 2011;76:93–109. https://doi.org/10.1007/s11085-011-9240-z. Niu Y, Zhu Y, Tan H, Hui S, Jing Z, Xu W. Investigations on biomass slagging in utility boiler: criterion numbers and slagging growth mechanisms. Fuel Process Technol 2014;128:499–508. https://doi.org/10.1016/j.fuproc.2014.07.038. Li L, Yu C, Huang F, Bai J, Fang M, Luo Z. Study on the deposits derived from a biomass circulating fluidized-bed boiler. Energy Fuels 2012;26:6008–14. https:// doi.org/10.1021/ef301008n. Niu Y, Tan H, Ma L, Pourkashanian M, Liu Z, Liu Y, et al. Slagging characteristics on the superheaters of a 12 MW biomass-fired boiler. Energy Fuels 2010;24:5222–7. https://doi.org/10.1021/ef1008055. Janz GJ, Allen Carolyn B, Bansal NP, Murphy RM, Tomkins RP. Physical Properties Data Compilations Relevant to Energy Storage. I. Molten Salts: Eutectic Data. National Bureau of Standards; 1978. Mlonka-Mędrala A, Gołombek K, Buk P, Cieślik E, Nowak W. The influence of KCl on biomass ash melting behaviour and high-temperature corrosion of low-alloy steel. Energy 2019;188:116062https://doi.org/10.1016/j.energy.2019.116062. Devasahayam S, Strezov V. Thermal decomposition of magnesium carbonate with biomass and plastic wastes for simultaneous production of hydrogen and carbon avoidance. J Clean Prod 2018;174:1089–95. https://doi.org/10.1016/j.jclepro. 2017.11.017. Magdziarz A, Gajek M, Nowak-Woźny D, Wilk M. Mineral phase transformation of biomass ashes – experimental and thermochemical calculations. Renew Energy 2018;128:446–59. https://doi.org/10.1016/j.renene.2017.05.057. Niu Y, Tan H, Liu Y, Wang X, Xu T. The effect of particle size and heating rate on pyrolysis of waste capsicum stalks biomass. Energy Sources, Part A Recover Util Environ Eff 2013;35:1663–9. https://doi.org/10.1080/15567036.2010.509084. Vassilev SV, Baxter D, Andersen LK, Vassileva CG. An overview of the composition and application of biomass ash: part 2. Potential utilisation, technological and ecological advantages and challenges. Fuel 2013;105:19–39. https://doi.org/10. 1016/j.fuel.2012.10.001. Magdziarz A, Wilk M, Gajek M, Nowak-Woźny D, Kopia A, Kalemba-Rec I, et al. Properties of ash generated during sewage sludge combustion: a multifaceted analysis. Energy 2016;113:85–94. https://doi.org/10.1016/j.energy.2016.07.029. Toscano G, Corinaldesi F. Ash Fusibility characteristics of some biomass feedstocks and examination of the effects of inorganic additives. J of Ag Eng 2010;4(2):13–9. https://doi.org/10.4081/jae.2010.2.13. Mlonka-Mędrala A, Magdziarz A, Kalemba-Rec I, Nowak W. The influence of potassium-rich biomass ashes on steel corrosion above 550 ° C. Energy Convers Manage 2019;187:15–28. https://doi.org/10.1016/j.enconman.2019.02.074.