Prediction of biomass corrosiveness over different coatings in fluidized bed combustion

Abstract

Energy production in biomass fired boilers is increasing rapidly due to the advantages of CO₂ neutrality and renewability, however damaging agents present in biomass composition accelerates power plant components corrosion. This study evaluates the influence of the biomass burned in fluidized bed combustion processes on high-temperature corrosion, by means of thermodynamic equilibrium modelling, considering those reactions occurring between the combustion atmosphere and different protective coatings (isFeAl, isNiAl and isSiCrAl). Fuels composition and operating conditions from a 10 kW BFB boiler were introduced as input data to improve the performance of the model. Representative samples from agricultural waste, industrial wood and forestry wood waste were selected for evaluation. Results showed industrial wood waste as highly damaging for most coatings studied, with high risk of salt stickiness, deposits formation and release of acidic gases. The elevated volatiles percentage together with significant ash content determined might lead to a major ash components release to the gas phase, available to later condense in the metals surfaces. Silication of alkali and deposited alkali chlorides were the dominant corrosion mechanisms observed for most cases. An increase in alloys corrosion resistance was detected through the model when nickel or chromium was present, showing isSiCrAl as the most resistant. However, alloys protection exhibited significant variations depending upon the biomass burned, thus materials selection should consider the compatibility with conditions for its final use. Thermodynamic modelling, based on real conditions and fuels composition, provides a useful tool to identify key factors for protective coatings design when employing new waste fuels.

Introduction

Energy from biomass, a renewable and CO₂-neutral source, has become the most globally used renewable energy source, corresponding to more than 60% of all renewable energy sources (RES) in Europe. Biomass includes a large variety of different fuels with different chemical composition and combustion characteristics [1].

In spite of the benefits, biomass composition could be a potential problem in the thermal conversion of low-grade fuels, as it contains an important concentrations of ash-forming elements (K, Na, S, Cl, Ca, Mg, Si, P, Al, and Fe) which can be released during the thermal conversion to form bottom and fly ashes, which at a later stage affect the thermal process, giving rise to operational problems such as deposit formation, corrosion, slagging and bed agglomeration in fluidized-bed boilers [1,2,3]. Deposit formation on the surface of the superheater tubes reduces the boiler efficiency. High concentrations of alkali and chlorine in the deposit also cause severe corrosion of the steel surfaces [2]. During biomass combustion, chlorine is released as HCl and Cl₂, reacting with the metal heating surface, causing severe active oxidation corrosion [3]. Also, it can promote the gasification of the alkali metal elements in the fuel and react with alkali metals to form volatile alkali chloride in the gas phase. The vast majority of gaseous alkali metals are deposited and condensed on the low-temperature metal heating surface. Furthermore, they trap solid particles in flue gas, which leads to serious slagging, fouling and corrosion [4].

High-temperature corrosion has always been a major obstacle to improve the boiler steam parameters [5]. With the increase in the boiler combustion temperature and the presence of corrosion elements, intense corrosion in the furnace has made the traditional alloys no longer satisfactory.

A corrosion attack can be mitigated in two ways: improving the materials or changing the surrounding environment of the materials. Several field and laboratory studies [6,7,8,9,10] have investigated both corrosion mechanisms and ways of minimizing this problem, such as the use of chemical additives [11,12,13]. A great deal of research has also been devoted to the development of new materials and technologies to meet the high-temperature environment in the furnace [14,15,16,17,18,19,20,21,22]. Modifying the surface of boiler components by depositing protective coatings based on the formation of corrosion-resistant oxide scales is also a possible mitigation route [23]. Coatings provide a way of extending the limits of use of materials at the upper end of their performance capabilities. The phases that constitute the coating layer should be thermodynamically stable in the operating environment. A large number of coating processes are available to provide surface protection [24]. Several authors have pursued efficient coatings to prevent the high-temperature corrosion of metallic materials in biomass fuelled boilers and waste incinerators [25,26,27].

Despite the growing interest in metallic coatings, a comprehensive treatment of the coatings from the experimental methodologies to the fundamentals and of their corrosion behaviour in real conditions is lacking. There are studies on different aspects of coatings, including wear [28] or erosion-corrosion performance [29], hot corrosion [30] low-temperature corrosion [31] and corrosion in supercritical boilers [32]. However, to date, studies developed exhibit difficulties in relating specific parameters to corrosion observed, as there are too many parameters such as temperature, ash deposit composition and gas composition which are continually fluctuating. For this reason, most studies carried out to check coatings efficiency are developed at laboratory reactors employing synthetic gas mixtures with addition of salts to simulate real boiler conditions, while there is a lack of research on the effect of coating composition in real corrosive environments typical of biomass energy power plants.

On the other hand, to understand corrosion mechanisms, it is necessary to know more about chemical reactions happening in the boiler. In this sense, it is fundamental to predict the ash transformation behaviour and also to find methods in order to estimate the degree of ash-related problems for specific fuels and fuel mixtures [2].

The first step to determine the adequacy of a given fuel, from the ash behaviour point of view corrosion, is the elemental analysis of the fuel ash content. Apart from the analytical methods, thermodynamic equilibrium modelling has also become a commonly used tool to better understand the ash transformation behaviour. Most of reported thermodynamic studies have been focused in various ash-related processes such as deposition related issues, slag formation in furnaces, bed agglomeration of fluidized-beds and smelt bed behaviour [33,34,35,36]. Other authors have also evaluated corrosion of heat transfer surfaces through the prediction of the potential reactions that may take place among the metals of the superheaters with the combustion atmosphere [1, 2, 23]. However, these studies are usually based on the evaluation of one specific fuel and, with respect to the composition of the combustion atmosphere (gases and ashes), representative compositions of biomass combustion atmospheres are often employed as input data for the model. Until the moment, there are not reported studies which evaluate and compare the affection of different biomass fuels over a specific material, employing real combustion conditions.

In this sense, this study aims to assess the risk of corrosion of representative biomass fuels over different protective coatings, considering the reactions that take place among the alloying elements with the combustion atmosphere and the deposits. Thermodynamic equilibrium modelling was used as a predictive tool, simulating different atmospheres arising from the combustion of representative biomasses from diverse categories (agricultural, forestry and industrial) over three different alloys (isFeAl, isNiAl and isSiCrAl).

The first part of this research includes a previous physicochemical characterization of the biomass fuels, in order to both anticipate the potential ash corrosiveness of the diverse biomass types and also, to better interpret the results obtained in the model. The second part evaluates the influence of the biomass composition on corrosion during FBC, by modelling the exposition of each selected alloy to the diverse combustion atmospheres. In this sense, modelling results have been divided into three main sections corresponding to three target coatings.

The final objective of this work consists of determining which of these alloys may involve a major protection in boilers when employing a specific type of biomass, thus last part of this article compares the potential resistance provided by the diverse coatings against a given biomass combustion atmosphere.

Material and methods

Biomass fuels analysis

The selection of the biomass fuels was conducted on the basis of covering the main biomass categories employed as power plant boiler fuels. In this sense, representative samples from agricultural waste (wheat straw, WS), industrial wood (industrial waste wood, IWW) and wood residues (forestry residues, FR, and eucalyptus wood, EW) were chosen for the study.

The first step in developing a thermodynamic equilibrium modelling is to carry out a complete characterization of the fuels. For effective utilization of biomass fuel, the knowledge of their characterization is essential. The constituents of biomass fuel vary from region to region and depend upon sources from which biomass is collected and method of preparation of biomass.

Characterization of the solid biofuels was carried out, based on the latest European standards for biomass materials. The proximate and ultimate analyses of the fuels were undertaken by utilizing LECO Truspec equipment. Minor elements content in fuels (ash analyses) was quantified by ICP-AES and ICP-MS techniques. Trace elements (TEs) concentrations were also determined through more specific techniques. For mercury case, a direct analyser (DMA 80) based on EPA Method 7473 was employed. As and Se were determined through HGAAS, while for Cd, GFAAS technique was used. The rest of TEs analyses were carried out through ICP-AES.

Coatings selection

The chemical composition of three promising coatings was used for the study. The selection includes the composition of two slurry aluminides coatings with and without silicon addition (isSiCrAl, isFeAl,) and one hybrid coating, consisting in a slurry aluminide coating combined with nickel electrodeposition (isNiAl). These slurry aluminide coatings have been produced on different substrates and have demonstrated excellent high-temperature corrosion resistance under steam, fire-side corrosion and metal dusting conditions among others.

In addition to the coating microstructure, its chemical composition is also of great relevance to the corrosion protection ability during biomass combustion. The chemical composition of selected alloys (Sect. 2.3) was used to introduce metallic elements contents.

Ni- and Fe-based coatings are the most commonly used alloy systems in boiler applications. The main motivation for using Fe-based alloys is that they are rather cheaper and more environmentally friendly compared to many other alloying elements. Ni-based thermal spray coatings often display good resistance to high-temperature corrosion. The Ni matrix is preferred over other metal matrices like Fe in Cl-containing environments as the formation of NiCl₂ is thermodynamically less favoured than other detrimental metallic chlorides such as FeCl₂.

Cr is a commonly used alloying element in materials for high-temperature applications (850 °C) that can form a stable oxide (Cr₂O₃).

Similar to Cr, Al is also widely used in thermal spray powder composition, particularly for high-temperature corrosion applications.

Low concentrations of Si are often used in high-temperature alloys. A small level of Si can also be found in thermal spray powder compositions to provide the high-temperature corrosion resistance by the formation of a protective SiO₂ layer on the coating surface. A small level of Si can also facilitate formation of other protective oxide scales such as Al-rich or Cr-rich oxides. If Cr is also available in the composition, the high O affinity of Si leads to the formation of a SiO2 layer beneath the chromia scale.

As was stated before, under oxidizing–chloridizing conditions, the situation is much more complex as an internal selective attack can occur, depending on the alloying elements present in the material. In this sense, detailed knowledge of the behaviour of different alloying elements in commercial materials, in particular, coatings, is important.

Modelling approach

The modelling was performed using HSC Chemistry 6.1 software. Thermodynamic equilibrium modelling is based on the Gibbs energy minimization, which calculates the chemical the most stable composition of the system (equilibrium composition) at specified conditions, considering thermodynamic data from all the phases and compounds. As a simplification, activity coefficients were taken as 1.0.

In the fluidized-bed environment, ash reactions could approach equilibrium due to a good gas–solid contact and mixing and a relatively long residence time for the solids. However, in spite of the advantages found for thermodynamic equilibrium modelling in combustion systems, it is important to consider the limitations of this program. Firstly, not all the possible species that may occur under different conditions are included in the thermodynamic databases. Secondly, physical processes are not taken into account, while in experimental conditions, reactions are highly dependent on the reactants mixing, residence time and on the kinetics of the reactions.

The first step in the equilibrium calculation is the definition of the system. In this study, calculations have been carried out considering combustion parameters obtained at a 10 kWth BFB pilot plant, located at Ciemat (Fig. 1), considering the combustion pattern in the fluidized-bed system and the temperature profile in the combustion zone, although results obtained are considered to be valid for bigger scales.

Fig. 1

Scheme of the equilibrium reactor. Inputs, outputs and operation conditions

As input data for the model, chemical composition of atmosphere (air) and biomass fuels (Table 1) as well as recorded operating data, such as temperature, pressure and air/fuel ratio, were introduced in the theoretical reactor, simulating the conditions in the real BFB boiler.

Table 1 Biomass fuels characterization

Secondly, the chemical compositions of selected alloys (Table 2) were used to introduce the values of the metallic elements.

Table 2 K moles of metallic elements introduced in the thermodynamic calculations

By crossing these coatings with the different biomass fuels, the theoretical matrix carried out for the study can be drawn as shown in Fig. 2. The study will thus compare both the effectiveness of the different coatings for a specific biomass and also the resistance of each coating versus the different types of biomasses.

Fig. 2

Matrix of the modelling tests

The maximum temperature that the flue gas reaches in the FBC system (850 °C) was considered; however, the main emphasis was to predict the condensation behaviour of the flue gas in the superheater section. Therefore, calculations were performed at atmospheric pressure into a global temperature range of 300–800 °C, giving a special attention to the temperature interval of 500–650 °C) where the flue gas meets the superheater steel tubes making it possible to evaluate the fouling tendency.

For discussion of the results, only species contributing more than 1% of total compounds were considered.

Results and discussion

Biomass fuels characterization

Table 1 shows physical and chemical characterization results achieved for selected biomass samples, through the methodology described in Sect. 2.1. Figures 3a and b shows, respectively, major and trace elements contents determined in the biomasses analyses, in order to compare and predict the potential corrosiveness of the samples.

Fig. 3

Biomass fuels characterization a Major elements concentrations; b Trace elements concentrations

Physicochemical characterization results have shown great variability among the different biomass fuels selected. From proximate analysis, it can be checked how the major ash content corresponds to forestry residues (FR) samples. The highest sulphur content is also detected for FR sample. However, the elevated calcium concentration present in this biomass may counteract sulphur released to the gas phase during combustion.

In addition, TEs content determined for this biomass together with that obtained for industrial waste wood (IWW) shows the highest values, with significant concentrations detected for elements such as Cu, Ti, Pb and Zn. In contrast, the lowest TEs contents were found for wheat straw (WS). These results indicate a more elevated TEs content (including mercury) in woody biomasses than in agro-residues. This statement is in agreement with previous studies of biomass materials characterization. Previous researchers [37] observed that fly ashes from wood samples have generally higher As, Cd, Pb and Hg contents than those of agricultural residues. Stromberg B [38] denoted a typical elevated content of impurities in waste wood materials.

It is also noteworthy ash analyses results obtained for eucalyptus wood samples, with significantly higher amount of ash-forming light metals (Al, Ca, Mg, Na and K) in contrast to the rest of the biomasses studied. These elements have great influence on ash-related problems in combustion plants.

With respect to agro-residues, wheat straw samples exhibit the most elevated chlorine concentration. This fact, joined to the important potassium content of this fuel, may lead to potential corrosive atmosphere during combustion.

Corrosiveness of biomass ashes

In order to anticipate and compare the potential risk of selected biomass materials in the combustion process, different fuel indexes (key numbers) were calculated for the diverse samples, based on the chemical composition results of biomass fuel ashes. These indexes relate the various substances in the ash to one another, being useful to find out the critical quantity relationships between different substances that can be directly related to risk components or risky mixtures of components [39]. A comprehensive overview on different fuel indexes has been provided in the literature [40, 41].

$$ {\text{Vaporization Ratio}}: {\text{Cl}}/ \, \left( {{\text{K}} + {\text{Na}}} \right) $$

(1)

$$ {\text{Salt}}\;{\text{Ratio}}: \left( {{\text{Cl}} + {\text{2S}}} \right)/ \, \left( {{\text{K}} + {\text{Na}}} \right) $$

(2)

$$ {\text{Sulphating}}\;{\text{number}}: {\text{2S}}/{\text{Cl}} $$

(3)

Table 3 shows key numbers values determined for biomass fuels ashes. The major vaporization ratio, which indicates high risk of corrosive chlorine-rich deposits formation, was obtained for industrial waste wood. Higher values for salt ratio index were also detected for this biomass together with forestry residues, predicting hazardous for salt stickiness, formation of deposits and free corrosive acidic gases. Hazardous for alkaline sintering is expected for wheat straw, with salt ratio values below 0.7.

Table 3 Key numbers for biomass materials studied

Sulphating number exhibit elevated values for all biomasses except for eucalyptus wood, with significantly major numbers calculated for forestry residues case. This key number shows whether there is sufficient surplus sulphur to reduce the risk of corrosion in conjunction with alkali chlorides. It has been found from experience that if the molar ratio S/Cl is between 2 and 4, the risk of chloride-induced corrosion will be reduced [38]. In this sense, chloride-induced corrosion may be offset by sulphur in forestry residues case. On contrast, low valued found for eucalyptus wood ashes denotes potential chloride-induced corrosion when employing this biomass.

Figure 4 summarizes the potential risks determined for the diverse biomass ashes studied, revealing that significant hazard might be caused if industrial waste wood or forestry residues were fed in a combustion process, leading to salt stickiness and free corrosive acidic gases.

Fig. 4

Potential risk determined for biomass ashes studied, based on fuel index calculations

Thermodynamic equilibrium modelling for isFeAl coating

Impact of wheat straw combustion on isFeAl coatings

Main solid species formed from Fe/Al interactions with ash-forming elements during conventional combustion of wheat straw were determined. Figure 5a shows equilibrium composition of main species determined at a wide temperature range from 300 to 800 °C. Only those species resulting from interactions towards isFeAl coating are shown in this figure, although other corrosive forms have also been detected to be originated in this atmosphere.

Fig. 5

Equilibrium composition and temperature influence of main species resulting from isFeAl–ash interactions during combustion of: a WS; b FR; c IWW; and d EW

In these conditions, results exhibit CaAl₂Si₃O₁₀(OH)₂ as the main compound formed, assuming more than 35% of total species, coming from the reaction of calcic components in fuel and being enhanced by both silica concentration and water content. Dodson JR research on wheat straw ashes use [42] also reported the occurrence of insoluble calcium silicate precipitates in wheat straw ashes combusted at 600 °C or above. Secondly, KFe₃(FeSi₃O₁₀)(OH)₂ accounts for approximately 20% of total species predicted to be formed in this scenery. Previous researchers have stated about the formation of insoluble potassium silicates and aluminosilicates during wheat straw combustion [43]. Nielsen et al. [44] gave a useful analysis of the deposition of potassium salts on superheater tubes during straw firing, showing how potassium was present as K₂OSiO₂(s), KCl(s) and K₂SO₄(s) and at low temperatures.

Results obtained (Fig. 5a) exhibits how both compounds are stable in a wide thermal range (300–780 °C). Dodson JR [42]also showed how up to 500 °C, potassium solubility remains constant irrespective of the extent of combustion or the cooling condition.

The presence of these compounds indicates corrosion processes associated with silication of alkali chlorides, which can be described through Eqs. 4–6:

$$ {\text{2MCl }}\left( {{\text{s}},{\text{ l}},{\text{ g}}} \right) \, + \, \left( {{\text{2SiO}}_{{2}} + {\text{ Al}}_{{2}} {\text{O}}_{{3}} } \right) \, \left( {{\text{s}},{\text{ l}}} \right) \, + {\text{ H}}_{{2}} {\text{O }}\left( {\text{g}} \right) \, \to {\text{2MAlSiO}}_{{4}} \left( {{\text{s}},{\text{ l}}} \right) \, + {\text{ 2HCl }}\left( {\text{g}} \right) $$

(4)

$$ {\text{2MCl }}\left( {{\text{s}},{\text{ l}},{\text{ g}}} \right) \, + \, \left( {{\text{4SiO}}_{{2}} + {\text{ Al}}_{{2}} {\text{O}}_{{3}} } \right) \, \left( {{\text{s}},{\text{ l}}} \right) \, + {\text{ H}}_{{2}} {\text{O }}\left( {\text{g}} \right) \, \to {\text{2MAlSi}}_{{2}} {\text{O}}_{{6}} \left( {{\text{s}},{\text{ l}}} \right) \, + {\text{ 2HCl }}\left( {\text{g}} \right) $$

(5)

$$ {\text{2MCl }}\left( {{\text{s}},{\text{ l}},{\text{ g}}} \right) \, + \, \left( {{\text{6SiO}}_{{2}} + {\text{ Al}}_{{2}} {\text{O}}_{{3}} } \right) \, \left( {{\text{s}},{\text{ l}}} \right) \, + {\text{ H}}_{{2}} {\text{O }}\left( {\text{g}} \right) \, \to {\text{2MAlSi}}_{{3}} {\text{O}}_{{8}} \left( {{\text{s}},{\text{ l}}} \right) + {\text{ 2HCl }}\left( {\text{g}} \right) $$

(6)

M in equations denotes the metal from alkali chlorides. HCl generated through Eqs. 4–6 is responsible for corrosion, as it diffuses towards the metal surface to form volatile metal chlorides, such as FeCl₂ leading to the continuous transport of the metal scale interface to the bulk gas [45, 46].

The bar graph in Fig. 5a compares the percentages of Fe/Al-ash compounds predicted to be originated at the selected temperature range of 500–700 °C. From the graphic, it can be checked how, apart from calcium, potassium also exerts a high influence in the formation of most deposits predicted for this biomass. On the other hand, silicon has been detected as an important precursor for most of determined chemical forms. Straw samples analyses (Fig. 3) exhibited great contents of silicon and potassium, in contrast to other biomasses, and calcium, which may have given rise to these chemical forms.

Berlanga et al. [1], in their thermodynamic calculations for wheat straw corrosion, detected metal oxidation initiated by the presence of potassium. During combustion, soluble K is released as K(g), KOH(g), KCl(g), and other species which may further react with other compounds of the flue gas. The most important secondary reaction is the conversion of alkali vapours (e.g. KOH(g), KCl(g), K2SO4(g)) into corresponding alkali silicates [28, 37]. Aluminium silicates may capture volatile potassium during combustion, thereby reducing the risk of deposit and corrosion. [39]. However, high contents of silica together with potassium are directly linked to the formation of tenacious surface deposits on firesides and heating surfaces [40].

In this sense, silica can modify the chemistry of other ash-forming elements, such as K, Na, and Ca, through secondary reactions and thus indirectly affect the deposition and corrosion properties of fly ash. Otherwise, the presence of hydroxyl groups in main chemical forms originated reveals that water content in fuel may increase the formation of main deposits.

With respect to minor components predicted, Fe oxides formation is observed to be strongly affected by temperature.

Impact of forestry residues combustion on is FeAl coatings

Figure 5b shows the equilibrium diagram of main compounds arising from reactions of forestry residues ash components with Fe/Al materials. The percentages of those species formed at 500–700ºC temperature range are exhibited in the bar graph from Fig. 5b.

Forestry residues was seen to contain significant concentrations of calcium (Fig. 3a), in comparison with the rest of the fuels studied (with the exception of eucalyptus). In this sense, main forms originated in the process (≥ 10%) are result of Al and Fe interactions with calcium components in the fuel, giving rise to CaAl₄Si₂O₁₀(OH)₂ and CaFe₅O₇. Silicon is also present in significant contents in this fuel, promoting Ca/Mg–aluminium silicates formation. As it was described for WS, the occurrence of alumina-silicates such as CaAl₄Si₂O₁₀(OH)₂ involves corrosion mechanisms associated with silication of alkali chlorides, (Eqs. 4–6) [45, 46]. CaFe₅O₇ indicates active oxidation of iron in these conditions, but the formation of this specie is seen to decrease at temperatures above 650 °C (Fig. 5b). Grabke et al. [47] gave the basis for understanding this mechanism.

A second group of aluminium compounds (5–10%) are result of aluminium interactions with Na, Mg, K and Cl (NaAlO₂, Mg*Al₂O₃, KAl₃(SO₄)₂(OH)₆ and AlClO). NaAlO₂ formation is related to corrosion associated with deposited alkali chlorides. Potential reactions for this corrosion mechanism are listed below (7, 8). M denotes the metal involved (i.e. Al, Fe, etc.):

$$ {\text{M}}_{{2}} {\text{O}}_{{3}} \left( {\text{s}} \right) \, + {\text{2NaCl }}\left( {\text{s}} \right) \, + {1}/{\text{2 O}}_{{2}} \left( {\text{g}} \right) \to {\text{ 2NaMO}}_{{2}} \left( {{\text{s}},{\text{ l}}} \right) \, + {\text{Cl}}_{{2}} \left( {\text{g}} \right) $$

(7)

$$ {\text{M}}_{{2}} {\text{O}}_{{3}} \left( {\text{s}} \right) \, + {\text{2KCl }}\left( {\text{s}} \right) \, + {1}/{\text{2 O}}_{{2}} \left( {\text{g}} \right) \to {\text{ 2KMO}}_{{2}} \left( {{\text{s}},{\text{ l}}} \right) \, + {\text{Cl}}_{{2}} \left( {\text{g}} \right) $$

(8)

These reaction products have been observed in previous corrosion testing reports [45, 48]. For this fuel, K and Na capture are seen to be more influenced by sulphur than by silica, as it occurs in wheat straw case. Sulphur reduces KCl concentration in the flue gas by sulphation. In addition, Kassman et al. [11] confirmed that the presence of gaseous SO₃ is of greater importance than that of SO₂ for the sulphation of gaseous KCl.

K/NaAl₃(SO₄)₂(OH)₆ observed in Fig. 5b, indicates corrosion processes associated with sulphation of alkali chlorides, which can be described by reactions 9, 10 [23]:

$$ {\text{2MCl}}\left( {{\text{s}},{\text{ l}},{\text{ g}}} \right) \, + {\text{ SO}}_{{2}} \left( {\text{g}} \right) \, + { 1}/{\text{2O}}_{{2}} \left( {\text{g}} \right) \, + {\text{H}}_{{2}} {\text{O}}\left( {\text{g}} \right) \leftrightarrow {\text{M}}_{{2}} {\text{SO}}_{{4}} \left( {{\text{s}},{\text{ l}}} \right) \, + {\text{ 2HCl}}\left( {\text{g}} \right) $$

(9)

$$ {\text{2MCl}}\left( {{\text{s}},{\text{ l}},{\text{ g}}} \right) \, + {\text{ SO}}_{{2}} \left( {\text{g}} \right) \, + {\text{O}}_{{2}} \left( {\text{g}} \right) \leftrightarrow {\text{M}}_{{2}} {\text{SO}}_{{4}} \left( {{\text{s}},{\text{ l}}} \right) \, + {\text{ 2Cl}}_{{2}} \left( {\text{g}} \right) $$

(10)

Different authors have shown how, during combustion, sulphur reduces the formation of KCl in the flue gas by sulphation [49,50,51]. The presence of gaseous SO₃ is of greater importance than that of SO₂ for the sulphation of gaseous KCl. This was confirmed by Kassman et al. [49, 50].

Impact of Industrial Waste Wood combustion on is FeAl coatings

Figure 5c presents equilibrium composition of main species determined for industrial waste wood combustion into Fe-Al system, at a wide temperature range from 300 to 800 ºC. Dominant compounds within the temperature range of 500–700 ºC are extracted in the bar chart from Fig. 5c.

Calcium and silicon are seen as the dominant major ash-forming elements content in industrial waste wood (Fig. 3a), thus CaFe₅O₇ and CaAl₄Si₂O₁₀(OH)₂) represent the main deposits (≥ 10 species %) formed during the combustion of this fuel. Na/Cl-aluminium oxides are also encountered in important amounts (NaAlO₂, AlClO).

In lower percentages (5–10%), aluminium is observed to react with magnesium and potassium components in fuel (MgO*Al₂O₃, KAlO₂, KAl₃(SO₄)₂(OH)₆). In this case, sulphur also participates in, primary, potassium and, secondly, sodium capture (K/Na–Al₃(SO₄)₂(OH)₆)_.

In a minor extent (˂5%), diverse Mg–Al silicates have also been determined (Mg₃5Al₁₈Si₇O₄₄(OH), Mg₇Al₉O_4*Al₉Si₃O₃₆). In this case, silicates and sulphates also react with alkali vapours, thus affecting the alkali chemistry in flue gas.

Predicted species calculated in these conditions may indicate that main corrosion mechanisms in this case are primary related to deposited alkali chlorides, 7, 8 (CaFe₅O₇, NaAlO₂), followed by silication of alkali chlorides, 4–6 (CaAl₄Si₂O₁₀(OH)₂) and direct corrosion by Cl₂(g) (AlClO). Sulphation of alkali chlorides (Eqs. 9–10) is also detected in minor concentrations (KAl₃(SO₄)₂(OH)₆).

Impact of Eucalyptus combustion on isFeAl coatings

Equilibrium composition of main compounds resulting from eucalyptus combustion into Fe-Al system is shown in Fig. 5d. Eucalyptus biomass was seen to contain major concentration for all major ash-forming elements than the rest of the biomasses studied, with significantly high silicon content (Fig. 3a), being silication of alkali chlorides the main corrosion mechanism involved with this fuel.

Main compounds within the temperature range of 500–700 ºC are exposed in the bar graph from Fig. 5d. (CaFe)_0.5SiO₃ is observed as the major compound originated, followed by KAlSi₃O₈ and NaFe(SiO₃)₂, which are also determined in important percentages (10–50%), as result of the high potassium and sodium concentration present in this biomass.

Other calcium silicates are detected in lower concentrations (1–5%), resulting from interactions with both Al and Fe (Ca₂Al₄Si₁₄O₃₆*14H₂O, CaFe(SiO₃)₂).

Thereby, in eucalyptus case, silica reactions are the governing factor affecting the chemistry of other ash-forming elements (K, Na, and Ca), which indirectly affects the deposition and corrosion properties of fly ash.

Conclusions of biomass influence on FeAl-based coating corrosion

When a FeAl-based coating is introduced in the hypothetical combustion reactor, different conclusions are drawn according to the biomass fed. Table 4 shows main compounds formed through Fe-Al/ biomass ashes interactions.

Table 4 Main species formed through Fe-Al/ biomass ashes interactions, at T = 500–700ºC, and corrosion mechanisms involved

In the case of wheat straw, both calcium and silicon present in the biomass tend to interact with aluminium to form Ca–Al–Si hydroxides, as dominant compounds. Potassium also exerts a high influence in the formation of most deposits predicted, with KFe₃(FeSi₃O₁₀)(OH)₂ as the second main compound (20% of total species) and diverse K/Na-Al-silicates in lower concentrations, which may reduce alkali vapours in the gas phase.

In forestry residues case, main forms originated in the process (≥ 10%) are also result of calcium interactions, but in this case, with both Al and Fe (CaAl₄Si₂O₁₀(OH)₂, CaFe₅O₇). Silicon is also present in significant contents in this fuel, promoting Ca/Mg–aluminium silicates formation.

On contrast, when eucalyptus is employed, iron is involved in main reactions, giving rise to diverse Ca/Na–Fe silicates. In addition, KAlSi₃O₈ is detected as the secondary dominant compound.

In industrial waste wood case, equal order of concentrations are determined for the different ash-forming elements present in this fuel, giving rise to similar interactions towards both iron and aluminium through diverse oxides, silicates, aluminates and sulphates compounds.

With respect to corrosion mechanisms towards this coating, silication of alkali and deposited alkali chlorides are dominant, followed by minor reactions of sulphation and oxidation by Cl₂(g).

Thermodynamic equilibrium modelling for isNiAl coating

Impact of wheat straw combustion on isNiAl coatings

Main chemical forms originated from isNiAl interactions with wheat straw ash-forming elements were determined. Figure 6a shows equilibrium composition of main species determined at a temperature range of 300–800 ºC. The bar chart in Fig. 6a shows Ni/Al-ash compounds predicted to be originated at 500–700ºC.

Fig. 6

Equilibrium composition and temperature influence of main species resulting from isNiAl–ash interactions during combustion of: a WS; b FR; c IWW; and d EW

When wheat straw composition is introduced in the combustion process, both aluminium and nickel are involved in main interactions, with silication of alkali chlorides (Eqs. 4–6) as the primary corrosion mechanism involved with this biomass. KAl₂Si₃O₈ is observed as the main specie predicted from metal interactions with wheat straw combustion ashes, which may be related by the important potassium and silicon concentration present in this fuel. The concentration of this compound (KAl₂Si₃O₈) is seen to be enhanced with potassium content in fuel.

2NiO*SiO₂ is the secondary compound formed in significant amounts. For this fuel, silicon is also observed as an important agent involved in the formation of main chemical species (KAl₂Si₃O₈, 2NiO*SiO₂ and NaAlSi₃O₈).

With regard to temperature effect, only minor compounds formed (under 1% of total species) are seen to be affected by this factor.

Impact of forestry residues combustion on isNiAl coatings

Chemical equilibrium compounds derived from forestry residues combustion is shown in Fig. 6b.

Percentages of main Ni/Al-ash species formed at different temperatures are plotted in the bar graph from Fig. 6b. In this scenery, there are few interactions with Ni materials, with only one compound predicted to be formed (NiO*Al₂O₃) in percentages around 10% of total compounds formed.

Dominant corrosion mechanisms in this case are related with silication processes (CaAl₄Si₂O₁₀(OH)₂) followed by deposited alkali chlorides (NaAlO₂). In this case, potassium seems to be removed from the combustion atmosphere by sulphation (KAl₃(SO₄)₂(OH)₆).

As in the previous case, minor compounds are more influenced by temperature than main forms.

Impact of industrial waste wood combustion on isNiAl coatings

Main chemical species produced from isNiAl reactions with waste wood ash-forming elements were determined. Figure 6c shows equilibrium composition of main species originated from 300 to 800 ºC.

In contrast to previous fuels, when waste wood is employed as fuel, high oxidation occurs towards nickel, as NiO (more than 45% of total compounds formed), although a significant number of interactions with aluminium are also observed, with CaAl₂Si₃O₁₀(OH)₂ as the dominant form, which is linked to silication reactions.

Nickel oxidation is seen to be slightly increased (around 4%) with temperature, but this factor exerts a major influence in the formation of other species which appear in lower concentrations (below 10%).

The bar chart in Fig. 6c shows main predicted compounds at selected temperatures. Major aluminium interactions are observed to be related with the presence of Ca, Si, K and Na. On the other hand, water content is seen to influence the formation of calcic species.

Impact of Eucalyptus combustion on isNiAl coatings

Equilibrium composition obtained, when eucalyptus composition is introduced in the isNiAl system, is charted in Fig. 6d. In this case, ash-forming elements primarily interact with aluminium, and main species formed are influenced by silicon presence. When this fuel is employed, corrosion is mainly associated with silication of alkali chlorides (4–6).

The bar graph in Fig. 6d presents Ni/Al-ash compounds predicted to be originated at selected temperatures. Dominant forms originated in the system are primary KAlSi₃O₈, (> 70%) and secondly, NaAlSi₃O₈ (12%) and Ca₂Al₄Si₁₄O₃₆*14H₂O (9%). The latest compound is highly influenced by temperature, being sharply reduced above 360ºC. Main nickel compound (2NiO*SiO₂) is produced in percentages around 7–8%, and is stable in the whole thermal range studied. The rest of predicted compounds are formed in concentrations under 1%.

Conclusions of biomass influence on NiAl-based coating corrosion

In summary, Sect. 3.2 indicates that the biomass type introduced in the system also influences in a high extent the results obtained for NiAl-based coatings.

In this case, biomass ashes are seen to react, in a major extent, with aluminium, giving rise to diverse compounds which are detailed in Table 5.

Table 5 Main species formed through Ni–Al/ biomass ashes interactions

Main differences are detected when industrial waste wood is employed. The composition of this fuel showed lower contents for most major ash-forming elements than the rest biomasses studied. The absence of these elements in the combustion atmosphere makes decrease the potential aluminium interactions, determined for other fuels. Lower silication and sulphation reactions are produced when this biomass is burned, which reduces alkali capture, increasing potential corrosion. For this reason, in this case, Ni oxides are the dominant predicted.

With regard to corrosion, silication of alkali chlorides is observed as the dominant mechanism involved, followed by deposited alkali chlorides and minor reactions of sulphation and oxidation by Cl₂(g).

Thermodynamic equilibrium modelling for isSiCrAl coating

Impact of wheat straw combustion on isSiCrAl coatings

Main chemical species formed from isSiCrAl interactions with wheat straw ashes were determined. Figure 7a exhibits the equilibrium chemical composition determined at the selected temperature range. The bar graph in Fig. 7a shows SiCr/Al-ash compounds predicted to be originated at selected temperatures. Major aluminium species detected (≥ 10%) are K-Al oxides and hydroxides (KAlO₂, KAl₂(AlSi₃O₁₀)(OH)₂, KAl₃(AlSi₃O₁₀)(OH)₂, which may be related by the high potassium content in this fuel. These chemical forms come from mechanisms of both silication of alkali chlorides and reactions with deposited alkali chlorides. In these conditions, equilibrium calculations exhibit how oxides and hydroxides formation is strongly affected by temperature, above 600 °C. On the other hand, hydroxides formation seems to be related with the water content in the biomass.

Fig. 7

Equilibrium composition and temperature influence of main species resulting from isSiCrAlAl–ash interactions during combustion of: a WS; b FR; c IWW; and d EW

Corrosion is also associated with Cl₂ present in the gas phase, giving rise to AlClO, as important percentages (˃ 10%).

In percentages of 5–10%, Cr₂O₃ and KAl₃(SO₄)₂(OH)₆ are observed as secondary forms produced with this fuel. With respect to the former, alloyed chromium is known to improve steel by forming a protective oxide layer consisting of Cr₂O₃ if the chromium content of the steel is sufficiently high. However, previous studies have shown how the protective chromium oxide may be destroyed by reaction with alkali chlorides (11, 12) and carbonates (13) with alkali chromate as the intermediate [48, 52]:

$$ {\text{2Cr}}_{{2}} {\text{O}}_{{3}} \left( {\text{s}} \right) \, + {\text{ 8MCl}}\left( {{\text{s}},{\text{ l}},{\text{ g}}} \right) \, + {\text{ 5O}}_{{2}} \left( {\text{g}} \right) \to {\text{4K}}_{{2}} {\text{CrO}}_{{4}} \left( {\text{s}} \right) \, + {\text{ 4Cl}}_{{2}} \left( {\text{g}} \right) $$

(11)

$$ {\text{2Cr}}_{{2}} {\text{O}}_{{3}} \left( {\text{s}} \right) \, + {\text{ 8MCl}}\left( {{\text{s}},{\text{ l}},{\text{ g}}} \right) \, + {\text{ 4H}}_{{2}} {\text{O}}\left( {\text{g}} \right) \, + {\text{ 3O}}_{{2}} \left( {\text{g}} \right) \to {\text{4K}}_{{2}} {\text{CrO}}_{{4}} \left( {\text{s}} \right) \, + {\text{ 8HCl}}\left( {\text{g}} \right) $$

(12)

$$ {\text{2Cr}}_{{2}} {\text{O}}_{{3}} \left( {\text{s}} \right) \, + {\text{ 4M}}_{{2}} {\text{CO}}_{{3}} \left( {\text{s}} \right) \, + {\text{ 3O}}_{{2}} \left( {\text{g}} \right) \to {\text{4K}}_{{2}} {\text{CrO}}_{{4}} \left( {\text{s}} \right) \, + {\text{ 4CO}}_{{2}} \left( {\text{g}} \right) $$

(13)

Other species such as CrO₂ and NaAlO₂ are also predicted to be originated in minor percentages (2–3%).

Impact of forestry residues combustion on isSiCrAl coatings

Main chemical compounds resulting from isSiCrAl interactions with forestry residues ashes were determined. Figure 7b shows equilibrium composition of main predicted species. Percentages obtained for each predicted compound at selected temperatures are shown in the bar chart from Fig. 7b. Silication reactions are seen as the dominant mechanisms leading to the major compounds formed in this case.

In this scenery, mainly deposits of, primary, SiO₂ and, secondly, CaAl₄Si₂O₁₀(OH)₂ are seen to be formed. Fast reactions are observed to occur at temperatures below 350 °C, with a sharp decrease in the concentration of diverse compounds (SiO₂(l), Al, CrO₂, CaAl₄Si₂O₁₀(OH)₂ and NaAl₂(AlSi₃O₁₀)(OH)₂), with temperature increase.

In minor percentages, an important number of alkali-Al silicates are detected, together with CrO₂, CrO_3, AlClO and different Al-sulphates (Al₂(SO₄)₃, Al₂(SO₄)₃*6H₂O). Chromium oxidation is detected at lower percentages as Cr₂O₃ and at temperatures above 420 °C. Al-sulphates reactions are observed at low temperatures (≤ 420 °C). Sulphur content in this fuel (Table 1) was determined to be quite low in comparison with the rest of the biomasses studied, which may have reduced the concentration of condensed sulphates formed. Ca, Na and water content are seen to favour silicates hydroxides formation at low T ≤ 300 °C.

Impact of industrial waste wood combustion on isSiCrAl coatings

Main condensed species formed from isSiCrAl interactions with industrial waste wood ashes were determined. Evolution of these compounds with temperature is shown in Fig. 7c.

In this case, CaAl₄Si₂O₁₀(OH)₂ is predicted as the dominant compound formed, which is stable with temperature increase. The second main effect observed in these conditions is chromium oxidation as Cr₂O₃, which would give rise to a protective oxide layer.

Other oxidized forms (CrO₂, NaAlO₂, AlClO, MgO*Al₂O₃) are predicted to appear in percentages over 5% at temperatures below 300 °C.

In minor percentages (˂ 5%), alkali sulphates (KAl₃(SO₄)₂(OH)₆), alkali carbonates (NaAlCO₃(OH)₂) and Ca-Al silicates (CaAl₂SiO₆) are also detected to be formed, which would reduce the concentration of available alkali, reducing corrosion risk.

Percentages obtained for each predicted compound at selected temperatures are shown in the bar graph from Fig. 7c, with CaAl₄Si₂O₁₀(OH)₂ as the main specie, followed by diverse aluminium interactions with potassium and calcium in form of oxides, sulphates and silicates (KAlO₂, KAl₃(SO₄)₂(OH)₆, CaAl₂SiO₆).

Diverse types of corrosion mechanisms are detected for this fuel, as it was also observed for wheat straw, which are primary related to by silication of alkali chlorides, Eqs. 4–6 (CaAl₄Si₂O₁₀(OH)₂), deposited alkali chlorides, Eqs. 7–8 (NaAlO₂) and direct corrosion by Cl₂(g) (AlClO). Sulphation, Eqs. 9–10 (KAl₃(SO₄)₂(OH)₆) of alkali chlorides, is also detected, but in minor concentrations (1–5% species).

Impact of Eucalyptus combustion on isSiCrAl coatings

Finally, eucalyptus composition was introduced in the isSiCrAl system. Figure 7d plots equilibrium diagrams calculated for main predicted species. The bar chart from Fig. 7d shows the percentages obtained for each compound at selected temperature range (500–700ºC).

Results obtained for dominant compounds in the equilibrium system show silicon oxidation, as oxides (87% species), as the main reaction involved.

In a lesser extent (≤ 2%), different compounds such as NaAlSi₃O₈, Mg₂Al₃(AlSi₅O₁₈)H₂O and KFe₃(FeSi₃O₁₀)(OH)₂ are also observed. Thereby silication of alkali chlorides is produced in much lower percentages than it was observed for other biomasses, so corrosion associated with this mechanism is not relevant in this case.

On the other hand, the formation of the protective chromium oxide (Cr₂O₃) is also detected, in minor percentages (0.5%).

Conclusions of biomass influence on SiCrAl-based coating corrosion

When isSiCrAl is introduced in the system, results obtained differ depending on the biomass used. Table 6 shows main interactions resulting from biomasses ash components with Si-Cr-Al-based coatings.

Table 6 Main species formed through SiCrAl/ biomass ashes interactions

When either forestry residues or eucalyptus are employed, silica and Ca-Al silicates are observed as the dominant forms, while alkali capture by silicates/sulphates formation occurs at much lower percentages than in the other biomasses. In addition, chromium oxidation as Cr₂O₃, which would involve the formation of a protective layer, is also produced in a minor extent.

On contrast, when wheat straw or industrial waste wood is introduced in the system, an important amount of aluminium interactions with K, Na and Ca are observed, as oxides, silicates and, in a minor extent, of sulphates, which would involve a lower concentration of alkali in the gas phase, thus reducing corrosion risk. On the other hand, for these fuels, chromium oxides (Cr₂O₃, CrO₂) are detected in significant percentages (˃5% species), mainly in the case of industrial waste wood.

With respect to corrosion mechanisms, silication reactions are dominant for all biomasses studied, while in wheat straw and industrial waste wood cases, mechanisms associated with solid alkali chlorides and direct oxidation by Cl₂(g) are also relevant. Chromium oxidation as Cr₂O₃ is mainly detected for industrial waste wood case and, secondly, for wheat straw, while for the rest of the biomasses this effect is negligible.

Comparison of results obtained through the global matrix of the modelling tests

By comparing the results obtained in the diverse modelling tests (Fig. 2), where different alloys were exposed to varied combustion atmospheres burning diverse biomass types, a definition of the most corrosive biomass fuels over a specific coating has been reached and schematized in Fig. 8. The corrosion mechanisms associated with each biomass, as well as the number of metals–ash compounds interactions determined through the model, were also included in this figure.

Fig. 8

Most damaging biomass fuels determined for each coating

In the case of isFeAl, major potential corrosion was observed when, primary, IWW or, secondly, FR were used as fuels in the combustion process, which is reflected by the significantly higher amount of interactions between the coating metal composition and the atmosphere chemical compounds when these biomasses are burned. The elevated volatiles content determined in these fuels might have led to a major release of ash components to the gas phase, available to later condense in the metals surfaces. The low sulphur concentration found in IWW samples also reduces ash-forming elements capture in the bottom ash thus increasing volatilization. On contrast, FR contains high sulphur content but this is counteracted by the important content of ash-forming elements. Ca-Al-Si hydroxides were seen as the main compounds formed in most sceneries tested for this coating.

With respect to isNiAl, main number of interactions was seen to be primary produced in the corrosive FR atmosphere, and secondly detected during EW combustion. EW also contains an elevated volatiles percentage and very important major elements content. However, the ash percentage in this fuel was determined in low values. In addition, results summarized in Table 5 for this coating, reflected how in IWW tests, isNiAl is affected by nickel oxidation reactions, while silication and sulphation reactions prevailed for the rest of the biomasses.

In the case of isSiCrAl, when either wheat straw or industrial waste wood compositions were introduced, significant number of reactions were observed related to diverse corrosion mechanisms (associated with Cl₂(g), with solid alkali chlorides and with silication processes).

In this sense, results reveal that the protective character of a specific alloy is highly depending upon the biomass burned. Figure 8 shows IWW and FR as highly damaging in most cases, which is in agreement with conclusions from fuels indexes calculations.

Predicting metal interactions, based on real combustion conditions and fuel composition, was seen to be useful to identify key factors for protective coatings design when employing new waste fuels. However, results obtained through this modelling require of field verification. In this sense, further research will be focused on coatings testing in real conditions through the insertion of coated coupons in the BFBC plant and the development of different combustion experiments operating with biomass fuels evaluated in this study.

Conclusions

A modelling of chemical interactions produced between the metals composing different coatings and the chemicals compounds released during the combustion of diverse biomass types was developed as a corrosion risk prediction strategy in biomass boilers.

Previous characterization of the biomass fuels showed great variability, with higher values obtained for both ash content and trace metals concentrations in FR case.

Fuels indexes calculations, based on ash chemical composition, have revealed that major hazard may be caused if IWW, in a major extent, or FR were fed in a combustion process, showing high risk of salt stickiness, deposits formation and release of free corrosive acidic gases.

Results obtained through the thermodynamic equilibrium model were in in agreement with conclusions from fuels indexes calculations, showing, primarily, IWW and secondarily, FR as highly damaging for most coatings studied. The significant ash content determined in FR samples together with the elevated volatiles content determined in both IWW and FR biomasses, might have led to a major release of ash components to the gas phase, available to later condense in the metals surfaces.

An increase in alloys corrosion resistance was detected through the model when nickel and/ or chromium was present, showing isSiCrAl as the most resistant followed by isNiAl. However, results showed how even highly alloyed materials could be severely corroded depending on the biomass composition, thus materials selection should be carried out regarding the compatibility with the environment conditions for its final use.

The model also provided information on the corrosion mechanisms occurring in each scenario, revealing both silication of alkali and deposited alkali chlorides as dominant processes, followed by minor reactions of sulphation and direct oxidation by Cl₂(g).

Predicting metal interactions, based on real combustion conditions and fuel composition, was seen as a useful tool to identify key factors for protective coatings design when employing new waste fuels.