Renewable Energy 128 (2018) 446e459

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Mineral phase transformation of biomass ashes e Experimental and

thermochemical calculations
Aneta Magdziarz a, *, Marcin Gajek a, Dorota Nowak-Wo

zny b, Małgorzata Wilk a
a
AGH University of Science and Technology, Mickiewicza 30 Av., 30 -056 Krakow, Poland
b
Wrocław University of Technology, 27 Wybrzez_ e Wyspian

skiego St., 50-370 Wrocław, Poland

a r t i c l e i n f o a b s t r a c t

Article history: The paper is focused on the thermal characterisation of biomass ashes especially of mineral phase
Received 23 January 2017 transformation. Ash generated during combustion (with high alkali metal concentration) can deposit and
Received in revised form consequently cause serious operating problems such as slagging, fouling and corrosion of metal surface
6 March 2017
limiting heat transfer. Four biomass ashes with different origin were investigated. The chemical
Accepted 18 May 2017
Available online 20 May 2017
composition of the mineral matter of ashes varied between the samples, but CaO, SiO2, K2O, MgO, P2O5
and Al2O3 are the main compounds of the ashes. The thermal behaviour of ashes was studied using TG-
DSC and Ash Fusion Temperatures (AFTs) techniques. In order to describe and understand the mineral
Keywords:
Biomass ash
matter transformation at high temperatures the FactSage Thermodynamics Model was used. FactSage
TGA calculations allowed to predict chemical composition in equilibrium, showed amount of liquid slag and
FactSage solid phases, and gave information about slagging properties of ashes. A corrosion study of the metal
Steel corrosion surface used for heat transfer biomass ashes was undertaken. The corrosion behaviour of steel
X10CrMoVNb9-1 in the present of biomass ash at 650
C in 1000 h in furnace. The degradation of the
metal surface was investigated by observation of surface morphology using the SEM method. The
presence of potassium and chlorine in deposits causes the acceleration of oxidation. They break and peel
off the surface easily, thus evidently accelerating corrosion.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction the fuel causes a serious problem in the combustion process of

Recently, fossil fuels have been the basis for the production of
energy. The use of biomass as a fuel has been increasing. The
development of processing technologies and energy use of
renewable fuels is taking place. The number of power plants using
biomass is on the increase, too. The thermal conversion techniques
of biomass are varied such as combustion, co-combustion, gasifi-
cation and pyrolysis. Combustion and co-combustion go in pulv-
erised coal boilers, fluidised bed boilers, and a grate furnace, whilst
further technologies are still developing. Biomass fuel is defined as
a green and zero CO2 e a natural energy source. The variety of
biomass is very wide; there are those that are wood-based (e.g.
forest and landscaping residues), agricultural (straw, grass, food
industry) and waste derived. Although biomass and sewage sludge
combustion processes are applied with good efficiency, ash deposit
is a new problem, which requires new solutions. The mineral part of
* Corresponding author.
E-mail address: amagdzia@agh.edu.pl (A. Magdziarz).
biomass [1e3] (Fig. 1).
The inorganic part of the fuel accelerates the risks associated
with the formation of ash deposits on the heating surface of the
boiler. The most important phenomena affecting the work condi-
tions of biomass and sewage sludge combustion devices are slag-
ging, bed agglomeration, fouling, and corrosion in the combustion
devices, which degrade their performance and severely damage the
firing equipment [4e9]. The deposit of ash on the heat transfer
surfaces of the boiler depends on the chemical composition of the
fuel and the combustion process parameters. The presence of ash
reduces the heat transfer and combustion efficiency, and damages
combustion chambers when large particles break off. The term
slagging is used to describe the formation and accumulation of
slags on the furnace sections (refractory walls, water walls and
grates). The structural forms of slags and bed agglomerations can
vary [10]. Their reactivity depends on the chemical reaction be-
tween ash components and other inorganic compounds present in
the environment. The term fouling is used to indicate the formation
of ash deposits on heat transfer surfaces in the convective parts of

http://dx.doi.org/10.1016/j.renene.2017.05.057
0960-1481/© 2017 Elsevier Ltd. All rights reserved.
 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459
Fig. 1. The scheme of biomass ash deposit problems.
the furnace. It causes insulation of the convection tubes and re-
duces heat transfer. Different correlations and methodologies have
been used for predicting the slagging and fouling tendencies of
solid fuel ashes. There are some empirical indices correlating the
standardized chemical composition of ashes, thus making it
possible to predict the ash behaviour and deposition tendencies.
The works of Pronobis [11,12] present empirical indices for the co-
combustion of coal and biomass from indices concerning coal
combustion.
The main elements in the biomass ash are calcium, silicon,
magnesium and potassium as well as sulphur, chlorine, and phos-
phorus. However, the biomass source determines the chemical
composition of ash, meaning that the generalisation of ash contents
and their composition is not possible and variable. The alkali metals
play an essential role in biomass combustion and next in slagging,
fouling and corrosion processes. The potassium content is impor-
tant because it indicates potential ash fusion and deposition by
vaporization and condensation. In biomass combustion K and Cl
can be released in gas such as HCl, KCl and potassium exists as
potassium silicate, aluminosilicate and sulphate (K2Ca(SO4)2,
K3Na(SO4)2). Depending on the temperature potassium can exist in
different forms; K2SO4 condenses on high temperature heating
surfaces, whereas KCl condenses on lower ones [1]. KCl is the most
stable gas-phase alkali metal that influences biomass slagging. It
should be emphasized that all forms of potassium play a significant
role in slagging and fouling. This was reported in experimental and
numerical calculations [2,13]. Additionally, in studies [10,14]
K2Ca(SO4)2 and K3Na(SO4)2 significant phases in slagging were
found.
The chemical and mineral compositions strongly influence on
the ash melting and fusion temperatures. Thermal techniques are
well known to study the characteristics of biomass and other fuels
[15,16] during the pyrolysis, gasification and combustion processes.
The thermal behaviour of ash can be successfully studied by ther-
mal analysis (TGAs: TG and DSC) connected with mass spectrom-
etry (MS), as well, to obtain information on ash melting, phase
transformation and release of gaseous phase [17,18]. TG and DSC
curves show the mass change via temperature and thermic effects
(endo-, exothermic) associated with the change of enthalpy.
Determination of ash fusion temperatures is important to study ash
deposit problems. Therefore, four temperatures are identified in
describing the ash melting process: initial deformation tempera-
ture (IDT), softening temperature (ST), hemispherical temperature
(HT) and flow temperature (FT) [19,20].
Generally, corrosion is the phenomenon of the destruction of
materials under the influence of the environment. The reaction of
the corrosion process is described in the following equation:
M(s) þ a/2 X2(g) 4 MXa (s,g)
where X can be: O2, Cl2, S.
Taking into account the corrosion which appears in the com-
bustion process, the most important is the degradation of the
447
metallic material of the boiler as well as the protective oxide scales
which take place under combustion and its associated products
(solid, liquid and gas) [21]. A high temperature corrosion process
influences negatively on the efficiency of the energy systems during
renewable fuels combustion. The mechanism of corrosion depends
on interactions between ash elements as well as the conditions of
the combustion process. The most aggressive agents existing in
biomass ashes are alkali metal compounds, especially potassium
and sodium. The reaction between the compounds in the ash lead
to products which can cause operating problems during combus-
tion. Detail description of this process have been widely studied in
the papers [22e25]. Some possible reactions (2e8) are listed
below:
2KCl(g) þ SO2(g) þ 1/2O2(g) þ H2O(g) 4 K2SO4(g) þ 2HCl(g) (2)
K2CO3(g) þ SO2(g) þ 1/2O2(g) 4 K2SO4(g) þ CO2(g) (3)
KOH(g) þ HCl(g) 4 KCl(g) þ H2O(g) (4)
K2SO4(g) þ SiO2(s,liq.) / K2O$SiO2(s,liq.) þ SO3(g) (5)
K2SO4(g) þ SiO2(s, / K2O$SiO2(s,liq.) þ SO3(g) (6)
liq.)
2KCl(g) þ (2Al2O3$SiO2(s,liq.)) þ H2O(g) / 2KAlSiO4(s,liq.) þ HCl(g) (7)
2KCl(s,liq) þ SO2(g) þ 1/2O2(g) þ H2O(g) 4 K2SO4(s,liq.) þ 2HCl(g) (8)
High-temperature heating surface corrosion during the com-
bustion of renewable fuels is due to chlorine deposition and re-
actions between chlorine and metal [26,27]. That is why the
chemical composition of metallic material plays a major role in
corrosion during combustion. In this context, stainless steel with a
high amount of chromium is advised. Chromium steel is charac-
terized by its high resistance of corrosion. Besides the chromium
other additions such as Ni, Nb, Ta and Mo are added to improve the
corrosion resistance [28,29].
The knowledge concerning the formation and transformation of
the mineral part of biomass ash is necessary to control problems
associated with fouling and slagging in boilers during biomass
combustion. In order to provide a deeper description of biomass
ash's thermal behaviour, TGA analysis was conducted, ash fusion
temperatures were examined under oxidizing atmosphere condi-
tions, and additionally the mineral phases were determined by
thermochemical calculations (FactSage). Additionally, a corrosion
study of the metal surface used for heat transfer under renewable
fuel ashes was undertaken. The work presents the character of
biomass ashes and the study of metallic surface morphology after
ashes influence. The tested steel was the type which is being
considered for use in super-heaters of stream boilers fired for
renewable fuel combustion.
2. Experimental
2.1. Material
Four different types of biomass ashes with different origin (three
wood and straw) were used in this study. The biomass materials are
typical renewable fuels used in energy units in the combustion
process. The ash samples have reflected biomass ashes in general.
(1) The ash samples were denoted as: OS e oat straw (agriculture
biomass) ash, BS e beech sawdust ash, LS - larch sawdust ash and
WS e wood sawdust ash (mixture of different kind of wood
biomass), respectively. The results and knowledge obtained in this
study can provide a better understanding of ash formation
448 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459

mechanisms and interactions between the ingredients to success-

fully predict and avoid slagging and fouling of ashes on boiler
surfaces.
The corrosion process was studied using steel type T91
(X10CrMoVNb9-1). This is a ferritic-martensitic (9% chromium, 1%
molybdenum) steel micro-alloyed with vanadium and niobium. A
higher chromium content should offer better corrosion and
oxidation resistance. This grade of steel is mainly used for high-
temperature applications in superheater and reheated tubes,
headers and steam piping. Table 1 presents the composition of the
studied steel.

2.2. Methods

The description of mineral phase transformation of biomass

ashes required using advanced instrumental methods as well as
thermochemical equilibrium calculations. Fig. 2 presents the
scheme of investigation work plan.
The main chemical composition of biomass ashes was deter-
mined by an X-ray fluorescence spectrometer (Riau ZSX Primus II).
The content of the major inorganic elements ashes in the form of Fig. 2. Scheme of investigation work plan.
oxides is given in Table 2.
The XRD (X-Ray Diffraction) analysis was carried out to identify
Table 2
the crystalline compounds present in the studied ashes using
The main elements of the ashes in the form of oxides, % (XRF method).
PANanalytical EMPYREAN DY 1061 with CuKa radiation in Bragg-
Brentano geometry. The operating conditions were 40 kV and % oxide in ash OS BS LS WS
40 mA CuKa (l ¼ 1.54 Å) radiation and step-scanned in the range SiO2 18.8 46.10 20.40 9.80
2Q ¼ 10e90
, step 0.06
, time 10 s. For the phase identification data Al2O3 1.13 14.4 4.46 2.51
Fe2O3 0.38 7.49 36.8 1.34
base PDF-4þ product of ICDD was used.
TiO2 e e 0.59 0.08
The TG/DSC technique (Thermogravimetry and Differential CaO 10.20 11.9 22.00 45.90
Scanning Calorimetry) was conducted using Netzsch STA 449 F3 MgO 2.73 4.68 4.28 10.70
Jupiter. The samples were heated in alumina crucibles from K2O 46.90 7.80 3.99 13.00
ambient temperature to 1200
C at a constant rate of 10
C/min in a Na2O 0.97 0.99 0.48 0.71
MnO 1.09 0.77 3.18
40 ml/min flow of air. The evolved gaseous products from de-
e
P2O5 14.50 3.62 2.93 10.00
compositions were identified on-line using a quadruple mass SO3 3.43 1.96 3.23 2.79
spectrometer (QMS Netzsch). The following ions which are char- Cl 0.89 e 0.04 e
acteristic of the molecules of interest were monitored: 18, 44, and RB 61.18 32.86 31.4 71.65
RB/A 3.07 0.54 1.29 5.78
64, H2O, CO2 and SO2, respectively. The combustion characteristics
RB/AþP 3.78 0.60 1.35 6.59
of biomass and its thermal behaviour have been studied using SR 58.55 65.69 24.43 14.47
thermal techniques as demonstrated [15,16]. In this paper, thermal RS 0.27 0 0.06 0
analysis was used to characterize the behaviour of ash samples due Fu 50.94 4.7 5.76 79.24
to study complex phase transformations related to the decompo- Fe2O3/CaO 0.03 0.63 1.67 0.03
sition of inorganic compounds. Additionally, these techniques was
used to obtain the information on ash melting temperatures and
gas-phase release. TGA advantages include a rapid assessment of characteristic temperatures are determined according to the
the fuel value, the temperatures where combustion starts and ends, change in the shape and size of the pellets. IDT is the temperature
the maximum reactivity temperature, the amount of ash and total where the first rounding of the apex of the cone occurs. ST is the
combustion time. temperature where the cone has fused down to a spherical lump
In order to know whether ash deposit problems could exist and the height becomes equal to the width of the base. At HT the
within the range corresponding to the combustion fusion temper- height becomes one-half of the width of the base. FT is the tem-
atures, the determination of ash fusion temperatures (AFTs) is perature where the fused mass has spread out in the nearly flat
necessary. Four temperatures are noted in describing the ash layer with a maximum height of 1/16 inch. The fusibility analysis
melting process: initial deformation temperature (IDT), softening was performed by a LECO AF700 Ash Fusion Determinator. Pre-
temperature (ST), hemispherical temperature (HT) and flow tem- prepared ash cones were mounted on a ceramic tray and placed
perature (FT) [18,19]. An ash fusion temperature experiment can be into a high-temperature and rampable furnace. The tests were done
performed in an oxidizing or a reducing environment. The specially under air and reducing atmospheres of up to about 1500
C. The
prepared pellets (pyramids) are heated up to 1500
C, and the heating process of the ash cones was observed and automatically

Table 1
Chemical composition of X10CrMoVNb9-1 steel.

Elements content, %

C Si Mn P S Cr Mo Ni Al N Nb V
0.1 0.33 0.41 0.01 0.003 8.66 0.94 0.27 0.011 0.049 0.076 0.02
 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459
Table 3
Crystalline compounds identified in studied ashes (XRD method).
Ash Crystalline compounds
OS KCl, KCaPO4, K2O, K3PO4, K3Fe(PO4)2, Ca2SiO4
LS Fe2O3, SiO2, CaCO3
WS MgO, Ca2SiO4, Ca3(PO4)2, Mg2P2O7, CaCO3, CaO, KCl, CaMnO3, SiO2
monitored (video camera) to determine deformation temperatures
(DT, ST, HT, FT).
The corrosion tests were done in the furnace, where metallic
material was covered with biomass ash. Each specimen was
exposed to the used deposit at a temperature 600
C for 1000 h.
After the exposure, the specimens were removed from the furnace,
cooled to an ambient temperature, and placed into a desiccator. As
a reference, a specimen without any deposits was heated in the
same conditions. Depiction of the surfaces before and after the
heating process are presented in Fig. 14. Next, corrosion was studied
by scanning electron microscopy (SEM) method in order to study
the degradation of the metallic surface. Micro- and nano-structural
analyses were conducted using a high resolution scanning electron
microscope (Merlin Gemini II of ZEISS), equipped with the FEG
electron gun and EDX detector with the Quantax 800 microanalysis
system of Bruker.
3. Results and discussion
3.1. Characterisation of biomass ashes mineral matter
Ash is the inorganic incombustible part of the fuel which is left
after combustion process. The ash content varies between the types
Fig. 3. X-ray patterns of biomass ashes.
449
of biomass and influences on ash melting temperatures, and
consequently ash agglomeration and corrosion. The chemical
composition of the mineral matter of biomass ashes is presented in
the oxide form in Table 2.
The concentration of the main oxides was different, because of
their different origin, but the main ash elements were SiO2, CaO,
Fe2O3, Al2O3, K2O, P2O5, and MgO, Na2O. The SiO2 and Al2O3 content
changed in a wide range and, accordingly the ratio SiO2 and Al2O3
varied from 3.2 to 16.63, for beech sawdust and oat straw, respec-
tively. The obtained XRF results confirmed that oat straw ash
(agriculture biomass) had characterized the highest alkali oxide
content (mainly potassium, calcium, magnesium and sodium).
Similar tendency of chemical composition of ash was reported by
Refs. [19,23,30], but direct comparison is difficult to do due to high
variety of biomass and its ash. Studied wood biomass ashes con-
tained high amount of alkali metals, particularly potassium, but the
concentrations were less than in oat straw ash. Alkali compounds
forming low-melting eutectics evaporate during combustion pro-
cess and agglomerate in the form of ash deposits on heat exchange
surface of energy unit. Additionally, the chlorine presence accel-
erates the high-temperature corrosion of metal surface [25] induce
by KCl present in combustion environment.
The crystalline mineral species in ash samples were identified
by XRD (Table 3) and are shown in Fig. 3 due to be more intuitive to
observe crystalline compounds. XRD analysis confirmed the exis-
tence of potassium phases, such as KCl, K3Fe(PO4)2, K3PO4.
The presence of high contents of K2O and P2O5 in oat straw ash
has decreased the ash fusion temperatures. The melting of potas-
sium salts is the most dominate factor influencing on the slagging
and agglomeration. Besides, potassium and sodium, with the
presence of phosphorus can reduce the melting points of ash. The
450 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459
Table 4
Characteristic fusion temperatures, AFTs (
C) of biomass ashes.
Temperature OS BS LS
Deformation (DT) 790 1190 1150
Softening (ST) 1140 1210 1190
Hemispherical (HT) 1230 1250 1390
Flow (FT) 1330 1470 1470
phosphorus phases were detected in the form of KCaPO4, K2O,
K3PO4, K3Fe(PO4)2, what can be caused by the fact that phosphorus
is more favourable for potassium than for silicon. Bostrom et al. [31]
confirm that the high amount of phosphorus is a distinctive feature
and they claim that ash-forming matter in agricultural biomass is
difficult to compare to wood biomass ashes. Moreover, phosphorus
phases were identified and discussed in detail by Steenari et al. [6].
The presence of CaO, MgO and Al2O3 may increase the ash melting
points, which was observed for woody biomass ash. The larch
sawdust ash had a very high content of Fe2O3 even visible in the
reddish-brown colour of ash. The hematite can be formed during
the oxidation of organic iron and iron carbonate what takes place in
combustion process [23]:
2FeCO3 þ 0,5 O2 / Fe2O3 þ 2CO2
Despite of significant content of silicon in crystalline form,
Ca2SiO4 and SiO2 were detected. More complex silicates were not
identified. Amorphous silicates with low melting temperatures can
be form in such processes, but using XRD method is not possible to
identified them. As well as aluminosilicates were difficult to detect.
Aluminosilicates have vary and variable composition. During the
combustion process the minerals undergo thermal reactions and
structural changes what causes that content of particular phases is
small and difficult to analyse.
As can be seen in Table 2 the highest RB and consequently RB/A
has wood ash (WS). The base-to-acid ratio describes the fouling
tendency of ashes. Oat straw and wood ashes characterize the high
tendency to sintering of deposits (0.6 < Fu  40 - a high fouling
inclination), whereas higher amounts of SiO2 and Al2O3 in BA and
LA ash decrease the fouling tendency. The slagging index RS
Fig. 4. TG, DSC, H2O (m/z ¼ 18), CO2 (m/z ¼ 44) and SO2 (m/z ¼ 64) evolution profiles of oat straw ash (OS) during the thermal process under air atmosphere.
calculated for ashes suggests a very low slagging inclination for
ashes. The slag viscosity index depends on SiO2 and Fe2O3, the
WS highest value of SR has BA ash. The low melting points eutectics of
1530
SiO2 and Fe2O3 might be a main cause of bed agglomeration. The
>1550 reactions between P, Mg and Ca with Si also enhance deposition of
>1550 ash. The ash composition influences the melting temperatures
>1550 affecting the heat transfer surface (Table 4). The ash deformation
temperatures (DT) verify the differences in character between
ashes. The OS has the lower whereas WS evidently has the highest
characteristic ash fusion temperatures. The presence of potassium,
sodium and the highest content of Cl reduce the melting temper-
ature due to greater ash volatility. The influence of Al2O3 and SiO2
content has been observed for BS and LS, which have relatively
similar AFTs. A very high melting temperature has WS, which was
the consequence of a high CaO. It should be emphasized that not
only chemical composition but also mineral phases influence on
ash fusion temperatures. The analysis of slagging and fouling
indices and ash fusion tests suggests that oat straw ash reveals a
higher tendency for the fouling thus accelerating the corrosion
process due to the existence of chlorine. As demonstrated in
Refs. [19,20,23,32] the agriculture biomass ashes with low con-
centration of Ca and high concentration of K have the tendency to
melt at lower temperatures than wood biomass ashes, which is in
(9) agreement with presented results.
3.2. Thermal analysis
Figs. 4e7 present the TG, DSC and MS curves during heating
the studied ashes under air atmosphere. TG corresponds to mass
change and DSC reflects heat flow. Additionally the emission of
H2O, CO2 and SO2 (MS results) is presented. TG and DSC results
show that thermal conversion of up to 1200
C of the studied
ashes takes place into three stages. The mass changes observed
on the TG curve, heat effects on the DSC curve and kind of
gaseous compounds are different for the studied ashes. The first
stage is observed to 250
C which corresponds to water evapo-
ration (m/z ¼ 18 line). This confirms the presence of a certain
amount of humidity in the ash samples. The highest mass loss of
5% was observed for OS ash, and then, respectively, 1% for BS and
LS, and 0.3% for WS ash. The noticeable endothermic peak at
 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459 451

Fig. 5. TG, DSC, H2O (m/z ¼ 18), CO2 (m/z ¼ 44) and SO2 (m/z ¼ 64) evolution profiles of beech sawdust ash (BS) during thermal process under air atmosphere.

100
C for OS ash reflected the water evaporation. At higher
temperatures, the second stage of thermal conversion was found.
This stage corresponds to the decomposition of carbonates,
mainly calcium carbonate. The evident emission of CO2 (m/z ¼ 40
line) is observed. The mass loss goes in one stage in the case of
OS, BS and LS ashes (5.5% - OS, 1.6% - BS and 3.0% - LS). A small
endothermic peak appears at DSC curves and consequently, at a
high CO2 peak, confirming the decomposition of carbonate. The
different thermal behaviour of ash can be observed for WS ash
(Fig. 7). These differences can be explained by their chemical
composition. A wide, fuzzy exothermic peak between the tem-
perature range of 300
C and 580
C and H2O and CO2 peaks
could result in the combustion of organic residue but also could
be caused by the decomposition of carbonates. The maximum of
temperature peak is 425
C and 1.6% mass loss. The temperature
range of 600e950
C presents two endothermic peaks at a
maximum of 760 and 890
C.
The first one corresponds with calcium carbonate decomposi-
tion and the second one appears because of the release of alkali
chlorides. The presence of KCl in WS was confirmed by XRD anal-
ysis. These effects reflect a mass loss of 7.2% and 2.8%. For BS ash the
third stage is observed. It can be connected with the decomposition
of sulphates, because evident emission of SO2 takes place. The
melting temperature of K2SO4 is 1069
C. For all of the ashes the
DSC curve present endotherm peak related to the appearance of
liquid phase. For OS ash it is observed at the lowest temperature,
c.a. 780
C, which was confirmed by ash fusion tests (DT ¼ 790
C).
The liquid phase of BS ash appears at 1100
C (Fig. 5). DSC analysis
did not prove the liquid phase for WS ash in the studied temper-
ature range (up to 1200
C). ASTs analysis has provided information
about very high ash fusion temperatures for WS ash. Regarding the
thermal analysis, the mass loss curves and exothermic/endo-
thermic peaks of biomass ashes during heating process, can be
attributed to the chemical composition of ash. The same statement

Fig. 6. TG, DSC, H2O (m/z ¼ 18), CO2 (m/z ¼ 44) and SO2 (m/z ¼ 64) evolution profiles of larch sawdust ash (LS) during thermal process under air atmosphere.
452 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459
Fig. 7. TG, DSC, H2O (m/z ¼ 18), CO2 (m/z ¼ 44) and SO2 (m/z ¼ 64) evolution profiles of wood sawdust ash (WS) during thermal process under air atmosphere.
could be found in the works of Vamvuka and Kakaras [23] and Du
et al. [19].
3.3. FactSage thermodynamic calculations
In order to describe and understand the mineral matter trans-
formation at high temperatures the FactSage Thermodynamics
Model was used. FactSage is a currently very popular software for
thermodynamic calculations which applies to problems associated
with renewable fuel ashes deposition [33,34]. FactSage is based on
minimization of the Gibbs energy of system. The transformation
and fusion of ash takes place at high temperatures, and kinetic
limitations, mass transport and chemical potentials have an influ-
ence on the calculated mineral phases. Thermochemical calcula-
tions can indicate chemical composition in equilibrium, show the
amount of liquid slag and solid phases, and give information about
the slagging properties of ashes. The chemical equilibrium was
calculated considering the Gibbs energy of all the phases and
minimizing the total Gibbs energy system. The Gibbs energy
minimization algorithms find the amount of various phases and the
composition of the solution phases. Gibbs energy can be calculated
from what is known of the chemical potential of the component by:
X the BS sample, and at 930
C for the LS sample, respectively. It is
G¼ ni Gi
i
where: the ni amount of the component i. Assuming some con-
straints such as p, T and overall composition, the Gibbs energy
minimization algorithms find the amount of various phases and the
composition of solution phases which give a global minimum in the
total Gibbs energy of the system. The Gibbs energy can be
expressed [33]:
0 1
ZT ZT
Cp ðTÞ A
GðTÞ ¼ DH298 þ Cp ðTÞdT  T D@S298 þ dT
T Based on Fig. 8 characteristic temperatures where the slag phase
298 298
The values of DH298 , S298 and Cp ðTÞ are from thermodynamic
databases.
The description of equilibrium of studied system allows to make
the thermochemical analysis of heterogeneous processes during
ash transformation. In the literature there are some reports con-
cerning thermochemical transformation processes of coal [35,36]
and biomass ashes [33,34,37].
The FactSage modelling program can be used to interpret the
processes taking place during the sintering of ash sample, to predict
the chemical composition at the equilibrium state and to predict
the AFTs. In particular the ‘equilibrium module’ and FToxid, FTmisc,
ELEM and FactPS databases were used. It was assumed that there
was a possibility of two immiscible slag phases. As an input data for
thermodynamic calculations the chemical composition of biomass
ash samples were used there (Table 2).
The thermodynamic calculations were performed at tempera-
ture range from 600 to 1500
C at 1 atm pressure. Fig. 8 shows the
relative amounts of liquid slag formation for beech sawdust (BS),
larch sawdust (LS), oat straw (OS) and wood sawdust (WS) ashes
using FactSage 6.4.
Figs. 9e12 show the mass present in the slag phases (relative to
ash amount) as a function of temperature, for beech sawdust ash
(BS), larch sawdust ash (LS), oat straw ash (OS) and wood sawdust
ash (WS) samples. It can be observed that the slag's liquid phase
starts to form below 600
C for OS and WS samples, at 920
C for
(10) interesting that the amount of the slag phase does not increase
significantly with the temperature for all samples and that the
maximum amount of slag phase is not correlated with the starting
temperature.
For example, for OS and WS samples, the slag liquid phase starts
to form at visible lower temperature than for samples LS and BS but
its maximum mass amount is also visible lower too. It means that
not only the starting temperature of slag formation is important but
likewise the kinetics of slag formation as well. Therefore the dif-
ference in the slag phase of forming processes was shown in the
form of some characteristic temperatures.
(summary of two immiscible phases) starts to form (t0), reaches a
(11) value of 15% wt. (t15), 25% wt.(t25) and 50% wt.(t50) of the total ash
mass can be defined. The lowest t0 and lowest t50 have OS sample
whereas WS has the lowest t0 but the highest t50 (above 1600
C).
 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459 453

Fig. 8. Mass percentage of slag liquid phases at different temperatures for sewage sludge ashes calculated by FactSage 6.4.

Fig. 9. Mineral phases in equilibrium with the slag phase for oat straw ash (OS).

The slag's liquid phases appeared at the highest temperatures for
LS. For the tested biomass samples there is a good correlation be-
tween the DT, FT (Table 4) and t50 (Table 5) temperatures. The re-
sults are shown in Fig. 13.
For the tested samples the dependences between t50 and DT or
FT are linear (12) and (13):
DT ¼ 178 þ 0.84, t50
FT ¼ 1110 þ 0.31, t50
The slag phase formation process depends on the processes in
the solid state that takes place at lower temperatures. Therefore,
FactSage 6.4 was used for the prediction of solid state phase
transitions and reactions. The characteristic temperatures together
with transitions in the solid state are collected in Table 6.
It can be observed that the slag phase formation (kinetics and
formation temperature) and phase transitions at solid state that
occur at lower temperatures, strongly depend on the ash compo-
sition. In beech sawdust ash (BS) there exist a relatively large
amount of potassium aluminum silicates and calcium magnesium
(12) silicates. In larch sawdust ash (LS) there exist a relatively large
amount of calcium iron silicates and calcium magnesium silicates
(13) but for oat straw ash (OS) sample the main components are calcium
phosphate and potassium silicate. Table 7 presents the trans-
formation of sodium and potassium phases during the sintering
process. The FactSage calculation suggests a large amount of cal-
cium phosphate, calcium oxide and calcium silicate and calcium
454 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459

Fig. 10. Mineral phases in equilibrium with the slag phase for beech sawdust ash (BS).

Fig. 11. Mineral phases in equilibrium with the slag phase for larch sawdust ash (LS).

magnesium silicate for wood sawdust ash (WS). Compared with the
experimental results FactSage is a good tool to study ash trans-
formation reactions. However, chemical equilibrium calculations
require proper information such as temperature, residence time,
flue gas velocity, gas conditions and others. Thermodynamic cal-
culations are a very useful technique to describe the biomass
combustion process especially ash transformation.
3.4. Corrosion tests
Corrosion takes place in an environment of reactive gases, and is
commonly known as the oxidation of metals. Taking into account
the combustion process the most important is corrosion caused by
the presence of ash deposits containing aggressive agents such as
alkali metals and chlorine. The acceleration of corrosion by biomass
ash is the result of its thermochemical properties and the in-
teractions between low-melting eutectics and oxide layer. The
problem of metallic material degradation during biomass com-
bustion is still not clearly solved, however, there are some inter-
esting investigations which discuss this problem [29,38,39].
Therefore, the corrosion tests are need to be study to enlarge the
knowledge the impact of biomass ash on metallic surface.
The performed multifaceted analysis of biomass ashes by
determination of chemical and phase compositions, fusion tem-
peratures, thermal behaviour and FactSage calculations allowed to
select the most appropriate ashes for the corrosion tests. Hence,
two biomass ashes, agriculture ash (OS) and wood ash (WS), were
chosen. Both of these ashes contained high concentration of alkali
 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459 455

Fig. 12. Mineral phases in equilibrium with the slag phase for wood sawdust ash (WS).

Table 5 Fig. 14 presents the SEM images of X10CrMoVNb9-1 steel after

Characteristic temperatures for slag phase formation. the exposure at 600
C for 1000 h, it is a reference, without the
Sample t0,
C t15,
C t25,
C t50,
C presence of ash comparing to the presence of biomass ash. Figs. 15
and 16 present the SEM images of X10CrMoVNb9-1 steel after the
OS <600 <600 <600 720
BS 920 920 1120 1170
exposure in the presence of biomass ashes. After the exposure of
LS 950 1120 1150 1210 the steel surface in the presence of biomass ashes, corrosive
WS <600 <600 1145 >1600 effects were observed and the formation of corrosion by-products

Table 6
metals (especially, potassium) based on XRF and XRD results. Main characteristic temperatures for the solid phase of ash transformations and
Additionally, the oat biomass ash characterizes by the lowest main processes (b-increase, a - decrease).
melting temperature resulting from AFTs analysis what is impor-
Temperature, Process
tant in corrosion process. Moreover, the wood biomass ash un-

C
derwent transformation up to 600
C (TGAs analysis). Such
OS (oat straw)
conclusions supported by FactSage calculations allowed to deter- 700e800 K2SiO3 phase disappearance
mine the final conditions of corrosion tests. 780e790 KCl solid-liquid phase transition
990e1000 CaO6P2 solid-liquid phase transition
1150e1160 Ca3 (PO4)2 solid-liquid phase transition
1060e1070 K2SO4 solid-liquid phase transition
1200e1400 KCl liquid phase disappearance
1340e1350 Mg3P2O8 solid-liquid phase transition
BS (beech sawdust)
680e690 KAlSi2O6 solid-solid phase transition
CaMgSi2O6 a and CaAl2Si2O8 b and MgSiO3 b
869e870 SiO2 solid-solid phase transition
920e980 NaAlSi3O8 phase disappearance
920e1600 KAlSi2O6 phase disappearance
980e990 MgSiO3 solid-solid phase transition
1020e1080 MgSiO3 a and Mg2SiO4 b
1140e1180 CaMgSi2O6 and CaAl2Si2O8 and CaSO4 disappearance
Ca3(PO4)2 solid-solid phase transition
1100e1310 Fe2O3 phase disappearance
LS (larch sawdust)
890e900 Ca3MgSi2O8 and CaMg2Al16O27 disappearance
890e1150 Ca3Fe2Si3O12 disappearance and Ca2MgSi2O7 appearance, Fe2O3
b
980e1080 K2SO4 a and KAlSi2O6 b and K2Ca2(SO4)3 b
and Ca2Al2SiO7 b
1040e1160 Ca2MgSi2O7 and Ca3Fe2Si3O12 transitions
WS (wood sawdust)
760e770 Ca2SiO4 disappears and Ca3MgSi2O8 appears and MgO a
890e900 Ca3MgSi2O8 disappears and Ca2SiO4(s2) appears and MgO b
1060e1070 K2SO4 solid-liquid phase transition
Fig. 13. Characteristic temperatures at which slag reaches value of 50% wt. (t50) vs. DT
1140e1160 Ca3(PO4)2 solid-solid phase transition and Ca2SiO4 disappear
and FT temperatures.
456 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459

Table 7
The transformation of sodium and potassium phases under high temperature conditions.

Temp. OS BS WS

600
C K2SiO3(s)
Y K2SO4(s2) KAlSi2O6(s)/KAlSi2O6(s2) K2SO4(s2)
KCl(s)/KCl(liq) NaAlSi3O8(s2)
Na3PO4(s)
800
C K2SO4(s2)
Y KCl(liq) KAlSi2O6(s2) K2SO4(s2)
K2SiO3(s)/K2O þ SiO2 NaAlSi3O8(s2)/NaAlO2þSiO2
Na3PO4(s)
1000
C K2SO4(s2)/K2SO4(liq)
Y KCl(liq) KAlSi2O6(s2) K2SO4(s2)/K2SO4(liq)
Na3PO4(s)


1200 C K2SO4(liq)
Y KCl(liq) KAlSi2O6(s2) e
Na3PO4(s)

Fig. 14. A photo of the steel surface, before (a) and after exposure to a high temperature (b): 600
C for 1000 h.

appeared on the surface of the steel. There are some cracks on
the scale, resulting from the destruction of subsequent layers of
steel. The degradation of the surface took place with varying
intensity and was highly dependent on the chemical composition
of the ash.
Fig. 15 presents the SEM images of the steel after its exposure
to agriculture ash (OS). The scale was in the form of flakes
covering the whole surface, but some chips of external layer also
appeared. By-products generated on the steel surface are similar
to those generated in the presence of synthetic KCl and K2SO4
salts, what was reported in Ref. [40]. Such behaviour of ash
indicates that potassium compounds play dominant role as cor-
rosive agents, because of formation of low melting temperatures
eutectics.
In the case of the wood ash layer scale it is heterogeneous, and a
slight delamination of steel was also observed (Fig. 16). This is a
serious problem because it leads to destruction, crack appear and
layers of metallic material are shed. Additionally, the use of
advanced electron microscopy method allowed to study the
morphology of the oxide scale and microstructure of material
(Figs. 17 and 18). A multilayered and multicomponent oxide scale
containing mainly Cr2O3 was found on the surface. Directly under

Fig. 15. SEM images of steel after exposure to agriculture ash (OS).
 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459
Fig. 16. SEM images of steel after exposure to wood ash (WS).
the oxide scale, a chromium-depleted zone with M23C6 carbides
was observed especially on the grain boundaries of the matrix.
The corrosion phenomena is very difficult to study, because of a
large variety of boiler construction and the most important the
variety of chemical composition of biomass. This problem could be
solved by: improvement of the fuel quality (using the low-chlorine
content fuel to combustion process), increasing the thickness of the
pipes and the use of overlays protection, the use of materials with
increased corrosion resistance (ferritic steels with high chromium
content instead of boiler steel) and the use of metallic, ceramic and
composites coatings on the surface of the higher temperature
[41,42].
4. Conclusions
This study focused on thermal the behaviour of four biomass
ashes with varying chemical compositions using thermal analysis
and the thermodynamic calculation method. The studied biomass
ashes contain silicon, potassium and calcium as principal ash-
forming constituents. The effect of potassium, silicon and
aluminium on the melting temperature was observed, the lowest
temperatures were found for oat straw ash (DT ¼ 790
C,
ST ¼ 1140
C, HT ¼ 1230
C, FT ¼ 1230
C), whereas the highest
Fig. 17. Morphology and metallographic cross-section of selected areas of the steel sample after corrosion test (600
C, 1000 h), covered by agriculture ash (OS) with the chemical
element distribution maps, SEM.
457
were obtained for wood biomass ash (DT ¼ 1530
C, ST, HT,
FT > 1550
C). The studied ashes were characterized by a high
tendency to fouling and slagging, which can limit the heat flow.
Additionally, the corrosion process can be intensive and problem-
atic, because of a high content of potassium, sodium and chlorine
especially in OS ash, where the KCl, KCaPO4, K2O, K3PO4, K3Fe(PO4)2
were identified by XRD. Thermal analysis has showed the thermal
behaviour resulting from the physical and chemical changes
(possible reactions) during heating. Important stages on the TG
results of the chemical composition of biomass ashes and decom-
position when contacted with the emission of gas phases were
identified. Fusion properties of calcium-rich biomass ash and
potassium-rich biomass ash confirm the TG results. FactSage was
used to predict the percentage of melting phases in a combustion
environment and possible solid phases existing at different tem-
peratures confirming possibility of potassium phases presence
(K2SiO3, K2SO4, KCl, KAlSi2O6) up to 1200
C. Finally, the results
obtained from the FactSage calculations agrees with XRD and ASTs
analysis and provide the information about the slagging properties
of biomass ashes. The biomass ashes containing alkali metals and
chlorine confirmed the corrosion properties during the experi-
mental studies. Hence, corrosion by-products appeared on the
surface of the steel. Concluding, the biomass ashes appeared to be
458 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459

Fig. 18. Morphology and metallographic cross-section of selected areas of the steel sample after corrosion test (600
C, 1000 h), covered by wood ash (WS) with the chemical
element distribution maps, SEM.

significant corrosive agents. The negative effects of corrosion and 4. SR e viscosity index:
deposition problems caused by biomass ashes can be limited by
using a special protective layer on the metal surface or by adding SR ¼ 100 SiO2/ (SiO2 þ eq(Fe2O3) þ CaO þ MgO) (17)
components which bind potassium and chlorine during the com-
bustion process. Unfortunately, both methods require additional
financial support.
5. RS e slagging index:

Acknowledgments RS ¼ RB/A Sd (18)

The research leading to these results has received funding from

the Polish-Norwegian Research Programme operated by the Na-
tional Centre for Research and Development under the Norwegian 6. Fu e fouling index:
Financial Mechanism 2009e2014 in the frame of Project Contract
No Pol_Nor/208189/105/2015. Fu ¼ RB/A (Na2O þ K2O) (19)

Appendix 1
References
Empirical indices which allow predicting the ash behaviour,
deposition tendencies [7,11,12]: [1] Y. Niu, Y. Zhu, H. Tan, S. Hui, Z. Jing, W. Xu, Investigations on biomass slagging
in utility boiler: criterion numbers and slagging growth mechanisms, Fuel
Process. Technol. 128 (2014) 499e508, http://dx.doi.org/10.1016/
1. RB/AþP (baseetoeacid) characterizes the ash fouling tendency: j.fuproc.2014.07.038.
[2] D. Nutalapati, R. Gupta, B. Moghtaderi, T.F. Wall, Assessing slagging and
RB/AþP ¼ (Fe2O3 þ CaO þ MgO þ Na2O þ K2O þ P2O5)/ fouling during biomass combustion: a thermodynamic approach allowing for
alkali/ash reactions, Fuel Process. Technol. 8 (2007) 1044e1052, http://
(SiO2 þ Al2O3 þ TiO2) (14) dx.doi.org/10.1016/j.fuproc.2007.06.022.
[3] L. Fryda, C. Sobrino, M. Cieplik, W.L. van de Kamp, Study on ash deposition
under oxyfuel combustion of coal/biomass blends, Fuel 89 (2010) 1889e1902,
http://dx.doi.org/10.1016/j.fuel.2006.01.020.
€
[4] M. Ohman, A. Nordin, B.-J. Skrifvars, R. Backman, M. Hupa, Bed agglomeration
2. RB/A (baseetoeacid): characteristics during fluidized bed combustion of biomass fuels, Energy Fuel
14 (2000) 169e178, http://dx.doi.org/10.1021/ef040093w.
[5] C. Luan, C. You, D. Zhang, Composition and sintering characteristics of ashes
RB/A ¼ (Fe2O3 þ CaO þ MgO þ Na2O þ K2O)/(SiO2 þ Al2O3 þ TiO2)(15) from co-firing of coal and biomass in a laboratory-scale drop tube furnace,
Energy 69 (2014) 562e570, http://dx.doi.org/10.1016/j.energy.2014.03.050.
[6] B.M. Steenari, A. Lundberg, H. Pettersson, M. Wilewska-Bien, D. Andersson,
Investigation of ash sintering during combustion of agricultural residues and
the effect of additives, Energy Fuel 23 (2009) 5655e5662, http://dx.doi.org/
3. RB e basic constituents, %: 10.1021/ef900471u.
[7] L. Wang, G. Skjevrak, J.E. Hustad, M.G. Gronli, Sintering characteristics of
RB/A ¼ (Fe2O3 þ CaO þ MgO þ Na2O þ K2O) (16) sewage sludge ashes at elevated temperatures, Fuel Process. Technol. 96
(2012) 88e97, http://dx.doi.org/10.1016/j.fuproc.2011.12.022.
[8] R.A. Antunes, M.C.L. de Oliveira, Corrosion in biomass combustion: a materials
selection analysis and its interaction with corrosion mechanisms and
 A. Magdziarz et al. / Renewable Energy 128 (2018) 446e459
mitigation strategies, Corros. Sci. 76 (2013) 6e26, http://dx.doi.org/10.1016/
j.corsci.2013.07.013.
[9] A. Magdziarz, A. K. Dalai, J. A. Kozin
ski, Corrosive properties of ash residues
generated during renewable fuels combustion, Fuel 176 (2016) 135e145,
http://dx.doi.org/10.1016/j.fuel.2016.02.069.
[10] Y.Q. Niu, H.Z. Tan, L. Ma, M. Pourkashanian, Z.N. Liu, Y. Liu, et al., Slagging
characteristics on the superheaters of a 12 MW biomass-fired boiler, Energy
Fuel 24 (2010) 5222e5227, http://dx.doi.org/10.1021/ef1008055.
[11] M. Pronobis, Evaluation of the influence of biomass co-combustion on boiler
furnace slagging by means of fusibility correlations, Biomass Bioenerg. 28
(2005) 375e383, http://dx.doi.org/10.1016/j.biombioe.2004.11.003.
[12] M. Pronobis, The influence of biomass co-combustion on boiler fouling and
efficiency, Fuel 85 (2006) 474e480, http://dx.doi.org/10.1016/
j.fuel.2005.08.015.
[13] M. Reinmo € ller, M. Klinger, M. Schreiner, H. Gutte, Relationship between ash
fusion temperatures of ashes from hard coal, brown coal, and biomass and
mineral phases under different atmospheres: a combined FactSage™
computational and network theoretical approach, Fuel 151 (2015) 118e123,
http://dx.doi.org/10.1016/j.fuel.2015.01.036.
[14] L. Li, C. Yu, F. Huang, J. Bai, M. Fang, Z. Luo, Study on the deposits derived from
a biomass circulating fluidized-bed boiler, Energy Fuel 26 (2012) 6008e6014,
http://dx.doi.org/10.1021/ef301008n.
[15] J.F. Saldarriaga, R. Aguado, A. Pablos, M. Amutio, M. Olazar, J. Bilbao, Fast
characterization of biomass fuels by thermogravimetric analysis (TGA), Fuel
140 (2015) 744e751, http://dx.doi.org/10.1016/j.fuel.2014.10.024.
[16] A. Magdziarz, M. Wilk, Thermal characteristics of the combustion process of
biomass and sewage sludge, J. Therm. Anal. Calorim. 114 (2013) 519e529,
http://dx.doi.org/10.1007/s10973-012-2933-y.
[17] A. Magdziarz, M. Wilk, M. Gajek, D. Nowak-Wo
zny, A. Kopia, I. Kalemba-Rec,
J.A. Kozin
ski, Properties of ash generated during sewage sludge combustion: a
multifaceted analysis, Energy 113 (2016) 85e94, http://dx.doi.org/10.1016/
j.energy.2016.07.029.
[18] L. Febrero, E. Granada, C. Perez, D. Patino, E. Arce, Characterisation and
comparison of biomass ashes with different thermal histories using TG-DSC,
J. Therm. Anal. Calorim. 118 (2014) 669e680, http://dx.doi.org/10.1007/
s10973-014-3717-3.
[19] S. Du, H. Yang, K. Qian, X. Wang, H. Chen, Fusion and transformation prop-
erties of the inorganic components in biomass ash, Fuel 117 (2014)
1281e1287, http://dx.doi.org/10.1016/j.fuel.2013.07.085.
[20] Q.H. Li, Y.G. Zhang, A.H. Meng, L. Li, G.X. Li, Study on ash fusion temperature
using original and simulated biomass ashes, Fuel Process. Technol. 107 (2013)
107e112, http://dx.doi.org/10.1016/j.fuproc.2012.08.012.
[21] P. Viklund, A. Hjo €rnhede, P. Henderson, A. Stålenheim, R. Pettersson, Corro-
sion of superheater materials in a waste-to-energy plant, Fuel Process.
Technol. 105 (2013) 106e112, http://dx.doi.org/10.1016/j.fuproc.2011.06.017.
[22] A. Pettersson, L.E. Amand, B.M. Steenari, Chemical fractionation for the
characterisation of fly ashes from co-combustion of biofuels using different
methods for alkali reduction, Fuel 88 (2009) 1758e1772, http://dx.doi.org/
10.1016/j.fuel.2009.03.038.
[23] D. Vamvuka, W. Kakaras, Ash properties and environmental impact of various
biomass and coal fuels and their blends, Fuel Process. Technol. 92 (2011)
570e581, http://dx.doi.org/10.1016/j.fuproc.2010.11.013.
[24] L.W. Sengelov, T.B. Hansen, C. Bartolome, H. Wu, K.H. Pedersen, F.J. Frandsen,
Sulfation of condensed potassium chloride by SO2, Energy Fuel 27 (2013)
3283e3289, http://dx.doi.org/10.1021/ef400405z.
[25] Y. Niu, H. Tan, S. Hui, Ash-related issues during biomass combustion: alkali-
induced slagging, silicate melt-induced slagging (ash fusion), agglomeration,
corrosion, ash utilization, and related countermeasures, Prog. Energy
Combust. Sci. 52 (2016) 1e61, http://dx.doi.org/10.1016/j.pecs.2015.09.003.
459
[26] M. Aho, Reduction of chlorine deposition in FB boilers with aluminium-
containing additives, Fuel 80 (2001) 1943e1951, http://dx.doi.org/10.1016/
S0016-2361(01)00049-7.
[27] X. Wei, U. Schnell, K.R.G. Hein, Behaviour of gaseous chlorine and alkali metals
during biomass thermal utilization, Fuel 84 (2005) 841e848, http://
dx.doi.org/10.1016/j.fuel.2004.11.022.
[28] R.A. Antunes, M.C.L. de Oliveira, Corrosion in biomass combustion: a materials
selection analysis and its interaction with corrosion mechanisms and miti-
gation strategies, Corros. Sci. 76 (2013) 6e26, http://dx.doi.org/10.1016/
j.corsci.2013.07.013.
[29] B.-J. Skrifvars, M. Westen-Karlsson, M. Hupa, K. Salmenoja, Corrosion of su-
perheater steel materials under alkali salt deposits. Part 1: SEM analyses of
different steel materials, Corros. Sci. 52 (2010) 1011e1019, http://dx.doi.org/
10.1016/j.corsci.2009.11.026.
[30] R. Saidur, E.A. Abdelaziz, A. Demibras, M.S. Hossain, S. Mekhilef, A review on
biomass as fuel for boilers, Renew. Sustain. Energy Rev. 15 (2011) 2262e2289,
http://dx.doi.org/10.1016/j.rser.2011.02.015.
[31] D. Bostrom, N. Skoglund, A. Grimm, C. Boman, M. Ohman, Brostrom,
M.R. Backman, Ash transformation chemistry during combustion of biomass,
Energy Fuel 26 (2012) 85e93, http://dx.doi.org/10.1021/ef201205b.
[32] C.H. Pang, B. Hewakandamby, T. Wu, E. Lester, An automated ash fusion test
for characterization of the behavior of ashes from biomass and coal at
elevated temperatures, Fuel 103 (2013) 454e466, http://dx.doi.org/10.1016/
j.fuel.2012.06.120.
[33] D. Lindberg, R. Backman, P. Chartrand, M. Hupa, Towards a comprehensive
thermodynamic database for ash-forming elements in biomass and waste
combustion d current situation and future developments, Fuel Process.
Technol. 105 (2013) 129e141, http://dx.doi.org/10.1016/j.fuproc.2011.08.008.
[34] T. Rizvi, P. Xing, M. Pourkashanian, L.I. Darvell, J.M. Jones, W. Nimmo, Pre-
diction of biomass ash fusion behaviour by the use of detailed characteriza-
tion methods coupled with thermodynamic analysis, Fuel 141 (2015)
275e284, http://dx.doi.org/10.1016/j.fuel.2014.10.021.
[35] S. Chakravarty, A. Mohanty, A. Banerjee, R. Tripathy, G.K. Mandal,
M.R. Basariya, M. Sharma, Composition, mineral matter characteristics and
ash fusion behavior of some Indian coals, Fuel 150 (2015) 96e101, http://
dx.doi.org/10.1016/j.fuel.2015.02.015.
[36] J.C. van Dyk, M.J. Keyser, Influence of discard mineral matter on slag-liquid
formation and ash melting properties of coal e a FACTSAGETM simulation
study, Fuel 116 (2014) 834e840, http://dx.doi.org/10.1016/j.fuel.2013.06.002.
[37] X.C. Baxter, L.I. Darvell, J.M. Jones, T. Barraclough, N.E. Yates, I. Shield, Study of
Miscanthus x giganteus ash composition e variation with agronomy and
assessment method, Fuel 95 (2012) 50e62, http://dx.doi.org/10.1016/
j.fuel.2011.12.025.
[38] I. Pisa, G. Lazaroiu, Influence of co-combustion of coal/biomass on the
corrosion, Fuel Process. Technol. 104 (2012) 356e364.
[39] M. Aho, P. Yrjas, R. Taipale, M. Hupa, J. Silvennoinen, Reduction of superheater
corrosion by co-firing risky biomass with sewage sludge, Fuel 89 (2010)
2376e2386, http://dx.doi.org/10.1016/j.fuel.2010.01.023.
[40] A. Magdziarz, M. Gajek, M. Wilk, I. Kalemba-Rec, P. Gara, Corrosion of metal
surface under renewable fuel ash deposits, SDEWES, in: Proceedings of 11th
Conference on Sustainable Development of Energy, Water and Environment
Systems, 2016 (Lisbon, Portugal).
[41] P. Viklund, A. Hjo € rnhede, P. Henderson, A. Stålenheim, R. Pettersson, Corro-
sion of superheater materials in a waste-to-energy plant, Fuel Process.
Technol. 105 (2013) 106e112, http://dx.doi.org/10.1016/j.fuproc.2011.06.017.
[42] R.A. Antunes, M.C.L. de Oliveira, Corrosion in biomass combustion: a materials
selection analysis and its interaction with corrosion mechanisms and miti-
gation strategies, Corros. Sci. 76 (2013) 6e26, http://dx.doi.org/10.1016/
j.corsci.2013.07.013.