Experimental evaluation of additives and K2O–SiO2–Al2O3 diagrams on hightemperature silicate meltinduced slagging during biomass combustion

Fuel 179 (2016) 52–59

Contents lists available at ScienceDirect

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

Experimental evaluation of additives and K2O–SiO2–Al2O3 diagrams
on high-temperature silicate melt-induced slagging during biomass
combustion
Yanqing Niu, Zhizhou Wang, Yiming Zhu, Xiaolu Zhang, Houzhang Tan, Shi’en Hui ⇑
Key Laboratory of Thermo-Fluid Science and Engineering of MOE, School of Energy and Power Engineering, Xi’an Jiaotong University, Xi’an 710049, China

h i g h l i g h t s
 Propose FT as an evaluation index of high-temperature silicate melt-induced slagging potential.
 Propose evaluation criteria on FT and high-temperature silicate melt-induced slagging.
 K2O–SiO2–Al2O3 diagram built on biomass ash undervalues FT of doped biomass 140–190 K.
 K2O–SiO2–Al2O3 diagram built on doped biomass over-predicts FT of pure biomass 200 K.
 FTs show ‘V’ shaped distributions with increased SiO2, Al2O3, and K2O, respectively.

a r t i c l e

i n f o

Article history:
Received 22 February 2016
Received in revised form 17 March 2016
Accepted 19 March 2016
Available online 28 March 2016
Keywords:
Biomass

Ash
Slagging
Silicate
Combustion
K2O–SiO2–Al2O3

a b s t r a c t
As one major barrier for biomass combustion, the high-temperature silicate melt-induced slagging is
studied by additions of SiO2, kaolin, and soil and two types of K2O–SiO2–Al2O3 ternary phase diagrams
constructed on basis of the real biomass ash properties and biomass by addition of K2O (in the form of
KOH), SiO2, and Al2O3, respectively. Results show that FT can be as the evaluate index for hightemperature silicate melt-induced slagging which increases with decreased FT. Meanwhile, a set of qualitative criteria on high-temperature silicate melt-induced slagging are proposed. However, because of the
refractory minerals originated from additives directly or alumina-silication reactions indirectly when biomass blended with additives, the quantitative prediction of pure biomass and the biomass added additives should be based on the K2O–SiO2–Al2O3 ternary phase diagrams build by pure biomass ash
properties and the biomass added Si/Al/K additives, respectively. Overall, FTs show ‘V’ shaped distributions with increased SiO2, Al2O3, and K2O in ash, respectively. Unlike SiO2 which exacerbates lowtemperature silicate melt-induced slagging, soil can substitute for expensive kaolin served as additives
during biomass combustion. The whole research provides useful guidelines for biomass selection,
improvement, and slagging prevention during combustion.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction
Biomass combustion has been developed around the world
because of the worsening environment and increasing energy crisis. In China, the biomass power generation capacity will reach
30 GW in 2020 and accounts for 3% of the total power installed
capacity, meanwhile, more than 130 dedicated biomass fired
power plants have been operated in the country. In Europe, biomass power generation capacity has taken up 70% of all generated

⇑ Corresponding author. Tel.: +86 13709181734.
E-mail address: (S. Hui).
/>0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.

renewable fuel power, and in USA the biomass power installed
capacity has reached 10 GW [1].

Unfortunately, severe slagging happened in both biomass-fired
fluidized bed (FB) and grate furnace [2,3] not only reduce heat
transfer efficiency but also damage super-heaters and eventually
lead to unscheduled shutdown frequently [3,4]. Consequently,
alkali metals (especially K) that have inescapable responsibilities
on slagging has been investigated widely, including the migration
and transportation behaviors during combustion [5–11], the existing forms in biomass [12,13], and the influence on ash fusion characteristics [5,6,14–16]. During combustion, alkali mainly released
as gaseous hydroxides, chlorides, and sulfates [7], but biomass
species [7], cultivated fields [17], combustion temperature and


53

Y. Niu et al. / Fuel 179 (2016) 52–59

atmosphere [6,7,18], and the concentrations of K/Cl/SO2 [8,9,11,19]
has significant effects on the concentration distribution. It goes
without saying that the widely studies on additives [20–23], leaching [24], and cofiring [2,18,25] that change the fuel properties
directly also have remarkably effects on the alkali migration and
transportation behaviors. Recently, on basis of the different slagging formation mechanisms alkali-induced slagging and silicate
melt-induced slagging (further classified into low-temperature
and high-temperature silicate melt-induced slagging) have been
proposed and investigated, respectively [5,17].
Simply, the alkali metals in biomass are mainly existed in the
forms of volatile, non-volatile water-soluble, and water insoluble
[26]. The volatile alkaline metal compounds such as KCl and
K2SO4 can serve as adhesive and bond fly ash themselves and fly
ash and heating surfaces together resulting in alkali-induced slagging [5,27]. Meanwhile, both the volatile alkaline metal compounds and the non-volatile but water-soluble alkaline metal
compound such as carbonates can react with SiO2 into lowmelting silicates causing low-temperature silicates melt-induced
slagging [5,28]. A global reaction of KCl and SiO2 resulting in the

generation of low-melting silicates can be expressed by (R1) where
when n equals 1, 2 and 4, the corresponding melting points of K2OÁnSiO2 are below 1073 K [29,30]. This is also a reason that
agglomeration formation in FB furnace (typical combustion temperature 1123–1223 K) where KCl released from biomass reacts
with quartz bed material into low-melting silicates.

2KCl þ nSiO2 þ H2 O ! K2 O Á nSiO2 þ 2HCl

ðR1Þ

The insoluble alkali silicates, alkali aluminum silicates, and
alkali calcium/magnesium silicates as refractory skeleton structure
in biomass ash dominate the high-temperature silicate meltinduced slagging [6,14]. When the temperature is above the melting
points of abovementioned substances or the fluidized temperature
(FT) of the ash, the ash will melt, bind other compounds, and
adhere on heating surface resulting in slagging. This is the reason
why kaolin and calcite have been served as effective additives used
to mitigate slagging during biomass combustion [21–23]. Through
Rs (2)–(5), kaolin not only suppresses the release of alkali chlorides
and sulfates consequently eliminating alkali-induced slagging but
also generates high-melting alkali aluminum silicates preventing
the occurrence of high-temperature silicate melt-induced slagging
[5], such as KAlSiO4 (kalsilite) and KAlSi2O6 (leucite) with the melting temperatures of greater than 1873 K and 1773 K, respectively
[20,31], which is obviously higher than the combustion bath temperatures either in FB or in grate furnace.

Al2 O3 Á 2SiO2 Á 2H2 O þ 2MCl ! 2MAlSiO4 þ 2HCl þ H2 O
ðR2Þ
ðR3Þ
Al2 O3 Á 2SiO2 Á 2H2 O þ M2 SO4 ! 2MAlSiO4 þ SO3 þ 2H2 O
Al2 O3 Á 2SiO2 Á 2H2 O þ 2MCl þ 2SiO2 ! 2MAlSi2 O6 þ 2HCl þ H2 O
ðR4Þ

Al2 O3 Á 2SiO2 Á 2H2 O þ M2 SO4 þ 2SiO2 ! 2MAlSi2 O6 þ SO3 þ 2H2 O
ðR5Þ
where M represents K and Na.
In comparison with experiments such as abovementioned
which provide detailed identification on special conditions, quantitative criterion numbers and evaluation index or simply qualitative
trendline based on statistic analysis of numbers of experiment data
can provide more general reference. For alkali-induced slagging
formed by the re-enrichment of fine particles primarily contained
high concentrations of K, Na, Cl, and S in the forms of KCl and
K3Na(SO4)2 and the re-capture of coarse large particles primarily
contained higher Si, Al and so on [17], gaseous alkali salts promote
the formation and development of the deposits, while Si and Al
play opposite functions by trapping the alkali salts before they
forms sticky deposits [32]. High Si and Al in the ash assist in the

alkali removal from the vapor phase and therefore reduce alkaliinduced slagging [5]. Furthermore, considering the formation
mechanisms of alkali-induce slagging, by means of a detailed analysis on the effects of S, Cl, Si, Al, and K on the distinct slagging characteristics of two cotton stalks burned in utility grate furnaces and
a series of statistical data from references, a quantitative criterion
number of alkali-induced slagging has been proposed recently as
follow[17].
While Cl ratio ðCl þ K2 O þ Na2 OÞ=ðSiO2 þ Al2 O3 Þ P 2:4
S ratio ðSVolatile þ K2 O þ Na2 OÞ=ðSiO2 þ Al2 O3 Þ P 1:9

While Cl ratio 6 1:0
slight slagging
S ratio 6 0:5


serious slagging

ðExp 1Þ

Due to the remarkably formation of vapor alkali chlorides and
sulfates, severe alkali-induced slagging occurs when the Cl ratio
and S ratio are greater than 2.4 and 1.9, respectively. By contrary,
slagging is slight when the both ratios are lower than 1.0 and
0.5, respectively. Slagging potential increases with increased Cl
ratio and S ratio in the ranges of 1.0–2.4 and 0.5–1.9, respectively.
This criterion number clearly certifies the exciting function of kaolin which mitigates slagging effectively [20–23,31].
For low-temperature silicate melt-induced slagging, Si2O and
Al2O3 can raise the initial deformation temperature (IDT) [16],
while the introduction of Al into potassium silicate melt lowers
the melting temperature for high K/Al and low (K2 + Al2)/Si ratio
[15]. Recently, the authors studied the low-temperature silicate
melt-induced slagging by additives and 30 biomasses, and found
that IDT can be used as the evaluation index for low-temperature
silicate melt-induced slagging, and high IDT reduces the potential
for low-temperature silicate melt-induced slagging occurrence.
IDT increases with increase in Al2O3 and SiO2/K2O, while decreases
with increase in K2O, SiO2, SiO2/Al2O3, and (SiO2 + K2O)/Al2O3. The
significant effects of the compounds on IDT follow: Al2O3 >
K2O > SiO2 > SiO2/K2O > SiO2/Al2O3 > (SiO2 + K2O)/Al2O3. Based on
the significant effects a set of criteria to evaluate the potential of
low-temperature silicate melt-induced slagging is proposed in
detailed [26].
In aspect of high-temperature silicate melt-induced slagging,
that is more depended on refractory minerals [33]. The refractory
minerals, which has been identified as quartz, potassium iron
oxide, and potassium magnesium silicate, potassium aluminum
silicate, potassium calcium silicate, calcium silicate, mullite, diopside, pyrope, and monticellite, etc. [6], provide structural support

for the skeleton-like structure in biomass ash [14]. Once the combustion temperature was above the melting point of the refractory
minerals or access the FT, they will melt and adhere and result in
high-temperature silicate melt-induced slagging, especially on
water wall in furnace where the combustion temperature is high.
Generally, it is commonly accepted that SiO2 can inhibit the slagging, whereas the high SiO2 in biomass may exacerbate slagging
[34] attributed to the unquestioning addition that just causes the
generation of low-melting K2OÁnSiO2 [29,30]. In addition, Xiong
et al. [35] pointed out that high K/(Ca + Mg) can inhibit slagging;
but the opposed trends that the slagging is strengthened with
increased K2O and decreased CaO and Al2O3 were found [14].
Although various research on high-temperature silicate meltinduced slagging have been conducted, a detailed criteria like the
proposed for alkali-induced slagging (Exp 1) [17] and lowtemperature silicate melt-induced slagging [26] and originated
from real biomass ash properties rather than simulated ash has
not be reported.
Drawing lessons from the previous research on lowtemperature silicate melt-induced slagging [26], this paper therefore focused on the effect of ash compounds on hightemperature silicate melt-induced slagging aims to provide a


54

Y. Niu et al. / Fuel 179 (2016) 52–59

Table 1
The distribution of inorganic elements in additives and biomass ashes at different temperatures, wt.%.

a
b

Material

Si

Al

Fe

Mg

Ca

Na

K

S

Cl

Others

Si/Al

Biomass
Kaolin
Soil
SiO2

16.9
22.9
23.4
46.6

2.1
24.7
13.5
0

1.9
0.4
4.6
0

4.3
0
0.3
0

7.9
0.2
0.1
0

0.3
0.1
0.1
0

12.8
0.2
1.7
0

1
0
0.1
0

1.7
0
0
0

51.3
51.4
56.2
53.4

7.8:1a
0.9:1a(2.6:1b)
1.7:1a(3.9:1b)
1a(13.4:1b)

The mole ratios of Si/Al for 100 wt.% materials.
The mole ratios of Si/Al for 97 wt.% biomass + 3 wt.% additives.

detailed method to evaluate the slagging potential, and the effects
of the concentrations of Si, Al, and K in ash, and SiO2, kaolin, and
soil additives on high-temperature silicate melt-induced slagging
are studied simultaneously. Firstly, the effects of SiO2, kaolin, and
soil on ash fusion characteristics are tested; secondly, through
the statistic analysis of the effects of ash compounds (Si, Al, and

K) from 30 biomass, a set of detailed evaluation criteria that can
be used to qualitatively guild biomass selection and improvement
by additives, leaching, and cofiring and consequently mitigating or
eliminating high-temperature silicate melt-induced slagging is
proposed; thirdly, two types of K2O–SiO2–Al2O3 ternary phase diagrams constructed on basis of the real biomass ash properties and

Fig. 1. Effects of SiO2, kaolin, and soil on ash fusion characteristics [26,36].

biomass by addition of K2O (in the form of KOH), SiO2, and Al2O3,
respectively, are compared in order to provide practice guideline
on biomass selection, improvement, and subsequent slagging predication and research.

2. Experiments
2.1. Experiment materials
In experiment, the wheat straw, an abundant agricultural residue in China, is selected as biomass representative. In order to
study the effects of different Si/Al compounds on the hightemperature silicate melt-induced slagging, pure biomass, biomass
blended with 3 wt.% SiO2, 3 wt.% kaolin and 3 wt.% soil additives,
respectively, are selected comparably. The corresponding inorganic
element compositions of the materials are listed in Table 1.
Seen from Table 1, the biomass contains high Si and K, and low
Al content, as well as high Si/Al mole ratio being around 7.8:1;
while it is around 0.9:1, 1.7:1, and infinite in kaolin, soil, and
SiO2 additive, respectively. In addition, the content of Si in SiO2 is
approximate two times of that in kaolin and soil; and the content
of Al in kaolin is about 2 times of that in soil. So the significant differences in the additives facilitate the efficiency comparisons on
the high-temperature silicate melt-induced slagging.
In order to systematically identify the effects of the concentrations of Si, Al, and K in ash on high-temperature silicate meltinduced slagging, the ash compositions and ash fusion characteristics of thirty biomasses fired in operating biomass power plants are
tested, and a K2O–SiO2–Al2O3 ternary phase diagram is built on
basis of the ash properties. Meanwhile, three K2O–SiO2–Al2O3
ternary phase diagrams (total K2O–SiO2–Al2O3, water insoluble

Fig. 2. SEM–EDS for the materials incinerated at 1088 K.


Y. Niu et al. / Fuel 179 (2016) 52–59

K2O–SiO2–Al2O3, and water soluble K2O–SiO2–Al2O3) are constructed by addition of K2O, SiO2, and Al2O3 into biomass. In comparison with K2O–SiO2–Al2O3 ternary phase diagrams constructed
by additions of K2O, SiO2 and Al2O3 oxides, the K2O–SiO2–Al2O3
ternary phase diagram built on basis of the thirty pure biomasses
are more comparable with the real biomass components, whereas
the K2O–SiO2–Al2O3 ternary phase diagrams constructed by additions of K2O, SiO2 and Al2O3 oxides may be more appropriate for
the prediction of improved biomass by additives and leaching.
2.2. Experiment apparatus
The fusion temperature testing on biomass ash is conducted in a
sintering instrument, which mainly consists of a temperature controllable electric heating furnace by program and a high-precision
digital read-out and photographic record camera (SJY, Xiangtan
Instrument Co., Ltd., China). Elements determination and main
compounds identification are accomplished with XRF (X-ray fluorescence, S4-Pioneer, Bruker Co., Germany) and XRD (X-ray powder diffractometry, Xpert pro, PANalytical B.V. Netherland)
respectively. Detailed descriptions on the instruments can be seen
in previous papers [3,14]. Meanwhile, the element distribution
outside the ash particle and the ash morphology analysis are performed by using SEM–EDS (scanning electron microscopy–energy
dispersive spectrometer, JSM-6390A, Japan), and ICP-AES (Inductively coupled plasma atomic emission spectroscopy, ICPE-9000,
Japan) is used to test the concentration of water soluble-K in biomass ash.

consistent with each other due to the silication reactions at
1088 K, and the distribution of Al does not match the distributions
of Si and K; while for addition of kaolin and soil, except few certain
zones which may be originated from the additives directly or generated by the insufficient silication and alumina-silication reactions at
1088 K, the distributions of Si, Al and K are not consistent with each
other. Thus, the sufficient silicate reactions lead to considerable

formation of low melting silicates which result in lower IDT and
ST of the ash of biomass adding SiO2 (especially IDT), and the
alumina-silication reactions of potassium chlorides and sulfates
(Rs (2)–(5)) and the generation of more high-melting substances
at elevated temperature result in increased FT.
Therefore, It can be deduced that the IDT affected by the significant formation of low melting silicates through the silication of
alkali chlorides and sulfates can be as an evaluate index for biomass low-temperature silicate melt-induced slagging [26]; while
the FT, which is mainly affected by the high temperature refractory
substances in biomass ash, can be as an evaluation index for hightemperature silicate melt-induced slagging mainly affected by high

3. Results and discussion
3.1. Effects of SiO2, kaolin and soil
Commonly, the high-temperature silicate melt-induced slagging is dominantly affected by the high-temperature refractory
materials which provide a supporting skeleton structure in the biomass ash [14]. Therefore, the effects of SiO2, kaolin, and soil on ash
fusion characteristics are performed.
It can be seen from Fig. 1 that except the IDT and soften temperature (ST) gained by addition of SiO2, in comparison with pure biomass the additions of SiO2, kaolin, and soil increase the ash fusion
temperatures as a whole, particular in FT, which considerably
increase because of the formation of more high-temperature
refractory materials from the additives directly and/or reactant
products indirectly. Meanwhile, the additions of kaolin and soil
present the almost same level increasing in the ash fusion temperatures. Therefore, it seems that the soil can substitute for kaolin
served as additives during biomass combustion. Whereas, further
study on soil need conducted because of the various compounds
of the different soil sources.
Both IDT and ST increase with the addition of kaolin and soil,
while decrease with the addition of SiO2 as singular points. The
decreased IDT and ST of biomass with the addition of SiO2 should
be caused by the significant silication of KCl (R1) [29,30]. Consequently, the considerable formation of K-silicates with melting
temperature below 1073 K results in the decreasing IDT and
ST, especially IDT. And later, along with the occurrence of

alumina-silication reactions of potassium chlorides and sulfates
(Rs (2)–(5)) and the generation of more high-melting substances,
FT increase.
This guesses that the decreasing IDT and ST of biomass with the
addition of SiO2 are really attributed to the silication reactions (R1)
can be verified from the SEM–EDS analysis on the ash generated by
incineration at 1088 K as shown in Fig. 2. The distributions of Si
and K in the ash of biomass with the addition of SiO2 are highly

55

Fig. 3. Statistic analysis on the effects of various components on FT.


56

Y. Niu et al. / Fuel 179 (2016) 52–59

temperature refractory skeleton structure constructed by alkali
calcium/magnesium silicates and alkali aluminum silicates originated from biomass contaminants directly and/or the reaction produces of the alumina-silication of potassium chlorides and sulfates
indirectly. Detailed investigation on low-temperature silicate
melt-induced slagging can be seen from previous reference [26].
In comparison with kaolin, soil which presents the almost same
effect on ash fusion temperatures can substitute for expensive kaolin served as additives during biomass combustion. However, SiO2
which exacerbates low-temperature silicate melt-induced slagging
by reacting with KCl into low melting silicates is not suitable for
additive.
3.2. Evaluation criteria
Once the temperature is above the melting points of the refractory ash compounds or the FT of the ash, the ash will undergo
deformation and melting, and then adhere on the heating-surface

by inertial impaction. That is the typically high-temperature silicate melt-induced slagging happened in furnace [5].
It can be seen from Fig. 3a that FT dramatically increases with
increase in K2O, and decreases with increase in Al2O3 and SiO2.
Al2O3 shows the highest effect on FT seen from the highest slope
(À462.8, 23.1 for 20Al2O3), followed by K2O (187.3) and SiO2
(À8.8) in turn. From Fig. 3b, it can be seen that FT decreases with
the increase in SiO2/Al2O3 (À9.9) and (SiO2 + K2O)/Al2O3 (À9.6),
and increases with the increase in SiO2/K2O (14.8). The slightly larger slope of (SiO2 + K2O)/Al2O3 than SiO2/Al2O3 indicates that K2O
has certain positive effect on FT. Moreover, from the slopes, it
can be concluded that the effects of above parameters on FT are
ordered as follow: Al2O3 > K2O > SiO2/K2O > SiO2/Al2O3 > (SiO2 + K2O)/Al2O3 > SiO2, and the positive effect orders are K2O > SiO2/
K2O, and the negative effect orders are Al2O3 > SiO2/Al2O3 >
(SiO2 + K2O)/Al2O3 > SiO2. Therefore, a detailed evaluation criterion
on high-temperature silicate melt-induced slagging based on FT is
described as follow and also illustrated in Fig. 4 clearly.
Route 1: If the biomass contains higher Al2O3, SiO2, and lower
K2O, it presents lower FT and higher high-temperature silicate
melt-induced slagging potential.
Route II: If the biomass contains higher Al2O3 and lower SiO2, it
needs to consider the combined parameter SiO2/Al2O3 due to

the negative effects of both Al2O3 and SiO2. Higher Al2O3 and
lower SiO2 lead to lower SiO2/Al2O3, so if the biomass possesses
higher K2O at the same time, then it presents higher FT and
lower high-temperature silicate melt-induced slagging potential; while if the biomass has lower K2O, (SiO2 + K2O)/Al2O3
must be considered because of the opposite effects of SiO2/
Al2O3 and K2O. When the biomass holds lower (SiO2 + K2O)/
Al2O3, it shows higher FT and lower low-temperature silicate
melt-induced slagging potential; Conversely, it shows lower
FT and higher high-temperature silicate melt-induced slagging

potential with higher (SiO2 + K2O)/Al2O3.
Route III: If the biomass contains higher Al2O3, SiO2 and K2O, the
combined parameter SiO2/K2O must be considered due to the
opposite effect of K2O relative to Al2O3 and SiO2. If the biomass
has lower SiO2/K2O, it possesses lower FT and higher hightemperature silicate melt-induced slagging potential. Similarly,
once the biomass contains higher SiO2/K2O, (SiO2 + K2O)/Al2O3
becomes the sole option because of the collision caused by
the opposite trends of higher Al2O3 and higher SiO2/K2O. The
higher the (SiO2 + K2O)/Al2O3 is, the lower the FT is, and the
easier the low-temperature silicate melt-induced slagging
becomes, and vice versa.

“■”Biomass+Kaolin; “●”Biomass+Soil; “▼”Biomass; “ƾ”Biomass+SiO2
; “·”Experimental value; unit: weight ratio
Fig. 5. K2O–SiO2–Al2O3 ternary phase diagrams based on 30 biomass ash
properties.

Fig. 4. Evaluation criteria on FT and high-temperature silicate melt-induced slagging potential.


Y. Niu et al. / Fuel 179 (2016) 52–59

Fig. 6. K2O–SiO2–Al2O3 ternary phase diagrams based on biomass by additions of K2O, SiO2 and Al2O3 oxides.

57


58

Y. Niu et al. / Fuel 179 (2016) 52–59

3.3. K2O–SiO2–Al2O3 ternary phase diagrams
Although abovementioned statistic analysis provides useful
qualitative guidelines for high-temperature silicate melt-induced
slagging and it is user-friendly by remembering the effect orders,
it unavoidably omits some key-points as shown in Fig. 4. Therefore,
in view of the limitation, and to compare the high-temperature silicate melt-induced slagging quantitatively, conveniently, and
directly, two sets of K2O–SiO2–Al2O3 ternary phase diagrams of
FT are constructed on basis of the thirty pure biomass ash properties (Fig. 5) and biomass by additions of K2O, SiO2 and Al2O3 oxides
(Fig. 6), respectively.
Fig. 5 shows the K2O–SiO2–Al2O3 (actually should be K2O–SiO2–
20Al2O3) ternary phase diagrams built on the ash properties of the
thirty pure biomasses. It can be seen that the predicted temperatures of the pure biomass is highly consistent with the measuring
value; while the measured FTs of the doped biomass are about
140–190 K higher than the predicted values. Therefore, it can be
concluded that even both pure biomass and the doped biomass
possess the same K2O–SiO2–Al2O3 constructions, in comparison
with pure biomass the doped biomass present higher FT due to
the newly generated high-temperature refractory silicates through
Rs (2)–(5) and/or the extra and un-reacted Si/Al compounds which
mainly exist in oxides or original refractory minerals increasing the
FT. Thus, the K2O–SiO2–Al2O3 ternary diagram built on pure biomass ash properties is improper for the FT predication of doped
biomass because of the existence of the excess oxide monomers
and/or refractory minerals originated from additives directly or
the reaction products through Rs (2)–(5) in the doped biomass.
Also, it can be seen from Fig. 5 that there exist some singular
zones where FT holds the relatively highest and lowest temperatures, i.e. where the occurrence of high-temperature silicate
melt-induced slagging is the hardest or easiest. One typical low
temperature zones is around where K2O:SiO2:20Al2O3 equals
0.3:0.55:0.15 more or less; and two high temperature zones are

around where K2O:SiO2:20Al2O3 is approximately (0.15–0.75):(0.
05–0.1):(0.25–0.75) and where K2O:SiO2:20Al2O3 equals
0.05:0.75:0.2 more or less. Moreover, the FTs show ‘V’ shapes with
increased SiO2, Al2O3, and K2O, respectively. This should be the reason why some conflicting results were reported when the research
were located on the two sides of the ‘V’ shapes, such as the reports
on SiO2 [34] and K/(Ca + Mg) [14,35] that have been described in
introduction.
Fig. 6 shows the three K2O–SiO2–Al2O3 ternary phase diagrams
constructed on basis of biomass by additions of K2O, SiO2, and
Al2O3. It can been seen that in either one of the three diagrams
(total K2O–SiO2–Al2O3, water insoluble K2O–SiO2–Al2O3, and water
soluble K2O–SiO2–Al2O3) the measured FT of pure biomass is 230–
245 K lower that the prediction value, while the measured FTs of
the doped biomass are well consistent with the predicted values.
Thus, it can be concluded that the FT prediction and comparison
of pure biomass should be according to the K2O–SiO2–Al2O3 ternary phase diagrams built on pure biomass ash properties (i.e.,
Fig. 5); whereas, the prediction and comparison of biomass
blended with Si/Al/K additives should be based on the K2O–SiO2–
Al2O3 ternary phase diagrams constructed on basis of biomass by
additions of K2O, SiO2 and Al2O3 oxides (i.e., Fig. 6), and anyone
of the three K2O–SiO2–Al2O3 ternary phase diagrams (either total
K2O–SiO2–Al2O3, or water insoluble K2O–SiO2–Al2O3, or water soluble K2O–SiO2–Al2O3) can provide high precision prediction.
Similarly, it can be seen from Fig. 6 that there also exist some
singular zones where FT holds the maximum or minimum temperature. On the whole, it is a low temperature zone when K2O:
SiO2:20Al2O3 is around (0.4–0.7):(0.3–0.6):(0–0.1), and there are
two high temperature zones where K2O:SiO2:20Al2O3 is around
(0–0.2):(0.7–1.0):(0–0.2) and where the normalized ratio of SiO2

is lower than 0.25. The distribution is similar with that presented
in Fig. 5, and the FTs show ‘V’ shapes with increased SiO2, Al2O3,

and K2O, respectively.
As a continuous research on biomass triple slagging (i.e., alkaliinduced slagging, low-temperature silicate melt-induced slagging,
and high-temperature silicate melt-induced slagging) [5,26], this
research focused on high-temperature silicate melt-induced slagging provides useful guidelines for biomass selection, improvement, and slagging prevention during combustion. Further study
will be focused on the acquisition of the quantitative criterion
number that can provide integration guidelines on the triple
slagging.
4. Conclusions
The high-temperature silicate melt-induced slagging during
biomass combustion is studied by additions of SiO2, kaolin, and soil
additives, statistic analysis on the ash properties of thirty biomass
fired in operating power plants, and K2O–SiO2–Al2O3 ternary phase
diagrams of FT constructed on basis of the thirty biomass ash properties and biomass by additions of K2O, SiO2, and Al2O3 oxides,
respectively. Results indicate that:
(1) For high-temperature silicate melt-induced slagging, FT can
be as the evaluate index. The higher the FT is, the lower the
high-temperature silicate melt-induced slagging potential is.
FT increases with increase in K2O and SiO2/K2O, and
decreases with increase in Al2O3, SiO2, SiO2/Al2O3, and
(SiO2 + K2O)/Al2O3. The significances are ordered as: Al2O3 > K2O > SiO2/K2O > SiO2/Al2O3 > (SiO2 + K2O)/Al2O3 > SiO2.
Meanwhile, on basis of the significance order a set of evaluation criteria which can provide qualitative comparison on
the potential of high-temperature silicate melt-induced
slagging is proposed as illustrated in Fig. 4.
(2) The K2O–SiO2–Al2O3 ternary phase diagrams built on pure
biomass ash properties (former, for short) and biomass
added Si/Al/K additives (latter, for short) provide high precision prediction on themselves respectively. However,
because of the oxide monomers and/or refractory minerals
originated from additives directly or generated from
alumina-silication reactions indirectly when biomass
blended with additives, the former K2O–SiO2–Al2O3 ternary

phase diagram underestimates the FT of doped biomass
about 140–190 K, and the latter over-predicts the FT of pure
biomass above 200 K.
(3) The FTs show ‘‘V” shapes with increased SiO2, Al2O3, and K2O
content in biomass ash, respectively. And in the K2O–SiO2–
Al2O3 ternary phase diagrams, there exist some singular
zones where FT is the relatively highest and lowest, i.e.
where the occurrence of high-temperature silicate meltinduced slagging is the hardest or easiest.
(4) In comparison with kaolin, soil which presents the almost
same effect on ash fusion temperatures can substitute for
expensive kaolin served as additives during biomass combustion. However, SiO2 which exacerbates lowtemperature silicate melt-induced slagging by reacting with
KCl into low melting silicates is not suitable for additive.

Acknowledgements
The present work was supported by the National Nature Science
Foundation of China (Grant No. 51406149) and the Fundamental
Research Funds for the Central Universities (Grant No.
2014gjhz08).


Y. Niu et al. / Fuel 179 (2016) 52–59

References
[1] Li L, Yu C, Huang F, Bai J, Fang M, Luo Z. Study on the deposits derived from a
biomass circulating fluidized-bed boiler. Energy Fuels 2012;26:6008–14.
/>[2] Elled AL, Davidsson KO, Amand LE. Sewage sludge as a deposit inhibitor when
co-fired with high potassium fuels. Biomass Bioenergy 2010;34:1546–54.
/>[3] Niu YQ, Tan HZ, Ma L, Pourkashanian M, Liu ZN, Liu Y, et al. Slagging
characteristics on the superheaters of a 12 MW biomass-fired boiler. Energy
Fuels 2010;24:5222–7. />[4] Reichelt J, Pfrang-Stotz G, Bergfeldt B, Seifert H, Knapp P. Formation of deposits

on the surfaces of superheaters and economisers of MSW incinerator plants.
Waste
Manage
2013;33:43–51.
/>wasman.2012.08.011.
[5] Niu Y, Tan H, Hui Se. Ash-related issues during biomass combustion: alkaliinduced slagging, silicate melt-induced slagging (ash fusion), agglomeration,
corrosion, ash utilization, and related countermeasures. Prog Energy Combust
Sci 2016;52:1–61. />[6] Niu Y, Du W, Tan H, Xu W, Liu Y, Xiong Y, et al. Further study on biomass ash
characteristics at elevated ashing temperatures: the evolution of K, Cl, S and
the ash fusion characteristics. Bioresour Technol 2013;129:642–5. http://dx.
doi.org/10.1016/j.biortech.2012.12.065.
[7] Wei XL, Schnell U, Hein KRG. Behaviour of gaseous chlorine and alkali metals
during biomass thermal utilisation. Fuel 2005;84:841–8. />10.1016/j.fuel.2004.11.022.
[8] Blomberg T. A thermodynamic study of the gaseous potassium chemistry in
the convection sections of biomass fired boilers. Mater Corros
2011;62:635–41. />[9] Li B, Sun Z, Li Z, Alden M, Jakobsen JG, Hansen S, et al. Post-flame gas-phase
sulfation of potassium chloride. Combust Flame 2013;160:959–69. http://dx.
doi.org/10.1016/j.combustflame.2013.01.010.
[10] Kokko L, Tolvanen H, Hamalainen K, Raiko R. Comparing the energy required
for fine grinding torrefied and fast heat treated pine. Biomass Bioenergy
2012;42:219–23. />[11] Yang TH, Kai XP, Sun Y, He YG, Li RD. The effect of coal sulfur on the behavior of
alkali metals during co-firing biomass and coal. Fuel 2011;90:2454–60. http://
dx.doi.org/10.1016/j.fuel.2011.02.031.
[12] Nutalapati D, Gupta R, Moghtaderi B, Wall TF. Assessing slagging and fouling
during biomass combustion: a thermodynamic approach allowing for alkali/
ash reactions. Fuel Process Technol 2007;88:1044–52. />10.1016/j.fuproc.2007.06.022.
[13] Liaw SB, Wu H. Leaching characteristics of organic and inorganic matter from
biomass by water: differences between batch and semi-continuous
operations. Ind Eng Chem Res 2013;52:4280–9. />ie3031168.
[14] Niu YQ, Tan HZ, Wang XB, Liu ZN, Liu HY, Liu Y, et al. Study on fusion

characteristics of biomass ash. Bioresour Technol 2010;101:9373–81. http://
dx.doi.org/10.1016/j.biortech.2010.06.144.
[15] De Geyter S, Ohman M, Bostrom D, Eriksson M, Nordin A. Effects of non-quartz
minerals in natural bed sand on agglomeration characteristics during fluidized
bed combustion of biomass fuels. Energy Fuels 2007;21:2663–8. .
org/10.1021/ef070162h.
[16] Li QH, Zhang YG, Meng AH, Li L, Li GX. Study on ash fusion temperature using
original and simulated biomass ashes. Fuel Process Technol 2013;107:107–12.
/>[17] Niu Y, Zhu Y, Tan H, Hui S, Jing Z, Xu W. Investigations on biomass slagging in
utility boiler: criterion numbers and slagging growth mechanisms. Fuel
Process
Technol
2014;128:499–508.
/>j.fuproc.2014.07.03.
[18] Fryda L, Sobrino C, Cieplik M, van de Kamp WL. Study on ash deposition under
oxyfuel combustion of coal/biomass blends. Fuel 2010;89:1889–902. http://
dx.doi.org/10.1016/j.fuel.2009.11.022.

59

[19] Kassman H, Bafver L, Amand LE. The importance of SO2 and SO3 for sulphation
of gaseous KCl – an experimental investigation in a biomass fired CFB boiler.
Combust
Flame
2010;157:1649–57.
/>j.combustflame.2010.05.012.
[20] Steenari BM, Lundberg A, Pettersson H, Wilewska-Bien M, Andersson D.
Investigation of ash sintering during combustion of agricultural residues and
the effect of additives. Energy Fuels 2009;23:5655–62. />10.1021/ef900471u.
[21] Bostrom D, Grimm A, Boman C, Bjornbom E, Ohman M. Influence of kaolin and

calcite additives on ash transformations in small-scale combustion of oat.
Energy Fuels 2009;23:5184–90. />[22] Davidsson KO, Amand LE, Steenari BM, Elled AL, Eskilsson D, Leckner B.
Countermeasures against alkali-related problems during combustion of
biomass in a circulating fluidized bed boiler. Chem Eng Sci
2008;63:5314–29. />[23] Xiong SJ, Burvall J, Orberg H, Kalen G, Thyrel M, Ohman M, et al. Slagging
characteristics during combustion of corn stovers with and without kaolin and
calcite. Energy Fuels 2008;22:3465–70. />[24] Dayton DC, Jenkins BM, Turn SQ, Bakker RR, Williams RB, Belle-Oudry D, et al.
Release of inorganic constituents from leached biomass during thermal
conversion. Energy Fuels 1999;13:860–70. />ef980256e.
[25] Wang L, Skjevrak G, Hustad JE, Gronli MG. Effects of sewage sludge and marble
sludge addition on slag characteristics during wood waste pellets combustion.
Energy Fuels 2011;25:5775–85. />[26] Niu Y, Zhu Y, Tan H, Wang X, Hui Se, Du W. Experimental study on the
coexistent dual slagging in biomass-fired furnaces: alkali- and silicate meltinduced slagging. Proc Combust Inst 2015;35:2405–13. />10.1016/j.proci.2014.06.120.
[27] Mu L, Zhao L, Liu L, Yin H. Elemental distribution and mineralogical
composition of ash deposits in a large-scale wastewater incineration plant: a
case study. Ind Eng Chem Res 2012;51:8684–94. />ie301074m.
[28] Brus E, Ohman M, Nordin A. Mechanisms of bed agglomeration during
fluidized-bed combustion of biomass fuels. Energy Fuels 2005;19:825–32.
/>[29] Pettersson A, Amand L-E, Steenari B-M. Chemical fractionation for the
characterisation of fly ashes from co-combustion of biofuels using different
methods for alkali reduction. Fuel 2009;88:1758–72. />10.1016/j.fuel.2009.03.038.
[30] Kai X. The influence of sulfur on the migration characteristics of alkali during
biomass and coal co-combustion. In: Safety Technology-Engineering. Shen
Yang Institute of Aeronautical Engineering; 2009.
[31] Wang L, Becidan M, Skreiberg O. Sintering behavior of agricultural residues
ashes and effects of additives. Energy Fuels 2012;26(59):17–29. .
org/10.1021/ef3004366.
[32] Aho M, Ferrer E. Importance of coal ash composition in protecting the boiler
against chlorine deposition during combustion of chlorine-rich biomass. Fuel
2005;84:201–12. />[33] Vassilev SV, Kitano K, Takeda S, Tsurue T. Influence of mineral and chemicalcomposition of coal ashes on their fusibility. Fuel Process Technol

1995;45:27–51. />[34] Lindstrom E, Larsson SH, Bostrom D, Ohman M. Slagging characteristics during
combustion of woody biomass pellets made from a range of different forestry
assortments. Energy Fuels 2010;24:3456–61. />ef901571c.
[35] Xiong SJ, Ohman M, Zhang YF, Lestander T. Corn stalk ash composition and its
melting
(slagging)
behavior
during
combustion.
Energy
Fuels
2010;24:4866–71. />[36] Niu Y, Zhu Y, Tan H, Hui Se, Du W. Experimental study on the synthetic effects
of kaolin and soil on alkali-induced slagging and molten slagging. Energy
Procedia 2014;61:756–9. />