Corrosion Science 76 (2013) 6–26

Contents lists available at SciVerse ScienceDirect

Corrosion Science
journal homepage: www.elsevier.com/locate/corsci

Review

Corrosion in biomass combustion: A materials selection analysis and its

interaction with corrosion mechanisms and mitigation strategies
Renato Altobelli Antunes a,⇑, Mara Cristina Lopes de Oliveira b
a
Engineering, Modeling and Applied Social Sciences Center (CECS), Federal University of ABC (UFABC), 09210-170 Santo André, SP, Brazil
b
Electrocell Ind. Com. Equip. Elet. LTDA, Technology, Entrepreneurship and Innovation Center (CIETEC), 05508-000 São Paulo, SP, Brazil

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

Article history: Corrosion induced by chlorine during biomass combustion is a major drawback to the consolidation of
Received 4 January 2013 this technology. The current literature brings valuable information on these issues. However, a complete
Accepted 7 July 2013 assessment of the corrosion mechanisms, mitigation strategies and materials selection for biomass com-
Available online 16 July 2013
bustion is not encountered. The aim of this work is to develop a materials selection strategy for super-
heater tubes used in biomass combustion, assessing its interaction with both the corrosion
Keywords: mechanisms and mitigation methods. The Ashby’s approach was used with this purpose.
A. Stainless steel
Ó 2013 Elsevier Ltd. All rights reserved.
A. Nickel
C. High temperature corrosion

1. Introduction processing are the most relevant sources for biomass energy from
solid fuels [11]. Demirbas [12] reported that direct combustion and
Fossil fuels are the basis of energy production worldwide. Burn- co-firing of these resources with coal are promising routes for gen-
ing of coal, oil and natural gas represents near 80% of the total eration of electricity. According to Dermibas [13], co-firing biomass
amount of energy used in the contemporary world [1]. The replace- with coal is an effective way of reducing greenhouse gas emissions
ment of fossil fuels by less pollutant alternatives is as an inevitable in comparison with the combustion of pure coal. Moreover, fuel
step into the establishment of sustainable development policies costs and waste generation are minimized whereas soil and water
based on a low-carbon economy. Biomass is a promising energy pollution can be diminished depending on the chemical composi-
source to reduce greenhouse gas emissions [2]. Attractive issues tion of the biomass source. For both direct combustion and co-fir-
regarding the use of biomass are its widespread availability and ing with coal, the preferred feedstocks for energy production are
possibility of production and consumption in a near-neutral CO2 forest residues, bagasse and other agricultural crops [11]. Khalil
basis [3]. According to Bhutto et al. [4], biomass represents 9% of et al. [14] highlighted that these types of biomass are continuously
the global primary energy demand. If one considers the energy de- produced at specific locations which can be possibly in the vicinity
rived only from renewable sources, biomass represents 75% of the of a power plant, being advantageous in regards to the cost reduc-
total amount [5,6]. Recently, Lim and Lee [7] pointed that the pro- tion of fuel transportation. Biomass combustion can be described
jections for the participation of biomass in the renewable energy by the following sequence of events: heating-up, drying and devol-
demand is expected to increase up to 2030. Kirkels [8] described atilization. During devolatilization, volatiles and char are produced.
a scenario of the last 30 years where the application of biomass Next, both the volatiles and char are combusted [15].
as an energy source in Western Europe sharply increased espe- Biomass composition plays a pivotal role in the combustion
cially after 2000, as well as the budget for research and develop- process. The relevance of this subject is thoroughly examined in
ment in this area. In Brazil, the energy production from biomass the excellent reports by Vassilev et al. [16,17]. Carbon, oxygen
increased 18.1% from 2009 to 2010 [9]. and hydrogen are the main constituents of all biomass with little
Energy outputs from biomass can be encountered in three dif- difference among different sources [18]. Inorganic elements such
ferent forms: solid fuels, liquid fuels such as biodiesel and bioeth- as N, P, K, Ca, Mg, Na, Si, S and Cl are always present. Heavy metals
anol and gas fuels produced through the gasification of solid such as cadmium and lead can also be found as trace elements [19].
biomass [10]. Wood, agricultural crops, municipal solid wastes, Biomass fuels such as straw are rich in alkali metals (K and Na)
animal wastes, algae and aquatic plants and waste from food which are mainly present as simple salts and organic compounds
[20] and chlorine. These species are promptly released to the gas
phase during combustion, forming HCl and KCl. High amounts of
⇑ Corresponding author. Tel./fax: +55 11 4996 8241. KCl in the combustion gases are frequently associated with
E-mail address: renato.antunes@ufabc.edu.br (R.A. Antunes).

0010-938X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.corsci.2013.07.013
 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26 7

enhanced deposit formation. This, in turn, will lead to corrosion of of HCl and O2. Next, solid metal chlorides are formed which evap-
superheater tubes in biomass-fired boilers [21]. Sulfur is also in- orates at high temperatures. The subsequent step is the formation
volved in the corrosion-related phenomena [22]. Johansson et al. of the corresponding metal oxide through the reaction of the metal
[23] considered superheaters corrosion as an important obstacle chloride with oxygen. By comparing the Gibbs free energies of
to the development of biomass as a competitive widespread source CrCl2 (286.0 kJ/mol), FeCl2 (232.1 kJ/mol), and NiCl2
for ‘‘green energy’’. (174.2 kJ/mol) at the reference temperature of 600 °C it is clearly
As a response to this complex technological barrier, corrosion seen that NiCl2 is the least reactive species. Regarding the conver-
studies regarding both biomass firing and co-firing have prolifer- sion of the metal chlorides into metal oxides, the equilibrium oxy-
ated in the past few years [24–27]. In spite of the growing interest gen partial pressures p(O2) for the different chloride/oxide
in these technologies, there is still much to be learned in order to reactions indicates the ease with which the reaction can proceed.
accomplish optimized materials performance and combustion NiO needs the highest oxygen partial pressure to be formed from
properties. In this context, corrosion phenomena are of crucial the nickel chloride whereas chromium oxide is more easily con-
importance. This work provides a thorough analysis of materials verted from CrCl2 at lower oxygen partial pressures. The conver-
selection for superheater tubes in biomass combustion, focusing sion of FeCl2 to Fe2O3 occurs at intermediate values of p(O2). The
its interaction with both the corrosion mechanisms and mitigation experimental data revealed that corrosion attack is more severe
methods involved in this field. The first part of the article gives an for iron and chromium than for nickel, confirming the validity of
overview of the corrosion mechanisms in biomass combustion and the active oxidation mechanism.
other related high temperature corrosion problems. In the second Andersson and Norell [31,32] carried out field tests in waste
part, mitigation methods are explored. In the last part, the Ashby’s boilers with the ferritic T22 steel, the austenitic 310 stainless steel,
approach is employed to propose a methodology for selecting a high alloy austenitic stainless steel (Sanicro 28) and a nickel-
materials for superheater tubes in biomass combustion. based alloy (Sanicro 65). The lowest corrosion loss was observed
for the Sanicro 28 alloy which presented the most compact oxide
2. Corrosion mechanisms scale. This was associated with the ability of the continuous oxide
layer at preventing the penetration of chlorine inwards into the
2.1. Active oxidation by chlorine metal surface which supports the active oxidation mechanism. In
this regard, the addition of chromium and nickel as alloying ele-
The deleterious action of chlorine on the operation of super- ments to iron seems to favor the formation of such a protective
heater tubes for combustion of biomass is well documented scale.
[20,28]. Active oxidation of iron and steels occurs when the oxide A materials selection strategy for superheater tubes in biomass
scale on the metallic surface has no protective character and chlo- combustion should entail the application of this knowledge. The
ride-based contaminants strongly accelerate its oxidation rate. chemical composition of the metallic alloy is essential to the for-
Grabke et al. [29] gave the basis for understanding this mechanism. mation of a stable and corrosion resistant oxide scale. Section 4
The main reactions involved in the active oxidation mechanism can deals with this problem by employing the Ashby’s philosophy to
be summarized as follows [29]: select materials for superheater tubes in biomass combustion.

2HCl þ 1=2 O2 $ Cl2 þ H2 O ð1Þ

2.2. Alkali-containing species
2NaCl þ Fe2 O3 þ 1=2 O2 $ Na2 Fe2 O4 þ Cl2 ð2Þ
2.2.1. Alkali chlorides
4NaCl þ Cr2 O3 þ 5=2 O2 $ 2NaCrO4 þ 2Cl2 ð3Þ The alkali content in the fuel has been has been recognized as
an important factor in the corrosion process of superheater tubes
Fe þ Cl2 $ FeCl2 ðsÞ ð4Þ used in biomass combustion and waste incineration plants. Krause
[33] early described the role of alkali metal oxides in the corrosion
FeCl2 ðsÞ $ FeCl2 ðgÞ ð5Þ of boiler tubes. These species, upon reaction with HCl, form alkali
chlorides which can condensate and form deposits on the metal
3FeCl2 þ 2O2 $ Fe3 O4 þ 3Cl2 ð6Þ surface. Corrosion reactions involving solid phase Cl-species in
deposits have been associated with alkali chlorides, especially
2FeCl2 þ 3=2 O2 $ Fe2 O3 þ 2Cl2 ð7Þ KCl in the case of biomass firing. Nielsen et al. [20] described a
mechanism based on the sulfation of alkali chlorides in deposits
Reactions (1)–(3) refer to the formation of chlorine from the
according to:
oxidation of HCl or upon reaction of condensed chlorides with
the oxide scale. In a next step, if chlorine penetrates the oxide scale 2KClðsÞ þ SO2 ðgÞ þ 1=2 O2 ðgÞ þ H2 OðgÞ ! K2 SO4 ðsÞ þ 2HClðgÞ ð8Þ
through pores or cracks, metal chlorides are able to form at the me-
tal/scale interface as shown in reaction (4) for FeCl2(s). This species
2KClðsÞ þ SO2 ðgÞ þ O2 ðgÞ ! K2 SO4 ðsÞ þ Cl2 ðgÞ ð9Þ
has a high vapor pressure at 500 °C, which is a typical temperature
of superheater materials employed for biomass combustion, and The HCl(g) formed through Eq. (8) can be further oxidized, pro-
volatilizes, according to reaction (5). When it diffuses through ducing Cl2(g). Another possibility is that HCl diffuses through the
the cracks and pores of the oxide scale, it can be oxidized, forming deposit, reaching the metal surface and forming volatile metal
Fe3O4 and/or Fe2O3, according to reactions (6) and (7), releasing chlorides, namely, FeCl2 or CrCl2. These species can diffuse out of
chlorine. The cycle can be repeated, thus sustaining the oxidation the deposit toward areas with high partial pressures of oxygen,
of the metallic surface beneath the non-protective oxide scale. forming metal oxides and releasing Cl2(g) or HCl(g). After this step,
The validity of the active oxidation mechanism has been exper- the corrosion process proceeds as in the case of gaseous Cl-species.
imentally investigated by Zahs et al. [30]. These authors performed A similar mechanism has been postulated by Montgomery et al.
a detailed study of the chloridation and oxidation of iron, chro- [34] after conducting field tests in straw-fired boilers.
mium, nickel and their alloys in chloridizing and oxidizing atmo- The generation of Cl2(g) from solid phase Cl-species can occur
spheres at temperatures ranging from 400 to 700 °C. Under by an alternative route. Alkali chlorides would react with the metal
oxidizing atmospheres, chlorine is formed through the reaction scale and not with SO2(g), as shown in Eqs. (10) and (11). Then,
8 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26

corrosion proceeds as described for the case of the sulfation of al- the most resistant alloys. In this scenario, the reactions of iron,
kali chlorides by the action of Cl2(g). chromium and nickel surfaces with alkali chlorides have been
investigated by Cha and Spiegel [47,48]. Alkali chlorides were
2NaClðs; lÞ þ 1=2 Cr2 O3 þ 5=4 O2 ðgÞ ! NaCrO4 ðs; lÞ þ Cl2 ðgÞ ð10Þ
shown to react with the iron surface at 300 °C, enhancing local oxi-
dation whereas this effect was not observed on the nickel surface.
2NaClðs; lÞ þ Fe2 O3 ðsÞ þ 1=2 O2 ðgÞ ! NaFe2 O4 ðs; lÞ þ Cl2 ðgÞ ð11Þ
No visible local reactions between KCl and chromium surfaces
Oxidation reactions involving molten alkali chlorides are also of were identified at 300 °C. When the testing temperature was raised
great relevance for the corrosion phenomena in biomass combus- to 500 °C, KCl particles were found to react locally with the chro-
tion. Nielsen et al. [20] stated that the corrosion rate increases in mium surface, accelerating its oxidation. This behavior was ex-
the presence of molten phases due to an acceleration of the chem- plained by an active oxidation process in which chloride diffused
ical reactions in comparison with solid–solid reaction and to the through the oxide layer to the chromium/oxide interface. These re-
formation of a conductive electrolyte for the onset of electrochem- sults imply that the presence of nickel as an alloying element
ical attack. The feasibility of a corrosion mechanism based on the would be beneficial to avoid excessive oxidation of superheater
metal attack by molten Cl-phases was confirmed by Nielsen et al. tubes during biomass combustion as the biomass-based fuels are
[35]. They conducted laboratory investigations of high tempera- typically rich in alkali species [49,50]. In fact, the protective char-
ture corrosion in straw-fired boilers, using ferritic alloys (X20 acter of oxide scale formed on nickel-containing alloys has been
and AISI 347) as superheater tubes. The samples were covered with highlighted by several authors [51,52].
synthetic (KCl and/or K2SO4) and real deposits and heated to Recent studies by Lehmusto et al. [53] attempted to elucidate
550 °C. The authors proposed that KCl forms a molten phase with the corrosion mechanisms of superheater materials used in bio-
K2SO4 and iron compounds such as FexOy and FeCl2 close to the mass combustion by investigating the role of different metal chlo-
oxide layer on the metal surface. Rapid sulfation of KCl in the melt rides. They studied the ability of BaCl2, CaCl2, KCl, LiCl, MgCl2, NaCl,
produces HCl or Cl2 close to the metal surface. Chlorine easily dif- PbCl2 and ZnCl2 to react with pure metallic chromium powder at
fuses through the oxide layer toward the metal surface, forming 400 °C, 500 °C, 550 °C and 600 °C. The results revealed that BaCl2,
volatile iron chlorides. From this point, the mechanism follows CaCl2 and MgCl2 did not react with chromium upon heating
the same steps of the gaseous Cl-species involved in the active oxi- whereas KCl, NaCl and PbCl2 reacted at 500 °C and at higher tem-
dation mechanism. van lith et al. [36] gave experimental support to peratures. LiCl reacted only at 600 °C. These indications suggested
this mechanism. The presence of KCl in the deposit formed on boi- that the presence of chlorine in the chloride form does not suffice
ler grade TP 347H FG stainless steel made the corrosion attack to explain the initiation of accelerated oxidation because not all the
more severe, resulting in the formation of a more uniform and dee- chlorides studied reacted with metallic chromium. The authors
per internal attack as the KCl content increased. Cha [37] reported highlight that the cation plays a role in the corrosion mechanism.
that the aggressiveness of deposited ashes on superheater materi- However, further studies are necessary to elucidate it.
als tested at 535 °C increased with the alkali content in the ash.
Similar findings have been reported by Sroda et al. [38] and Bros- 2.2.2. Potassium carbonate and hydroxide
sard et al. [39]. Jonsson et al. [40] showed that the corrosion pro- KCl has been unequivocally correlated with the major corrosion
cess of a low alloyed steel was initiated below 400 °C, leading to mechanisms operating during biomass combustion. Recent re-
a fast redistribution of KCl particles and iron ions on the metal sur- search activity has been devoted to the study of other potas-
face. The rapid transport of KCl and corrosion products was due to sium-containing species such as carbonates, hydroxides and
the formation of a thin liquid layer of a KCl/FeCl2 mixture whose sulfates whose corrosion mechanisms are less explored in the
eutectic temperature is at 355 °C. The authors confirmed the very literature.
corrosive nature of KCl, observing that significant areas of the steel The corrosive action of potassium carbonate (K2CO3) on the
surface were covered by thick corrosion products after only 1 h of 304L stainless steel at 500 °C and 600 °C has been found to be very
exposure, while in the absence of this species only a thin oxide film similar to that of KCl. Both agents lead to chromium depletion of
was formed. Pan et al. [41] observed that KCl increased the corro- the protective oxide scale by the formation of potassium chromate
sion rate of Fe–Ni–Al–Cr alloys in air at 650 °C due to the formation according to Eqs. (12) and (13), thus making the steel surface less
of potassium chromate and not of a protective Cr2O3 oxide. Petters- resistant to oxidation [54].
son et al. [42] identified a similar mechanism for the high temper-
1=2 Cr2 O3 ðsÞ þ 2KClðsÞ þ H2 OðgÞ þ 3=4 O2 ðgÞ ! K2 CrO4 ðsÞ þ 2HClðgÞ ð12Þ
ature corrosion of 304 stainless steel samples at 600 °C and
exposed to KCl. However, at 400 °C the chromium oxide layer is
1=2 Cr2 O3 ðsÞ þ K2 CO3 ðsÞ þ 3=4 O2 ðgÞ ! K2 CrO4 ðsÞ þ CO2 ðgÞ ð13Þ
still protective during experiments conducted for 168 h [43]. This
mechanism was also shown to be active for the high-nickel alloy Lehmusto et al. [55] confirmed the aggressive character of
Sanicro 28 [44,45]. The reaction of KCl with the protective chro- K2CO3 to Cr2O3-protected stainless steels. The proposed mecha-
mium oxide led to the acceleration of the corrosion process under nism for the active oxidation of Cr2O3-protected stainless steels
the formation of potassium chromate. Ishitsuka and Nose [46] ob- initiates with the destruction of this passive layer through the
served that the protective Cr2O3 scale can be easily dissolved to reaction with KCl or K2CO3, forming K2CrO4 or K2Cr2O7. In the next
CrO24 in alkali chloride mixtures, thus leading to accelerated corro- step, oxidation continues if there is chloride available to react with
sion of boiler tubes in waste incineration environments. Moreover, the exposed metal but it does not proceed in the absence of chlo-
molybdenum and silicon improved the corrosion resistance of Fe– ride. The investigation has been extended by the same authors by
Cr–Ni during hot corrosion experiments. It is suggested that these considering the combined effect of water vapor and potassium car-
elements and vanadium or tungsten can be effectively used as bonate [56] on the high temperature corrosion behavior of 304L
alloying elements to improve the high temperature corrosion per- stainless steel which is a more realistic environment for biomass
formance of steels used in the manufacturing of boiler tubes. combustion processes than under the dry conditions employed in
The remarkable effect of alkali chlorides in the corrosion pro- their previous work [55]. The main effect of humidity was related
cess of superheater materials raises an issue on how different met- to the increase of the oxide layer thickness in comparison with the
als behave in contact with deposits consisting of these species. The dry conditions whereas the oxide structure was not significantly
propensity to form corrosion products in typical biomass combus- affected. In another publication [56] the same group showed that
tion atmospheres can provide useful guidelines to properly select the corrosive character of KCl was higher than that of K2CO3 for a
 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26 9

high nickel alloy (Alloy 625) and a ferritic stainless steel (10CrMo). Petersson et al. [54] observed that the reaction of K2SO4 with
The corrosion resistance of the nickel-based alloy was higher than the protective Cr2O3 scale formed on 304L stainless steel probes
that of the ferritic stainless steel, showing the beneficial effect of is not thermodynamically favored under simulated biomass com-
nickel as an alloying element in order to produce a more homoge- bustion conditions. This would preserve the protective character
neous and protective oxide scale. The relevance of alkali carbonate of the oxide scale, preventing the depletion of chromium within
to the high temperature corrosion during combustion processes it by the formation of potassium chromate according to reactions
have also been evidenced by other authors [57,58]. (12) and (13) which typically occur when KCl and K2CO3 are in con-
Blomberg [59] extended the analysis of the corrosion mecha- tact with the stainless steel surface. This result points to a low
nisms involved in biomass combustion by considering the rele- aggressive character of K2SO4 when compared to KCl and K2CO3.
vance of KOH in this process. The author suggests that KOH In this regard, the sulfation reaction shown in Eq. (19) could be ex-
condensation and heterogeneous reactions with HCl, SO2, SO3 plored to reduce the corrosion rate of superheater tubes consisting
and CO2 would take place and should be included in experimental of stainless steels.
works and theoretical calculations concerning the high tempera- Yin and Wu [64] have found that SO2 had a positive effect on the
ture corrosion in biomass fired boilers. When KOH reacts in the control of the corrosion rate of a 316L stainless superheater tube in
gaseous phase forming K2CO3, KCl formation on the superheater a biomass-fired boiler operating at 500 °C. This result was ex-
tube surface would occur by condensation of KCl(g) or according plained by the formation of the less aggressive K2SO4, according
to: to the same route shown in Eq. (19). This reaction is accompanied
by the formation of a compact scale consisting of FeSO4 which re-
K2 CO3 ðsÞ þ 2HClðgÞ ! 2KClðsÞ þ 2CO2 ðgÞ þ H2 OðgÞ ð14Þ
duces the mass gain at high temperatures. Davidsson et al. [65]
The proposed route is the initial condensation of KOH on the confirmed the validity of this mechanism in a commercial scale
tube surface, followed by its reaction with the flue gases, according boiler. The addition of SO2(g) to the burning atmosphere led to
to: the formation of K2SO4 according to Eq. (19) whereas the addition
of HCl shifted this accelerated corrosion of the superheater tubes
KOHðlÞ þ HClðgÞ ! KClðsÞ þ H2 OðgÞ ð15Þ
by the reaction shown in Eq. (12).
The beneficial effect of sulfur to the reduction of the corrosion
2KOHðlÞ þ SO2 ðgÞ þ 1=2 O2 ! K2 SO4 ðsÞ þ H2 OðgÞ ð16Þ
rate of superheater materials in biomass combustion has practical
implications. Mitigation methods based on this mechanism can be
2KOHðlÞ þ SO3 ðgÞ ! K2 SO4 ðsÞ þ H2 OðgÞ ð17Þ
advantageously explored by the co-firing of biomass and coal as
will be detailed in Section 3.1.
2KOHðlÞ þ CO2 ðgÞ ! K2 CO3 ðsÞ þ H2 OðgÞ ð18Þ
Yet, KOH(g) is stable at typical superheater surface tempera- 2.4. Temperature-related issues
tures (400 °C to 600 °C) if the K(g)/Cl(g) ratio in the gas phase ex-
ceeds unity. The condensation of KOH(g) to KOH(l) is likely to Corrosion of superheater materials in biomass combustion or
occur on the surface of the superheater tubes which is followed waste incineration plants can be affected by both the metal and
by a heterogeneous formation of K2CO3(s). This, in turn, can possi- flue gas temperatures [66]. The thermodynamic stability of the cor-
bly induce the formation of basic alkali melts on the tube surface rosive species has a remarkable dependence on the combustion
[60,61]. This implies that the protective Cr2O3 film on the surface temperature and, therefore, strongly affects the corrosion process
of stainless steels would be dissolved in contact with alkaline solu- of superheater materials.
tions typically formed in biomass combustion. In this regard, this Nielsen et al. [67] gave a useful analysis of the deposition of
mechanism would be of high relevance to the corrosion processes potassium salts on superheater tubes during straw firing. The ther-
of superheater materials. modynamic stability of potassium species in straw combustion
(Fig. 1) was shown to support and explain the experimental results.
2.3. The role of sulfur As seen in Fig. 1, potassium is present as KCl(s), K2SO4(s) and K2-
OSiO2(s) at low temperatures. KCl(g) and KOH(g) are the stable
In addition to chlorine, sulfur is also released during biomass species at higher temperatures. K2SO4(s) would be the unique con-
combustion. Alkali metal sulfates or calcium sulfates are the main densed species below 1080 °C when sufficient amounts of sulfur
species in the condensed phase while SO2(g) and SO3(g) are found are present in the system. K2SO4 condensed first and, then, KCl
in the gas phase [62]. The effects of sulfur-containing species on condensed on these particles and directly on the deposition probe.
the corrosion mechanisms involved in biomass combustion have
been investigated by several authors.
Hansen et al. [63] assessed the influence of deposit formation
on the corrosion of a straw-fired boiler consisting of stainless steel
tubes. Straw is a chlorine-rich fuel which has been found to cause
severe corrosion problems in boilers used for biomass firing
[49,50]. The formation of K2SO4 was due to the condensation of
gaseous KCl which is sequentially sulfated to K2SO4. In this regard,
KCl is initially formed and then reacts with a sulfur-containing
compound (SO2 or SO3), yielding K2SO4. The sulfation reaction
can be expressed as shown in:
2KCl þ SO2 þ 1=2 O2 þ H2 O ! K2 SO4 þ 2HCl ð19Þ
The thin and dense K2SO4 layer was believed to act as an effec-
tive barrier to progressive corrosion of the superheater. The com-
pact nature of this layer would avoid the diffusion of gaseous
chloride species (HCl or Cl2) to the metal surface. This reaction Fig. 1. Thermodynamically stable potassium compounds in straw combustion
would be favored at high steam temperatures (520 °C). (reproduced from Nielsen et al. by permission of Elsevier Science, UK [67]).
10 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26

According to the experiments conducted at high temperatures available in the flue gas and not SO2(g). Thus, the rate limiting fac-
(1200–1400 °C), the addition of SO2 greatly increased the amount tor would be the oxidation of SO2(g) toSO3(g) [72,75,76]. Additives
of K2SO4 in the condensed deposit with a proportional reduction that promote the formation of SO3(g) would therefore be beneficial
of the KCl particles. As depicted in Sections 2.2 and 2.3, the corro- to slow down the corrosion processes of superheater materials
sion rate of superheater materials depends on the formation of KCl used in biomass combustion. This mitigation method will be dealt
or K2SO4, the lowest rates being associated with the formation of with in Section 3.2.
the latter. In this context, the corrosiveness of the combustion
atmosphere is a function of its temperature. 2.6. Comparison to other types of hot corrosion
Theoretical thermodynamic calculations by Otsuka [68] showed
that chloride salts condense from the vapor phase form the tube Combustion processes associated with coal firing and gas tur-
deposits on superheaters and melt at relatively low temperatures. bine operation lead to hot corrosion of metallic materials. The cor-
When the tube wall temperature exceeds the melting point of the rosion mechanisms entailed in these cases have similarities with
chloride salts corrosion of the superheater tubes becomes more in- those typical of biomass firing. This section aims at highlighting
tense. The most severe corrosion process occurs when the waste such similarities in order to gain a deeper insight into the corrosion
chemistry consists of high Cl, low S, moderate contents of Na and mechanisms involved in biomass combustion.
K. It is noticeable that the flue gas temperature influence prevails
over the tube wall temperature regarding the formation of poten- 2.6.1. Coal combustion
tially corrosive salts on the surface of superheater tubes. The Corrosion problems of metallic components used in coal utiliza-
molecular quantity of vapor-condensed KCl and NaCl increased tion and conversion systems arises from impurities typically
with the flue gas temperature at 550–750 °C whereas it remained encountered in this carbon-based fuel such as mineral ash, sulfur
unchanged with the increase of the tube wall temperature. The and chlorine [77]. The basic mechanisms of fireside corrosion in
study reveals that the severe fireside corrosion of superheater coal-fired boilers have been described by Nelson and Cain [78]
tubes is a consequence of enhanced condensation of KCl and NaCl and later reviewed by Harb and Smith [79]. Sulfidation or sulfida-
salts at high flue gas temperatures. tion-oxidation the main reactions involved in the corrosion attack
Experimental evidence for the strong effect of flue gas temper- [80]. The presence of sulfur at low oxygen partial pressures favors
ature on the corrosion rate of boiler tubes made of carbon steel the corrosion process of superheater tubes because of the difficult
used in waste combustion plants has been given by Brossard formation of protective scales under this condition [79]. Low melt-
et al. [39]. The corrosion rate increases with the flue gas tempera- ing point deposits consisting of sodium iron trisulfate, potassium
ture, especially above 650 °C. Two concomitant corrosion mecha- iron trisulfate or mixtures thereof form and become molten at
nisms act in this situation. First, the formation of molten alkali the typical operating temperatures during coal combustion, form-
phases above the protective oxide scale, which is dissolved in a ing slagging deposits on the surface of the superheater tubes [81].
fluxing mechanism. Second, active oxidation by chlorine should The trisulfates would form by reaction of the alkali sulfate with the
not be disregarded. metal oxide and sulfur trioxide, according to:
The melting temperature of the fly ash particles produced dur-
3Na2 SO4 þ Fe2 O3 þ 3SO3 ! 2Na3 FeðSO4 Þ3 ð20Þ
ing biomass combustion is also of prime importance [62] due to
the increased corrosion rate in the presence of molten phases
3K2 SO4 þ Fe2 O3 þ 3SO3 ! 2K3 FeðSO4 Þ3 ð21Þ
[20]. In this context, the composition of the fly ash and its corre-
sponding deposit on the surface of the superheater material deter- The major fraction of SO3 in the combustion chamber arises
mine if molten or solid phases will be present at a specific from the oxidation of SO2. In this regard the concentration of SO2
temperature. A survey of melting temperatures of relevant com- would control the extent to which reactions (20) and (21) occur.
pounds and eutectic mixtures for biomass combustion can be The alkali sulfates (sodium or potassium) would be formed by
found in [20,69]. The prediction of the melting properties of differ- the reaction of NaCl or KCl from the coal with H2O and SO2 and
ent fly ashes is an important step toward the proper control of mol- O2 in the gas phase, releasing HCl [82]. In this regard, the content
ten salt corrosion of superheater tubes [59]. of Na, K, Cl and S in the coal is of prime importance for the corro-
sion processes in coal-fired boilers. The potassium and sodium
2.5. Formation of fine particles content of coals have been correlated with increasing corrosion
rates of superheater materials by the mechanisms depicted above
Biomass combustion releases fine particles (<1 lm) which con- [83]. This deleterious action of alkali metals in coal-fired boilers
tribute to the formation of deposits on the surface of superheater is, therefore, similar to that of biomass combustion in regard to
materials. The well-known high contents of alkali and chlorine in the high temperature corrosion of superheater materials. The alkali
biomass fuels accelerate the corrosion rate of boiler tubes [70]. content, especially potassium, of biomass is higher and the species
Hence, the aerosols produced during biomass combustion are clo- formed during biomass combustion are different than those
sely related to the corrosion processes of superheater materials formed during coal combustion. Nevertheless, low melting tem-
ensuing great interest in its formation mechanisms. Jimenez and perature phases consisting of alkali-based species are active corro-
Ballester have deeply explored this subject [71–74]. The basic sive agents in both cases. Sulfur and SO2 have been reported to
assumption is that particles form by nucleation/condensation of accelerate the corrosion of superheater steels due to the formation
volatilized matter during the cooling process. Experimental evi- of non-protective oxide scales [84]. Chlorine, even though recog-
dence has been found that K2SO4 nucleates between 1300 °C to nized as playing a part in the corrosion process during coal firing,
900 °C and KCl is not found in this temperature range. KCl was de- has not been clearly correlated with increasing corrosion rates
tected at 560 °C, forming a condensed layer on the already existing [83]. The effects of S and Cl in biomass combustion are opposite
K2SO4 particles [71]. The composition of the deposits originated to these.
from fine particles can be tailored to yield less corrosive species. The stability of the oxide layer formed on the metal surface un-
This can be accomplished by properly controlling the relative der combustion conditions is very important to withstand high
SO2(g) to O2(g) concentration in the flue gas as this ratio strongly temperature corrosion either in coal- or biomass-fired boilers.
affects the relative fraction of alkali chloride and sulfate in the fine The simultaneous addition of chromium and nickel as alloying
particles. The formation of K2SO4 depends on the amount of SO3(g) elements has been shown to improve the oxidation resistance of
 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26 11

ferrous alloys [79]. Even high corrosion resistant alloys can be at- of superheater tubes, depending on its thermodynamic sta-
tacked depending on the flue gas temperature and concentration bility. The corrosion process will proceed with different
of impurities in the coal [77]. Thus, a proper control of coal compo- kinetics in each case.
sition and a conscientious materials selection procedure must be (c) The formation of either aggressive species such as KCl and
implemented to avoid premature tube failure during coal combus- K2CO3 or more protective ones such as K2SO4 in solid depos-
tion. The same recommendation is valid for biomass combustion, its can be controlled by the flue gas composition. However,
as will be discussed in Sections 3 and 4. reliable thermodynamic models to predict the intermediate
gas phase reactions that lead to the formation of each spe-
cies are not yet readily available for real biomass combus-
2.6.2. Hot corrosion in gas turbines
tion conditions. Thus, empirical data have to be obtained
Hot corrosion has been defined as the oxidative attack of metals
in order to confirm the formation of specific compounds.
and alloys upon contact of a thin film of fused salt in an oxidizing
(d) The metallic alloy used as the superheater material gives rise
atmosphere [85]. This process is of particular interest for gas tur-
to different interactions with the combustion environment
bines [86]. The corrosion mechanism is associated with the reac-
depending on its chemical composition. Alloying elements
tion of the metallic surface with a thin film of molten Na2SO4
such as chromium, nickel and molybdenum are known to
which is the dominant compound in the salt due to its high ther-
have strong effects on the corrosion behavior of superheater
modynamic stability [87,88]. The sulfide compounds resulting
materials in biomass-fired boilers. However, the effect of
from this reaction are only produced by the interaction between
other elements with potential to form stable oxide layers
the metal and the fused salt and not upon reaction with sulfur spe-
such niobium, tantalum, vanadium, titanium and aluminum
cies in the gas phase. This mechanism diverges from those ob-
is much less investigated.
served for coal and biomass combustion in which the sulfur
species (SO2 and SO3) in the gas phase markedly participate in
In spite of various unresolved questions ensuing from the points
the corrosion process. The hot corrosion mechanism proposed by
depicted above, the literature provides well-established data
Rapp and Gotto [89] is based on the fluxing of the protective oxide
regarding the corrosion mechanisms in biomass combustion. Ac-
scale. The protective oxide scale dissolves at the oxide/salt inter-
tive oxidation by chlorine, the pernicious influence of alkali chlo-
face and reprecitates within the salt film in the form of non-protec-
rides and alkali carbonates, the beneficial effect of adding SO2 to
tive particles. In this context, oxide solubilities in Na2SO4 have to
the flue gas, the protective character of the Cr2O3-based scales
be carefully assessed. Hot corrosion will proceed if a negative sol-
and the beneficial action of nickel as an alloying element have been
ubility gradient is established by the reduction of the solubility of
shown by several authors as described in the previous sections.
the oxide in the salt/gas interface, thus leading to the precipitation
The main question which is pursued in this work is how this
of the non-protective particles within the salt film [90]. Chromium
information can be used to support materials selection for super-
has been found to play a major role in the protection ability of the
heater tubes used in biomass combustion. Furthermore, how mit-
oxide scale formed during hot corrosion. Cr2O3 does not satisfy the
igation methods derive from the knowledge accumulated so far
negative solubility gradient criterion in the salt film and does not
regarding the corrosion mechanisms involved in biomass corro-
support the sustained hot corrosion process. Moreover, the forma-
sion? Next section deals with this question. It can be anticipated
tion of solid deposits consisting of Cr2O3 can block reducing sites
that the beneficial effect of SO2 resulted in the development of
such as grain boundaries and flaws at the salt/metal interface, pro-
co-combustion processes based on the concomitant burning of
tecting the metallic material from hot corrosion [90].
coal and biomass or on the addition of sulfur-based additives.
In spite of the different mechanisms involved in the high tem-
Furthermore, modifying the surface of the superheater materials
perature corrosion resulting from biomass combustion and the
by depositing protective coatings based on the formation of cor-
hot corrosion described by the Rapp–Gotto criterion, one common
rosion resistant oxide scales is also a possible mitigation route.
feature must be highlighted. The formation of Cr2O3 scales on the
This route is based on the fact the Cr2O3-forming alloys and nick-
surface of the metallic alloys increases the ability to withstand
el-containing alloys have increased corrosion resistance under
the operating conditions in both cases. This feature can be explored
biomass firing conditions. Hence, the use of co-combustion, addi-
to properly design and select the engineering alloys which will be
tives and coatings as mitigation methods for preventing corrosion
used in these environments, irrespective if they are Fe-based or Ni-
during biomass combustion is explored in the next section. The
based alloys. This issue will be developed in Section 4.
analysis of the materials selection problem will be developed in
Section 4.
2.7. Overview and the need for protection methods

The aim of the foregoing sections was to give the reader an 3. Mitigation strategies
overview of the main corrosion mechanisms involved in biomass
combustion and other related high temperature processes. The 3.1. Co-combustion
complexity of this scenario can be promptly realized by consider-
ing the following points: Avoiding chlorine deposition on the surface of the superheater
tubes is a traditional method of preventing corrosion during bio-
(a) Although potassium plays a major role in the corrosion of mass combustion [91]. The sulfation of metal chlorides is an
metallic materials in biomass-fired boilers, several other important reaction in this scenario. Sulfur dioxide reacts with alka-
species can participate in this process such as Na, S, Cl, Si, li chlorides, entrapping the alkali and releasing HCl according to
Ca and Mg. The chemical composition of biomasses from dif- Eq. (8). In this regard, sulfur-rich fuels can be used to prevent chlo-
ferent sources and locations can vary significantly, leading to rine deposition by properly mixing with chlorine-rich biomasses.
the formation of different compounds during the burning Aluminum–silicates acts through a similar mechanism.
operation. Aho and Silvennoinen [92] studied the prevention of biomass
(b) The temperatures at which the combustion process is car- ash-related corrosion by co-combusting biomass fuels. The fuels
ried out will determine if a species will be released into were chosen based on the elemental composition, aiming at achiev-
the gas phase or will be in the molten state at the surface ing stoichiometric relations that would favor the sequestration of
12 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26

alkalis from the alkali chlorides, according to Eqs. (22) and (23), et al. [101] have also pointed out the effectiveness of this treat-
where M = K, Na: ment at reducing chloride-induced corrosion of superheater mate-
rials in wood-fired boilers. Recently, Kassman et al. [102,103]
2MCl þ SO2 þ 1=2 O2 þ H2 O ! M2 SO4 þ 2HCl ð22Þ
confirmed that the use of ammonium sulfate as an additive re-
duced the content of chlorine in deposits to negligible amounts
Al2 O3  SiO2 þ 2MCl þ H2 O ! M2 O  Al2 O3 :2SiO2 þ 2HCl ð23Þ
during biomass combustion in a large-scale circulating fluidized
Pine bark, agricultural waste (AGW) and pulp sludge were se- bed boiler. Viklund et al. [104] have found that ammonium sulfate
lected to the study. Depending on the initial contents of alkali, sul- greatly reduced the amounts of chlorine in the flue gas and depos-
fur, Al2O3, chlorine and SiO2 in the fuels the authors showed that its on cooled superheater probes during combustion experiments
co-combustion favored the inhibition of chlorine deposition espe- in which mixtures of municipal and industrial wastes were burned.
cially by the mechanism shown in Eq. (23). In another investiga- Schofield [105] showed that molybdenum salts can avoid the for-
tion, Aho and Ferrer [93] studied the effect of coal ash mation of pernicious alkali chlorides deposits in combustors by
composition in protecting the boiler against chlorine-induced cor- the formation of the more stable Na2Mo2O7 or K2Mo2O7
rosion during combustion of meat and bone meal (MBM). They ob- compounds.
served that high aluminum and silicon concentrations in the coal Kaolin has been reported as an additive for biomass combustion
were beneficial to prevent chlorine deposition. The formation of al- by several authors [106–108]. The aluminosilicates in kaolin will
kali aluminosilicates (Eq. (23)) was considered to be the active path bind potassium according to the possible reactions shown below
of alkali sequestration, avoiding alkali chloride deposition. The [109]:
presence of sulfur did not prevent chlorine deposition evidencing
Al2 Si2 O5 ðOHÞ4 þ 2KCl ! 2KAlSiO4 þ H2 O þ 2HCl ð27Þ
the little effect of the sulfation reaction in comparison with the for-
mation of alkali aluminosilicates.
Al2 Si2 O5ðOHÞ4 þ 2KCl þ 2SiO2 ! 2KAlSi2 O6 þ H2 O þ 2HCl ð28Þ
The protection mechanism based on Eq. (23) was also observed
by other authors [94–97]. Zheng et al. [98] have also confirmed the
Al2 Si2 O5 ðOHÞ4 þ K2 SO4 ! 2KAlSiO4 þ 2H2 O þ SO3 ð29Þ
viability of this co-combustion strategy. They gave a further under-
standing on this mechanism by showing that Ca and Mg play
Al2 Si2 O5 ðOHÞ4 þ K2 SO4 þ 2SiO2 ! 2KAlSi2 O6 þ 2H2 O þ SO3 ð30Þ
important roles in the alkali sequestration ability of the coal/bio-
mass blend. The Ca/Si mole ratio determines the availability of The compounds KAlSiO4 (kalsilite) and KAlSi2O6 (leucite) have
Ca to react with silicates according to the following reactions: high melting points (1600 °C for pure kalsilite and 1500 °C for pure
leucite), leading to less corrosion problems in comparison with the
CaOðsÞ þ SiO2 ðsÞ ! CaO  SiO2 ðsÞ ð24Þ
low melting point KCl and K2SO4 compounds that are likely to cre-
ate deposits on the surface of superheater [110].
CaOðsÞ þ SiO2  Al2 O3 ðsÞ ! CaO  SiO2  Al2 O3 ðsÞ ð25Þ
The high chemical inertness and relatively small specific surface
A high content of calcium and magnesium has a negative effect of kaolinite are limiting factors for binding the alkalis present in
on the binding of potassium and, consequently, on the formation of biomasses. Mroczek et al. [111] gave a valuable contribution to-
the potentially corrosive KCl compound. This occurs because of the ward the solution of this challenge. They replaced kaolinite by hal-
possible reaction of calcium and magnesium with aluminosilicates, loysite as an additive during combustion of different biomasses.
so that this last species becomes less available to react with KCl, The advantage of using halloysite, an aluminosilicate clay mineral
releasing chlorine as HCl to the gas phase. In this regard, K/Si (Al4(OH)8Si4O1010H2O), instead of kaolinite resides in its mineral-
and K(S + Si) mole ratios are very important to reduce the chlo- ogical structure. This structure is represented in Fig. 2 and consists
rine-induced corrosion problems of the co-combustion process. of Si tetrahedra and Al octahedra, forming a single plate which is
The increase of the S/Cl mole ratio in the fuel favors the formation separated from the next layer by a space in which potassium cat-
of the less sticky and higher melting point K2SO4 instead of KCl due ions can get intercalated. The high reactivity of halloysite can be
to the sulfation reaction of KCl. These observations reveal that a explained by the presence of two hydroxyl groups: one located
proper control of the fuel composition is imperative to avoid chlo- internally between tetrahedral and octahedral layers and another
rine-induced corrosion problems during the co-combustion of coal superficial group located on the octahedral layer. Such disposition
and biomass. gives rise to an interesting bonding structure where the bonds be-
tween ions are allowed to occur both on the surface layers and in-
3.2. Additives side the crystal. Mroczek et al. showed that suitable blending of
halloysite with biomass can reduce the high temperature corrosion
Additives can reduce chlorine-induced corrosion during bio- during combustion, effectively diminishing KCl and NaCl concen-
mass combustion by two different routes: (i) prevention of gaseous trations in the ash deposits.
KCl release or (ii) reaction with KCl forming less corrosive species
[99]. Sulfur-containing additives are suitable to this purpose as 3.3. Coatings
they can react with alkali chlorides yielding less corrosive alkali
sulfates. The traditional concept based on this approach is a pat- The troublesome situation of high temperature corrosion of
ented process known as ChlorOut [100]. In this process an aqueous metallic materials induced by chlorine during biomass combustion
solution of ammonium sulfate is sprayed in the turbulent zone at and co-combustion can be effectively managed by employing
the entrance of superheaters to induce the sulfation of gaseous al- protective coatings. Schütze et al. [112] identified key factors for
kali chlorides in the flue gas. The reactions proceed firstly by protective coatings used in high temperature applications. The
decomposition of ammonium sulfate into NH3 and SO3, according phases that constitute the coating layer should be thermodynami-
to the reaction shown below: cally stable in the operating environment. Interdiffusion between
coating and substrate plays a key role in the protective character
ðNH4 Þ2 SO4 ðaqÞ ! 2NH3 ðgÞ þ SO3 ðgÞ þ H2 OðgÞ ð26Þ
of the coating layer. It should occur as slowly as possible to avoid
The next step is the sulfation of alkali chlorides by the reactive premature coating failure. Moreover, the coefficients of thermal
SO3(g). The effectiveness of this method at reducing the corrosion expansion of the substrate, coating and scale should be as close as
rate and deposit was confirmed by Broström et al. [99]. Henderson possible in order to prevent thermal stresses during cooling/heating
 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26 13

Fig. 2. Scheme showing the structure of halloysite with the possible sites for absorbed ions (reproduced from Mroczek et al. by permission of Elsevier Science, UK [111]).

cycles. In addition to these aspects, the coating process is also of mechanisms involved in biomass combustion as depicted in Sec-
paramount importance to the performance of the protective layer. tion 2 of the present work.
The deposited film should have a dense and flat splat structure to Aluminide coatings produced by pack cementation have been
resist corrosive environments at high temperatures. As the corro- explored as another protection route for superheater alloys [127].
sive species mostly diffuse through pores and along splat bound- This method is an in situ chemical vapor deposition (CVD) process
aries a dense and flat splat structure increases the distance from in which the substrate surface is enriched with elements that are
the coating surface to the coating/substrate interface thereby slow- able to form a protective oxide layer in service such as Al, Cr, Si
ing down the corrosion process of the coated substrate [113]. These and combinations thereof. The coating is produced by diffusion
general guidelines constitute a complex materials selection prob- of these elements at high temperatures. The pack cementation
lem. The designer must deal with the interaction between substrate mixture consists of the metal elements to be deposited, halide acti-
and coating, seeking for a compatible thermal expansion behavior vators and inert filler such as alumina [128]. The formation of
without neglecting the thermodynamic stability of the phases that Al2O3 on the surface of the treated substrate accounts for its en-
constitute the coating. Furthermore, the selected coating material hanced oxidation resistance.
must be produced as densely as possible to be an effective barrier Films obtained by physical vapor deposition (PVD) methods are
against corrosion. Thus, a proper deposition method should be also hardly considered for the high temperature corrosion protection of
carefully selected to achieve the required morphological architec- metallic materials used in biomass firing or coal-biomass co-firing.
ture for the coating material. Leyens et al. [129] showed that, depending on the microstructure
In light of this challenging scenario, several authors have pur- obtained after deposition, the anticorrosion performance of the
sued efficient coatings to prevent the high temperature corrosion PVD layer can be sustained for long periods in contact with Na2SO4.
of metallic materials in biomass fueled boilers and waste incinera- However, intrinsic defects such as pinholes and macroparticles
tors [114–116]. Hearley et al. [117] showed that coatings produced [130] which are typical of conventional PVD processes should be
by the high velocity oxygen fuel (HVOF) thermal spray process avoided. In this regard, multilayered coatings should provide en-
form effective barrier against high temperature corrosion. This hanced resistance to high temperature corrosion. This coating
deposition method is frequently reported for the production of architecture allows the formation of tailored films with a dense
coatings designed to resist high temperature oxidation in a variety and compact structure [131], improving the barrier effect against
of industrial applications [118–120]. The attractive corrosion resis- corrosive species. This approach can be advantageously explored
tance of HVOF coatings comes from the intrinsic film characteris- to develop effective coatings to protect metallic materials against
tics such as high density, low oxide levels and increased high temperature corrosion in biomass combustion applications.
thickness [121] which make them superior to coatings obtained
by other spraying processes such as plasma spraying [122]. The 3.4. Nanostructured materials
protection ability of HVOF coatings can be further improved by
post-treatments that remelt the deposited layer, removing inter- Nanotechnology has emerged as a hot topic in corrosion re-
connected porosity and oxides from the splat boundaries. Laser search over the last two decades. The interest of developing nano-
melting is reported to be an effective post-treatment of HVOF- structured materials comes from the possibility of achieving
deposited Ni-based coatings under biomass combustion conditions improved mechanical and physical properties in comparison with
[123,124]. the conventional polycrystalline counterparts. Fatigue and wear
In addition to the coating microstructure, its chemical composi- properties, as well as static mechanical strength have been re-
tion is also of great relevance to the corrosion protection ability ported to be enhanced by nanostructuring [132–134]. The corro-
during biomass combustion. Studies by Mayoral et al. [125] and sion resistance of metallic materials can be also positively
Torrell et al. [126] showed that the simultaneous presence of chro- affected by nanostructuring. This can be achieved by two different
mium and nickel in the coating layer is beneficial to confer a reli- approaches. The first one is based on the top-down method in
able protective character. Coatings that allow the formation of which an existing coarse-grained material is processed to produce
homogeneous oxide scales consisting of Cr2O3 and Ni-based oxides a nanostructured alloy by means of a significant grain refinement.
decreases the corrosion rate of the metallic substrate. This is in The second one is based on a bottom-up method in which a nano-
accordance with oxidation behavior described by the corrosion structured material is assembled from nanoscale building blocks
14 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26

such as nanoparticles [135]. This is the case of nanostructured bulk diffusion [150]. Atomic diffusion is facilitated for nanocrystal-
coatings. Both approaches are discussed inthis section, focusing line materials [151]. Consequently, creep is likely to occur at lower
the effect on the corrosion resistance of metallic alloys and their temperatures and stress levels [152]. The lack of investigations on
possible applicability for producing improved superheater materi- the creep behavior of nanocrystalline metals has been highlighted
als for biomass combustion. by several authors [153,154]. These reports generally confirm that
the creep strength decreases with the grain size reduction whereas
3.4.1. Top-down approach the ductility has an opposite trend. This behavior constitutes a fur-
The top-down approach has been developed mainly by severe ther concern for the application of nanocrystalline metals as super-
plastic deformation (SPD) processes such as equal channel angular heater tubes in biomass-fired boilers. In addition to high
pressing (ECAP), accumulative roll-bonding (ARB) and high pres- temperature corrosion, experiments and reliable models of the
sure torsion (HPT) [136,137]. These processes are capable of pro- creep behavior of nanocrystalline metallic alloys are still of limited
ducing bulk nanocrystalline metallic alloys for structural availability in the current literature. In one hand, this scarceness of
applications [138]. The conventional feedstock for the SPD meth- information may constitute a barrier for the prompt adoption of
ods is in the form of bars, rods or sheets. However, tubular compo- nanocrystallization through SPD processes as a viable method to
nents have also been produced using SPD processes [139,140]. avoid high temperature corrosion of superheater tubes. But, on
Recently, Faraji et al. [141,142] proposed new SPD methods for the other hand, it also provides an irrefutable opportunity of con-
obtaining tubular parts called tubular channel angular pressing ducting unprecedented research studies on the structural applica-
(TCAP) and parallel tubular channel angular pressing (PTCAP). tions of bulk nanocrystalline metals.
These processing routes would be of interest to the production of In parallel to the development of bulk nanocrystalline materials
superheater tubes for application in biomass-fired boilers. Not- by SPD processes, the concept of surface nanocrystallization has
withstanding, it is important to notice that most part of the inves- been explored as well [155]. Different techniques are employed
tigations regarding the production of bulk nanocrystalline metallic to produce nanocrystalline surfaces in metallic materials such as
materials by SPD methods are focused on magnesium or aluminum high energy shot peening, ultrasonic shot peening and surface
alloys [143,144]. As will be highlighted in Section 4, the super- mechanical attrition treatment (SMAT) [156–158]. These pro-
heater alloys employed for biomass combustion purposes are cesses, generally designated as surface severe plastic deformation
mostly stainless steels or nickel alloys. If one considers the pro- (S2PD) methods, have been developed as a tentative to overcome
cessing of these materials by SPD methods, the available reports manufacturing obstacles to the widespread industrial production
are much scarcer than those for either Al or Mg alloys. This is espe- of bulk nanocrystalline metals regarding both the size and cost of
cially true for nickel alloys or even for pure nickel [145]. One finished parts. SMAT has been particularly applied to attain surface
exception can be found in the report by Kim and Kwun [146] nanocrystalline aluminum, magnesium, nickel, titanium and stain-
who successfully processed a Ni-based alloy (IN718) by ECAP. less steel components [159–163]. In this technique surface nano-
SPD-processed stainless steels, in turn, are more frequently re- crystallization is achieved by shots or milling balls impacts, thus
ported [147,148]. Zheng et al. [149] prepared nanocrystalline 304 promoting grain refinement induced by plastic deformation
stainless steel by ECAP and evaluated the resulting corrosion resis- [164]. The effects of SMAT on the room temperature corrosion
tance in comparison to a conventional microcrystalline alloy. The resistance of metallic alloys have received much attention
authors underlined that corrosion studies of SPD-processed stain- [159,165,166]. Depending on the processing parameters corrosion
less steels are very scarce in the open literature. The ECAPed sam- resistance can be either improved or diminished [167,168]. Surface
ples showed higher corrosion resistance than the as-received alloy. defects induced by plastic deformation can decrease corrosion
This result was attributed to the improvement of the compactness resistance whereas the stability of the passive film formed on the
and stability of the passive film after the ECAP processing. Enrich- nanocrystalline surface can be enhanced due to the faster diffusion
ment of chromium in the passive film was not observed. However, of passivating elements, thus increasing the corrosion resistance of
it is important to notice that the experiments were conducted at the alloy. While room temperature investigations are commonly
room temperature. Thus, if high temperature tests had been per- carried out to assess the performance of nanocrystalline metallic
formed it would be possible that diffusion of chromium through surfaces against corrosion, high temperature studies have been
the nanocrystalline grains would had been favored, thereby hardly reported. It is generally agreed that the properties of nano-
increasing the concentration of this element in the passive film. crystalline surfaces depend on their thermal stabilities [169]. It is
Whilst room temperature corrosion is of little relevance to the well-known that nanocrystalline materials have an excess of free
metallic alloys employed as superheaters in biomass-fired boilers, energy which is associated with the high numbers of interfaces
the improved corrosion resistance of SPD-processed metals can be at grain boundaries. Thus, they are in a metastable state and, upon
regarded as an opportunity to develop new and optimized materi- heating, thermally induced grain growth is likely to occur, affecting
als for this application. In this context, high temperature corrosion the physical properties of the treated surface [170]. This effect is
studies of SPD-processed metals are a fertile research field yet to be dependent on both the metallic material and the method em-
explored. Since the compactness and stability of the passive film is ployed for nanocrystallization. To the best of our knowledge, these
altered by the severe plastic deformation process, it is likely that aspects are not investigated in the current literature related to the
the corrosion behavior of the metallic alloy can be improved even superheater materials used in biomass combustion.
at high temperatures. Moreover, diffusion of passivating elements Attempts of clarifying the high temperature stability of S2PD-
toward the passive film could give a further enhancement for the treated metals can be found in the recent report by Wen et al.
corrosion resistance. [171]. These authors obtained nanocrystalline surface in commer-
Nevertheless, in addition to corrosion resistance, the mechani- cial titanium plates using SMAT and studied their oxidation behav-
cal strength at elevated temperatures is an important issue to be ior between 500 and 700 °C. The SMAT treatment formed
addressed when designing metallic materials for structural appli- numerous grain boundaries and dislocations on the surface of the
cations. Creep strength and ductility are the key properties for this titanium plates. The presence of these crystalline imperfections al-
situation. Creep deformation is a time-dependent phenomenon lowed the formation of a thicker and denser oxide layer in compar-
which is active at temperatures above 50% of the material’s melt- ison to that formed on the coarse grained material. The formation
ing point (Tm). It occurs by diffusion and dislocation motion of a thick and dense oxide layer can be beneficial to avoid corrosion
through mechanisms associated with grain boundary sliding or induced by aggressive ions such as chlorine, provided that it does
 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26 15

not spall from the metallic surface. This effect is similar to that of a crystalline phases. Thus, the microstructure of the coating can be
protective coating with a dense and compact structureas those de- easily tailored during the deposition process. This can be advanta-
scribed in Section 3.3 for the protection of superheater tubes geously used to protect superheater materials with nanocrystalline
against chlorine-induced high temperature corrosion during bio- dense PVD films used in biomass fired-boilers.
mass combustion. However, these results are referred to titanium
alloys. Typical superheater materials for biomass combustion are
mainly stainless steels or nickel-based alloys. For these materials 3.5. Relationship between corrosion mechanisms and mitigation
high temperature corrosion studies are not encountered for methods: the outputs for materials selection
SMAT-treated samples. In this context, research activity in this
area is needed, especially focusing typical biomass combustion At this point, it is important to identify how the mitigation
environments. The potential of S2PD processes to improve the oxi- methods described in this section are related to the corrosion
dation resistance of superheater materials used in biomass com- mechanisms reviewed in Section 2. Ultimately, this information
bustion is yet to be explored. will serve as the basis for the materials selection methodology
developed in Section 4.
3.4.2. Bottom-up approach In this respect, two major aspects regarding the corrosion
The bottom-up approach is focused on the development of mechanisms involved in biomass combustion have to be high-
nanocrystalline coatings from nanocrystalline powders such as in lighted. The most critical issue is the remarkable effect of alkali
the HVOF process or spray-based processes or on nanocrystalline chlorides, especially potassium chloride, in the enhancement of
films grown directly on the substrate such as physical vapor depo- the corrosion rate of the metallic alloys used as superheater tubes.
sition or chemical vapor deposition methods (CVD). Nanostruc- Reactions between the metallic surface and this species, or other
tured films based on Ni–Cr alloys have been reported to increase alkali-containing species such as carbonates, have to be avoided
the oxidation resistance of metallic substrates. This is due to the in order to increase the corrosion resistance of the superheaters.
development of a dense and compact film structure which acts The second aspect to be emphasized is that the formation of a
as an effective barrier against penetration of corrosive compounds dense and compact oxide film decreases the corrosion rate of
such as Na2SO4 or Na2SO4–K2SO4 mixtures. Moreover, the fast for- superheater materials under biomass combustion conditions and
mation of a protective Cr2O3 oxide scale on the surface of the nano- this is favored for Cr2O3-forming alloys. A further contribution to
crystalline coating is believed to be responsible for its increased decreased corrosion rates can be obtained when nickel and chro-
hot corrosion resistance [172–174]. According to Yu et al. [175] mium are concomitantly present as alloying elements.
this is also valid for Al-containing Ni-based nanocrystalline coat- Mitigation methods based on the co-combustion of coal with
ings which form a Al2O3 protective scale more rapidly than their biomasses are designed to prevent the reaction of the metallic sur-
conventional microcrystalline counterparts. Furthermore, rare- face with alkali chlorides by the formation of less aggressive alkali
earth elements such as La and Ce and their oxides are reported sulfates. The effectiveness of this method depends mainly on the
to increase the hot corrosion resistance of Al2O3 and Cr2O3-forming sulfur content of the coal and also on the alkali and chlorine con-
coatings by the refinement of the microstructure which facilitates tent of the biomass. The use of additives is also based on this phi-
the formation of protective scales [176,177]. losophy. Either Sulfur-containing or aluminosilicates-containing
Rahman et al. [178] studied the hot corrosion behavior of nano- additives can sequestrate the alkali species during biomass com-
crystalline Cr/Co–Al coatings on a nickel super alloy substrate. The bustion, forming alkali sulfates or aluminosilicates, instead of alkali
coatings were deposited using PVD-based method. The nanostruc- chlorides, thus avoiding corrosion.
tured coatings inhibited the movement of anions and cations, thus Coatings, in turn, rely on the formation of a dense layer on the
providing high hot corrosion resistance. This high performance was surface of the metallic substrate to protect it from corrosion. The
also supported by the dense structure of the coating. These results dense film avoids penetration of aggressive species such as alkali
evidence the beneficial effect of nanostructuring for improving the chlorides that would otherwise react with the substrate. Further-
corrosion resistance of the Ni-based alloy. more, the stability of the coating layer depends on its chemical
Another route for obtaining dense nanocrystalline coatings on composition. In this regard, nickel-based coatings and Cr2O3-form-
metallic substrate has been developed by Mohammadnezad et al. ing alloys are preferable to protect superheater materials against
[179]. They employed mechanical alloying to obtain nanostruc- corrosion during biomass combustion.
tured Ni–Al coatings on carbon steel substrates. The results Nanostructured materials obtained from either the top-down or
showed that the proper control of processing parameters such as bottom-up approaches can also be used to mitigate corrosion prob-
milling time can lead to the formation of dense nanostructured lems in biomass combustion. In this case, the increased corrosion
films with enhanced hardness. The high temperature corrosion resistance of nanostructured surfaces arises from the more rapid
resistance of the coated-steel was not evaluated in this work. How- formation of protective oxide scales. Obviously, the protective
ever, the dense structure of the nanostructured film points toward character of the scale will depend on its chemical composition
a possible positive effect of the mechanical alloying process on this and compactness. Once more, nickel and chromium would be man-
property. In the same way, DC/RF magnetron sputtering was suc- datory for a satisfactory performance against corrosion.
cessfully used by Rahman et al. [180] to produce nanocrystalline The effectiveness of protection methods can be improved if they
Ni–Al coatings with high corrosion resistance at elevated temper- are combined with the proper selection of superheater materials.
atures. The excellent performance of the Ni–Al was ascribed to a This, in turn, can be affected by the combustion operating condi-
dense morphology of oxide layer developed during oxidation of tions, such as fuel type, fuel composition and temperature. In this
the nanosized grains, forming a protective scale consisting of regard, corrosion mechanisms, mitigation methods and materials
Cr2O3, Al2O3, NiO, Fe2O3 and NiCr2O4. Zhong et al. [181] showed selection are correlated. Indeed, materials selection itself could
that not only nanocrystalline Ni–Al but also nanocrystalline Ni– be envisaged as a mitigation method, as a proper choice of materi-
Al–N coatings can be produced by RF magnetron sputtering with als can greatly decrease the corrosion problems in specific environ-
dense structures tailored by the proper control of the substrate ments. A direct consequence of the important points raised in this
bias and showing increased oxidation resistance. It is interesting section, is that any materials selection strategy should encompass
to note that substrate bias influences the ion bombardment during the chemical composition of the metallic alloy, in order to seek for
deposition process and, consequently, controls the grain size of the those which will allow the formation of stable oxide scales under
16 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26

biomass combustion conditions. This topic will be unraveled in the the material selection chart. The materials which do not meet the
next section. technical requirements established by the constraints are thereby
easily screened.
The objective(s) is used to rank the remaining candidates, al-
4. Materials
ways seeking to maximize a given performance. Ashby proposes
that the objectives define material indices, for which extreme val-
The proper selection of metallic materials is a must-attend issue
ues are sought. The definition of the material indices is done by the
focusing the successful long-term operation of structural engineer-
derivation of an objective function. This function has the general
ing parts in a variety of industrial applications. One of the main
form presented in:
concerns regarding biomass combustion resides in the corrosion
behavior of superheater tube materials. In this context, we firstly P ¼ f ðF; G; MÞ ð31Þ
present a list of the metallic alloys employedby different authors
P is the performance of the material which is a function (f) of three
as superheater materials for biomass combustion, displaying the
different elements: functional requirements (F) that are often some
chemical composition of each alloy. This information, shown in Ta-
measurable quantity not directly expressing a material property.
ble 1, is important to guide the selection of the best candidates for
Examples are the heat flux through a bar, the load it carries, the
the application. Next, a selection strategy based on technical crite-
temperature at which it is exposed, the electrical current flowing
ria to the design of metallic alloys with improved corrosion resis-
through it and so on. The second element encompasses the geomet-
tance under the typical conditions of biomass combustion is
ric parameters (G) of the component such as length, diameter, thick-
discussed.
ness or width. The last element comprises the materials properties
(M), forming the so-called material index. Deriving M involves the
4.1. Materials selection strategy identification of proper equations relating the materials properties
that are relevant to the application under study. Ashby gives a deep
The development of a materials selection strategy depends on understanding of this procedure in [196]. Moreover, it is important
the definition of reliable technical criteria which will allow to to realize that the three elements in Eq. (31) are indeed separable
clearly identifying the best candidates for a specific engineering and it can be written as follows:
application. Condensing the data reviewed above and establishing
direct guidelines to properly select superheater materials for bio- P ¼ f1 ðFÞ  f2 ðGÞ  f3 ðMÞ ð32Þ
mass combustion is a challenging task. The Ashby approach was According to Eq. (32) f1, f2 and f3 are separate functions that can
chosen as the strategy to meet this goal. This methodology was be simply multiplied. Then, it is assumed that the materials prop-
developed by Prof. Ashby at the Cambridge University. It is based erties are independent from the design-related F and G parameters.
on the use of the CES (Cambridge Engineering System) software Consequently, the selection problem is greatly simplified as it is
to construct materials property charts. The relevant aspects of this not necessary to solve for the complete picture, only for the mate-
methodology are described below, supporting the arguments dis- rials attributes. When conflicting objectives are involved, a trade-
cussed thereafter. A detailed view of this strategy is found in [196]. off strategy can be undertaken.
The chart method is classified as a screening method for mate-
4.1.1. Materials selection using the Ashby approach rials selection. It advantageously applies as an initial screening of
The Ashby method is design-driven. It starts by identifying the materials for a given engineering application. Moreover, it can be
desired materials properties profile. The next step is to compare easily coupled to the field of mechanical design, being considered
this profile with real engineering materials, thus highlighting the a simple and quick method of evaluating the performance of spe-
best match. The identification of the attributes which are impor- cific candidates. However, it is frequently regarded as a useful tool
tant to an specific application is based on a translation step consist- for selecting materials when only two or three criteria govern the
ing of four different actions: (i) to express the function(s) that the selection process. If, though, multiple criteria are necessary to ex-
material will perform; (ii) to label the constraint(s) for the applica- ploit the case under study, then the method is of limited applicabil-
tion; (iii) to define the objective(s); (iv) to recognize which vari- ity. Multi-criteria decision making methods can be effectively used
ables the designer is free to decide. After translation, the next in these cases. An excellent introduction to such methods can be
step is to screen all the available materials using the constraints. found in [197]. Although the Ashby approach has limitations at
By doing so, the number of candidates is significantly reduced in dealing with situations involving multiple objectives [197], it suc-
comparison with the initial materials universe. Then, candidates cessfully applies to a variety of components [198–200]. This meth-
that passed the screening step are ranked using the objective(s). Fi- odology is developed in the present section to screen and rank
nally, the designer should search for supporting information candidates for superheater materials for biomass combustion
regarding, for instance, the history of the top-ranked candidates plants.
owing to the application under consideration or some not-exam-
ined aspects such as recyclability or availability. This strategy is 4.1.2. Selection strategy for superheater materials for biomass-fired
schematized in Fig. 3. boilers
The constraints are defined based on relevant materials proper- 4.1.2.1. Translation stage: function and constraints. Applying the
ties to the focused application and also by dimension requirements procedure described above the first step of the Ashby method is
of the component. The Ashby method proposes the graphical rep- the translation stage, in which we define the function, constraints,
resentation of these properties in Cartesian axes, using logarithmic objective(s) and free variables of the materials selection process.
scales. By doing so, two different properties can be simultaneously The function is promptly defined as superheater tubes for bio-
explored and the performance of a given material in respect to mass-fired boilers. Next, the constraints must be specified. The first
these properties can be promptly evaluated. This graphical repre- constraint can be specified considering that the maximum service
sentation is known as Ashby chart or material property chart. It temperature (Tmax) of the superheater material should be at least
is considered a powerful tool to the selection of engineering mate- 400 °C. This criterion is envisaged as a must-pass stage. All the
rials, granting for the analysis of different inter-related properties materials that do not pass this stage must be screened from the
in a simple and rapid way. Then, the constraints are translated into allowable alternatives. A second constraint is defined based on
attribute limits which are plotted as horizontal or vertical lines on the fracture toughness of the material given as K1C values. This
Table 1
List of superheater materials tested by different authors.

Reference Alloy steels

Material Composition (wt.%)
Cr Ni Mo Nb Ti Al C Si Cu V W Mn Fe N B
[29,35,37,38,31,69,99,101,182,183] X20 (X20CrMoV121) 10.3 0.47 0.80 – – – 0.18 0.30 0.20 0.30 – 0.80 Bal. – –
[25,26,29,31,32,37,38,40,99,101,104,182,184,185] 2.25Cr-1Mo (T22) 2.29 0.44 0.96 – – – – 0.23 – 0.01 – 0.59 Bal. – –
[25] T91 8.0–9.5 0.40 0.85–1.05 0.06–0.10 – 0.03 0.08–0.12 0.50 – – – 0.30–0.60 Bal. – –
[101,186] T92 9.15 0.26 0.5 0.6 – – 0.11 0.22 – 0.20 1.70 0.46 Bal. 0.05 0.0003
[99,101,183] 15Mo3 – – 0.30 – – – 0.20 0.20 – – – 0.60 Bal. – –
[99,101,182,187] 13CrMo44 0.90 – 0.50 – – – 0.10 0.30 – – – 0.70 Bal. – –

Reference Stainless steels

Material Composition (wt.%)
Cr Ni Mo Nb + Ta Ti Al C Si Cu V W Mn Fe N
[31,32,125,188] 310 25.4 21.1 0.10 – – – 0.02 0.40 0.10 – – 1.70 Bal. –

R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26

[31,42–45,54,56] 304L 18.3 10.1 – – – – 0.01 0.40 – – – 1.20 Bal. –
[189,125,188] 304 17.8 8.16 – – – – 0.02 0.28 – – – 1.03 Bal. 0.04
[36–38,69,101,190–192] TP 347H 17.6 10.7 – 0.56 – – 0.05 0.42 – – – 1.84 Bal. –
[34] X3CrNiMoN17-13 17.01 12.5 1.90 – – – – 0.45 – – – 1.91 Bal. –
[34,128,187] HCM12 11.67 0.15 0.78 – – – – 0.18 – 0.28 1.72 0.44 Bal. –
[35,186] AISI 347 FG 18.0 12.0 – <1.2 – – 0.07 <0.75 – – – 2.0 Bal. –
[25,26,34,37,38,99,101,183,188,187,192] Esshete 1250 15.0 9.65 0.94 – – – 0.08 0.58 – 0.22 – 6.25 Bal. –
Super 304 18.0 9.0 – 0.40 – – 0.03 0.20 3.0 – – 0.80 Bal. –
[187] 317L 18.5 14.5 3.1 – – – 0.03 0.40 – – – 0.40 Bal. –
[186] HR3C 25.0 20.0 – 0.40 – – 0.10 0.75 – – – 62.0 Bal. 0.20
[193] X8CrNiMoNb 16 16 15.5–17.5 15.5–17.5 1.6–2.0 1.2 – – 0.04–1.0 0.30–0.60 – – – 1.5 Bal. –
[193] X8CrNiNb 16 13 15.0–17.0 12.0–14.0 – 1.2 – – 0.04 0.30–0.60 – – – 1.5 Bal. –

Reference Ni-based alloys

Material Composition (wt.%)
Cr Ni Mo Nb + Ta Ti Al C Si Cu V W Mn Fe N Co
[25,31,32,38,44,45,183,187] Sanicro 28 27.6 30.4 3.3 – – – 0.01 0.40 0.90 – – 1.60 Bal. –
[31,32] Sanicro 65 20.7 61.3 8.30 – – – 0.01 0.40 – – – 0.40 Bal. –
[30,31] Alloy 825 22.4 39.5 3.2 – 0.7 0.10 0.01 0.40 2.0 – – 0.20 Bal. –
[26,37,184,187,194–196] Alloy 625 21.5 61.0 9.0 3.60 0.2 0.2 0.05 0.20 – – – 0.20 2.0 –
[25] HR11N 27.0–30.0 38.0–42.0 0.5–1.5 – – – 60.03 60.60 – – – 62.0 Bal. 0.10–0.20
[25] Sanicro 63 21.0 Bal. 8.5 3.5 – – 60.03 60.50 – – – 60.50 3.0 –
[30,193] Alloy 800 20.3 30.3 – – – – – – – – – – Bal. –
[30] Alloy 600 16.0 Bal. – – – – – – – – – – 8.3 –
[128] IN617 22.0 Bal. 9.0 – – – 0.07 – – – – – <1.0 –
[195] Alloy 59 22.0–24.0 Bal. 15.0–16.5 – – 0.10–0.40 – <0.10 – – – <0.50 <1.50 –
[195] Alloy 556 22.0 20.0 3.0 0.60 – – 0.10 0.40 – – 2.5 1.0 Bal. –
[187] Hastelloy C-2000 22.5 Bal. 15.7 – – – 0.01 – 1.50 – – 0.20 1.20 –

17
18 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26

temperature. The materials that meet both constraints simulta-

All available materials neously are concentrated in the upper right part of the chart. These
candidates are mainly ferrous metals and some non-ferrous alloys. In
order to facilitate the visual inspection of the remaining candidates,
the chart K1C–Tmax was replotted in Fig. 4b, showing only the mate-
rials that survived the screening stage. Some representative materi-
als are identified in the chart. Refractory metals such as tantalum
Translation (definition of
alloys, nickel alloys, stainless steels, cast irons, low alloy and high al-
function, constraints, objectives loy steels are the surviving candidates.
and free variables)
4.1.2.2. Translation stage: objectives. The next step of the translation
process is to define the objective(s) of the selection process. In or-
der to do this, the following question should be made: what should
be minimized or maximized? The answer to this question is not
Screening (materials which do absolute and different designers can choose different criteria. In
not meet the constraints are the present selection process, we have based our reasoning on
screened) the information collected from the relevant literature concerning
the biomass combustion research field.
In this context, a must-attend issue to validate our selection
process is to derive our objectives from materials attributes clearly
related to the corrosion mechanisms reviewed in Section 2 of the
Ranking (classification of the present text. As highlighted in Section 2.7 the chemical composi-
surviving materials according to tion of the metallic alloy plays a major role in the corrosion process
the objectives) during biomass combustion. The formation of a stable oxide scale
decreases the corrosion rate in biomass combustion environments.
This is mainly accomplished with Cr2O3-forming alloys and nickel-
based alloys. The simultaneous addition of Cr and Ni as alloying
elements is reported to greatly enhance the corrosion resistance
Seek documentation (further of superheater materials [79]. Thus, our first objective will encom-
information about the top-ranked pass this well-established experimental evidence. Hence, from the
information reviewed so far, our first objective is to maximize the
candidates)
high temperature corrosion resistance in typical biomass combus-
tion environments. A reliable technical criterion is needed to guide
Fig. 3. Materials selection based on the Ashby’s approach.
the selection and allow different candidates to be compared. At a
first approximation, the pitting resistance equivalent number
property is a measure of the resistance to brittle fracture. Accord- (PREN) is a common parameter used to rank different stainless
ing to Ashby [196], 20 MPa m1/2 is considered a minimum value of steels and nickel-based alloys according to their resistance to pit-
K1C for conventional engineering design. Materials with K1C values ting corrosion by chloride attack [201]. This number is calculated
below this limit must be screened from the selection process. This from the chemical composition of the passivating alloy and de-
stage is performed using the information shown in Fig. 4. This is pends on the contents of the three major elements associated with
the Ashby chart for fracture toughness (K1C) versus maximum ser- the formation of a passive oxide film with high resistance to local-
vice temperature. The chart was constructed using the CES Edu- ized corrosion, that is, Cr, Mo and N for stainless steels and Cr, Mo
pack 2009 software developed by Granta Design. The constraints and W for nickel alloys. The expressions used to calculate PREN for
are marked as solid lines in Fig. 4a which shows the materials uni- stainless steels and nickel alloys are shown in Eqs. (33) and (34),
verse available in the database of the CES software. The materials respectively [202,203]. In both equations the percentages refer to
classes are represented as envelopes with specific colors. In this re- the mass percentages of the alloying elements.
gard, technical ceramics are represented as yellow,1 vitreous PREN ¼ %Cr þ 3:3%Mo þ 16%N ð33Þ
ceramics as magenta, non-technical ceramics as dark yellow, elasto-
mers as cyan, thermosetting resins as dark-blue (navy), thermoplas- PREN ¼ %Cr þ 3:3ð%Mo þ 0:5%WÞ ð34Þ
tics as blue, polymer foams as green, wood and natural materials as
olive, ferrous metals as dark cyan, and non-ferrous metals as red It is important to realize, though, that PREN is an empiric
expression, resulting from data obtained under specific experimen-
bubbles. The bigger envelopes represent each family of materials
tal conditions such as immersion in ferric chloride solutions with
in a more general way and can be used to guide the reader even in
controlled concentrations of Cl- [204]. Moreover, the expressions
the absence of colors. The chart shows a range of values for each
used to calculate PREN do not take the nickel content into account.
property and material. Such spanning results from different porosity
As reviewed in Section 2, the addition of nickel is necessary to de-
levels for ceramics, different heat treatments and cold working con-
velop a more stable oxide scale under biomass combustion condi-
ditions for metals, different degree of crystallinity for polymers and
tions [31,32]. Furthermore, Phongphiphat et al. [195] showed that
so on. As clearly seen, most materials fail at meeting both constraints
alloying elements which avoid the loss of chromium due to inter-
simultaneously. All the plastics and polymer foams are below the
granular corrosion, such as Nb, Ta and Ti, are also desirable to im-
limit for maximum service temperature. Ceramics (both technical
prove the corrosion resistance of superheater materials during
and non-technical) are below the limit for fracture toughness and
biomass combustion. The corrosion rate was found to be reduced
various non-ferrous metals are below the limit for maximum service
up to six times in the presence of these elements. Yet, according
to Ishitsuka and Nose [46] Mo and W could also improve the sta-
1
For interpretation of color in Fig. 4, the reader is referred to the web version of bility of the oxide scale. In this regard, we will define a modified
this article. criterion, based on the same philosophy used for PREN, i.e., to give
 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26 19

Fig. 4. Fracture toughness – maximum service temperature chart.

a relative measure of the oxide layer stability, but focused on the CI ¼ %Ni þ %Cr þ 5:0ð%Nb þ %Ta þ %TiÞ þ %W þ %Mo ð35Þ
specific case of biomass combustion and use it as one objective
of our materials selection process. Thus, we define this corrosion Then, CI must be maximized to maximize the high tempera-
resistance index (CI) as follows where the percentages are referred ture corrosion resistance in typical biomass combustion
to the mass percentages of each element in the alloy: environments.
20 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26

Table 2 resistance of metallic materials to plastic deformation than the

Desgin requirements for superheater tubes for biomass-fired boilers. creep strength. This means that the yield strength of different can-
Design requirements didates which are available in different sources will be comparable,
Function Superheater tubes for biomass-fired boilers
since they are obtained according to a unique procedure indicated
Constraints Maximum service temperature above 400 °C; fracture in ASTM E8. This led us to choose the yield strength as our second
toughness (K1C) above 20 MPa m1/2 objective. It will be used as a merit index to rank the candidates of
Objectives To maximize the corrosion resistance index (CI); to the selection process. However, creep strength will not be ne-
maximize the yield strength (0.2% offset limit)
glected. As described in Section 4.1.1 the Ashby’s approach recom-
Free variables Tube wall thickness; choice of material
mends that, after ranking, supporting information should be seek
in order to cope with some not-examined characteristics of the
Now, our second objective must be defined. According to Jones top-ranked materials. In our case, the creep strength of the top-
[205], creep fractures of boiler tubes cause 10% of all power-plant ranked materials will be examined after the ranking stage. This
breakdowns and 30% of all tube failures in boilers. Creep is defined should give consistent supporting to the conclusions obtained
as a time-dependent plastic deformation under static stresses usu- throughout the whole selection process.
ally significant at temperatures above one 0.3–0.5Tm where Tm is The translation stage can be now completed by defining the free
the melting temperature of metals [206]. The relevance of creep variables of the selection process. In our case, these will be the wall
for the mechanical stability of boiler tubes has been highly recog- thickness of the superheater tube and the material used to manu-
nized by many authors [207–209]. Hence, creep strength can be a facture it. The design requirements defined in the translation stage
valuable criterion to rank the best candidates for superheater tubes are summarized in Table 2.
in biomass-fired boilers. Nevertheless, comparing different materi-
als based on this property is a challenging task. Creep behavior de- 4.1.2.3. Screening and ranking stages. The screening stage has al-
pends on both the temperature and the static stress level employed ready been done, as shown in Fig. 4. The materials which passed
during the tests [210]. Thus, in spite of the availability of engineer- this stage were easily identified, using the constraints presented
ing data regarding the creep strength of typical boiler materials, in Table 2. The possible candidates have been identified as refrac-
several data are obtained according to different experimental con- tory metals such as tantalum alloys, nickel alloys, stainless steels,
ditions regarding either the temperature or the static stress or both cast irons, low alloy and high alloy steels. To proceed with the
parameters. Consequently, creep data of different superheater selection process, these candidates must be ranked according to
materials obtained under the same experimental conditions are the objectives defined in the translation stage. Before this step,
not readily available from different sources. In this scenario, the we will further restrict the surviving materials, thus obtaining a
tensile yield strength at 0.2% offset obtained according to the pro- minor set of candidates to the ranking stage. In order to carry
cedure described in the ASTM E8 standard [211] can be regarded as out this further restriction we will consider only the materials
a reliable parameter for comparing the strength of different metal- shown in Table 1 to our ranking stage. These materials have been
lic materials to the onset of plastic deformation. It must be clear, collected from the relevant literature reviewed in this article and
though, that the yield strength and the creep strength are different can be envisaged as traditional options for superheater tubes.
properties, quantifying the response of the metallic material to dis- Hence, cast irons and refractory metals will not be considered to
tinct phenomena. Creep relates to time-dependent plastic defor- our purposes since these materials are not reported as viable can-
mation at high temperatures whereas the yield strength relates didates to superheater tubes of biomass-fired boilers. This can be a
to plastic deformation at room temperature and higher strain rates. consequence of a relatively low K1C in the case of cast irons or high
Thus, different deformation mechanisms apply to each case. How- cost in the case of refractory metals. Thus, only the nickel alloys,
ever, the room temperature yield strength is determined directly stainless steels, low alloy steels and high alloy steels shown in Ta-
according to the standardized procedure indicated in ASTM E8. It ble 1 will be ranked according to the objectives defined in the
does not depend on experimental parameters such as the stress le- translation stage.
vel and temperature employed in creep tests. Thus, it can be con- Fig. 5 is an Ashby chart of the CI values of the alloys shown in
sidered as a more easily comparable parameter to indicate the Table 1 as the ordinates whereas the yield strength (0.2% offset

120
Sanicro 63 224 Alloy 625 223
X8CrNiNb 16 13 232
Hastelloy C-2000 221 Alloy 59 222
100 X8CrNiMoNb 16-16 232
Sanicro 65 224

Alloy 600 223 IN617 223

80 HR11N 230
Alloy 825 223

Sanicro 28 225
CI

60 310 228 Alloy 556 223

Alloy 800 223
317L 228

347FG 228 TP347H 228

40 HR3C 220
X3CrNiMoN17-13 232
T92 220
HCM12 229
20 304 228 Super 304 231
T91 220 X20 233
Esshete 1250 220
304L 228 13CrMo44 226
15Mo3 227 T22 234
0
180 280 380 480 580
Yield strength (MPa)

Fig. 5. CI – yield strength chart (See above-mentioned references for further information.).
 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26 21

limit) is plotted in the abscissas. These values were obtained from high temperature applications with an outer layer (based on Alloy
the references indicated as superscripts, next to the alloy designa- 625) co-extruded to inner ferritic steel [212]. Sanicro 65 is similar
tion shown in the chart. The CI values were calculated using Eq. to Sanicro 63 except that it does not contain niobium in its chem-
(35). It is worth mentioning that the chart in Fig. 5 is based on ical composition [213]. We did not find information regarding the
the same principles of the CES software but was not plotted using creep behavior of these materials in the current literature. How-
it. The materials displayed in the chart are those investigated by ever, they are recommended for high temperature applications
top research groups throughout the world. The CES software ver- [214] and can be regarded as having at least creep strength compa-
sion used to plot the chart in Fig. 4 is an educational version and rable to Alloy 625. Finally, comparable creep data could be found
does not contain such information. Then, we chose to use the rep- for IN617 and Alloy 625 [215,216]. The static tensile stress to pro-
resentation shown in Fig. 5. It follows the same philosophy of the duce rupture of Alloy 625 within 100 h at 649 °C is reported to be
CES software and is perfectly suited to the materials selection pro- 490 MPa whereas for IN617 it is only 379 MPa. Furthermore, the
cess developed in this text. high creep strength of these materials has been evidenced in the
The disposition of the different candidates shown in Fig. 5 re- literature [217,218]. Thus, the supporting information focused on
veals that the alloys can be ranked according to three distinct ap- creep data confirmed that the best candidate to superheater tubes
proaches. If one considers that the most important objective is to for biomass-fired boilers is Alloy 625.
maximize the yield strength, the best candidates could be ranked In light of the results encountered in the current literature, it is
according to the following order, considering only the top ten evident that the corrosion performance of the superheater materi-
materials: Alloy 625 > X20 > HCM12 > T92 > Sanicro 63  Sanicro als used in biomass combustion or co-combustion is strongly re-
65 > T91  T22 > Alloy 556 > IN617. If one considers that the most lated to the alloy composition. High levels of nickel and
important objective is to maximize the CI value, the best candi- chromium increase the stability of the oxide films and the corro-
dates are ranked according to the following order, considering only sion resistance of the alloys. Molybdenum and carbide former ele-
the top 10 materials: Alloy 625 > Sanicro 63 > Hastelloy C- ments (Nb, Ta and Ti) are also highly desirable. However, it should
2000  Alloy 59 > Alloy 600 > Sanicro 65 > IN617 > HR11N > Alloy be noted that, depending on the metal temperature and fuel com-
825 > Sanicro 28. The third approach is based on the candidates position, even highly alloyed materials can be severely corroded. In
that give the best compromise between both objectives, but is this regard, any selection strategy for superheater materials should
not necessarily optimal by either of the objectives. In this case, encompass the previous analysis of existing data regarding the
the best candidate is Alloy 625 which is the ultimate choice from compatibility of the intended superheater material with the envi-
both objectives, followed by Sanicro 63 and Sanicro 65 which lie ronment conditions envisaged for its final use. Frequently, a corro-
more internally in the chart. It is concluded, thus, that the best can- sion protection method will have to be employed to guarantee that
didates based on the CI values are not those that maximize the the superheater will have a longer service life. Coatings are attrac-
yield strength. This configures an approach based on the conflict- tive options in this case. Coated alloys can be regarded as new
ing objectives philosophy developed by Ashby [196]. In our selec- materials, giving rise to a broader set of candidates for a given
tion process we will consider that both objectives have the same engineering application. The materials selector should be aware
relative importance. Hence, the top-ranked candidates are those of this possibility. Yet, in addition to the objectives considered in
which give the best compromise between the two objectives. Con- our selection process, a cost analysis is also mandatory. A highly al-
sequently, Alloy 625 is the best solution to the selection process, loyed material can withstand corrosion during an extended life-
followed by Sanicro 63, Sanicro 65. IN617 is also an interesting time but its cost is frequently high. A low alloyed material with a
solution, regarding its excellence in corrosion resistance based on proper protective coating could resist in a similar or even better
the CI value. manner with a more attractive final cost. The literature shows that
This result reveals the remarkable influence of nickel and chro- bare metallic alloys are affected by high temperature corrosion.
mium in the formation of a stable oxide layer capable of with- The extent of corrosion damage can be low for the highly alloyed
standing the corrosion processes taking place during biomass materials but it is always present. Only coated alloys have been re-
combustion. This is in agreement with results published bydiffer- ported to be unaffected by high temperature corrosion [192].
ent authors. Moreover, our materials selection strategy is further
supported by the corrosion mechanisms by Phongphiphat et al. 5. Conclusions
[195] owing to the beneficial role of elements that avoid the loss
of chromium due to intergranular corrosion such as niobium and A materials selection strategy for superheater materials used in
titanium, thus forming a more stable and protective Cr2O3 scale. biomass combustion has been developed in this work. Corrosion
The importance of considering the presence of such alloying ele- mechanisms and mitigation methods regarding corrosion in bio-
ments is promptly recognized when the relative positions of other mass combustion were reviewed to support the selection proce-
high-CI alloys such as Hastelloy C-2000 and Alloy 59 (which do not dure. A direct interaction between these subjects could be
contain either Nb or Ti but are highly alloyed with Ni, Mo and Cr) identified. The main findings of this work can be summarized as
are compared with the positions of Alloy 625 and Sanicro 63 which follows:
contain Nb and are also highly alloyed with Ni, Cr and Mo. Alloy
625 and Sanicro 63 have a higher corrosion resistance index (CI) (a) Alkali chlorides, especially potassium chloride, increase the
and should be preferable to the manufacturing of superheater corrosion rate of the materials used as superheater tubes.
tubes used in biomass combustion. Reactions between the metallic surface and this species, or
other alkali-containing species such as carbonates, have to
4.1.2.4. Supporting information. The last stage of the Ashby method be avoided in order to increase the corrosion resistance of
is to seek supporting information of the top-ranked materials. As the superheaters.
mentioned above, creep strength is an important property for (b) The formation of a dense and compact oxide film decreases
superheater materials used in biomass-fired boilers. Thus, creep the corrosion rate of superheater materials under biomass
data of Alloy 625, Sanicro 63, Sanicro 65 and IN617have been con- combustion conditions. This is favored for Cr2O3-forming
sulted. Sanicro 63 and Sanicro 65 are trademarks of Sandvik Mate- alloys. A further contribution to increased corrosion resis-
rials Technology. They consist of composite tubes developed for tance can be obtained when nickel and chromium are
22 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26

concomitantly present as alloying elements. Moreover, nio- [13] A. Demirbas, Combustion characteristics of different biomass fuels, Prog.
Energy Combust. Sci. 30 (2004) 219–230.
bium, tantalum and titanium prevent the loss of chromium
[14] R.A. Khalil, E. Houshfar, W. Musinguzi, M. Becidan, O. Skreiberg, F. Goile, T.
due to intergranular corrosion further improving the corro- Lovas, L. Sorum, Experimental investigation on corrosion abatement in straw
sion resistance of superheater alloys. combustion by fuel mixing, Energy Fuels 25 (2011) 2687–2695.
(c) Mitigation methods are designed to avoid the reaction of the [15] A. Williams, J.M. Jones, L. Ma, M. Pourkashanian, Pollutants from the
combustion of solid biomass fuels, Prog. Energy Combust. Sci. 38 (2012)
metallic surface with the aggressive alkali-containing con- 113–137.
taining species. This can be accomplished by co-combustion [16] S.V. Vassilev, D. Baxter, L.K. Andersen, C.G. Vassileva, An overview of the
strategies based on the simultaneous burning of coal and chemical composition of biomass, Fuel 89 (2010) 913–933.
[17] S.V. Vassilev, D. Baxter, L.K. Andersen, C.G. Vassileva, T.J. Morgan, An overview
biomass or by the use of sulfur-containing additives or spe- of the organic and inorganic phase composition of biomass, Fuel 94 (2012) 1–
cies capable of sequestrating the alkalis in the biomass, 33.
forming less aggressive compounds such as alkali sulfates [18] A. Faaij, J. van Doorn, T. Curvers, L. Waldheim, E. Olsson, A. van Wijk, C. Daey-
Ouwens, Characteristics and availability of biomass waste and residues in the
of aluminosilicates, instead of alkali chlorides. Coatings Netherlands for gasification, Biomass Bioenergy 12 (1997) 225–240.
based on nickel and chromium as the major elements can [19] S. Vassilev, C. Braekman-Danheux, P. Laurent, Characterization of refuse-
form stable oxide layers to withstand the corrosion pro- derived char from municipal solid waste: 1. Phase mineral and chemical
composition, Fuel Process. Technol. 59 (1999) 95–134.
cesses during biomass combustion. [20] H.P. Nielsen, F.J. Frandsen, K. Damp-Johansen, L.L. Baxter, The implications of
(d) Nanostructured materials, although little explored in the chlorine-associated corrosion on the operation of biomass-fired boilers, Prog.
current literature, are promising alternatives to mitigate cor- Energy Combust. Sci. 26 (2000) 283–298.
[21] M. Öhman, A. Nordin, B.J. Skrifvars, R. Backman, M. Hupa, Bed agglomeration
rosion problems in biomass combustion due to the more
characteristics during fluidized bed combustion of biomass fuels, Energy
rapid formation of a stable oxide layer. Fuels 14 (2000) 169–178.
(e) The Ashby’s approach successfully applied to the selection of [22] L.L. Baxter, Influence of ash deposit chemistry and structure on physical and
materials for superheater tubes used in biomass combus- transport properties, Fuel Process. Technol. 56 (1998) 81–88.
[23] L.-G. Johansson, J.-E. Svensson, E. Skog, J. Pettersson, C. Pettersson, N.
tion. A corrosion resistance index (IC) was proposed as a Folkeson, H. Asteman, T. Jonsson, M. Halvarsson, Critical corrosion
technical criterion to selection the best candidates for this phenomena on superheaters in biomass and waste-fired boilers, in:
application. This index was derived based on the informa- Proceedings of the Sino-Swedish Structural Materials Symposium, 2007, pp.
35–39.
tion reviewed from the literature concerning the corrosion [24] N. Folkeson, J. Pettersson, C. Pettersson, L.-G. Johansson, E. Skog, J.-E.
mechanisms involved in biomass combustion. The best can- Svensson, Fireside corrosion of stainless and low alloyed steels in a waste-
didates identified in this work were those which gave the fired CFB boiler: the effect of adding sulphur to thefuel, Mater. Sci. Forum
595–598 (2008) 289–297.
best compromise between corrosion resistance (determined [25] B.-J. Skrifvars, M. Westen-Karlsson, M. Hupa, K. Salmenoja, E. Vakkilainen,
using the IC value of the metallic alloy) and the mechanical Corrosion of superheater steel materials under alkali salt deposits. Part 1: the
strength (considering the tensile yield strength at 0.2% offset effect of salt deposit composition and temperature, Corros. Sci. 50 (2008)
1274–1282.
obtained according to the procedure described in the ASTM [26] B.-J. Skrifvars, M. Westen-Karlsson, M. Hupa, K. Salmenoja, Corrosion of
E8 standard). The relevance of the chemical composition of super-heater steel materials under alkali salt deposits, Corros. Sci. 52 (2010)
the superheater material was evidenced. In this context, 1011–1019.
[27] K.O. Davidsson, L.-E. Amand, B.-M. Steenari, A.L. Elled, D. Eskilsson, B. Leckner,
nickel, chromium and niobium have the most striking influ-
Contermeasures against alkali-related problems during combustion of
ence on the final choice. The development of metallic alloys biomass in a circulating fluidized bed boiler, Chem. Eng. Sci. 63 (2008)
to withstand corrosion in biomass combustion can be suc- 5314–5329.
cessfully designed based on the materials selection strategy [28] D.A. Tillman, D. Duong, B. Miller, Chlorine in solid fuels fired in pulverized
fuel boilers – sources, forms, reactions and consequences: a literature review,
described in this work. Energy Fuels 23 (2009) 3379–3391.
[29] H.J. Grabke, E. Reese, M. Spiegel, The effects of chlorides, hydrogen chloride
and sulfur dioxide in the oxidation of steels below deposits, Corros. Sci. 37
(1995) 1023–1043.
References [30] A. Zahs, M. Spiegel, H.J. Grabke, Chloridation and oxidation of iron, chromium,
nickel and their alloys in chloridizing and oxidizing atmospheres at 400–
700 °C, Corros. Sci. 42 (2000) 1093–1122.
[1] A.A. Khan, W. de Jong, P.J. Jansens, H. Spliethoff, Biomass combustion in
[31] P. Andersson, M. Norell, Field test of superheater corrosion in a CFB waste
fluidized bed boilers: potential problems and remedies, Fuel Process. Technol.
boiler: part 1 – metal loss characteristics, Mater. Corros. 56 (2005) 449–
90 (2009) 21–50.
458.
[2] G. Taylor, Biofuels and the biorefinery concept, Energy Policy 36 (2008) 4406–
[32] P. Andersson, M. Norell, Field test of superheater corrosion in a CFB waste
4409.
boiler. Part II – scale formation characteristics, Mater. Corros. 56 (2005) 550–
[3] K. Openshaw, Biomass energy: employment generation and its contribution
560.
to poverty alleviation, Biomass Bioenergy 34 (2010) 365–378.
[33] H.H. Krause, High temperature corrosion problems in waste incineration
[4] A.W. Bhutto, A.A. Bazmi, G. Zahedi, Greener energy: issues and challenges for
systems, J. Mater. Energy Syst. 7 (1986) 322–332.
Pakistan – biomass energy prospective, Renew. Sustain. Energy Rev. 15 (2011)
[34] M. Montgomery, A. Karlsson, O.H. Larsen, Field test corrosion experiments in
3207–3219.
Denmark with biomass fuels. Part 1: straw-firing, Mater. Corros. 53 (2002)
[5] H. Haberl, T. Beringer, S.C. Bhattacharya, K.-H. Erb, M. Hoogwijk, The global
121–131.
technical potential of bio-energy in 2050 considering sustainability
[35] H.P. Nielsen, F.J. Frandsen, K. Dam-Johansen, Lab-scale investigations of high-
constraints, Curr. Opin. Environ. Sustain. 2 (2010) 394–403.
temperature corrosion phenomena in straw-fired boilers, EnergyFuels 13
[6] K. Srirangan, L. Akawi, M.-Y. Murray, C.P. Chou, Towards sustainable
(1999) 1114–1121.
production of clean energy carriers from biomass resources, Appl. Energy
[36] S.C. van Lith, F.J. Frandsen, M. Montgomery, T. Vilhelmsen, S.A. Jensen, Lab-
100 (2012) 172–186.
scale investigation of deposit-induced chlorine corrosion of superheater
[7] S. Lim, K.T. Lee, Leading global energy and environmental transformation:
materials under simulated biomass-firing conditions. Part 1: exposure at
unified ASEAN biomass-based bio-energy system incorporating the clean
560 °C, Energy Fuels 23 (2009) 3457–3468.
development mechanism, Biomass Bioenergy 35 (2011) 2479–2490.
[37] C.C. Cha, High temperature corrosion of superheater materials below
[8] A.F. Kirkels, Discursive shifts in energy from biomass: a 30 year European
deposited biomass ashes in biomass combusting atmospheres, Corros. Eng.
overview, Renew. Sustain. Energy Rev. 16 (2012) 4105–4115.
Sci. Technol. 42 (2007) 50–60.
[9] Brasil. Empresa de Pesquisa Energética. Balanço Energético Nacional 2011 –
[38] S. Sroda, M. Mäkipãa, S.-C. Cha, M. Spiegel, The effect of ash deposition on
Ano base 2010: Resultados Preliminares Rio de Janeiro: EPE, 2011.
corrosion behaviour of boiler steels in simulated combustion atmospheres
[10] P. McKendry, Energy production from biomass (part 2): conversion
containing carbon dioxide (CORBI PROJECT), Mater. Corros. 57 (2006) 176–
technologies, Bioresour. Technol. 83 (2002) 47–54.
181.
[11] M.F. Demirbas, M. Balat, H. Balat, Potential contribution of biomass to the
[39] J.-M. Brossard, I. Diop, X. Chaucherie, F. Nicol, C. Rapin, M. Vilasi, Superheater
sustainable energy development, Energy Convers. Manage. 50 (2009) 1746–
fireside corrosion mechanisms in MSWI plants: lab-scale study and on-site
1760.
results, Mater. Corros. 62 (2011) 543–548.
[12] A. Demirbas, Potential applications of renewable energy sources, biomass
[40] T. Jonsson, N. Folkeson, J.-E. Svensson, L.-G. Johansson, M. Halvarsson, An
combustion problems in boiler power systems and combustion related
ESEM in situ investigation of initial stages of the KCl induced high
environmental issues, Prog. Energy Combust. Sci. 31 (2005) 171–192.
 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26 23

temperature corrosion of a Fe–2.25Cr–1Mo steel at 400 °C, Corros. Sci. 53 [70] B.M. Jenkins, L.L. Baxter, T.R. Miles Jr, T.R. Miles, Combustion properties of
(2011) 2233–2246. biomass, Fuel Proc. Technol. 54 (2002) 17–46.
[41] T.J. Pan, Y.S. Li, Q. Yang, R.F. Feng, A. Hirose, Internal oxidation and phase [71] S. Jiménez, J. Ballester, Formation and emission of submicron particles in
transformations of multi-phase Fe–Ni–Al and Fe–Ni–Al–Cr alloys induced by pulverized olive residue (Orujillo) combustion, Aerosol Sci. Technol. 38
KCl corrosion, Corros. Sci. 53 (2011) 2115–2121. (2004) 707–723.
[42] J. Pettersson, H. Asteman, J.-E. Svensson, L.-G. Johansson, KCl induced [72] S. Jiménez, J. Ballester, Influence of operating conditions and the role of sulfur
corrosion of a 304-type austenitic stainless steel at 600 °C; the role of in the formation of aerosols from biomass combustion, Combust. Flame 140
potassium, Oxid. Met. 64 (2005) 23–41. (2005) 346–358.
[43] J. Pettersson, J.-E. Svensson, L.-G. Johansson, KCl-induced corrosion of a 304- [73] S. Jiménez, J. Ballester, Particulate matter formation and emission in the
type austenitic stainless steel in O2 + H2O environment: the influence of combustion of different pulverized biomass fuels, Combust. Sci. Technol. 178
tremperature, Oxid. Met. 72 (2009) 159–177. (2006) 655–683.
[44] C. Pettersson, J. Pettersson, H. Asteman, J.-E. Svensson, L.-G. Johansson, KCl- [74] S. Jiménez, J. Ballester, Formation of alkali sulphate aerosols in biomass
induced high temperature corrosion of the austenitic Fe–Cr–Ni alloys 304L combustion, Fuel 86 (2007) 486–493.
and Sanicro 28 at 600 °C, Corros. Sci. 48 (2006) 1368–1378. [75] K.A. Christensen, M. Stenholm, H. Livbjerg, The formation of submicron
[45] C. Pettersson, L.-G. Johansson, J.-E. Svensson, The influence of small amounts aerosol particles, HCl and SO2 in straw-fired boilers, J. Aerosol Sci. 29 (1998)
of KCl (s) on the initial stages of the corrosion of alloy Sanicro 28 at 600 °C, 421–444.
Oxid.Met. 70 (2008) 241–256. [76] K.A. Christensen, H. Livbjerg, A plug flow model for chemical reactions and
[46] T. Ishitsuka, K. Nose, Stability of protective oxide films in waste incineration aerosol nucleation and growth in an alkali-containing flue gas, Aersol Sci.
environment – solubility measurement of oxides in molten chlorides, Corros. Technol. 33 (2000) 470–489.
Sci. 44 (2002) 247–263. [77] H.S. Hsu, Review of the development and breakdown of protective oxide
[47] S.C. Cha, M. Spiegel, Local reactions between NaCl and KCl particles and metal scales on alloys exposed to coal-derived atmospheres, Oxid. Met. 28 (1987)
surfaces, Corros. Eng. Sci. Technol. 40 (2005) 249–254. 213–235.
[48] S.C. Cha, M. Spiegel, Local reactions of KCl particles with iron, nickel and [78] W. Nelson, C. Cain, Corrosion of superheaters and reheaters of pulverized-
chromium surfaces, Mater. Corros. 57 (2006) 159–164. coal-fired boilers, Trans. ASME, Ser. A., J. Eng. Power 82 (1960) 194–204.
[49] H.P. Michelsen, O.H. Larsen, F.J. Frandsen, K. Dam-Johansen, Deposition and [79] J.N. Harb, E.E. Smith, Fireside corrosion in PC-fired boilers, Prog. Energy
high temperature corrosion in a 10 MW straw fired boiler, Fuel Process. Combust. Sci. 16 (1990) 169–190.
Technol. 54 (1998) 95–108. [80] J. Stringer, A.J. Minchener, High temperature corrosion in fluidized bed
[50] C. Yin, L.A. Rosendahl, S.K. Kaer, Grate-firing of biomass for heat and power combustion systems, J. Mater. Energy Syst. 7 (1986) 333–344.
production, Prog. Energy Combust. Sci. 34 (2008) 725–754. [81] E.M. Labuda, D.A. Cline Jr., K.J. Shields, Fireside corrosion in coal and oil-fired
[51] Y. Kawahara, Y. Kaihara, Recent Trends in Corrosion-resistant Tube Materials units: failure mechanisms and methods of prevention, in: NACE Conference,
and Improvements of Corrosion Environments in WTE Plants, Corrosion 2001, Paper 00234, Corrosion 2000, 2000.
Paper No. 01173, Houston, TX, 2001. [82] A.J.B. Cutler, E. Raask, External corrosion in coal-fired boilers: assessment
[52] B.A. Baker, G.D. Smith, L.D. Shoemaker, Performance of Commercial Alloys in from laboratory data, Corros. Sci. 21 (1981) 789–800.
Simulated Waste Incineration Environments, Corrosion 2001, Paper 01183, [83] L. Li, Y. Duan, Y. Cao, P. Chu, R. Carty, W.-P. Pan, Field corrosion tests for a low
Houston, TX, 2001. chromium steel carried out at superheater area of a utility boiler with three
[53] J. Lehmusto, B.-J. Skrifvars, P. Yrjas, M. Hupa, High temperature oxidation of coals containing different chlorine contents, Fuel Process. Technol. 88 (2007)
metallic chromium exposed to eight different metal chlorides, Corros. Sci. 53 387–392.
(2011) 3315–3323. [84] X. Peng, K. Liu, W.-P. Pan, J.T. Riley, High-temperature corrosion of A210-C
[54] J. Pettersson, N. Folkeson, L.-G. Johansson, J.-E. Svensson, The effects of KCl, carbon steel in simulated coal-combustion atmospheres, Oxid. Met. 60 (2003)
K2SO4 and K2CO3 on the high temperature corrosion of a 304-type austenitic 117–135.
stainless steel, Oxid. Met. 76 (2011) 93–109. [85] R.A. Rapp, Chemistry and electrochemistry of hot corrosion of metals, Mater.
[55] J. Lehmusto, D. Lindberg, P. Yrjas, B.-J. Skrifvars, M. Hupa, Thermogravimetric Sci. Eng. 87 (1987) 319–327.
studies of high temperature reactions between potassium salts and [86] J. Stringer, Hot corrosion of high-temperature alloys, Ann. Rev. Mater. Sci. 7
chromium, Corros. Sci. 59 (2012) 55–62. (1977) 477–509.
[56] J. Lehmusto, B.-J. Skrifvars, P. Yrjas, M. Hupa, Comparison of potassium [87] R.A. Rapp, Hot corrosion of materials, High Temp. Sci. 27 (1990) 355–367.
chloride and potassium carbonate with respect to their tendency to cause [88] R.A. Rapp, Y.S. Zhang, Hot corrosion of materials: fundamental studies, JOM
high temperature corrosion of stainless 304L steel, Fuel Process. Technol. 105 46 (1994) 47–55.
(2013) 98–105. [89] R.A. Rapp, K.S. Goto, in: J. Braunstein, J.R. Selman (Eds.), Hot Corrosion of
[57] M. Steinberg, K. Schofield, The controlling chemistry of surface deposition Metals by Molten Salts, Molten Salts 1, Electrochem. Soc., Pennington, NJ,
from sodium and potassium seeded flames free of sulfur or chlorine 1981, pp. 159–177.
impurities, Combust. Flame 129 (2002) 453–470. [90] R.A. Rapp, N. Otsuka, The role of chromium in the hot corrosion of metals, ECS
[58] K. Schofield, The chemical nature of combustion deposition and corrosion: Trans. 16 (2009) 271–282.
the case of alkalichlorides, Combust. Flame 159 (2012) 1987–1996. [91] M. Aho, P. Vainikka, R. Taipale, P. Yrjas, Effective new chemicals to prevent
[59] T. Blomberg, Which are the right test conditions for the simulation of high corrosion due to chlorine in power plant superheaters, Fuel 87 (2008) 647–
temperature alkali corrosion in biomass combustion?, Mater Corros. 57 654.
(2006) 170–175. [92] M. Aho, J. Silvennoinen, Preventing chlorine deposition on heat transfer
[60] T. Blomberg, A thermodynamic study of the gaseous potassium chemistry in surfaces with aluminium-silicon rich biomass residue and additive, Fuel 83
the convection sections of biomass fired boilers, Mater. Corros. 62 (2011) (2004) 1299–1305.
635–641. [93] M. Aho, E. Ferrer, Importance of coal ash composition in protecting the boiler
[61] T. Blomberg, Correlation of the corrosion rates of steels in a straw fired boiler against chlorine deposition during combustion of chlorine-rich biomass, Fuel
with the thermodynamically predicted trend of KOH(g) in the flue gases, 84 (2005) 201–212.
Biomass Bioenergy 39 (2012) 489–493. [94] M. Aho, A. Gil, R. Taipale, P. Vainikka, H. Vesala, A pilot-scale fireside deposit
[62] T. Lind, E.I. Kauppinen, J. Hokkinen, J.K. Jokiniemi, M. Orjala, M. Aurela, R. study of co-firing Cynara with two coals in a fluidised bed, Fuel 87 (2008) 58–
Hillamo, Effect of chlorine and sulfur on fine particle formation in pilot-scale 69.
CFBF of biomass, Energy Fuels 20 (2006) 61–68. [95] C. Bartolomé, A. Gil, I. Ramos, Ash deposition behavior of cynara–coal blends
[63] L.A. Hansen, H.P. Nielsen, F.J. Frandsen, K. Dam-Johansen, S. Horlyck, A. in a PF pilot furnace, Fuel Process. Technol. 91 (2010) 1576–1584.
Karlsson, Influence of deposit formation on corrosion at a straw-fired boiler, [96] M. Aho, P. Yrjas, R. Taipale, M. Hupa, J. Silvennoinen, Reduction of superheater
Fuel Process. Technol. 64 (2000) 189–209. corrosion by co-firing risky biomass with sewage sludge, Fuel 89 (2010)
[64] J. Yin, Z. Wu, Corrosion behavior of TP316L of superheater in biomass boiler 2376–2386.
with simulated atmosphere and deposit, Chin. J. Chem. Eng. 17 (2009) 849– [97] M. Aho, T. Envall, J. Kauppinen, Corrosivity of flue gases during co-firing
853. Chinese biomass with coal at fluidized bed conditions, Fuel Process. Technol.
[65] K.O. Davidsson, L.-E. Amand, B. Leckner, Potassium, chlorine, and sulfur in 105 (2013) 82–88.
ash, particles, deposits, and corrosion during wood combustion in a [98] Y. Zheng, P.A. Jensen, A.D. Jensen, B. Sander, H. Junker, Ash transformation
circulating fluidized-bed boiler, Energy Fuels 21 (2007) 71–81. during co-firing coal and straw, Fuel 86 (2007) 1008–1020.
[66] G. Sorell, The role of chlorine in high temperature corrosion in waste-to- [99] M. Broström, H. Kassman, A. Helgesson, M. Berg, C. Andersson, R. Backman, A.
energy plants, Mater. High Temp. 14 (1997) 137–150. Nordin, Sulfation of corrosive alkali chlorides by ammonium sulfate in a
[67] H.P. Nielsen, L.L. Baxter, G. Sclippab, C. Morey, F.J. Frandsen, K. Dam-Johansen, biomass fired CFB boiler, Fuel Process. Technol. 88 (2007) 1171–1177.
Deposition of potassium salts on heat transfer surfaces in straw-fired boilers: [100] ChlorOut, European Patent EP 1354167, International Patent Application:
a pilot-scale study, Fuel 79 (2000) 131–139. PCT/SE 02/00129, 2002.
[68] N. Otsuka, A thermodynamic approach on vapor-condensation of corrosive [101] P. Henderson, P. Szakálos, R. Pettersson, C. Andersson, J. Högberg, Reducing
salts from flue gas on boiler tubes in waste incinerators, Corros. Sci. 50 (2008) superheater corrosion in wood-fired boilers, Mater. Corros. 57 (2006) 128–
1627–1636. 134.
[69] M. Montgomery, T. Vilhelmsen, S.A. Jensen, Potential high temperature [102] H. Kassman, M. Broström, M. Berg, L.-E. Ämand, Measures to reduce chlorine
corrosion problems due to co-firing of biomass and fossil fuels, Mater. Corros. in deposits: application in a large-scale circulating fluidised bed boiler firing
59 (2008) 783–793. biomass, Fuel 90 (2011) 1325–1334.
24 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26

[103] H. Kassman, J. Pettersson, B.-M. Steenari, L.-E. Ämand, Two strategies to [132] T. Roland, D. Retraint, K. Lu, J. Lu, Fatigue life improvement through surface
reduce gaseous KCl and chlorine in deposits during biomass combustion— nanostructuring of stainless steel by means of surface mechanical attrition
injection of ammonium sulphate and co-combustion with peat, Fuel treatment, Scr. Mater. 54 (2006) 1949–1954.
Process. Technol. 105 (2013) 170–180. [133] M. Jafari, M.H. Enayati, M.H. Abbasi, F. Karimzadeh, Compressive and wear
[104] P. Viklund, R. Pettersson, A. Hjörnhede, P. Henderson, P. Sjövall, Effect of behaviors of bulk nanostructured Al2024 alloy, Mater. Des. 31 (2010) 663–
sulphur containing additive on initial corrosion of superheater tubes in waste 669.
fired boiler, Corres. Eng. Sci. Technol. 44 (2009) 234–240. [134] K.A. Padmanabhan, Mechanical properties of nanostructured materials,
[105] K. Schofield, New method to minimize high temperature corrosión resulting Mater. Sci. Eng. A 304–306 (2001) 200–205.
from alkali sulfate and chloride deposition in combustion systems. II. [135] Y.T. Zhu, T.C. Lowe, T.G. Langdon, Performance and applications of
Molybdenum salts, Energy Fuels 19 (2005) 1898–1905. nanostructured materials produced by severe plastic deformation, Scr.
[106] X. Wei, U. Schnell, K.R.G. Hein, Behaviour of gaseous chlorine and alkali Mater. 51 (2004) 825–830.
metals during biomass thermal utilization, Fuel 84 (2005) 841–848. [136] R.Z. Valiev, R.K. Islamgaliev, I.V. Alexandrov, Bulk nanostructured materials
[107] M. Öhman, D. Boström, A. Nordin, H. Hedman, Effect of kaolin and limestone from severe plastic deformation, Prog. Mater. Sci. 45 (2000) 103–189.
addition on slag formation during combustion of wood fuels, Energy Fuels 18 [137] V. Maier, T. Hausöl, C.W. Schmidt, W. Böhm, H. Nguyen, M. Merklein, H.W.
(2004) 1370–1376. Höppel, M. Göken, Tailored heat treated accumulative roll bonded aluminum
[108] M. Aho, J. Silvennoinen, Preventing chlorine deposition on heat transfer blanks: microstructure and mechanical behavior, Metall. Mater. Trans.A 43
surfaces with aluminium–silicon rich biomass residue and additive, Fuel 83 (2012) 3097–3107.
(2004) 1299–1305. [138] Z. Horita, T. Fujinami, T.G. Langdon, The potential for scaling ECAP: effect of
[109] B.M. Steenari, A. Lundberg, H. Pettersson, M. Wilewska-Bien, D. Andersson, sample size on grain refinement and mechanical properties, Mater. Sci. Eng. A
Investigation of ash sintering during combustion of agricultural 318 (2001) 34–41.
residues and the effect of additives, Energy Fuels 23 (2009) 5655– [139] M.S. Mohebbi, A. Zkbarzadeh, Accumulative spin-bonding (ASB) as a novel
5662. SPD process for fabrication of nanostructured tubes, Mater. Sci. Eng. A 528
[110] M. Theis, B.J. Srifvars, M. Zevenhoven, M. Hupa, H. Tran, Fouling tendency of (2010) 180–188.
ash resulting from burning mixtures of biofuels. Part 2: deposit chemistry, [140] L.S. Toth, M. Arzaghi, J.J. Fundenberger, B. Beausir, O. Bouaziz, R. Arruffat-
Fuel 85 (2006) 1992–2001. Massion, Severe plastic deformation of metals by high-pressure tube
[111] K. Mroczek, S. Kalisz, M. Pronobis, J. Soltys, The effect of halloysite additive on twisting, Scr. Mater. 60 (2009) 175–177.
operation of boilers firing agricultural biomass, Fuel Process. Technol. 92 [141] G. Faraji, M.M. Mashhadi, H.S. Kim, Tubular channel angular pressing (TCAP)
(2011) 845–855. as a novel severe plastic deformation method for cylindrical tubes, Mater.
[112] M. Schütze, M. Malessa, V. Rohr, T. Weber, Development of coatings for Lett. 65 (2011) 3009–3012.
protection in specific high temperature environments, Surf. Coat. Technol. [142] G. Faraji, A. Babaei, M.M. Mashhadi, K. Abrinia, Parallel tubular channel
201 (2006) 3872–3879. angular pressing (PTCAP) as a new severe plastic deformation method for
[113] T.S. Sidhu, S. Prakash, R.D. Agrawal, Hot corrosion studies of HVOF NiCrBSi cylindrical tubes, Mater. Lett. 77 (2012) 82–85.
and Stellite-6 coatings on a Ni-based superalloy in an actual industrial [143] G.B. Hamu, D. Eliezer, L. Wagner, The relation between severe plastic
environment of a coal fired boiler, Surf. Coat. Technol. 201 (2006) 1602– deformation microstructure and corrosion behavior of AZ31 magnesium
1612. alloy, J. Alloys Compd. 468 (2009) 222–229.
[114] D. Bendix, G. Tegeder, P. Crimmann, J. Metschke, M. Faulstich, Development [144] E.A. El-Danaf, M.S. Soliman, A.A. Almajid, M.M. El-Rayes, Enhancement of
of thermal sprayed layers for high temperature areas in waste incineration mechanical properties and grain size refinement of commercial purity
plants, Mater. Corros. 59 (2008) 389–392. aluminum 1050 processed by ECAP, Mater. Sci. Eng. A 458 (2007) 226–234.
[115] R.A. Mahesh, R. Jayaganthan, S. Prakash, Evaluation of hot corrosion [145] K. Neishi, Z. Horita, T.G. Langdon, Grain refinement of pure nickel using
behaviour of HVOF sprayed Ni–5Al and NiCrAl coatings in coal fired boiler equal-channel angular pressing, Mater. Sci. Eng. A 325 (2002) 54–58.
environment, Surf. Eng. 26 (2010) 413–421. [146] I.-H. Kim, S.I. Kwun, Precipitation behavior of c00 in severely plastic deformed
[116] H. Naganuma, N. Ikeda, T. Ito, H. Satake, M. Matsuura, Y. Ueki, R. Yoshiie, I. Ni-base alloys, Mater. Sci. Forum 534–536 (2007) 213–216.
Naruse, Control of ash deposition in solid fuel fired boiler, Fuel Process. [147] H. Ueno, K. Kakihata, Y. Kaneko, S. Hashimoto, A. Vinogradov, Enhanced
Technol. 105 (2013) 77–81. fatigue properties of nanostructured austenitic SUS 316L stainless steel, Acta
[117] J.A. Hearley, C. Liu, J.A. Little, A.J. Sturgeon, Corrosion of Ni–Al high velocity Mater. 59 (2011) 7060–7069.
oxyfuel (HVOF) thermal spray coating by fly ash and synthetic biomass ash [148] C.X. Huang, G. Yang, Y.L. Gao, S.D. Wu, Z.F. Zhang, Influence of processing
deposits, Br. Corros. J. 36 (2001) 111–120. temperature on the microstructures and tensile properties of 304L stainless
[118] H.Y. Al-Fadhli, J. Stokes, M.S.J. Hashmi, B.S. Yilbas, The erosion-corrosion steel byECAP, Mater. Sci. Eng. A 485 (2008) 643–650.
behavior of high velocity oxy-fuel (HVOF) thermally sprayed Inconel-625 [149] Z.J. Zheng, Y. Gao, Y. Gui, M. Zhu, Corrosion behaviour of nanocrystalline 304
coatings on different metallic surfaces, Surf. Coat. Technol. 200 (2006) 5782– stainless steel prepared by equal channel angular pressing, Corros. Sci. 54
5788. (2012) 60–67.
[119] M. Mohammadi, S. Javadpour, S.A.J. Jahromi, K. Shirvani, A. Kobayashi, [150] P.C. Millett, T. Desai, V. Yamakov, D. Wolf, Atomistic simulations of
Characterization and hot corrosion performance of LVPS and HVOF- diffusional creep in a nanocrystalline body-centered cubic material, Acta
CoNiCrAlYSi coatings, Vacuum 86 (2012) 1458–1464. Mater. 56 (2008) 3688–3698.
[120] M.M. Verdian, K. Raeissi, M. Salehi, Characterization and electrochemical [151] I.A. Ovid’ko, A.G. Sheinerman, Enhanced ductility of nanomaterials through
properties of Ni(Si)/Ni5Si2 multiphase coatings prepared by HVOF spraying, optimization of grain boundary sliding and diffusion processes, Acta Mater.
Appl. Surf. Sci. 261 (2012) 493–498. 57 (2009) 2217–2228.
[121] T.S. Sidhu, R.D. Agrawal, S. Prakash, Hot corrosion of some superalloys and [152] B. Arab, S. Sanjabi, A. Shokuhfar, Size dependency of self-diffusion and creep
role of high-velocity oxy-fuel spray coatings—a review, Surf. Coat. Technol. behavior of nanostructured metals, Mater. Lett. 65 (2011) 712–715.
198 (2005) 441–446. [153] S. Gollapudi, K.V. Rajulapati, I. Charit, C.C. Koch, R.O. Scattergood, K.L. Murty,
[122] H. Herman, S. Sampath, R. Mccune, Thermal spray: current status and future Creep in nanocrystalline materials: role of stress assisted grain growth,
trends, MRS Bull. 25 (2000) 17–25. Mater. Sci. Eng. A 527 (2010) 5773–5781.
[123] M.A. Uusitalo, P.M.J. Vuoristo, T.A. Mäntylä, High temperature corrosion of [154] V. Slenicka, J. Dvorak, P. Kral, M. Svoboda, M. Kvapilova, T.G. Langdon, Factors
coatings and boiler steels in reducing chlorine-containing atmosphere, Surf. influencing creep flow and ductility in ultrafine-grained metals, Mater. Sci.
Coat. Technol. 161 (2002) 275–285. Eng. A 558 (2012) 403–411.
[124] M.A. Uusitalo, P.M.J. Vuoristo, T.A. Mäntylä, High temperature corrosion of [155] K.Y. Zhu, A. Vassel, F. Brisset, K. Lu, J. Lu, Nanostructure formation mechanism
coatings and boiler steels in oxidizing chlorine-containing atmosphere, of a-titanium using SMAT, Acta Mater. 52 (2004) 4101–4110.
Mater. Sci. Eng. A 346 (2003) 168–177. [156] G. Liu, S.C. Wang, X.F. Lou, J. Lu, K. Lu, Low carbon steel with nanostructured
[125] M.C. Mayoral, J.M. Andrés, J. Belzunce, V. Higuera, Study of sulphidation and surface layer induced by high-energy shot peening, Scr. Mater. 44 (2001)
chlorination on oxidized SS310 and plasma-sprayed Ni–Cr coatings as 1791–1795.
simulation of hot corrosion in fouling and slagging in combustion, Corros. [157] G. Liu, J. Lu, K. Lu, Surface nanocrystallization of 316L stainless steel induced
Sci. 48 (2006) 1319–1336. by ultrasonic shot peening, Mater. Sci. Eng. A 286 (2000) 91–95.
[126] M. Torrell, S. Dosta, J.R. Miguel, J.M. Guilemany, Optimisation of HVOF [158] B. Arifvianto, M. Mahardika, P. Dewo, P.T. Iswanto, U.A. Salim, Effect of
thermal spray coatings fortheir implementation as MSWI superheater surface mechanical attrition treatment (SMAT) on microhardness, surface
protector, Corros. Eng. Sci. Technol. 45 (2010) 84–93. roughness and wettability of AISI 316L, Mater. Chem. Phys. 125 (2011) 418–
[127] J. Kalivodova, D. Baxter, M. Schütze, V. Rohr, High temperature corrosion of 426.
coatings and boiler steels in oxidizing chlorine-containing atmosphere, [159] L. Wen, Y. Wang, Y. Zhou, L.X. Guo, J.H. Ouyang, Iron-rich layer introduced by
Mater. Corros. 56 (2005) 882–889. SMAT and its effect on corrosion resistance and wear behavior of 2024 Al
[128] V. Rohr, M. Schütze, Diffusion coatings for heat exchanger materials, Surf. alloy, Mater. Chem. Phys. 126 (2011) 301–309.
Eng. 20 (2004) 266–274. [160] M. Laleh, F. Kargar, Effect of surface nanocrystallization on the
[129] C. Leyens, I.G. Wright, B.A. Pint, Hot corrosion of an EB-PVD thermal-barrier microstructural and corrosion characteristics of AZ91D magnesium alloy, J.
coating system at 950 °C, Oxid. Met. 54 (2000) 401–424. Alloys Compd. 509 (2011) 9150–9156.
[130] P. Panjan, M. Cekada, M. Panjan, D. Kek-Merl, Growth defects in PVD hard [161] L.L. Shaw, J.-W. Tian, A.L. Ortiz, K. Dai, J.C. Villegas, P.K. Liaw, R. Ren, D.L.
coatings, Vacuum 84 (2010) 209–214. Klarstrom, A direct comparison in the fatigue resistance enhanced by surface
[131] M. Bromark, M. Larsson, P. Hendenqvist, S. Hogmark, Wear of PVD Ti/TiN severe plastic deformation and shot peening in a C-2000 superalloy, Mater.
multilayer coatings, Surf. Coat. Technol. 90 (1997) 217–223. Sci. Eng. A 527 (2010) 986–994.
 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26 25

[162] M. Wen, G. Liu, J.-F. Gu, W.-M. Guan, J. Lu, The tensile properties of titanium [192] A. Nafari, A. Nylund, Field study on superheater tubes in the loop seal of a
processed by surface mechanical attrition treatment, Surf. Coat. Technol. 202 wood fired CFB plant, Mater. Corros. 55 (2004) 909–920.
(2008) 4728–4733. [193] I. Klevtsov, H. Tallermo, T. Bojarinova, R. Crane, High temperature corrosion
[163] T. Balusamy, S. Kumar, T.S.N.S. Narayanan, Effect of surface of boiler steels under chlorine-containing surface deposits, J. Press. Ves.
nanocrystallization on the corrosion behaviour of AISI 409 stainless steel, Technol. 127 (2005) 106–111.
Corros. Sci. 52 (2010) 3826–3834. [194] C. Cuevas-Arteaga, D. Verhelst, A. Alfantazi, A. Alfantazi, Performance of alloy
[164] K. Lu, J. Lu, Nanostructured surface layer on metallic materials induced by 625 under combustion gas environments: a review, ECS Trans. 28 (2010) 61–
surface mechanical attrition treatment, Mater. Sci. Eng. A 375–377 (2004) 76.
38–45. [195] A. Phongphiphat, R. Ryu, Y.B. Yang, K.N. Finney, A. Leyland, V.N. Sharifi, J.
[165] Y.-W. Hao, B. Deng, C. Zhong, Y.-M. Jiang, J. Li, Effect of surface mechanical Swithenbak, Investigation into high-temperature corrosion in a large-scale
attrition treatment on corrosion behavior of 316 stainless steel, J. Iron Steel municipal waste-to-energy plant, Corros. Sci. 52 (2010) 3861–3874.
Res. Int. 16 (2009) 68–72. [196] M.F. Ashby, Materials Selection in Mechanical Design, fourth ed., Elsevier,
[166] H. Han, L. Zhang, J. Lu, W. Zhang, Thermal stability and corrosion resistance of Oxford, UK, 2010.
nanocrystallized zirconium formed by surface mechanical attrition [197] A. Jahan, M.Y. Ismail, S.M. Sapuan, F. Mustapha, Material screening and
treatment, J. Mater. Res. 24 (2009) 3136–3145. choosing methods – a review, Mater. Des. 31 (2010) 696–705.
[167] A.Q. Lu, Y. Li, G. Liu, Q.H. Zhang, C.M. Liu, Effect of nanocrystalline and twin [198] A.K. Sharma, N. Gupta, Material selection of RF-Mems switch used for
boundaries on corrosion behavior of 316L stainless steel using SMAT, Acta reconfigurable antenna using Ashby’s methodology, Prog. Electromagn. Res.
Metall. Sin. 19 (2006) 183–189. Lett. 31 (2012) 147–157.
[168] T. Wang, J. Yu, B. Dong, Surface nanocrystallization induced by shot peening [199] A. Chauhan, R. Vaish, A comparative study on material selection for micro-
and its effect on corrosion resistance of 1Cr18Ni9Ti stainless steel, Surf. Coat. electromechanical systems, Mater. Des. 41 (2012) 177–181.
Technol. 200 (2006) 4777–4781. [200] B.N. Aditya, N. Gupta, Material selection methodology for gate dielectric
[169] H.-W. Chang, P.M. Kelly, Y.-N. Shi, M.-X. Zhang, Thermal stability of material in metal–oxide–semiconductor devices, Mater. Des. 35 (2012) 696–
nanocrystallized surface produced by surface mechanical attrition 700.
treatment in aluminum alloys, Surf. Coat. Technol. 206 (2012) 3970–3980. [201] Z. Szklarska-Smialowska, Pitting and Crevice Corrosion of Metals, NACE Intl.,
[170] Y.J. Mai, X.H. Jie, L.L. Liu, N. Yu, X.X. Zheng, Thermal stability of Houston, TX, 2005.
nanocrystalline layers fabricated by surface nanocrystallization, Appl. Surf. [202] H.-B. Li, Z.-H. Jiang, Y. Yang, Y. Cao, Z.-R. Zhang, Pittingcorrosion and crevice
Sci. 256 (2010) 1972–1975. corrosion behaviors of high nitrogen austenitic stainless steels, Int. J. Miner.
[171] M. Wen, C. Wen, P. Hodgson, Y. Li, Thermal oxidation behaviour of bulk Metall. Mater. 16 (2009) 517–524.
titanium with nanocrystalline surface layer, Corros. Sci. 59 (2012) 352–359. [203] N.S. Zadorozne, C.M. Giordano, M.A. Rodríguez, R.M. Carranza, R.B. Rebak,
[172] J. He, J.M. Schoenung, Nanostructured coatings, Mater. Sci. Eng. A 336 (2002) Crevice corrosion kinetics of nickel alloys bearing chromium and
274–319. molybdenum, Electrochim. Acta 76 (2012) 94–101.
[173] K. Tao, X.-L. Zhou, H. Cui, J.-S. Zhang, Oxidation and hot corrosion behaviors of [204] ASTM G48, Standard Test Methods for Pitting and Crevice Corrosion
HVAF-sprayed conventional and nanostructured NiCrC coatings, Trans. Resistance of Stainless Steels and Related Alloys by Use of Ferric Chloride
Nonferrous Met. Soc. China 19 (2009) 1151–1160. Solution 03, 2009.
[174] A. Keyvani, M. Saremi, M.H. Sohi, An investigation on oxidation, hot corrosion [205] D.R.H. Jones, Creep failures of overheated boiler, superheater and reformer
and mechanical properties of plasma-sprayed conventional and tubes Eng, Fail. Anal. 11 (2004) 873–893.
nanostructured YSZ coatings, Surf. Coat. Technol. 206 (2011) 208–216. [206] H.-T. Yao, F.-Z. Xuan, Z. Wang, S.-T. Tu, A review of creep analysis and design
[175] P. Yu, W. Wang, F. Wang, S. Zhu, High-temperature corrosion behavior of under multi-axial stress states, Nucl. Eng. Des. 237 (2007) 1969–1986.
sputtered K38 nanocrystalline coatings with and without yttrium addition in [207] K. Maruyama, H.G. Armaki, R.P. Chen, K. Yoshimi, M. Yoshizawa, M. Igarashi,
molten sulfate at 900 °C, Surf. Coat. Technol. 206 (2011) 68–74. Cr concentration dependence of overestimation of long term creep life in
[176] T. Sundararajan, S. Kuroda, J. Kawakita, S. Seal, High temperature corrosion of strength enhanced high Cr ferritic steels, Int. J. Press. Ves. Piping 87 (2010)
nanoceria coated 9Cr–1Mo ferritic steel in air and steam, Surf. Coat. Technol. 276–281.
201 (2006) 2124–2130. [208] M.W. Spindler, S.L. Spindler, Creep deformation, rupture and ductility of
[177] H. Wang, D. Zuo, G. Chen, G. Chen, G. Sun, X. Li, X. Cheng, Hot corrosion Esshete 1250, Int. J. Press. Ves. Piping 85 (2008) 89–98.
behaviour of low Al NiCoCrAlY cladded coatings reinforced by nano-particles [209] P.P. Psyllaki, G. Pantazopoulos, H. Lefakis, Metallurgical evaluation of creep-
on a Ni-base super alloy, Corros. Sci. 52 (2010) 3561–3567. failed superheater tubes, Eng. Fail. Anal. 16 (2009) 1420–1431.
[178] A. Rahman, V. Chawla, R. Jayaganthan, R. Chandra, R. Ambardar, Study of [210] W.F. Hosford, Mechanical Behavior of Materials, second ed., Cambridge
cyclic hot corrosion of nanostructured Cr/Co–Al coatings on superalloy, University Press, New York, 2010.
Mater. Chem. Phys. 126 (2011) 253–261. [211] ASTM E8/E8M-11 Standard Test Methods for Tension Testing of Metallic
[179] M. Mohammdnezhad, M. Shamanian, M.H. Enayati, Formation of Materials, ASTM International, 2011.
nanostructured NiAl coating on carbon steel by usingmechanical alloying, [212] G.Y. Lai, High Temperature Corrosion and Materials Applications, ASM
Appl. Surf. Sci. 263 (2012) 730–736. International, 2007.
[180] A. Rahman, R. Jayaganthan, S. Prakash, V. Chawla, R. Chandra, High [213] J.R. Kish, D.L. Singbeil, J.R. Keiser, North american experience with composite
temperature oxidation behavior of nanostructured Ni–Al coatings on tubes in kraft recovery boilers, Pulp Paper Canada 106 (2005) T75–T81.
superalloy, J. Alloys Compd. 472 (2009) 478–483. [214] M. Hamaguchi, E. Vakkilainen, Corrosion of superheater tubes in recovery
[181] D. Zhong, J.J. Moore, E. Sutter, B. Mishra, Microstructure, composition and boilers: a challenge, O Papel 71 (2010) 57–71.
oxidation resistance of nanostructured NiAl and Ni–Al–N coatings produced [215] <www.haynes.ch/doc/HAYNES_617.pdf> (accessed 22.05.13).
by magnetron sputtering, Surf. Coat. Technol. 200 (2005) 1236–1241. [216] <www.hightempmetals.com/techdata/hitempInconel625data.php>
[182] I. Pisa, G. Lazaroiu, Influence of co-combustion of coal/biomass on the (accessed 22.05.13).
corrosion, Fuel Process. Technol. 104 (2012) 356–364. [217] R.W. Swindeman, M.J. Swindeman, A comparison of creep models for nickel
[183] K. Persson, M. Broström, J. Carlsson, A. Nordin, R. Backman, High temperature base alloys for advanced energysystems, Int. J. Press. Ves. Piping 85 (2008)
corrosion in a 65 MW waste to energy plant, Fuel Process. Technol. 88 (2007) 72–79.
1178–1182. [218] M.D. Mathew, K.B.S. Rao, S.L. Mannan, Creep properties of service-exposed
[184] J. Lehmusto, P. Yrjas, B.-J. Skrifvars, M. Hupa, High temperature corrosion of Alloy 625 after re-solution annealing treatment, Mater. Sci. Eng. A 372 (2004)
superheater steels by KCl and K2CO3 under dry and wet conditions, Fuel 327–333.
Process. Technol. 104 (2012) 253–264. [219] ASTM A213/213M Standard Specification for Seamless Ferritic and Austenitic
[185] K. Chandra, V. Kain, G.K. Dey, Failure of 2.25Cr–1Mo steel superheater tubes Alloy-steel Boiler, Superheater, and Heat-exchanger Tubes, ASTM
in a fluidized bed combustor due to fireside corrosion, Mater. Charact. 62 International, 2011.
(2011) 62–69. [220] ASTM B564-11 Standard Specification for Nickel Alloy Forgings, ASTM
[186] A.U. Syed, N.J. Simms, J.E. Oakey, Fireside corrosion of superheaters: effects of International, 2011.
air and oxy-firing of coal and biomass, Fuel 101 (2012) 62–73. [221] ASTM B626-10 Standard Specification for Welded Nickel and Nickel–Cobalt
[187] P. Viklund, A. Hjörnhede, P. Henderson, A. Stálenheim, R. Pettersson, Alloy Tube, ASTM International, 2010.
Corrosion of superheater materials in a waste-to-energy plant, Fuel Process. [222] ASTM International Volume 2: Properties and Selection: Nonferrous Alloys
Technol. 105 (2013) 106–112. and Special Purpose Materials, 2000.
[188] P. Viklund, R. Pettersson, HCl-induced high temperature corrosion of stainless [223] <www.smt.sandvik.com/en/materials-center/material-datasheets/welding-
steels in thermal cycling conditions and the effect of peroxidation, Oxid. Met. products/welding-wire/sanicro-60> (accessed 22.05.13).
76 (2011) 111–126. [224] ASTM B668-05 Standard specification for UNS N08028 Seamless Pipe and
[189] H. Asteman, M. Spiegel, Investigation of the HCl(g) attack on pre-oxidized Tube, ASTM International, 2010.
pure Fe, Cr, Ni and commercial 304 steel at 400 °C, Corros. Sci. 49 (2007) [225] ASTM A387/A387M-11 Standard Specification for Pressure Vessel Plates,
3626–3637. Alloy Steel, Chromium–Molybdenum, ASTM International, 2011.
[190] C. Liu, J.A. Little, P.J. Henderson, P. Ljung, Corrosion of TP347H FG stainless [226] ASTM A53/A53M-12 Standard Specification for Pipe, Steel, Black and Hot-
steel in a biomass fired PF utilityboiler, J. Mater. Sci. 36 (2001) 1015–1026. dipped, Zinc-coated, Welded and Seamless, ASTM International, 2012.
[191] M. Montgomery, S.A. Jensen, U. Borg, O. Biede, T. Vilhelmsen, Experiences [227] ASTM A240/A240M-12a Standard Specification for Chromium and
with high temperature corrosion at straw-firing power plants in Denmark, Chromium–Nickel Stainless Steel Plate, Sheet, and Strip for Pressure Vessels
Mater. Corros. 62 (2011) 593–605. and for General Applications, ASTM International, 2012.
26 R.A. Antunes, M.C.L. de Oliveira / Corrosion Science 76 (2013) 6–26

[228] A. Hernas, G. Moskal, K. Rodak, J. Pasternak, Properties and microstructure of [231] EN 10216-5 European Standard. Seamless Steel Tubes for Pressure Purposes –
12% Cr–W steels after long-term service, J. Achiev. Mater. Manufact. Eng. 17 Technical Delivery Conditions – Part 5: Stainless Steel Tubes, European
(2006) 69–72. Committee for Standardization, 2004.
[229] ASTM B407-08a Standard Specification for Nickel–Iron–Chromium Alloy [232] Z.-F. Hu, Heat-resistant steels, microstructure evolution and life assessment
Seamless Pipe and Tube, ASTM International, 2008. in power plants, in: M. Rasul (Ed.), Thermal Power Plants, In Tech
[230] K. Zhan, C.H. Jiang, V. Ji, Surface mechanical properties of S30432 austenitic Publications, Croatia, 2012, pp. 195–226.
steel after shot peening, Appl. Surf. Sci. 258 (2012) 9559–9563. [233] ASTM A250/A250M-05 Standard Specification for Electric-resistance-welded
Ferritic-alloy Steel Boiler and Superheater Tubes, ASTM International, 2009.