Volume 6 Preprint 74
Erosion Corrosion of Coatings Exposed in a Fluidised Bed Test Rig
Anders HjÃƒÂ¶rnhede and Anders Nylund
Keywords: thermal spraying, laser cladding, erosion corrosion
Because you are not logged-in to the journal, it is now our policy to display a 'text-only' version of the preprint. This version is obtained by extracting the text from the PDF or HTML file, and it is not guaranteed that the text will be a true image of the text of the paper. The text-only version is intended to act as a reference for search engines when they index the site, and it is not designed to be read by humans!
If you wish to view the human-readable version of the preprint, then please Register (if you have not already done so) and Login. Registration is completely free.
Volume 6 Paper H018
EROSION CORROSION OF COATINGS EXPOSED
IN A FLUIDISED BED TEST RIG
Anders Hjörnhede and Anders Nylund
Materials Science and Engineering, Chalmers University of Technology
SE 412 96 Göteborg, Sweden, firstname.lastname@example.org,
Low alloyed steel tubes deposited with 18 different coatings were
exposed in a fluidised bed test rig. The material temperature was
550°C, the erodent silica (SiO2) and the gas environment air. The
coatings used in this study were commercially available ones based on
Fe, Ni, Co and carbides deposited with the HVOF (High Velocity OxyFuel), arc-spray and laser cladding techniques. After 30 days of
exposure the coated tubes were dismounted and the material
degradation quantified in terms of diameter reduction and roundness
measurements. Further, the tubes were sectioned and the coating
processing the data a classification was done. The performance of the
thermally sprayed coatings was better compared to what was shown by
the laser deposited ones. It is suggested that the larger erosion
resistance is due to the presence of hard carbides and oxide inclusions
in the coatings.
Keywords: thermal spraying, laser cladding, erosion corrosion
The importance of the fluidised bed combustion (FBC) technology
during production of heat and electricity is steadingly increasing, due
to its economical and environmental advantages. The FBC:s in service
can be either bubbling or circulating and they may work under
atmospheric or pressurised conditions. A drawback with the technique
is however that components such as heat exchangers and superheater
tubes are susceptible to erosion-corrosion.
A solution to the problem is to cover exposed low alloyed steel
components with protective coatings. Field studies have shown that
the coating performance is strongly dependent on the type of fuel
used and the atmospheric conditions developed [ref. 1]. The corrosion
resistance of Co- and Fe-based coatings in power plants fired with
coal is good, while in biomass fired plants they are subjected to
corrosion and erosion-corrosion, respectively [ref. 1]. The best overall
performance when comparing different types of plants and fuels is
usually shown by Ni-based coatings and the worst by carbidecontaining coatings [ref. 1]. In order to obtain a deeper understanding
of the coating degradation processes taking place in commercial
thermal plants, there is a need of a reference study under wellcontrolled conditions. The aim of this paper is therefore to determine
the erosion-corrosion resistance of coatings deposited with the arcspray, High Velocity Oxyfuel (HVOF) and laser techniques during
exposure in a laboratory test rig with air as the fluidising gas and a
controlled particle flow.
Figure 1a) Illustration of the test rig with the exposure positions of the test tubes.
b) The lengthwise measurement positions on the tube surface.
A specially designed erosion-corrosion test rig made of stainless steel
has been used for the exposures, Fig. 1a. The cross-sectional
dimension of the bed vessel is 0.2 m x 0.3 m and the bed height is
0.86 m. Electrical heat radiant elements are mounted around the rig.
The erodent material was the commercial product “Silversand 90”,
which consists of SiO2 with an average particle diameter of 0.7 mm, a
sphericity of 0.8 and a density of 2.6 g/cm3. The volume of the
erodent material in the bed was 48 dm3. Air enters the bed through
eleven evenly distributed nozzles located at the bottom of the vessel.
The minimum fluidisation velocity (Umf) was calculated to be 0.25 m/s
and the superficial velocity (Ufl) is measured to be 0.4 m/s resulting in
an excess gas velocity (Uex) of 0.15 m/s at the test temperature
546±1°C. Three probes at different positions (see Fig. 1a) are
continuously monitoring the bed pressure and temperature.
The tube set consists of 59 tubes, of which 21 are exchangeable and
used for the experiments. The tube length is 260 mm. In order to
calibrate the material wastage at these positions reference tubes made
(Fe18.1Cr10.1Ni1.3Mn0.5Si0.3Mo0.2Cu0.02C) were exposed at 550°C
(Fe0.15C1Cr0.5Mo) with the arc-spray, High Velocity Oxy-fuel (HVOF)
and laser cladding techniques were then exposed in the same
positions. The coating compositions are given in Table 1. In order to
reflect the commercial use of coated tubes in FBC boilers no
pretreatment of the tube surfaces was performed prior to exposure.
The exposure time was 720 h at the same operational conditions as
used for the reference tubes.
Table 1 Coatings used in this study; coating techniques and compositions.
Before exposure the reference tubes were finish-turned to an outer
Metco 443 NS
Metco 43C NS
diameter of 26.4 mm to obtain a smooth surface. The tube diameter
was measured at every 30th degree around the tubes at three different
lengthwise positions as illustrated in Fig. 1b. In order to compensate
for surface irregularities the surface topography was continuously
recorded in a roundness profilometer (Mitutoyo Roundtest RA-116).
The topography was then superimposed on the diameter and
integration gives the area of the cross-section. By repeating the same
measurement procedure after exposure and subtracting the crosssectional area of the exposed tube from the corresponding area of the
non-exposed tube the material wastage could thereby be estimated.
Exactly the same characterisation was made on the coated tubes
before and after exposure.
Results and discussion
Figure 2 shows the material wastage with the circumferential angle as
recorded on five 304L tubes positioned at different heights along a
vertical line in the test rig (tubes 2, 7, 10, 18 and 21 according to Fig.
1a). The measurements were made in position B on the tube surfaces
as defined in Fig. 1b. The appearance of the wastage profiles and the
degree of material degradation is the same for tubes 10, 18 and 21
with two wastage maxima observed at the circumferential angles 150°
and 225°. The material wastage on the tube at the lowest position
Wastage rate [µm/1000h]
Circumferential angle [º]
(tube 2) is much less and only one wastage maximum is recorded on
the central part on the tube underside. Tube 7 shows a transient
behaviour in between that of tube 2 and the tubes in the upper part of
the bed. The erosion pattern shown by tube 2 is usually referred to as
Type B behaviour, while the wastage profiles of the upper tubes
display a typical Type A behaviour [ref. 2]. The difference in wastage
behaviour is due to changes in the hydrodynamic conditions [ref 3,4].
It has been suggested [ref. 5] that the Type B behaviour observed in
the lower region of the bed is a result of erosion from bubbles that are
small in comparison to the bed cross-section and therefore are not
constrained by the bed walls. In the upper regions of the bed
coalescence into single bubbles, creates a more constrained flow
pattern and the wastage profile changes to Type A.
The absolute material loss in position B on the tube surface as a
function of exposure position in the test rig is given in Fig. 3 for the
Figure 2. Material wastage profiles for selected tubes at different heights above
the air distributor plate.
reference 304L- tubes (green colour). An inspection of the figure bars
reveals that on the same level above the air distributor plate (e.g.
tubes 10, 11 and 12) the material degradation is less on the tubes
closest to the vessel walls. The red coloured bars show the same
parameter as measured for the coatings. The type of coatings located
at the specified exposure positions can be seen from Table 1. The
fluidisation conditions have been identical in both cases except for the
exposure time which was twice as long during the reference
Material loss [µm /1000h]
Exposure position in test rig
Figure 3. Absolute material loss on reference tubes [green] and coatings [red] as a
function of exposure position.
In order to obtain a comparison with regard to the different exposure
positions in the test rig a normalisation was done. The absolute
material loss of each coating was divided by the material loss of the
reference tube at the corresponding position. The difference in
exposure time was accounted for. This relative material loss is
presented in Fig. 4 as an average value between the relative material
losses as calculated at the three lengthwise tube positions; 80, 130
and 180 mm, (see Fig 1b). It is clearly seen that the coatings deposited
with the laser technique are subjected to a higher degree of material
loss than the thermally sprayed ones. The erosion rate is 4 (Duroc
5177, Metco 43C NS, Duroc 17x1%C, Stellite 6 and Amdry 995C) to 9
times higher (Metco 8443) than for the reference tubes exposed at the
Relataive Material Loss
6 7+8 9 10 11 12 13 14 15 16 17 18 19 20 21
Exposure position in test rig
Figure 4. Relative material loss for coatings in the test rig.
losses of the thermally sprayed coatings are all within the interval 1
(Metcoloy 2, HVOF) to 3.5 (HMSP 1616, HVOF). The Ni-based
materials, HMSP 1616, HMSP 1660 and Inconel 625 along with the Cobased Eutronic 508 show the highest erosion rates in this category.
There is no observed correlation between the Cr-concentration and
the erosion resistance. Despite the same Cr-content (21.5 wt%) and
almost the same composition the degree of material loss is 2.5 times
larger on the laser deposited Duroc 5171 coating than on the HVOF
deposited Inconel 625 coating.
The microstructure from the laser deposited coatings is mainly free
from pores and oxides, while the thermally sprayed coatings contain
significant amounts of oxide inclusions and pores, Table 2. The
content of oxide phases in the arc-sprayed Metcoloy 2 coatings is
about 15% and the porosity about 3%. The corresponding values for
respectively. A cross-section of a Metcoloy 2 coating, deposited with
the arc-spray technique is shown in Fig. 5a. The layered splatter
structure which is typical for arc-sprayed coatings is clearly seen. The
black areas are pores and the brownish are Cr – rich oxides, [ref. 6].
Figure 5b shows the microstructure of a HVOF deposited HMSP-1660
coating. The porosity is somewhat larger compared to the arc-sprayed
coatings, but the oxide phase content is reduced. The microstructure
of laser coated Stellite 21 is homogenous and virtually free from pores
and oxides, Fig. 5c.
Table 2. Oxide phase content and porosity for some selected coatings.
Oxide phase content [%]
Duroc 5177, laser
Metco 443, laser
Stellite 21, laser
Metcoloy 2, arc-spray, air as
Metcoloy 2, arc-spray, N2 as
HMSP 1660, HVOF
Hardness measurements were performed on some coatings which had
not been exposed in the test rig. In Table 3 the results are compared
with the ones obtained for the exposed tubes. The results show that
except for the Metcoloy 2 coating which was arc-sprayed in N2–gas the
coatings are slightly softer after exposure. A suggestion for the
softening is that the coatings are subjected to recovery and stress
relief processes which usually take place at temperatures just below or
around the exposure temperature. Considering the Metcoloy 2
coatings the hardness of the oxide phase is larger than that of the
metal phase. Thus, the coating is strengthened by the oxide phase.
Figure 5. Cross-sections of exposed coatings a) Metcoloy 2, Arc-sprayed in air b) HMSP 1660, HVOF
deposited c) Stellite 21, laser deposited (Optical microscopy).
From the experimental results shown above it is clear that the Metco
3007 coating performed very well. The high erosion corrosion
resistance has been attributed to the formation of a skeleton network
of hard oxide/carbides within a ductile binder [ref. 7]. The coating
contains 80 wt% Cr3C2 particles, of which the hardness is about 1000-
1300 HV at 550ºC [ref. 8]. The hardness decrease of Al2O3 and ZrO2 at
the same temperature is small compared to that of the metal matrix
[ref. 8]. Most likely Cr2O3 shows the same behaviour. It is therefore
suggested that the erosion resistance of the thermally sprayed
coatings at elevated temperatures is associated with the presence of
hard carbide or oxide phases. The correlation between hardness and
erosion rate is not straightforward even if the erosion rate is
considered to decrease with the coating hardness. It has been stated
that the oxides must be small and well dispersed [ref. 9]. The results
in the present study contradict this statement, since the oxide
inclusions in the arc–sprayed coatings are relatively large and non-
dispersed. An improved erosion performance at high temperatures due
to the reinforcement of a metal coating with a secondary hard phase
has also been observed in a previous study [ref. 10]. The requirement
is however that the inter–droplet cohesion is high, either by choice of
spray technique or quantity of particles, else the opposite effect is
Table 3 Hardness of the coatings in unexposed and exposed conditions at room
Duroc 5177, laser
Metco 443, laser
Stellite 21, laser
Metcoloy 2, air as carrier gas
Metcoloy 2, N2 as carrier gas
HMSP 1660, HVOF
The coating performance during exposure in commercial fluidised bed
boilers differs from the laboratory studies [ref. 1]. Metco 3007 tends to
oxidise and delaminate from the substrate when exposed in biomass
fired boilers at material temperatures of 420ºC and 550ºC. Arc-
sprayed Metcoloy 2 – coatings are also subjected to severe high
temperature corrosion during combustion of biofuel. The corrosion
attack results in an increased material loss rate and the performance
of thermally sprayed coatings in biomass fired boilers is generally
weak. However, in the same environments, Stellite 21 and Duroc
17x1%C coatings deposited with the laser technique showed no
significant corrosion attacks and only a low degree of erosion. Hence,
the results from the boiler exposures in comparison with the results
from this study points out that the chemical environment is decisive in
terms of material loss.
Twenty coating qualities as deposited with the arc spray, HVOF and
laser cladding techniques have been exposed in an erosion test rig
under controlled conditions. The conclusions are drawn as follows;
Laser deposited coatings are subjected to a higher degree of
erosion than thermally sprayed coatings.
The coating hardness at the exposure temperature is of major
Coatings containing sufficient amounts of oxides or carbides
which are relatively hard at the exposure temperature are
subjected to a less degree of erosion.
The Cr- and Ni- contents seem to be of minor importance for
the erosion resistance.
The coating performance at elevated temperatures is strongly
dependent on the chemical environment.
The KME (Consortium for Material Technology directed towards
Thermal Energy Processes) are acknowledged in financing this project.
Midroc Metalock AB and Duroc Energy AB are also acknowledged for
production of the coatings.
1. Hjörnhede A. and Nylund A., ”Performance of erosion corrosion
resistant coatings in different combustion environments”, 7th Liege
Conference Materials for Advanced Power Engineering, Materials for
Advanced Power Engineering, eds. J- Lecomte-Beckers, M. Carton, F.
Schubert and P. J. Ennis, Part II, (2002), 979-988.
2. Stringer J., Macadam S. S., Wright I. G. and Sethi V. K., Effect of
process variables on wastage in fluidized-bed combustors- criteria for
test procedures, Jornal De Physique IV, 3(C9), 797-805.
3. Macadam S. S. and Stringer J., The circumferential distribution of
wastage on in-bed tubes in fluidized-bed combustors, Wear, 186(1),
4. Hou P. Y., Macadam S. S., Zhang H. and Stringer J., Summary of
results from the Berkely in-tube erosion simulator, Materials at High
Temperatures, 14(3), 325-335.
5. Johansson K., Norling R., Hjörnhede A., Almstedt A-E., Johnson F.
and Nylund A., “Hydrodynamics and erosion on horizontal tubes in a
fluidized bed at elevated temperature”, Submitted to Chemical
6. Hjörnhede A. and Olefjord I., Erosion Corrosion of Thermally
Sprayed Fe12Cr and Cr3C2 – NiCr Coatings, Proc. 5th Int. Symp. On
High-Temperature Corrosion and Protection of Materials, Les Embiez,
France, May 22 – 26, 2000, eds R. Streiff, I.G. Wright, R.C. Krutenat, M.
Caillet and A. Galerie, (2001), 507 – 514.
7. Wang B. Q., Luer K., The erosion-oxidation behaviour of HVOF Cr3C2
– NiCr cermet coating, Wear, 174 (1994) 177-185.
8. Materials Science and Technology: Structure and Properties of
Ceramics. Swain m. (Ed), vol 11, VCH (1994).
9. Hoop P. J., Allen C., The high temperature erosion of commercial
thermally sprayed metallic and cermet coatings by solid particles,
Wear, 233-235 (1999), 334-341.
10. Haeth G. R., Kammer P. A. and Stack M.M., The High Temperature
Erosion/Corrosion Behavior of Industrial Thermally Sprayed Coatings,
Thermal Spray: Practical Solutions for Engineering Problems, ASM
International (1996), 29-37.