Enhanced high-temperature corrosion resistance of (Al
2O
3–Y
2O
3)/Pt micro-laminated coatings on 316L stainless steel alloy
Xiaoxu Ma, Yedong He
⇑, Deren Wang, Jin Zhang
Beijing Key Laboratory for Corrosion, Erosion and Surface Technology, University of Science and Technology Beijing, 100083 Beijing, China
a r t i c l e i n f o
Article history:
Received 11 June 2011 Accepted 10 September 2011 Available online 19 September 2011
Keywords:
A. Platinum A. Stainless steel A. Sputtered films B. SEM
C. Hot corrosion C. Oxidation
a b s t r a c t
A 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating was successfully prepared on 316L stainless steel alloy by magnetron sputtering. High-temperature cyclic oxidation and hot corrosion tests were adopted to investigate the high-temperature corrosion resistance of the coating. It is revealed that the (Al2O3– Y2O3)/Pt micro-laminated coating which effectively suppressed the inward diffusion of oxygen and corrosive fused salt to an extremely low level can significantly improve the high-temperature corrosion resistance of alloy substrate. The great mechanical properties of such coating were attributed to the brit- tle/ductile laminated composite structure by means of multilayer toughening and release mechanisms.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
Stainless steel covers a wide range of steel types and grades for corrosion or oxidation resistance applications. The main require- ment for stainless steels is that they should be corrosion resistance for a specified application or environment[1]. Austenitic stainless steels are used in numerous industrial applications, mainly due to their great corrosion resistance. And 316L austenitic stainless steel alloy offers the most resistance to corrosion in numerous standard services [2]. However, engineering components made of these steels are in many applications operated in such extreme condi- tions that improvement of their already outstanding high-temper- ature oxidation and hot corrosion resistance would be desirable. It is well known that the high oxygen affinity elements such as Y, Ce, La, Er and other rare earth elements added to steels in small amounts can improve their corrosion and high-temperature oxida- tion resistance because of the ‘‘reactive element effects’’ (REEs)[3].
Besides, recent research has revealed that the use of high-temper- ature protective coatings is one of the most practical methods to protect the stainless steel alloys against corrosion[4,5]. The design of high-temperature protective coatings should considering the following requirements: great oxidation and hot corrosion resis- tance, low oxygen permeability, low solubility in fused salts, similar thermal expansion to that of the substrate and strong adhe- sion to the substrate[6]. By now, alloy coatings[7–9]and ceramic
coatings[10,11]are the two most important developed practical protective coatings.
It was well known that the oxidation resistance of alloy coatings is determined by the formation of Al2O3, Cr2O3or SiO2scales by selective oxidation[12]and their failures caused by the stresses generated in oxide scales. Moreover, it is reported that the doped reactive elements (REs) in the alloy coatings can significantly im- prove the oxidation and spallation resistance of the alloy substrate owing to the REE[13]. While, from the composite materials point of view, the beneficial effects on the mechanical properties of the formed oxide scales doped with RE can be attributed to the multi- phase composite structures by means of the particle toughening and release mechanism as cracking and spalling of oxide scales and ceramic coatings with a single phase cannot be avoided completely.
With ceramic coatings, the high-temperature corrosion resis- tance of superalloys mainly depends on the integrality of ceramic coatings. Therefore, the processes of cracking and spallation are the key factors affecting the service lifetime[14]. As the ceramic composites possess superior mechanical characteristics at high temperatures compared with monophasic ceramics [15], it is reasonable to propose a ceramic coating with composite struc- tures should provide better mechanical properties. He, Gao and their teams [16–18] have developed various types of ceramic–
ceramic composite coatings since 1990s to be used to improve the resistance of superalloys to severe service conditions involv- ing high-temperature oxidation, corrosion, and wear. In such coatings, the fracture resistance is increased by the toughening effect, the thermal stress is decreased owing to the increase of
0010-938X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.corsci.2011.09.014
⇑ Corresponding author. Tel./fax: +86 10 62332715.
E-mail address:htgroup@mater.ustb.edu.cn(Y. He).
Contents lists available atSciVerse ScienceDirect
Corrosion Science
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / c o r s c i
thermal-expansion coefficients (CTEs), and Al2O3 phase can seal the alloy substrate well.
Noble metal has always been considered an ideal barrier mate- rial because it has plenty of attractive properties that no other materials can be compared to, such as thermodynamic stability, low oxygen diffusion rate and low solubility in molten salts. It is summarized from the literature that TiAl-based alloy coated with a Au coating (20
l
m thick) exhibits great hot corrosion resistance in NaCl-induced salt [19] and a non-carbon metal-based inert anode covered with a Pt layer (10l
m thick) provides good corro- sion resistance in cryolite–alumina molten salts[20]. These have indicated a feasible way using noble metals as barrier materials in high-temperature corrosion protection. However, the mutual diffusion at high temperatures might be harmful to the mechanical properties of the alloy substrate. Also, the noble metal (especially Au) protective coatings with low wear resistance are vulnerable to damage under erosion attack, resulting in the structural defects which will cause the coating early failure. Consequently, the recently developed noble metal coatings are difficult to be used in industrial production. Accordingly, a novel ceramic/noble metal composite coating system has been proposed and investigated in our latest works. It is shown that the Au nano-particles (NPs) doped Al2O3composite coating[21]providing enhanced oxidation and spallation resistance owing to the Au NPs (<2.0 wt.%) can effec- tively improve the fracture toughness and the adhesion of the coat- ing by means of NPs toughening, crack bridging and stress release.The aluminum electrolysis tests shown that the
a
-Al2O3(>85.0 wt.%) embedded Au (Au–Pt, Au–Pd, Au–Rh) matrix compos- ite coatings [22] exhibit superior erosion–corrosion resistance owing to the toughening of the composite structure and the inert- ness of the noble metals in cryolite molten salts. More importantly, a sealing mechanism is concluded that in these ceramic/noble metal composite coatings at least one phase with lowest oxygen diffusion coefficient and solubility in molten salts should seal the alloy substrate.
In this work, a novel composite coating with alternative 0.5 wt.% Y2O3 doped Al2O3 and Pt layers for high-temperature corrosion protection is designed and prepared on 316L stainless steel alloy substrates by magnetron sputtering (MS) method, which has become the popular choice for the deposition of a wide range of industrially important high-quality, well-adhered coat- ings[23,24]. The basic concept in improving the high-temperature corrosion resistance of the protective coating is to significantly re- duce the oxygen (corrosive agents) permeation through the coat- ing to the substrate material and remarkably increase the mechanical properties by introducing the noble metal materials to the ceramic matrix in form of multi-layers. The Y2O3 doped Al2O3 composite ceramic coatings are the preferable candidate for the ceramic layers as the small amount of doped RE and the multiphase composite structure have beneficial effects on the oxidation resistance and mechanical properties of the designed protective coating [25–27]. Moreover, it has been proved in our work[18]that the 0.5 wt.% Y2O3doped Al2O3coating having the Al2O3–YAG eutectic structure exhibits great resistance to high- temperature oxidation and spallation. Pt with low oxygen perme- ability[28]is the best oxygen diffusion barrier (ODB) and is highly considered as an ideal corrosion protection material because of its attractive properties such as high ductility, high melting tempera- ture, thermodynamic stability, and excellent chemical and electro- chemical inertness [29,30]. Based on the proposed sealing mechanism, the sealed Pt layers should significantly reduce the oxygen (or corrosive agents) permeation through the coating to the substrate. Meanwhile, the sealed Y2O3doped Al2O3layers can also provide preferable oxidation resistance because of their great high-temperature performance. In addition, the influence on the oxidation, spallation and hot corrosion resistance of 316L stainless
steel alloy was also investigated and mechanisms accounting for such effects are discussed in this paper.
2. Experimental procedures
2.1. Preparation of the (Al2O3–Y2O3)/Pt micro-laminated coating
The (Al2O3–Y2O3)/Pt micro-laminated composite coatings (Al2O3–Y2O3stands for Y2O3doped Al2O3, the same below) were fabricated in an opposite-targets magnetron sputtering (MS) sys- tem (Model TUS-800MP, Technol Ltd. Co. Beijing, as shown in Fig. 1.) on AISI 316L commercial stainless steel alloy substrates by layers. All the surfaces of the alloy substrates were mechanically ultra-precision polished and the average surface roughness (Ra, 0.03
l
m) was measured by using a surface roughness tester (Surf- test SJ-400, Mitutoyo, Japan). The base pressure of the chamber and the working pressure were 5.0 104Pa and 1.0 Pa. And the alloy substrates (20 10 2 cm3) were fixed on a rotating specimen holder (the rotation speed is 20 rpm) to ensure that all the surfaces can be deposited uniformly. First, the bottom (Al2O3–Y2O3) ceramic layer was deposited on the substrates at 300 °C, 50 W by Radio Fre- quency MS using a 0.5 wt.% Y2O3 dopeda
-Al2O3 ceramic target.Then, the Pt layer was deposited by Direct Current MS using a Pt target (purity >99.99%) at 300 °C, 150 W. In this work, the thick- ness of the laminated coating was determined by the sputtering time because the work pressure, sputtering power and work temperature are constant. Previous work has shown that the sput- tering rate of Pt and Y2O3doped Al2O3ceramic are approximate 20 nm/min and 30 nm/h in the above opposite-targets MS system.
In addition, a single Y2O3doped Al2O3coating (1-layer Y2O3doped Al2O3 coating, the same below) was prepared on 316L stainless steel alloy by Radio Frequency MS. The high-temperature corrosion resistance of the as-prepared coatings was investigated by cyclic oxidation test and hot corrosion test. Right before the tests, the coatings were heat pretreated at 1000 °C for 20 h with a heating rate of 10 °C/min.
2.2. High-temperature cyclic oxidation test and hot corrosion test
In this work, cyclic oxidation test was performed in a horizontal furnace at 1000 °C in static air for 200 h. After a certain oxidation period of 10 h, samples were taken out and rapidly cooled down to room temperature. Then, the mass gain (specimen + crucible)
Fig. 1. Schematic diagram of the opposite-targets magnetron sputtering system.
and spallation mass (crucible) of samples were weighed using an electronic balance with an accuracy of 105g. After the measure- ment, samples were put back to the heating zone immediately.
The test providing 20 times thermal cycles was applied to evaluate the thermal shock resistance of the prepared coatings. Otherwise, the hot corrosion test was carried out for 100 h exposure on coated samples that have been given a pre-oxidation treatment. Prior to hot corrosion test, all test samples were placed in a muffle furnace at 300 °C for 10 min and then sprayed with an aqueous solution of 75 wt.% Na2SO4+ 25 wt.% NaCl generated by an ultrasonic nebu- lizer, which resulted in about 3 mg/cm2 of solid salt deposit on the given sample. Then, the samples were taken to a horizontal furnace at 900 °C in air for 100 h. After a certain period of 10 h, samples were taken out and rapidly cooled to room temperature.
Then, the mass gains of samples were weighed using an electronic balance with an accuracy of 105g and the corrosion kinetics was recorded. After the measurement, the residual salts on samples were cleaned in the boiled water for 30 min and then the above process of salt depositing was repeated before the samples were put back to the heating zone.
2.3. Characterization
Phases of the samples were characterized by X-ray diffraction (XRD) (Rigaku-D/max-PB) on a Rigaku-DMax 2400 diffractometer equipped with a graphite monochromatized Cu K
a
radiation flux at a scanning rate of 0.02° s1 in the 2h range 10–100°. The morphology and composition of the as-prepared (Al2O3–Y2O3)/Pt micro-laminated coating before and after the cyclic oxidation and the hot corrosion tests were characterized by high-resolution field emission scanning electron microscopy (FE-SEM) with an energy- dispersive spectroscopy (EDS) system. The mechanical properties of the (Al2O3–Y2O3)/Pt micro-laminated coating were measured using Nanoindentation measurements (made with a Nanoindenter DCM, MTS Systems Corp., Eden Prairie, MN).3. Results
3.1. Morphologies of the prepared coatings
Fig. 2shows the cross-sectional and surface morphologies of the prepared (Al2O3–Y2O3)/Pt micro-laminated coating and the 1-layer Y2O3 doped Al2O3 coating obtained by means of magnetron sputtering.Fig. 2a (the backscattered cross-sectional SEM image) presents seven alternate layers of (Al2O3–Y2O3) and Pt which is consistent with deposition process and the average thickness of each layer is 150 nm, approximately. Besides, it is seen that no cracks or flaws are observed at the interfaces between layers and each layer is continuous, compact, bonding with the others tightly.
Moreover, the deposited micro-laminated coating has good adhe- sion with substrate as no micro-/nano-cracks are detected at the coating–substrate interface. Fig. 2b illustrates the secondary electron surface SEM images of the micro-laminated coating. It is seen that the coating has very smooth and compact surface with refined nanostructure. Correspondingly, cross-sectional and sur- face images of the 1-layer Y2O3 doped Al2O3 composite coating are shown inFig. 2c and d. It is viewed that the ceramic composite coating with dense and refined structure approximately has the same thickness as the 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating. The XRD results (not shown) of the deposited micro-lam- inated coating indicated the amorphous alumina phase has completely transformed to crystalline
a
-Al2O3after the heat treat- ment. In addition, the element area distributions of the (Al2O3– Y2O3)/Pt micro-laminated coating after heat treated at 1000 °C for 10 h is shown inFig. 3. It is observed that the microstructureof the micro-laminated coating is dense and integrity, without cracks or spallation. To protect the thin sputtered laminated coating from breakage and spalling under polishing, a thick Ni–P coating was deposited on the samples by electroless plating as it often provides great adhesion with various substrates[31].
3.2. High-temperature cyclic oxidation kinetics
Fig. 4shows the mass gain and spallation mass per unit area (DM/A, mg/cm2) as a function of time for the cyclic oxidation of samples in air at 1000 °C. It is shown that the 1-layer Y2O3doped Al2O3coating exhibits limited improvement for the oxidation and spallation resistance of the alloy substrate. In comparison, the oxi- dation kinetics results of samples coated with the 7-layer (Al2O3– Y2O3)/Pt micro-laminated coating revealed that the oxidation and spallation resistance of the 316L stainless steel alloy have been sig- nificantly improved, exhibiting the optimal protection effects against the oxidation and thermal cyclic spallation with less than 0.08 mg/cm2mass gain and 0.1 mg/cm2spallation mass after cyclic oxidation for 200 h. This implies that the (Al2O3–Y2O3)/Pt mciro- laminated coating can effectively improve the spallation resistance of 316L stainless steel and suppress the inward diffusion of oxygen to an extremely low level. Besides, the mass gain and spallation mass increase very slowly during the oxidation tests, suggesting that this novel coating has good long-term durability.
3.3. Surface and cross-sectional morphologies of samples after high- temperature cyclic oxidation
Surface images of samples after oxidation test are shown in Fig. 5. It is seen that the oxide scale of blank sample seriously peel off (Fig. 5a1) due to the strong interface stress caused by the mismatch of the thermal-expansion coefficient (CTE) between the oxide scale and alloy substrate. Detailed picture (Fig. 5a2) shows a very rough surface composed of loose oxide scale particles (Fe2O3, Fe3O4, MnCr2O4 and Cr2O3 particles). In comparison, the surface of the sample coated with a 1-layer (Al2O3–Y2O3) coating is much denser and smoother as shown inFig. 5b1after oxidation test.Fig. 5b2exhibits that less spallation is observed and the most region of the surface is continuous
a
-Al2O3layer instead of porous oxide scales. That is mainly because the Y2O3 can reduce the growth rate of thermal growth oxide scale and maintain finea
- Al2O3grains which can improve the coating’s mechanical property according to Hall–Petch strengthening mechanism[26,27]. How- ever, cracks and spallation ofa
-Al2O3still exist, which may lead to the further oxidation of the alloy substrate.Fig. 5c1and c2indi- cate that the (Al2O3–Y2O3)/Pt micro-laminated coating can provide superior oxidation and spallation resistance under thermal cycling because the outmosta
-Al2O3 layer (with average grain size of 200 nm) of the micro-laminated coating is compact and no obvious deep cracks, spallation or thermal growth oxides are observed on the surface.Moreover, backscattered cross-sectional SEM images of the blank sample and sample coated with the 7-layer (Al2O3–Y2O3)/
Pt micro-laminated coating after oxidation are shown inFig. 6. It is seen that the blank sample has a very thick thermal growth oxide scale (>20
l
m thick) which mainly composed of porous and discontinuous Fe2O3, Fe3O4, Cr2O3, and MnCr2O4according to the EDS results (not shown) and the XRD spectra of the samples (Fig. 7). As no continuously protective oxide scale is formed, further oxidation will occur. While, the (Al2O3–Y2O3)/Pt micro- laminated structure of the novel protective surface coating still keeps its integrity as shown inFig. 6b. The thermal growth oxide scale at the coating–substrate interface is only about 200 nm thick, which implies the oxygen diffusion rate of alloy substrate has been reduced to an extremely low level. Furthermore, it is observed thatthe generated nano-cracks in the outermost
a
-Al2O3 nano-layer are effectively shielded by the lower Pt nano-layer (enlarged view:detailed image inFig. 6b). Therefore, the channels for the oxygen inward diffusion can be decreased, resulting in low oxygen diffu- sion rate.
3.4. Hot corrosion behavior
Under certain conditions, the hot corrosion attack of salt-depos- ited materials would become serious and accelerate the degrada- tion of materials. Protective coatings which can provide good high-temperature oxidation and spallation resistance can still suffer from serious Na2SO4-induced hot corrosion attacks
[8,32,33]. In the present study, hot corrosion tests of the novel (Al2O3–Y2O3)/Pt micro-laminated coating was conducted.
Fig. 8shows the hot corrosion kinetics of (NaCl–Na2SO4)-coated samples corroded at 900 °C for 100 h. The huge weight loss of the blank sample is observed, which reveals that serious spallation and vaporization of the scales happened. In comparison, the prepared protective coatings can remarkably improve the spallation resis- tance of the alloy substrate under hot corrosion as the weight changes are above zero. In addition, the 1-layer (Al2O3–Y2O3) coating has limited improvement for the hot corrosion resistance as the sudden surge of the kinetics at 40 h reveals the accelerating corrosion of alloy and failure of the protective coating. However, the novel 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating can Fig. 3. Element area distributions of the prepared 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating.
Fig. 2. SEM images of prepared coatings: (a) cross-section of the (Al2O3–Y2O3)/Pt micro-laminated coating, (b) surface of the (Al2O3–Y2O3)/Pt micro-laminated coating, (c) cross-section of the 1-layer (Al2O3–Y2O3) coating, and (d) surface of the 1-layer (Al2O3–Y2O3) coating.
effectively avoid the further hot corrosion attack of the alloy sub- strate as the weight gain increases very slowly and reaches only 0.82 mg/cm2after 100 h hot corrosion test.
Surface and cross-sectional images of the blank sample and sample coated with (Al2O3–Y2O3)/Pt micro-laminated coating after hot corrosion test are shown inFig. 9a and b, respectively. Sever spallation and deep holes in oxide scale are observed inFig. 9a1 and a2, which is in good agreement with the XRD result of showing the strong substrate peaks (Fig. 10). Besides, the alloy sulfides
mainly distributed along the alloy grain boundaries were found in the deep zone of the 316L stainless steel alloy substrate (about 30
l
m below the surface). Besides, the surface morphology (Fig. 9b1) shows a very smooth and dense surface, without obvious defects. The cross-sectional backscattered SEM image (Fig. 9b2) shows the most distinguished feature is that the (Al2O3–Y2O3)/Pt micro-laminated structure of the protective surface coating still keeps its integrity and density after a long term hot corrosion test.It is seen that no alloy sulfides were found. Moreover, a compact Fig. 4. Oxidation kinetic curves of different samples: (a) mass gain per unit area versus time and (b) spallation mass per unit area versus time (A: Al2O3; Y: Y2O3, the same below).
Fig. 5. Surface morphologies of samples after cyclic oxidation test: (a1and a2) blank, (b1and b2) coated with the 1-layer (Al2O3–Y2O3) coating, and (c1and c2) coated with the (Al2O3–Y2O3)/Pt micro-laminated coating.
and continuous protective Cr2O3thermal growth scale has formed.
The main reason responsible for the formation of Cr2O3layer is that the deposited novel protective coating could promote the selective oxidation of Cr in alloy substrate when the content of Cr in alloy is too low to form a continuous protective scale on the surface during the oxidation as shown inFig. 9a.
4. Discussion
4.1. Detailed mechanisms for the enhanced high-temperature mechanical properties
With practical high-temperature protective coatings, the corro- sion resistance of alloy substrates mainly depends on the integral- ity of protective coatings. In this study, the processes of protective coating cracking and spallation are the key factor to influence the lifetime of 316L stainless steel alloy and the prepared protective coating. In other words, the high-temperature oxidation resistance and hot corrosion resistance of the protective coatings are mainly determined by their mechanical properties including spallation and crack resistances. Therefore, it is vital to understand the mech- anism for the special mechanical properties of the (Al2O3–Y2O3)/Pt micro-laminated coating.
Cracking and spalling of oxide scales and coatings are often caused by thermal stresses, which arise from differential thermal expansion between the alloy substrate and surface coating. Timo- shenko [34], derived the thermal stress,
r
, in oxide and coating during cooling as:r
¼Ecða
ca
mÞDT1
m
ð1Þwhere subscript c is coating (the same below), Ec is the elastic modulus of surface coating,
a
canda
mare the linear thermal-expan- sion coefficients (CTEs) for the surface coating and metal substrate, andDT is the change in temperature. As analyzed by Evans[14], spallation of a coating will occur when the elastic strain energy stored in the coating exceeds the fracture resistance, G, of the inter- face. The criterion for failure is given as the following equation:ð1
m
Þr
2ch Ec>G ð2Þ
where
m
is the Poisson’s ratio of the coating, h is the scale thickness, andr
cis the equal biaxial residual stress in the coating.According to Eq.(2), there are two ways to avoid coating spall- ation: one is decreasing the stress,
r
c, by decreasing the (a
ca
m) and the Ec. The other one is increasing the fracture resistance by the toughening effects of composite structures.In a laminated composite with 3D axes, we define the longitu- dinal CTE (al) as the CTE in the x or y direction and the transverse CTE (at) as the CTE in the z (thickness direction). The Schapery equations[35]modified for laminated composites give the CTE in the longitudinal and transverse directions as:
a
x¼a
y¼a
1¼a
1E1V1þa
2E2V2E1V1þ E2V2
a
z¼a
t ffi ð1 þm
1Þa
1V1þ ð1 þm
2Þa
2V2a
1m
ð3:1Þ
where
m
¼m
1V1þm
2V2 ð3:2Þand E is Young’s modulus, V is the volume fraction. The subscripts 1 and 2 indicate the components of the composite, and the subscripts Fig. 6. FE-SEM cross-sectional micrographs of samples after cyclic oxidation at 1000 °C for 200 h: (a) blank and (b) coated with the 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating.
Fig. 7. XRD spectra of samples after the cyclic oxidation test: (a) blank and (b) coated with the 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating.
Fig. 8. Mass change per unit area versus time for different samples, which were coated with a 3.0 ± 0.1 mg cm2NaCl–Na2SO4salt film, after hot corrosion at 900 °C for 100 h.
l and t indicate the longitudinal and transverse directions, respectively.
The longitudinal CTE (al) and the transverse CTE (at) of the 7- layer (Al2O3–Y2O3)/Pt micro-laminated coating at various temper- atures are shown in Fig. 11 [36]. Research [37]has shown that the elastic modulus of thin film Pt is similar to the bulk Pt. So, the elastic modulus of Pt nano-layers is considered as the elastic modulus of bulk Pt in this work.Fig. 11shows that the CTE of 7- layer (Al2O3–Y2O3)/Pt laminated coating increases with the increasing of temperature. Moreover, compared with the CTE of the 1-layer Al2O3 coating, the CTE of the (Al2O3–Y2O3)/Pt lami- nated coating is much closer to that of the 316L stainless steel al- loy (the average CTE is 19.9 106K, 273–1200 K). Consequently, the thermal stress at the coating–substrate interface will decrease because of the decreasing of (
a
ca
m). Besides, previous research [22]has shown that the single Pt coating has been destroyed and a Pt–based intermetallic layer will form at the coating-substrate after the oxidation because of the interdiffusion between Pt and alloy substrate at high temperatures. Consequently, the CTE ofthe single Pt coating is no need to calculate and compare with the prepared laminated coating due to the vanishment of the sin- gle Pt coating after the cyclic oxidation and hot corrosion tests.
And the CTE of the single Pt coating inFig. 11is used to calculate the CTE of the 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating by Eq. (3). Moreover,Fig. 12a shows the elastic modulus of the 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating versus displace- ment into surface before high-temperature corrosion tests and the mean elastic modulus is listed inTable 1. Considering the to- tal thickness of the 7-layer (Al2O3–Y2O3)/Pt micro-laminated coat- ing (1020 nm) and the influence on the modulus by the surface of the coating, the measured elastic modulus from 100 nm to 1000 nm of displacement into surface were chosen to calculate the mean elastic modulus of coatings. In Table 1, it is indicated that the elastic modulus of 1-layer (Al2O3–Y2O3) coating is about 231.0 GPa, which is bigger than that of the 7-layer (Al2O3–Y2O3)/
Pt micro-laminated coating (160.0 GPa). Thus, according to Eq.
(1), less thermal stress (
r
) is generated in the novel 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating because of the smaller (a
ca
m) and Ec.Fig. 9. SEM images of samples after hot corrosion test at 900 °C for 100 h: (a1) surface of the blank sample, (a2) cross-sectional image of the blank sample, (b1) surface of the sample coated with the 1-layer (Al2O3–Y2O3) coating, and (b2) cross-section image of sample coated with the 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating.
Fig. 10. XRD spectra of samples after the hot corrosion test: (a) blank and (b) coated with the 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating.
Fig. 11. Thermal expansion coefficient of Pt, Al2O3and the 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating as a function of temperature (al: the longitudinal CTE; at: the transverse CTE).
A brief analysis on the increasing fracture resistance by the mi- cro-laminated composite structures is given as follow. Firstly, the brittle Y2O3doped Al2O3composite nano-layers can provide suffi- cient strength for the coating. The plastic deformation of ductile metal ligaments (Pt nano-layers) and the Y2O3particles distributed along
a
-Al2O3grain boundaries can absorb the impact stress caused by the unavoidable mismatch in CTE and the growth stress during cyclic oxidation, exhibits effective toughness. Secondly, Ho and Suo[38]found that decreasing the thickness of layers is very help- ful to suppress crack extension in the laminated coating because below the critical thickness for the constrained brittle layer bonded between tougher substrates under residual and applied stresses, no tunneling cracks occurred regardless of the size of original cracks.Thus, the (Al2O3–Y2O3)/Pt micro-laminated coating with nano-lay- ers would provide superior crack resistance and it would be reason- able to expect excellent mechanical properties of this novel composite coating. Thirdly, as the thickness of each layer reaches nanoscale with refined structure, it may give rise to the superplas- ticity of the laminated coating, which can also relax the thermal stress and improve the spallation resistance[38]. Therefore, the multilayer toughening of the brittle/ductile laminated structure en- sures the novel coating with higher fracture resistance (G), great flaw tolerance and crack resistance, which make it be able to with- stand bigger stresses. In conclusion, the (Al2O3–Y2O3)/Pt micro- laminated coating can withstand bigger external stress and effec- tively protect the 316L stainless steel alloy from spallation under thermal cycling by its special mechanical properties.
Furthermore, scratching test of samples after the oxidation has been adopted. The adhesions of the formed oxide scales or the protective coating were analyzed by a scratching adhesion tester system at room temperature. The frictional force (Ff) and acoustic emission (AE) signals generated during the tests are recorded.
Fig. 12b shows the test result of the 7-layer (Al2O3–Y2O3)/Pt mi- cro-laminated coating. It is seen that the first huge flex points (a–b) of the AE signal curve and the inflection point (P) of the slope of the Ff curves approximately coincide at the load of 30 N. That
is because the sound generated by the remarkable breakage of the surface protective coating leads to the abrupt change of the received AE signal. Also, the spallation of the (Al2O3–Y2O3)/Pt mi- cro-laminated coating will change the friction force of the sample surface because of the different friction ratios between the surface coating and the 316L stainless steel alloy substrate. So, the critical load of the oxide scale could be determined as 30.0 N. The results revealed that the 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating can withstand the biggest critical load (30.0 N) without spallation (Fig. 12b andTable 1), which directly demonstrates the proposed mechanisms for the improved high-temperature mechanical properties.
4.2. Detailed mechanisms for the great high-temperature corrosion resistance
Detailed mechanisms for the great hot corrosion resistance and high-temperature oxidation resistance of the new developed (Al2O3–Y2O3)/Pt micro-laminated coating are discussed.
It has been well-known that the solubility of protective coatings or oxide scales in fused sodium sulfate and the ‘‘negative solubility gradient’’ will cause continuing hot corrosion of the protective coatings [39]. However, the proposed mechanism as shown in Fig. 13a exhibits a ‘‘no solubility gradient’’ is found in the novel laminated coating. This can be explained in following three as- pects. Firstly, it has been demonstrated that the Pt with sealed structure can effectively suppress the diffusion of O2, S, Cl, Na+ and even Fowing to its excellent inertness in fused salts Fig. 12. (a) Elastic modulus of the 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating versus displacement into surface before high-temperature corrosion tests measured by Nano Indenter DCM and (b) The friction force curve (Ff) and the acoustic emission signal curve (AE) of the (Al2O3–Y2O3)/Pt micro-laminated coating after oxidation test.
Table 1
Mean modulus and critical load of the test coatings.
Test coatings Mean modulus (GPa) Critical load (N)
Blank – Barely none
1-layer(A + Y) 230 2.2
7-layer(A + Y)/Pt 160 30
Note: Mean modulus was measured before oxidation and hot corrosion test while the critical loads were measured after oxidation test; ‘‘–’’ means no need to measure the critical load.
Fig. 13. Schematic representation of the mechanisms for the: (a) hot corrosion resistance and (b) high-temperature oxidation resistance of the 7-layer (Al2O3– Y2O3)/Pt micro-laminated coating.
[22]. Therefore, the ‘‘basic and acidic fluxings’’ in fused salt can be stopped as the (Al2O3–Y2O3)/Pt micro-laminated coating has extre- mely low solubility in fused sodium sulfate because of the sealed Pt nano-layers in the laminated coating. Once the ‘‘negative solubility gradient’’ no longer exists, the salt film should just saturate to the solute concentration established at the coating–salt interface and the reaction should stop; no continuing hot corrosion[39]and sub- strate sulfation should happen[40]. Secondly, as the (Al2O3–Y2O3)/
Pt micro-laminated coating has great crack and spallation resis- tance, the nano-cracks tip can be shield and the crack propagation can be effectively suppressed. Without the diffusion tunnel, the electrochemical reactions on hot corrosion will not happen[41].
Thirdly, it is known that the chlorides such as NaCl not only desta- bilize Al2O3and Cr2O3protective oxides into less protective spinels, but also release chlorine, which is detrimental to the stability of the traditional protective coatings. As opposed to Al2O3 and Cr2O3 phases, noble metals have been found to be stable in the sulfate–chloride melt. Therefore, the (Al2O3–Y2O3)/Pt laminated coatings can provide super hot corrosion resistance in the sul- fate–chloride melt owing to the Pt layers. In all, great hot corrosion resistance can be acquired by the new coating system.
Fig. 13b shows the model of oxygen diffusion in the 7-layer (Al2O3–Y2O3)/Pt protective coating. Generally, the thermodynami- cally stable structure at high temperatures is the fundamental aspect on the great oxidation resistance of the laminated coating.
More importantly, Based on the proposed sealing mechanism [18]and the discussions about the great oxidation resistance of Al2O3–Y2O3 coating system and Pt, oxygen partial pressure (POxygen) can be suppressed in stages by the multi-sealed alternat- ing Al2O3–Y2O3and Pt nano-layers from the coating–hot air inter- face to the coating–substrate interface. And it is observed that in Fig. 6b, a very thin (only about 200 nm thick) and adhesive thermal growth oxide scale formed on the alloy after 200 h oxidation under thermal cycling, suggesting that the oxygen partial pressure (P0) has decreased to an extremely low level at the coating–substrate interface. Consequently, it is demonstrated that the proposed 7-layer (Al2O3–Y2O3)/Pt micro-laminated coating can effectively suppress the inward diffusion of oxygen and can reduce further oxidation of alloy substrate in a long term service.
5. Conclusions
In conclusion, the novel 7-layer (Al2O3–Y2O3)/Pt micro-lami- nated coating prepared by magnetron sputtering significantly im- proved the high-temperature corrosion resistance of 316L stainless steel alloy due to its special mechanical properties under thermal cy- cling and its extremely low corrosion rate at high temperatures. It has been demonstrated that the brittle/ductile micro-laminated structure which provides good coating–substrate compatibility can effectively enhance the strength and toughness in combination with improved damage resistance of the composite coatings. It is also concluded that the sealed structure (sealed multi-layers) of the coating is the key factor that can effectively suppress the diffu- sion of oxygen or other corrosive agents to the substrate alloys, which results in the significant improvement of the high-tempera- ture corrosion resistance. Therefore, the easily-made and corro- sion–protection coating with (Al2O3–Y2O3)/Pt laminated composite structure prepared in this work opens a brand new oppor- tunity to solve the fundamental issues of the hot corrosion and high- temperature oxidation of the 316L stainless steel alloy.
Acknowledgments
The authors thank the financial support from the Chinese National Nature Science Foundation (Grant 51071030).
References
[1] D.J. De Renzo, I. Mellan, Corrosion Resistant Materials Handbook, fourth ed., 1985.
[2] T. Michler, Y. Lee, R.P. Gangloff, J. Naumann, Influence of macrosegregation on hydrogen environment embrittlement of SUS 316L stainless steel, Int. J.
Hydrogen Energy 34 (2009) 3201.
[3] H. Yang, K. Yang, B. Zhang, Pitting corrosion resistance of La added 316L stainless steel in simulated body fluids, Mater. Lett. 61 (2007) 1154–
1157.
[4] H.B. Qi, D.G. Lees, Y.D. He, Effect of surface-applied rare earth containing thin oxide films on high-temperature oxidation of Fe25Cr, Corros. Sci. Protect.
Technol. 11 (1999) 193–201 (in Chinese).
[5] J.G. Gao, Y.D. He, D.R. Wang, Preparation of YSZ/Al2 O3 micro-laminated coatings and their influence on the oxidation and spallation resistance of MCrAlY alloys, J. Eur. Ceram. Soc. 31 (2011) 79–84.
[6] H. Hosono, Y. Mishima, H. Takezoe, K.J.D. Mackenzie, Nanomaterials—from Research to Applications, first ed., 2006.
[7] B.D. Prasad, S.N. Sankaran, K.E. Wiedemann, D.E. Glass, Platinum substitutes and two-phase-glass overlays as a low cost alternatives to platinum aluminide coatings, Thin Solid Films 345 (1999) 255–262.
[8] V. Deodeshmukh, N. Mu, B. Li, B. Gleeson, Hot corrosion and oxidation behavior of a novel Pt + Hf-modifiedc0-Ni3Al +c-Ni-based coating, Surf. Coat. Technol.
201 (2006) 3836–3840.
[9] X. Yang, X. Peng, F. Wang, Hot corrosion of a novel electrodeposited Ni–6Cr–
7Al nanocomposite under molten (0.9Na, 0.1K)2SO4at 900 °C, Scripta Mater. 56 (2007) 891–894.
[10] M. Medraj, R. Hammond, M.A. Parvez, R.A.L. Drew, W.T. Thompsond, High temperature neutron diffraction study of the Al2O3–Y2O3system, J. Eur.
Ceram. Soc. 26 (2006) 3515–3524.
[11] S.M. Lakiza, L.M. Lopato, Stable and metastable phase relations in the system Alumina–Zirconia–Yttria, J. Am. Ceram. Soc. 80 (1997) 893–902.
[12] C. Wagner, Theoretical analysis of the diffusion processed determining the oxidation rate of alloys, J. Electrochem. Soc. 99 (1952) 369–380.
[13] P.Y. Hou, J. Stringer, The effect of reactive element additions on the selective oxidation, growth and adhesion of chromia scales, Mater. Sci. Eng. A 202 (1995) 1–10.
[14] H.E. Evans, Cracking and spalling of protective oxide layers, Mater. Sci. Eng. A 120 (1989) 139–146.
[15] A.R. Boccaccini, V. Winkler, Fracture surface roughness and toughness of Al2
O3-platelet reinforced glass matrix composites, Composites: Part A 33 (2002) 125–131.
[16] Y. He, F.H. Stott, The selective oxidation of Ni–10%Cr alloys promoted by surface-applied thin oxide films, Corros. Sci. 36 (1994) 1869–1884.
[17] Z. Li, W. Gao, M. Yoshihara, Y. He, Improving oxidation resistance of Ti3Al intermetallic compounds with electro-spark deposit coatings, Mater. Sci. Eng.
A347 (2003) 243–252.
[18] Theoretical consideration on oxidation behavior of composite oxide scales and coatings, High Temp. Mater. Process, in press.
[19] M.P. Bacos, M. Thomas, J.-L. Raviart, A. Morel, S. Mercier, P. Josso, Influence of an oxidation protective coating upon hot corrosion and mechanical behaviour of Ti–48Al–2Cr–2Nb alloy, Intermetallics 19 (2011) 1120–1129.
[20] E. S. Scherba, US Patent No. 6551476, Noble-Metal Coated Inert Anode for Aluminum Production, April 22, 2003.
[21] X. Ma, Y. He, D. Wang, Preparation and high-temperature properties of Au nano-particles dopeda-Al2O3composite coating on TiAl-based alloy, Appl.
Surf. Sci. 257 (2011) 10273–10281.
[22] X.X. Ma, Y.D. He, D.R. Wang, Inert anode composed of Ni–Cr alloy substrate, intermediate oxide film and a-Al2O3/Au(Au–Pt, Au–Pd, Au–Rh) surface composite coating for aluminium electrolysis, Corros. Sci. 53 (2011) 1009–1017.
[23] J. Musil, P. Baroch, J. Vlcek, K.H. Nam, J.G. Han, Reactive magnetron sputtering of thin films: present status and trends, Thin Solid Films 475 (2005) 208–218.
[24] P.J. Kelly, R.D. Arnell, Magnetron sputtering: a review of recent developments and applications, Vacuum 56 (2000) 159–172.
[25] C.T. Yang, C.H. Koo, Improving microstructure and high temperature properties of the Ti–40Al–16Nb alloy by the addition of a minor Sc or La–
rich Misch metal, Intermetallics 12 (2004) 235–251.
[26] E.O. Hall, The deformation and ageing of mild steel: III discussion of results, Proc. Phys. Soc. B 64 (1952) 747–753.
[27] N.J. Petch, The cleavage strength of polycrystals, J. Iron Steel Inst. 174 (1953) 25–28.
[28] M.E. Eberhart, M.M. Donovan, R.A. Outlaw, Ab initio calculations of oxygen diffusivity in group-IB transition metals, Phys. Rev. B 46 (1992) 12744–
12747.
[29] J. Thonstad, Anode overvoltage on platinum in cryolite–alumina melts, Electrochim. Acta 13 (1968) 449–456.
[30] J. Merker, D. Lupton, M. Töpfer, H. Knake, High temperature mechanical properties of the platinum group metals, Platinum Met. Rev. 45 (9) (2001) 74–82.
[31] Jiangqiang Kang, Yifu Yang, Huixia Shao, Comparing the anodic reactions of Ni and Ni–P amorphous alloy in alkaline solution, Corros. Sci. 51 (2009) 1907–1913.
[32] T.S. Sidhu, S. Prakash, R.D. Agrawal, Hot corrosion performance of a NiCr coated Ni-based alloy, Scripta Mater. 55 (2006) 179–182.
[33] Takehiko Hirata, Satoshi Ota, Tatsuo Morimoto, Influence of impurities in Al2
O3ceramics on hot corrosion resistance against molten salt, J. Eur. Ceram. Soc.
23 (2003) 91–97.
[34] S.P. Timoshenko, Analysis of bi-metal thermostats, J. Opt. Soc. Amer. 11 (1925) 233–255.
[35] R.A. Schapery, Thermal expansion coefficient of composite materials based on energy principles, J. Comp. Mater. 2 (3) (1968) 380–404.
[36] B. Tryon, T.M. Pollock, M.F.X. Gigliotti, K. Hemker, Thermal expansion behavior of ruthenium aluminides, Scripta Mater. 50 (2004) 845–848.
[37] H.-J. Choi, J.-H. Kim, H.-J. Lee, S.-A. Song, H.-J. Lee, J.-H. Han, M.-W. Moon, Wrinkle-based measurement of elastic modulus of nano-scale thin Pt film
deposited on polymeric substrate: verification and uncertainly analysis, Exp.
Mech. 50 (2010) 635–641.
[38] S. Ho, Z. Suo, Tunneling cracks in constrained layers, J. Appl. Mech. Trans. ASME 60 (1993) 890–894.
[39] R.A. Rapp, K.S. Goto, in: J. Braunstein, J.R. Selman (Eds.), Hot corrosion of metals by molten salts, Molten Salts 1, Electrochem. Soc. Pennington, NJ, 1981, p. 159.
[40] J.A. Goebel, F.S. Pettit, Na2SO4-induced accelerated oxidation (hot corrosion) of nickel, Metall. Trans. 4 (1970) 1943–1954.
[41] X. Zheng, R.A. Rapp, Electrochemical impedance study of Platinum electrode in fused Na2SO4melt in SO2–O2environments, J. Electrochem. Soc. 140 (1993) 2857–2862.
