1.Introduction

With the increase of energy demand,it becomes more important to improve the thermal efficiency and economics of power plants.Compared to conventional liquid fluids,power cycle with supercritical fluids can realize higher thermal efficiency up to ～50%[1,2].Among several interesting supercritical fluids,supercritical carbon dioxide(SCO2)has a number of advantages,such as high density,good stability,low viscosity,high specific heat and low critical temperature(31°C,which allows the S-CO2 power cycles operating in a wide temperature range)[3–5].The S-CO2 Brayton cycle is considered to have a wide range of applications in nuclear and fossil fuel plants,waste heat recovery,concentrated solar power and so on[1,4–8].Therefore,investigating the corrosion behaviors of candidate materials and selecting suitable materials for the high temperature S-CO2 environment are important issues for design and construction of the power conversion system with S-CO2 Brayton cycle.

The corrosion behaviors of many steels and alloys in the high temperature S-CO2 environment have been evaluated in previous studies[9–20].In general,the traditional ferritic/martensitic steels usually experience larger weight gain and exhibited poor oxidation resistance,while austenitic stainless steels are moderate and Ni-base alloys possess superior resistance to oxidation.Microstructure observation reveals that a thick and porous oxide layer with outer Fe3O4 and inner spinel oxide forms on the surface of traditional ferritic/martensitic steels,and this oxide structure leads to poor protection[11–13].For the Ni-base alloys,a continuous chromia(Cr2O3)layer can be formed on the exposed surface,which effectively prevented further oxidation of alloy substrate[17–19].Based on the existing research results,Ni-base alloys seem to be the best choice for the key components from the point of view of oxidation resistance.However,Ni-base alloys are more expensive compared to stainless steels.Moreover,for the nuclear cladding materials,neutron economy is also an important factor and Ni-base alloys have poor neutron economy due to the high Ni content[21].As a consequence,Ni-base alloys should be excluded from the cladding materials.On the other hand,austenitic stainless steels have inferior irradiation resistance to swelling compared to the ferritic/martensitic stainless steels[22].Therefore,ferritic/martensitic stainless steels may be a better choice for the nuclear cladding materials providing that the oxidation resistance of this steel can be significantly improved.

针对InSAR干涉图局部噪声斑块情况，对噪声集中的区域先进行掩膜滤波处理，最大程度保留噪声集中区域的真实相位，然后利用多次迭代的解缠算法对掩膜滤波区域进行填充，保证较大噪声处的误差在解缠时不被传递，保持与原相位拟合.

FeCrAl alloys are one of the most promising candidates for the cladding materials because of their reasonable formability,moderate strength and excellent oxidation resistance at high temperatures[23,24].The excellent oxidation resistance is mainly attributed to the alumina(Al2O3)scale,which not only protects the alloy substrate from further oxidizing,but also has superior resistance to carburization compared to chromia layer[25].Previous study showed that the carburized depth of Kanthal APM(one of FeCrAl alloys)was significantly lower than those of Fe-35Ni-25Cr and Fe-25Ni-20Cr alloys after these alloys were exposed to high temperature and high carbon-potential atmospheres[26].Considering the fact that the corrosion of structural materials in the high temperature S-CO2 environment often involves both oxidation and carburization[27,28],FeCrAl alloys are expected to possess better performance in the high temperature S-CO2 environment compared to conventional stainless steels.

Recently,McCarroll et al.investigated the isothermal and cyclic performance of Kanthal APM in CO2 environment at 1100°C,and they found that this alloy performed well and a thin alumina scale was developed on the alloy surface[29].However,the oxidation conditions in this study are quite different from the design parameters of S-CO2 Brayton cycle.Gibbs conducted a systematic design for the S-CO2cooled nuclear reactors,and obtained the relatively good parameters of 650°C inlet temperature and 20 MPa cycle pressure by weighing the thermal efficiency and materials limitations[3].This design temperature is far below 1100°C.Moreover,the CO2 pressure in that study seems to be atmospheric pressure and CO2 state is not supercritical.Thus,there are few reports about the oxidation behaviors of FeCrAl alloys in the S-CO2 environment at temperature of interest(500–750°C)for nuclear reactors.

Under the oxide scales of the high-strength FeCrAl alloys,many white particles precipitate in the matrices and the average size of these white phases is significantly larger than that of precipitates in the central regions of matrices,as shown in the microstructure images of Figs.5–7.The corresponding EDS line scanning results show these white phases are rich in Nb and Mo.Table 3 shows the EDS point results of these white phases and the surrounding matrices.Although the carbon intensity was similar between the white phase and matrix from the EDS line scanning results,the EDS point results indicate that the C content of white phases is higher than the surrounding matrices,and is quite similar to the intergranular precipitates of the un-oxidized alloy(shown in Fig.1).Therefore,it can be inferred that these white phases are(Nb,Mo)C carbides.In the isothermal oxidation process,the free carbon generated by the above reactions can penetrate into the matrix and reacts with the alloying elements(Nb and Mo)and/or the intragranular Fe2(Nb,Mo)Laves phases to form the MC-type carbides because Nb and Mo are the strong carbide-forming elements[50].Hence many(Nb,Mo)C carbides are present in the sub-scale matrices after oxidation in the high-temperature S-CO2 environment.The enrichment of Nb/Mo-containing phases has also been observed under the oxide scale during the high temperature oxidation process in other studies[51–53].Thermodynamic and kinetic calculations suggest that for the chromia-forming alloys,the enrichment of Nb-containing phase results from the Cr depletion in the sub-scale matrix,which causes the activity gradient of Nb and induces its outward diffusion from the inner matrix[53].

On the other hand,high-temperature mechanical properties are an important factor to determine the application of FeCrAl alloys to the SCO2 Brayton cycle.But FeCrAl alloys usually exhibit lower strengths than the austenitic stainless steels and Ni-base alloys due to the ferritic matrix.Recent studies show that the addition of Mo and Nb could improve the strengths of FeCrAl alloys by means of solid-solution strengthening and/or second-phase precipitate strengthening[30,31].These high-strength FeCrAl alloys are more suitable for the nuclear cladding.Therefore,the present paper aims to investigate the oxidation behavior of the high-strength FeCrAl alloys(Mo,Nb-containing)in the S-CO2 environment at 650°C and 20 MPa.The oxidation resistance is evaluated by weight gains and microstructural characterization.Considering that the Al content has a great influence on the properties of alloys,the effect of Al content on the oxidation behavior of the highstrength FeCrAl alloys is also discussed.

2.Experimental procedures

Three high-strength FeCrAl alloys with different Al content was investigated in this study.The chemical compositions of these FeCrAl alloys were detected by the inductively coupled plasma mass spectrometry,and are given in Table 1.The alloys were prepared by vacuum induction melting and homogenized at 1100°C.Then these alloys were hot-rolled at 800°C with an Engineering strain of 70%and subsequently annealed at 800°C for 30 min.For the purpose of oxidation testing,coupons with a dimension of 10×10×1 mm were cut from the rolled plates by spark machining and a small hole with a diameter of 2 mm was drilled at the corner of the coupons.Before the oxidation testing,the coupons were mechanically ground with 1200 grit SiC papers.The polished coupons were then degreased and cleaned ultrasonically in ethanol.

The coupons of three FeCrAl alloys were subjected to the isothermal oxidation testing in an S-CO2 corrosion testing facility.The detailed description and schematic of the S-CO2 corrosion testing facility could be found in our previous work[18].The oxidation tests were conducted in a high temperature S-CO2 environment at 650°C and 20 MPa for 500 h.Prior to heating up the furnace,research-grade CO2(99.999%)was fed into the testing system to remove the residual air for 5 min.Then the liquid CO2 was pressurized into the testing system with a5 ml/min by a high pressure supercritical pump supplied by Scientific Systems Incorporation.When the liquid CO2 passed through the preheating furnace and arrived at autoclave,it was vaporized and became a supercritical fluid.In the meantime,the autoclave system was slowly heated up to the target temperatures of 650°C.The pressure of S-CO2 was regulated by using a back pressure regulator located at the end of the loop.During the whole isothermal oxidation testing,the temperature and pressure of S-CO2 were maintained at 650±2°C and 20±1 MPa,respectively.

Table 1 Chemical compositions(wt%)of three high-strength FeCrAl alloys.The x in FeCrxAl represents x wt%Al in these FeCrAl alloys.

Specimens Fe Cr Al Mo Nb C N FeCr3Al Bal. 12.82 3.10 1.96 1.03 <0.01 <0.005 FeCr4Al Bal. 12.46 4.03 2.04 0.98 <0.01 <0.005 FeCr5Al Bal. 12.59 4.96 1.94 0.99 <0.01 <0.005

After the oxidation tests,the weight gains were measured using an electronic microbalance(Mettler Toledo AT21 Comparator)with a resolution of 0.001 mg.Three coupons for each alloy were weighed to acquire accurate value of weight gains.The microstructures and chemical compositions of the high-strength FeCrAl alloys before and after oxidation test were examined by a scanning electronic microscopy(SEM,FEI Magellan400)equipped with energy dispersive X-ray spectroscopy(EDS,Oxford Instruments).In order to better electrical conductivity,the specimens were sprayed Pt at 5 V for 60 s before SEM observation. The surface oxides were analyzed by an X-ray diffractometer(XRD,Rigaku D/Max-2500)using Cu K α radiation,which was performed in the 2θ range of 20–80°with a diffraction rate of 4°/min.

3.Results and discussion

3.1.Microstructures of the high-strength FeCrAl alloys

Although the carbon contents obtained by EDS are not very reliable because of carbon contamination in the vacuum chamber and the obtained values will be slightly higher than the actual contents,the EDS carbon data are still meaningful by comparing them in the same condition.The C contents of intragranular precipitates(micro-regions 3,6 and 9)are comparable to that of the matrices,while these values of intergranular precipitates(micro-regions 2,5 and 8)are obviously higher than that of the matrices.It should be noted that,as the white precipitates are too small,the measured EDS values of precipitates are contaminated with those underlying matrices,thus a large amount of Fe,Cr and Al are also detected.The EDS results suggest that the intragranular precipitates are likely to be Fe2(Nb,Mo)Laves phases,while the intergranular precipitates may consist mainly of the(Nb,Mo)-rich MC-type precipitates,or primary carbides(Nb,Mo)C.Earlier studies indicated that the Laves phases were usually precipitated in the matrix and at grain boundaries in the Nb-containing steels[31–33].Therefore, except the (Nb,Mo)C carbides, some Fe2(Nb,Mo) Laves phases may be also distributed at the grain boundaries.The formation of the intergranular(Nb,Mo)C carbide is probably due to a higher C content at grain boundaries and the continuous supply of C atoms through the grain boundaries which serve as a fast diffusion channel[34].

Fig.1 shows the typical microstructures of the un-oxidized highstrength FeCrAl alloys with different Al content.Although the Al content increases,their microstructures have no obvious difference.From the high-magnification micrographs,it can be seen that many white precipitates are randomly distributed in grains while some are arranged along grain boundaries.The precipitates located at grain boundaries are distinctly larger than those within the grains.These precipitates are further analyzed by EDS and the results of chemical compositions of micro-regions marked by numbers 1–9 are presented in Table 2.The EDS results of the matrices(micro-regions 1,4 and 7)are generally consistent with the bulk chemical compositions of high-strength FeCrAl alloys with higher C values and lower Nb contents.Meanwhile,precipitates(micro-regions 2,3,5,6,8 and 9)are rich in Nb and Mo.Therefore,lower Nb contents in the matrices could be attributed to the formation of Nb-rich precipitates.

Fig.1.Back-scattered electron SEM morphologies of the un-oxidized highstrength FeCrAl alloys:(a)FeCr3Al alloy;(b)FeCr4Al alloy and(c)FeCr5Al alloy.Inset images are the high-magnification micrographs of the regions marked by white rectangles,the micro-regions marked by numbers are analyzed by EDS and the results are listed in Table 2.

3.2.Weight gains of the high-strength FeCrAl alloys after S-CO2 oxidation

The effect of Al content on the weight gains of the high-strength FeCrAl alloys is shown in Fig.2.The largest weight gain is approximately 0.15 mg/cm2 for the FeCr3Al alloy.With the increase of Al content,the oxidation resistance of the high-strength FeCrAl alloys is gradually enhanced.The FeCr5Al alloy exhibits the best oxidation performance in the high temperature S-CO2 environment with weight gain of only about 0.05 mg/cm2.When compared with literature data[15],the weight gain of FeCr5Al alloy is at least two orders of magnitude lower than those of typical ferritic/martensitic steels(8.18 mg/cm2 for F91 steel,13.80 mg/cm2 for HCM12A steel)and comparable to that of 800 H austenitic steel(～0.06 mg/cm2)in the same oxidation condition (at 650°C and 20 MPa for 500 h).Also,compared withvarious alloys in the same oxidation condition[35],measured weight gain of FeCr5Al alloy is also lower than that of 310SS,Alloy 800 H,Haynes 230 and Alloy 625,and is comparable to that of PM2000 ODS steel which is also an alumina-forming steel with 5.5 wt%Al content.These comparisons indicate that the alumina-forming alloys have better oxidation resistance than the chromia-forming alloys in the high temperature S-CO2 environment.

Table 2 EDS results of chemical compositions (wt%) of micro-regions marked by numbers 1–9 in Fig.1.

a Note:the EDS values are the mixture of white precipitates and underlying matrix because of the small precipitate size.

Regions Fe Cr Al Mo Nb C Possible phases 1 80.42 12.19 3.20 1.41 0.26 2.52 Matrix 2a 59.85 7.30 2.28 6.30 16.77 7.50 (Nb,Mo)C 3a 71.63 12.24 3.27 2.84 7.66 2.36 Fe2(Nb,Mo)4 79.18 12.57 4.22 1.63 0.21 2.19 Matrix 5a 57.35 9.47 3.23 4.58 17.05 8.32 (Nb,Mo)C 6a 69.32 12.52 3.96 2.95 8.76 2.49 Fe2(Nb,Mo)7 77.90 12.46 5.47 1.75 0.13 2.29 Matrix 8a 54.46 7.17 2.92 6.16 19.82 7.47 (Nb,Mo)C 9a 71.66 10.49 4.99 2.54 7.87 2.45 Fe2(Nb,Mo)

Fig.2.Weight gains of the high-strength FeCrAl alloys as a function of Al content after oxidization in the high-temperature S-CO2 environment for 500 h.

In general,it has been known that alumina developed at the temperatures above 1000°C are stable and protective α-Al2O3,whereas they are thermodynamically metastable“transient alumina”(γ-Al2O3,δ-Al2O3 and θ-Al2O3)at lower temperatures[36–38].Therefore,the alumina formed on the high-strength FeCrAl alloys at 650°C,in the present study,is considered to be the transient alumina.Also,it is well known that the transient alumina has high epitaxial growth rates and possess fairly poor protective properties due to the high defect concentrations in these oxides[36,39].This then leads to a question why the high-strength FeCrAl alloys show good oxidation resistance at the intermediate temperature of 650°C where transient alumina is known to be unstable.To answer this question,the surface oxides formed on the high-strength FeCrAl alloys was investigated in details.

3.3.Characterization of the surface oxides

Fig.3.Secondary electron SEM surface morphologies and corresponding EDS results of the oxidized high-strength FeCrAl alloys:(a)FeCr3Al alloy;(b)FeCr4Al alloy and(c)FeCr5Al alloy.

The surface microstructures of the oxidized high-strength FeCrAl alloys are shown in Fig.3.It can be seen that the oxide scales mainly consist of a large number of fine grains and some larger white particles randomly distributed on the oxide scales.Both the number and size of these white particles decrease progressively with the increase of Al content.In view of the change of weight gains shown in Fig.2,the alloy with more white particles exhibits higher weight gain,indicating that the formation of these white particles is strongly associated with the weight gains of the high-strength FeCrAl alloys.The corresponding EDS results shows that the white particles(spectrums 1,3 and 5 in Fig.3)are mainly composed of Fe,O,Al and C,and the atomic ratio of(Fe+Al)to O is approximately 2:3.In consideration of their typical morphologies,the white particles are likely to be α-Fe2O3 with incorporation of high content of Al3+,which needs to be confirmed by XRD.

Fig.4.XRD patterns of the oxidized high-strength FeCrAl alloys:(a)FeCr3Al alloy;(b)FeCr4Al alloy and(c)FeCr5Al alloy.

In addition,element C is detected in the oxide scales and sub-scale matrix of the high-strength FeCrAl alloys from the EDS data(Fig.3 and Figs.5–7).It is generally accepted that C could be produced in the high temperature S-CO2 environment by the Boudouard reaction and the reactions of metallic elements with CO and CO2, as listed below[11,18]:

Meanwhile,it is also shown in Fig.3 that the compositions of the dark oxide scales(spectrums 2,4 and 6)are different from those of the white particles,such that the oxide scales also contain a certain amount of Cr,in addition to Fe O,Al and C.Considering the compositions and the XRD patterns shown in Fig.4,it can be inferred that the dark oxide scales are mixture of Al2O3and spinel oxides(Fe3-xCrxO4).These oxidation products are consistent with the description of literatures[35,43],in which Al2O3 and spinel oxides formed on the aluminaforming steels in the high temperature S-CO2 environment.By further observing the XRD patterns,the Al2O3mainly consists of α-Al2O3 with a small amount of γ-Al2O3.It is easy to understand the formation of γ-Al2O3 because this alumina is reported to form in the temperature range of 500–840°C[44].It is interesting to note that a large amount of α-Al2O3 also forms on the high-strength FeCrAl alloys at 650°C.At present,it is reported that there are two ways to form the α-Al2O3 at lower temperatures[40,44].One is the transformation of metastable Al2O3 to α-Al2O3.This transformation usually induces a whisker or blade-type morphology[36,40,45],which is not observed in the present study.In addition,the transformation sequences are known to be γ-Al2O3→δ-Al2O3→θ-Al2O3→α-Al2O3 or γ-Al2O3→θ-Al2O3→α-Al2O3[44,46].If α-Al2O3 transforms from γ-Al2O3,some intermediate alumina should be remained in the oxide scale during the transformation process.However,except for α-Al2O3 and γ-Al2O3,other transient alumina was not detected in the oxide scale of the high-strength FeCrAl alloys.Therefore,it can be concluded α-Al2O3 does not form through these transformations in the present study.

从2011年起，阜外医院外科手术量及内科PCI数量就已双双破万，且逐年递增。医院相关负责人介绍，如此庞大的手术工作量，为手术室及导管室使用管理工作带来了很大的挑战。传统管理模式手术人员进入手术室后先登记后换鞋、取衣、更衣穿戴整齐后才能到手术室工作，服装管理和人员管理面临量大、繁琐的困境。

Alternatively,the direct development of α-Al2O3 is possible with the assistance of crystallographic templates.Previous studies have shown that the precursor oxides(α-Fe2O3 and α-Cr2O3)with the same crystal structure of α-Al2O3,can serve as a crystallographic template for the nucleation and growth of α-Al2O3[40,41].Moreover,these precursor oxides with incorporation of Al3+,can increase the nucleation of α-Al2O3 and accelerate the growth of the protective layer[47,48].As mentioned before,our experiments have observed certain amounts of α-Fe2O3 with high Al3+content.Therefore,α-Al2O3 in the oxide scales probably nucleates and grows on the Al3+-doped α-Fe2O3particles during S-CO2 oxidation in this study.Then,the nucleation and growth of α-Al2O3 can in turn suppress the formation of metastable Al2O3[37,49],thus the amount of γ-Al2O3in the oxide scales of high-strength FeCrAl alloys is very small.

where M represents the metallic elements Fe,Al and Cr in the present system.The C produced by above reactions can be deposited on the oxide surface and at the oxide/metal interface.Furthermore,C can diffuse into the matrix and reacts with the alloying elements.By comparing Fig.5c,Fig.6c and Fig.7c,it can be seen that the C intensity is very high in oxide scale of the FeCr3Al alloy and it reduces gradually with the increase of Al content.This phenomenon can be explained by two reasons.Firstly,there are more α-Al2O3 generated in the oxide scale for the alloy with higher Al content,and α-Al2O3 scale acts as barrier to prevent the further oxidation of the alloy matrix.The total amount of oxides will decrease in the alloy with higher Al content.According to the reactions(1–3),fewer oxides are associated with the less C product.Therefore,the production of C is lower in the FeCrAl alloy with higher Al content.Secondly,the α-Al2O3 scale also suppresses the inward diffusion of C which is deposited on the oxidesurface.Previous study has indicated that this protective Al2O3 scale can effectively inhibit the penetration of free carbon into the alloy matrix[25,26].Thus the amount of C in the oxide scales gradually decreases with the increase of Al content in the high-strength FeCrAl alloys.

3.4.Cross-sectional microstructures of the oxidized high-strength FeCrAl alloys

Figs. 5–7 exhibit the cross-sectional microstructures and corresponding EDS line scanning results of three high-strength FeCrAl alloys after oxidation in the high-temperature S-CO2 environment.All the oxide scales of the high-strength FeCrAl alloys are very thin and their thicknesses are less than 500 nm.Some oxide particles(Al3+-doped α-Fe2O3)are also present in the oxide layers from the cross-sectional microstructures.It can be seen that the size of these oxide particles is obviously larger than the thickness of oxide scales.These results also imply that the weight gains of high-strength FeCrAl alloys are mainly attributed to the formation of these oxide scales.Therefore,the alloy with more oxide particles possesses higher weight gain,as shown in Figs.2 and 3.

Fig.5.Back-scattered electron SEM cross-sectional morphologies and corresponding EDS line scanning results of the oxidized FeCr3Al alloy:(a)lowmagnification micrograph;(b)high-magnification micrograph and(c)EDS line scanning results.

The EDS line scanning results show that the intensities of Al and O are very high near the oxide layer regions.By careful comparison,it can be seen that the intensity of Al in the oxide layer increases gradually with the increase of Al content in the alloys.These results suggest that the high-strength FeCrAl alloy with higher Al content can form more α-Al2O3 in the oxide layer,and oxide layer becomes more protective.This protective α-Al2O3 scale can impede the outward diffusion of iron ions and the associatively growth of oxide particles(α-Fe2O3).Therefore,the oxidized high-strength FeCrAl alloy with higher Al content has fewer oxide particles with smaller size,and exhibits less weight gain.

Fig.6.Back-scattered electron SEM cross-sectional morphologies and corresponding EDS line scanning results of the oxidized FeCr4Al alloy:(a)lowmagnification micrograph;(b)high-magnification micrograph and(c)EDS line scanning results.

Fig.4 shows the XRD patterns of three high-strength FeCrAl alloys after oxidation in the high-temperature S-CO2 for 500 h.Many α-Fe2O3 peaks presented in the XRD patterns,verifying these white particles were α-Fe2O3.As mentioned previously,the oxidation products of FeCr ferritic/martensitic steels developed in the S-CO2 environment are usually outer Fe3O4 and inner Fe3-xCrxO4 spinel oxides[11–13],while α-Fe2O3 is not easy to be formed on these steels unless the oxygen potential is too high.However,in the present study,α-Fe2O3 formed on the high-strength FeCrAl alloys and no Fe3O4 was detected by XRD.Previous studies have indicated that Fe can react with oxygen before Al and Cr to form α-Fe2O3 at the early oxidation stage of the aluminaforming steels at the intermediate temperatures[40,41].It can be assumed that the pure α-Fe2O3 also forms on the surface of the highstrength FeCrAl alloys at the initial oxidation stage.In the next oxidation stage,the outward diffusing Al atoms can incorporate into the further growth of α-Fe2O3 and the Al3+ions will substitute for Fe3+in α-Fe2O3,finally resulting in α-Fe2O3doped with Al3+.Because of the incorporation of Al3+,the equilibrium oxygen dissociation pressure of α-Fe2O3 decreases and Al3+-dopedα-Fe2O3 can be preserved in the oxide layer during the subsequent oxidation. Furthermore, the formation of Al3+-doped α-Fe2O3 is relatively rapid.Brito et al.[40]found that when the Fe-Al alloys were isothermally oxidized at 700°C,Al3+-dopedα-Fe2O3 formed in the oxide layer after only 2 min oxidation,and the molar concentration of Al3+in α-Fe2O3 gradually increased with increasing oxidation times.Although the solubility limit of α-Al2O3 is about 10 at%in α-Fe2O3 at 650°C according to the equilibrium α-Fe2O3–α-Al2O3 phase diagram[42],the experimental results have shown that the supersaturation of Al3+in α-Fe2O3 may occur in the oxide layer[40],which can explain the formation of α-Fe2O3 with high Al3+content on the high-strength FeCrAl alloys in the present study.

Based on the description in the above sections,the oxidation mechanism of the high-strength FeCrAl alloys in the high-temperature SCO2 environment at 650°C can be speculated,as schematically shown in Fig.8.This oxidation process is supposedly divided into three stages:initial stage,transient stage and steady stage.

Fig.7.Back-scattered electron SEM cross-sectional morphologies and corresponding EDS line scanning results of the oxidized FeCr5Al alloy:(a)lowmagnification micrograph;(b)high-magnification micrograph and(c)EDS line scanning results.

It is noteworthy to mention that the positions of XRD peaks of α-Fe2O3 and α-Al2O3in the present study are not in full compliance with the standard peaks of pure α-Fe2O3 and α-Al2O3.The α-Fe2O3 peaks are slightly shifted to the left of the standard peaks while the α-Al2O3 peaks are shifted to the right.The reason is that the ionic radius of Al3+is smaller than Fe3+,the high Al3+content in α-Fe2O3 can reduce the lattice parameters,resulting in the left shift of XRD peaks.Similarly,the dissolved Fe3+increases the lattice parameter of α-Al2O3 and induces the XRD peaks shifting right.Similarly,the in-situ synchrotron XRD has verified the decrease of lattice parameter in the Al3+-doped α-Fe2O3 and the increase of lattice parameter in the Fe3+-doped α-Al2O3[40].

Table 3 EDS results of chemical compositions (wt%) of micro-regions marked by numbers 1–6 in Fig.5,Fig.6 and Fig.7.

a Note:the EDS values are the mixture of white precipitates and underlying matrix because of the small precipitate size.

Regions Fe Cr Al Mo Nb O C Possible phases 1 77.14 11.68 3.38 0.96 0.00 0.89 5.95 Matrix 2a 55.99 8.91 1.94 12.94 11.27 0.82 8.12 (Nb,Mo)C 3 76.54 11.83 4.07 0.30 0.00 0.52 6.37 Matrix 4a 63.49 9.92 3.20 6.79 8.84 0.61 7.15 (Nb,Mo)C 5 74.96 11.79 5.71 1.10 0.00 0.50 5.59 Matrix 6a 66.62 10.32 3.60 5.86 5.58 0.53 7.49 (Nb,Mo)C

陆九渊和王阳明创立的理学框架的核心以及最高本体是“心”，他们论证“性”和“理”的原因是要证明宇宙的本体是心，同时心也是认识外界的主体、伦理道德的本体。心包括道、性、理，心就是理和心就是性，这两者的本质是一样的，且两者能够一起实现。

Furthermore, it is interesting to note that a large number of(Nb,Mo)C carbides are randomly distributed in the sub-scale matrix of the FeCr3Al alloy(Fig.5),while with the increase of Al content,their number gradually decreases and they tend to be arranged in a line parallel to the surface in the high-strength FeCrAl alloys(Fig.6 and Fig.7).These phenomena may be attributed to the different degrees of depletion of the oxidized elements(Fe,Cr and Al)in the sub-scale matrix.For the FeCrAl alloy with lower Al content,more elements are oxidized because of its poor oxidation resistance,thus the depletion degree of the oxidized elements in the sub-scale matrix should be significant.Then,this depletion will induce more Nb and Mo diffusing form the center region into the sub-scale matrix.On the other hand,as mentioned above,more C is present in the oxide scale and sub-scale matrix for the FeCr3Al alloy.Therefore,the nucleation of(Nb,Mo)C carbides may occur in the sub-scale matrix and many(Nb,Mo)C nuclei can incessantly grow in the FeCr3Al alloy.To the contrary,because of the small amount of Nb,Mo and C elements in the sub-scale matrix for the FeCrAl alloy with higher Al,only a few of(Nb,Mo)C carbides can successfully grow up at some preferential sites. Consequently, the amount of(Nb,Mo)C carbides in the sub-scale matrix decreases for the FeCrAl alloys with higher Al content.

3.5.Oxidation mechanism in high-temperature S-CO2

资料来源该院诊治的神经外科疾病合并糖尿病患者，共120例，按照随机数字表法将患者分成两组，对照组60例，男性36例，女性24例，年龄21～79岁，平均（49.35±8.34）岁。神经外科疾病：开放性颅脑损伤20例，脑出血12例，脑膜瘤11例，颅骨骨折9例，神经纤维瘤和颅内动脉瘤各4例；观察组60例，男性和女性患者的例数分别为34例和26例，年龄23～75岁，平均（48.23±8.23）岁。神经外科疾病：开放性颅脑损伤19例，脑出血13例，脑膜瘤12例，颅骨骨折7例，神经纤维瘤5例，颅内动脉瘤4例。对两组患者的临床资料进行统计分析，差异无统计学意义（P＞0.05），具有可比性。

·In the initial stage,many small α-Fe2O3 particles form on the alloy surface before Al and Cr react with oxygen.The matrix of the highstrength FeCrAl alloys is hardly affected by the oxidation reactions because the initial stage is very short. A large amount of intragranular Fe2(Nb,Mo)precipitates and intergranular(Nb,Mo)C carbides exist in the entire matrix of the high-strength FeCrAl alloys.

·In the transient stage,Al incorporates into the growth of α-Fe2O3 particles and the Al3+ions substitute for Fe3+in the α-Fe2O3 crystal lattice to form the Al3+-doped α-Fe2O3 particles.The Al3+-doped α-Fe2O3 particles serve as crystallographic template to assist the formation of α-Al2O3 phase.Free carbon is continually generated by these oxidation reactions.In the matrix near surface,new(Nb,Mo)C carbides form as a result of the reaction of Fe2(Nb,Mo)precipitates and the inwardly diffusing carbon.

Fig.8.Schematic diagram showing the oxidation mechanism of the high-strength FeCrAl alloys in the high-temperature S-CO2 environment.

·In the steady stage,some Al3+-doped α-Fe2O3 particles further grow to form the larger α-Fe2O3 particles on the oxide scale.The oxide scale mainly consists of α-Al2O3 and spinel oxides.Based on the previous experiences[40,41],this oxide scale is likely to be a twolayer structure with outer spinel oxides layer and inner α-Al2O3 layer.In the sub-scale matrix,larger(Nb,Mo)C carbides form by the growing and/or dissolving of many small(Nb,Mo)C carbides and by the outward diffusion of Nb and Mo from the center region into the sub-scale matrix.The Nb and Mo-depleted zone is developed due to the formation of larger(Nb,Mo)C carbides.

4.Conclusions

The oxidation behavior of high-strength FeCrAl alloys has been investigated in the S-CO2 environment at 650°C and 20 MPa for 500 h and the effect of Al content on the oxidation behavior was evaluated in the present study.

1.Many intragranular Fe2(Nb,Mo) Laves precipitates and intergranular(Nb,Mo)C carbides are present in the un-oxidized highstrength FeCrAl alloys,while Al content has no obvious effect on the microstructures of the alloys.

具体的检测过程及检测结果是：对电磁阀电阻进行测量，电阻为30Ω，新电磁阀电阻为26Ω，没有明显异常。电磁阀插头1#端子，试灯点亮，说明供电正常。在电磁阀插头13和2#端子之间连接试灯，GDS2作动电磁阀工作，试灯不能点亮，晃动线束，试灯依然不会点亮。使用专用诊断仪GDS2检测，读取故障码P201000(进气歧管通路控制阀控制电路电压过高)。测量2#端子对地电压3.56V，这是来自ECM的监测电压，说明线路是导通的。KEY OFF时测量对地电阻，正向15.6MΩ，反向OL，对比进气歧管调谐阀控制电路的测量，结果一致。

对于我们来说，共享经济的监督管理是一种比较重要的方式。因为共享经济产品主要的生产目的就是为人们进行服务，人们通过使用共享经济产品不仅仅可以提高生活质量还可以满足人们的日常生活需求。但是到目前为止即使政府加入了监管，但是很多方面还是出现了监管不力的情况。在我们的实际生活中发现，这种新型的管理模式随处可见，但是由于这种模式刚刚开始运行，所以很多人对这种共享模式的认知还不够全面。而且共享经济作为一种新生事物，很多时候政府没有全面的履行自己的职责。例如：现阶段城市中出现的共享单车，很多人在使用之后都不会将车停到停车位，导致车辆的随意摆放，给城市造成了交通影响。

2.With the increase of Al content,the weight gains of the highstrength FeCrAl alloys gradually decrease.This decrease of weight gain is mainly attributed to the reduction in number and size of α-Fe2O3 particles which possibly form on the oxide scales at the early oxidation stage.

经评分，模型组小鼠阴道病变情况普遍严重，病变小鼠数量较多，而给药各组均得到有效改善，其中以黄柏碱（40 mg/kg）与加替沙星改善作用最为明显。经观察，模型组小鼠阴道黏膜上皮坏死脱落、溃烂，出现炎性细胞浸润、黏膜充血及大量炎性物渗出等。给予不同剂量黄柏碱或加替沙星治疗后，BV小鼠阴道病变得到不同程度的缓解，充血水肿、细胞浸润、组织溃烂、上皮坏死等均得到改善。经HE染色发现，模型组小鼠阴道切片中有大量炎性细胞分布，而给药组均得到不同程度缓解，HE染色结果见图1，病理评分结果见表2。

3.The oxide scales are primarily composed of Al2O3(α-Al2O3 with a small amount of γ-Al2O3)and spinel oxides.The amount of α-Al2O3 in the oxide scales increases with the increase of Al content.

4.Less C is present in the oxide scale and sub-scale matrix for the highstrength FeCrAl alloys with higher Al content,because α-Al2O3 scale serves as barrier to prevent the further oxidation and to suppress the inward diffusion of C element.

5.Larger(Nb,Mo)C carbides form in the sub-scale matrix of the highstrength FeCrAl alloys and the number of these(Nb,Mo)C carbides decreases with increasing Al content.

建筑电气工程需要考虑使用者的经济条件和使用要求。在施工的过程中，需要首先考虑的就是用户的使用要求，要使安装的电气能够满足用户的使用要求，并使其在使用过程中方便、安全。同时，电气施工还要考虑到用户的经济条件，在满足安全适用条件的基础上，为用户节省资源，也要坚持我国节约型社会的建设。

SF用它的经营理念更快帮助客户、对市场更快更好地做出反应：为了提高企业的市场竞争力不断地推出新产品，使服务周期能够缩短；调整竞争策略，使经营成本能够降低。SF为国民经济的持续健康发展贡献了税收做出了积极贡献，它的企业规模也在缓解社会就业压力上做出了应有的积极贡献。

Acknowledgements

This study was supported by the Engineering Research Center Program(No.2016R1A5A1013919)and the Nuclear R&D Program(No.2018M2A8A4023311)of MSIP/NRF of Rep.of Korea.The authors are grateful for the financial support of the Sichuan Science and Technology Programm(No.2018JY0155)of P.R.of China and the BKPlus Program of the MSIP/NRF of Rep.of Korea.H.Chen acknowledges support from the China Scholarship Council.

References

[1] K.Brun,P.Friedman,R.Dennis,Fundamentals and Applications of Supercritical Carbon Dioxide(sCO2)Based Power Cycles,Woodhead Publishing,2017.

[2] R.Moore,T.Conboy,Metal corrosion in a supercritical carbon dioxide-liquid sodium power cycle.Sandia National Laboratories Reports,SAND2012-0184,2012.

[3] J.P.Gibbs,Power conversion system design for supercritical carbon dioxide cooled indirect cycle nuclear reactors.Doctor Thesis of Massachusetts Institute of Technology,2008.

[4] Y.Ahn,Bae,S.J.Bae,M.Kim,S.K.Cho,S.Baik,J.I.Lee,J.E.Cha,Nucl.Eng.Technol.47(2015)647–661.

[5] P.Garg,P.Kumar,K.Srinivasan,J.Supercrit.Fluid.76(2013)54–60.

[6] H.C.No,J.H.Eoh,Y.H.Yoo,S.O.Kim,Nucl.Technol.173(2011)99–114.

[7] B.D.Iverson,T.M.Conboy,J.J.Pasch,A.M.Kruizenga,Appl.Energ.111(2013)957–970.

[8] S.A.Wright,C.S.Davidson,W.O.Scammell,Thermo-economic analysis of four wasteheat recovery power systems,in:Proceedings of the 5th International,Symposium–Supercritical CO2 Power Cycles,San Antonio,Texas,March 28–31,2016.

[9] G.O.Subramanian,H.J.Lee,S.H.Kim,C.Jang,Oxid.Met.89(2018)683–697.

[10] H.J.Lee,S.H.Kim,H.Kim,C.Jang,Appl.Surf.Sci.388(2016)483–490.

[11] F.Rouillard,T.Furukawa,Corros.Sci.105(2016)120–132.

[12] T.Furukawa,Y.Inagaki,M.Aritomi,Prog.Nucl.Energ.53(2011)1050–1055.

[13] F.Rouillard,G.Moine,L.Martinelli,J.C.Ruiz,Oxid.Met.77(2012)27–55.

[14] G.Cao,V.Firouzdor,K.Sridharan,M.Anderson,T.R.Allen,Corros.Sci.60(2012)246–255.

[15] L.Tan,M.Anderson,D.Taylor,T.R.Allen,Corros.Sci.53(2011)3273–3280.

[16] V.Firouzdor,G.Cao,K.Sridharan,M.Anderson,T.R.Allen,Corros.Sci.69(2013)281–291.

[17] H.J.Lee,G.O.Subramanian,S.H.Kim,C.Jang,Corros.Sci.111(2016)649–658.

[18] H.J.Lee,H.Kim,S.H.Kim,C.Jang,Corros.Sci.99(2015)227–239.

[19] H.J.Lee,S.H.Kim,C.Jang,Mater.Charact.138(2018)245–254.

[20] R.I.Olivares,D.J.Young,P.Marvig,W.Stein,Oxid.Met.84(2015)585–606.

[21] C.R.F.Azevedo,Eng.Fail.Anal.18(2011)1943–1962.

[22] R.L.Klueh,A.T.Nelson,J.Nucl.Mater.371(2007)37–52.

[23] D.J.Park,H.G.Kim,J.Y.Park,Y.I.Jung,J.H.Park,Y.H.Koo,Corros.Sci.94(2015)459–465.

[24] Z.Sun,Y.Yamamoto,Mater.Sci.Eng.A 700(2017)554–561.

[25] S.H.Kim,G.O.Subramanian,C.Kim,C.Jan,K.M.Park,Surf.Coat.Tech.349(2018)415–425.

[26] Kanthal,Kanthal powder metallurgy high-temperature tubes.Sandvik Materials Technology Report,2011.

[27] S.H.Kim,C.Kim,G.O.Subramanian,C.Jang,Corrosion and carburization behaviour of Ni-Cr-Mo-Nb superalloys in a high temperature supercritical-CO2 environment,in:Proceedings of the 9th International Symposium on Superalloy 718&Derivatives:Energy,Aerospace,and Industrial Applications,The Minerals,Metals&Materials Series,Springer,Cham,2018.

[28] F.Rouillard,G.Moine,M.Tabarant,J.C.Ruiz,Oxid.Met.77(2012)57–70.

[29] I.E.McCarroll,A.La Fontaine,T.D.Nguyen,A.F.Smith,J.Zhang,D.J.Young,J.M.Cairney,Corros.Sci.139(2018)267–274.

[30] Y.Yamamoto,Y.Yang,K.G.Field,K.Terrani,B.A.Pint,L.L.Snead,Letter Report Documenting Progress of Second Generation ATF FeCrAl Alloy Fabrication(ORNL/LTR-2014/219),Oak Ridge National Laboratory(ORNL/LTR-2014/219),2014.

[31] Z.Sun,H.Bei,Y.Mater.Charact.132(2017)126–131.

[32] G.R.Speich,Trans.Metall.Soc.AIME 224(1962)850–858.

[33] Z.Sun,P.D.Edmondson,Y.Yamamoto,Acta Mater.144(2018)716–727.

[34] D.J.Siegel,J.C.Hamilton,Acta Mater.53(2005)87–96.

[35] V.Firouzdor,G.P.Cao,K.Sridharan,M.Anderson,T.R.Allen,Mater.Corros.66(2015)137–142.

[36] H.E.Kadiri,R.Molins,Y.Bienvenu,M.F.Horstemeyer,Oxid.Met.64(2005)63–97.

[37] Y.Kitajima,S.Hayashi,T.Nishimoto,T.Narita,S.Ukai,Oxid.Met.75(2011)41–56.

[38] C.Jang,D.Lee,D.Kim,Int.J.Pres.Ves.Pip.856(2008)368–377.

[39] H.Götlind,F.Liu,J.E.Svensson,M.Halvarsson,L.G.Johansson,Oxid.Met.67(2007)251–266.

[40] P.Brito,H.Pinto,A.Kostka,Corros.Sci.105(2016)100–108.

[41] B.Pöter,F.Stein,R.Wirth,M.Spiegel,Z.Phys.Chem.219(2005)1489–1503.

[42] A.C.Turnock,H.P.Eugster,J.Petrol.3(1962)533–565.

[43] L.F.He,P.Roman,B.Leng,K.Sridharan,M.Anderson,T.R.Allen,Corros.Sci.82(2014)67–76.

[44] W.H.Gitzen,Alumina as a Ceramic Material,Wiley,1970.

[45] E.N’Dah,A.Galerie,Y.Wouters,D.Goossens,D.Naumenko,V.Kochubey,W.J.Quadakkers,Mater.Corros.56(2005)843–847.

[46] K.Reszka,J.Morgiel,Z.Zurek,A.Jaron,Arch.Metall.Mater.59(2014)77–81.

[47] H.Asteman,M.Spiegel,Corros.Sci.50(2008)1734–1743.

[48] S.Hayashi,Y.Takada,I.Saeki,A.Yamauchi,Y.Nishiyama,T.Doi,S.Kyo,M.Sato,Mater.Corros.63(2012)862–868.

[49] Y.Kitajima,S.Hayashi,T.Nishimoto,T.Narita,S.Ukai,Oxid.Met.73(2010)375–388.

[50] V.Kuzucu,M.Aksoy,M.H.Korkut,J.Mater.Process.Tech.82(1998)165–171.

[51] F.Delaunay,C.Berthier,M.Lenglet,J.M.Lameille,Microchim.Acta 132(2000)337–343.

[52] H.J.Lee,H.Kim,D.Kim,C.Jang,Mater.Charact.106(2015)283–291.

[53] A.Chyrkin,P.Huczkowski,V.Shemet,L.Singheiser,W.J.Quadakkers,Oxid.Met.75(2011)143–166.