Title of Invention: METHOD OF SEPARATELY VISUALIZING AUSTENITE PHASE, MARTENSITE PHASE AND BAINITIC-FERRITE MATRIX IN BAINITIC STEEL AND BAINITIC STEEL SPECIMEN FOR MICROSTRUCTURE OBSERVATION

Technical Field

[0001] The disclosure generally relates to a method of separately visualizing an austenite phase, a martensite phase and a bainitic-ferrite matrix in a bainitic steel and a bainitic steel specimen for microstructure observation.

Background

[0002] Progress in bainitic steels development is inseparably connected with precise and accurate knowledge of their structure. Mechanical properties of these steels are strongly affected by carbon enriched residual martensite-austenite (MA) constituents. Characterization of nano-bainitic steels is particularly challenging. These steels contain very fine secondary phases, which are inhomogeneously dispersed within a bainitic-ferrite matrix such as a bainitic matrix. Detailed knowledge of MA constituents fraction, size and morphology play a crucial role in understanding of the mechanism influencing properties of bainitic steels.

[0003] There are several conventional methods enabling characterization of MA constituents in bainitic steels. Using of these methods is inseparably connected with a dilemma: either high special resolution needful for visualization of nano-phases or possibility to investigate large area necessary in the case of low fraction inhomogeneously distributed MA constituents. There is not any technique enabling combination of both of the above-mentioned requirements.

[0004] Conventional and also advanced techniques for an optical microscopy are described in NPL 1. The objective of this extensive study is to develop the methodology for the quantitative and qualitative determination of phases in multiphase steels. There are described a wide range of etch methods for an optical microscopy (OM), such as conventionally used etching techniques (i.e. simple electro-etching, Nital, Picral, le Pera and Klemm etching) and also advanced two step etch methods. All of the above-mentioned techniques have been applied on TRIP (Transformation Induced Plasticity) steels and it can be concluded that: >

- conventional electro-polishing, Nital etching and Picral etching are not able to provide a contrast between secondary phases and matrix in the

OM;

- although the conventional methods using the Le Pera and Klemm etchants enable to differentiate between the secondary phases and matrix based on their specific color in the OM, separation of austenite and martensite is impossible (but NPL 2 describes that only Klemm etchants can be utilized for the retained austenite visualization in the OM); and

- two steps etching techniques utilizing two reagents (i.e. Picral + sodium bisulfate, Nital + sodium bisulfate and V2A + Klemm) are able to differentiate the secondary phases from the matrix and moreover there is a possibility to distinguish an austenite phase from a martensite phase because of their different color in the OM micrographs.

[0005] This work also contains the experiments, which combine the color etching techniques with a scanning electron microscopy technique (SEM). The SEM technique offers high spatial resolution and nano-scale features can be visualized. Unfortunately, the experiments shown that it is not possible to separate martensite from austenite.

[0006] The above-mentioned techniques utilize the optical microscope, but its spatial resolution is limited (about 200 nm). These techniques despite possibility of visualize and even separate the phases in MA constituents are not sufficient for characterization of advanced bainitic steels containing nano-sized MA constituents.

[0007] PTL 1 describes a color etching technique for visualization of MA constituents in a laser confocal microscope, which improved the spatial resolution. Unfortunately, the technique doesn't enable to possibility to separate martensite and austenite phase in MA constituents. The technique described in PTL 1 offers a possibility to visualize MA constituents with higher spatial resolution than the techniques utilizing the optical microscope, but it is still insufficient for characterization at the nano-scale.

[0008] The next possibility how to visualize MA constituents in bainitic steel is to use an electron backscattered technique (EBSD) in the SEM. This technique enables to differentiate austenite phase due to its different crystal lattice (face-centered cubic (FCC) type). Martensite phase is possible to separate based on its very low quality of diffraction pattern. The EBSD technique offers to visualize MA constituents in bainitic steels and separate martensite and austenite phase with high spatial resolution (about 20 nm).

Citation List

Patent Literature

[0009] PTL 1 : CN 102175191 A

Non Patent Literature

[0010] NPL 1 : E. Leunis et al., Quantitative phase analysis of multi-phase steels - PHAST, Final Report, EUR 22387 EN, European Communities, 2006 NPL 2: K. Radwanski et al. : Role of the advanced microstructure characterization in modeling of mechanical properties of AHSS steels, Materials Science and Engineering A 639 (2015) 567-574

Summary

Technical Problem

[0011] There are two main disadvantages of the EBSD technique. Firstly, spatial resolution of about 20 nm becomes insufficient for characterization for nano-sized MA constituents, such as foil-like austenite between the ferritic laths and fine austenite phases inside MA constituents. Secondly, this technique is not optimal for large area mapping due to very slow data collection in the case of nano-scale mapping.

[0012] Transmission electron microscopy offers the better spatial resolution, but, on the other hand, a carefully prepared sample is required and there is an absence of large area mapping and representative and statistically significant data cannot be obtained.

[0013] Thus, there are not any technique that can separately visualize MA constituents including an austenite phase and a martensite phase in bainitic steels, which offers high spatial resolution (less than 10 nm) and simultaneously fast data collection enabling large area mapping in suitable time. [0014] An object of the present disclosure is, therefore, to provide a method of separately visualizing an austenite phase and a martensite phase in a bainitic steel capable of clearly and visually identifying the austenite and martensite phases in the bainitic steel individually at the nanoscale. Another object of the present disclosure is to provide a bainitic steel specimen for microstructure observation capable of clearly and visually identifying the austenite and martensite phases in the bainitic steel individually at the nanoscale.

Solution to Problem

[0015] As a result of conducting intensive study, the present inventors have discovered that, when a surface of a specimen of a bainitic steel is electro-polished in a certain electrolyte solution and thereafter the specimen of the bainitic steel is continuously retained in the electrolyte solution, corrosion product layers are formed selectively on martensite and austenite phases of the surface of the specimen; and that the thickness of the corrosion product layer depends on the carbon content of the corroded phase. The present inventors have thus found that, when the surface of the specimen of the bainitic steel thus treated is observed with a scanning electron microscopy (SEM) and a backscattered electron (BSE) detector, the difference in thickness of the corrosion product layers formed on the martensite and austenite phases of the surface of the specimen appears as a contract in an SEM BSE image, which enables to visually observe the austenite phase and the martensite phase individually.

[0016] The present disclosure has been made based on these findings, and its gist is as follows:

(1) A method of separately visualizing an austenite phase, a martensite phase and a bainitic-ferrite matrix in a bainitic steel comprising the steps of: a) electro-polishing a surface of a specimen of a bainitic steel in an electrolyte solution containing a corrosive acid and an organic solvent to form a native oxide layer on the surface of the specimen;

b) after completion of the electro-polishing, continuously retaining the specimen in the electrolyte solution to form first and second corrosion product layers selectively on domains of the native oxide layer covering martensite and austenite phases, respectively; and c) subjecting the surface of the specimen to scanning electron microscopy observation by means of a backscattered electron detector and providing a contrast between the first and second corrosion product layers and remaining surface to visualize the austenite phase as a darkest area, the bainitic-ferrite matrix as a brightest area, and the martensite phase as an intermediate brightness area;

wherein the step c) is performed under a condition that a landing energy of the primary electrons is 3 keV or less.

[0017] (2) The method according to foregoing (1), wherein the backscattered electron detector is an in-lens type detector.

[0018] (3) The method according to foregoing ( 1), wherein the backscattered electron detector is located below a pole-piece and consists of individually operable concentric segments.

[0019] (4) The method according to foregoing (1), wherein the backscattered electron detector is located below a pole-piece and has a working distance longer than 10 mm.

[0020] (5) The method according to any one of foregoing (1 ) to (4), wherein the landing energy of the primary electrons is 1 keV or less.

[0021] (6) The method according to any one of foregoing ( 1 ) to (5), wherein the landing energy of the primary electrons is 0.5 keV or more.

[0022] (7) The method according to any one of foregoing (1) to (6), wherein a mean thickness of the first corrosion product layer is between 1 nm and 9 nm.

[0023] (8) The method according to any one of foregoing (1) to (7), wherein a mean thickness of the second corrosion product layer is between 2 nm and 10 nm.

[0024] (9) The method according to any one of foregoing (1 ) to (8), wherein the corrosive acid is selected from a group consisting of an oxidizing acid and an organic acid.

[0025] (10) The method according to any one of foregoing (1) to (9), further comprising: d) evaluating quantitative and/or qualitative characterization of the visualized austenite and martensite phases.

[0026] (11 ) A bainitic steel specimen for microstructure observation comprising:

a bainitic steel base having a plurality of phases including an austenite phase, a martensite phase and a bainitic-ferrite matrix,

a native oxide layer formed on a surface of the bainitic steel base, a first corrosion product layer formed on a domain of the native oxide layer covering the martensite phase, and

a second corrosion product layer formed on a domain of the native oxide layer covering the martensite austenite phase and having a thickness larger than a thickness of the first corrosion product layer.

[0027] (12) The bainitic steel specimen for microstructure observation according to foregoing ( 11), wherein a mean thickness of the first corrosion product layer is between 1 nm and 9 nm.

[0028] (13) The bainitic steel specimen for microstructure observation according to foregoing (1 1 ) or (12), wherein a mean thickness of the second corrosion product layer is between 2 nm and 10 nm. Advantageous Effect

[0029] The method of visualizing an austenite phase, a martensite phase and a bainitic-ferrite matrix in a bainitic steel and the bainitic steel specimen for microstructure observation according the present disclosure enable to clearly and visually identify the austenite and martensite phases in the bainitic steel individually at the nanoscale.

Brief Description of Drawings

[0030] In the accompanying drawing:

Fig. 1 (a) is an SEM BSE image of a bainitic steel containing MA constituents prepared by a conventional electro-polishing and captured at 1 keV landing energy of the primary electrons using a BSE detector located below a pole-piece, and Fig. 1 (b) is an SEM BSE image of a bainitic steel containing MA constituents prepared according to the present disclosure and captured at 1 keV landing energy of the primary electrons using a BSE detector located below a pole-piece.

Figs. 2(a)-(c) are SEM BSE images of an austenite phase inside the bainitic-ferrite matrix obtained at 1 keV, 3 keV, and 5 keV landing energies, respectively, of the primary electrons.

Figs. 3(a)-(c) are SEM images obtained at 1 keV landing energy of the - - primary electrons by an SE in-lens detector, an SE out-lens Everhart-Thornley (ET) detector, and a BSE detector, respectively.

Fig. 4(a) is an electron backscatter diffraction (EBSD) phase map in which the green parts represent the austenite phase and the red parts represent the martensite phase and the bainitic-ferrite matrix, Fig. 4(b) is an EBSD image quality map in which the black parts represent the martensite phase, and Fig. 4(c) is an SEM BSE image of the same area obtained at 1 keV landing energy of the primary electrons by a BSE detector.

Fig. 5(a) is an EBSD image quality map, Fig. 5(b) is a phase map, and Fig. 5(c) is a BSE image of the same area according to the present disclosure.

Fig. 6 shows area fractions of the austenite and martensite phases acquired by the method according to the present disclosure, and a volume fraction of the austenite phase obtained by the XRD technique.

Fig. 7 shows the effects of the annealing temperature on size, shape and morphology of the austenite and martensite phases.

Fig. 8 is an SEM BSE image obtained at 1 keV landing energy of the primary electrons by an BSE in-lens detector.

Detailed Description

[0031] A method of separately visualizing an austenite phase, a martensite phase and a bainitic-ferrite matrix in a bainitic steel according to one embodiment of the present disclosure includes: processing a surface of a bainitic steel base with certain treatments (steps a and b) to give a specimen to be observed, and then subjecting the surface of the specimen to SEM observation by means of a back scatter electron detector to separately visualize an austenite phase, a martensite phase and a bainitic-ferrite matrix under a certain condition (step c). The method may also include evaluating quantitative and/or qualitative characterization of the visualized austenite and martensite phases (step d).

[0032] The bainitic steel specimen is prepared by the following manner. First, a piece of a bainitic steel is provided. A surface of the bainitic steel piece is then mechanically polished to mirror finish. Any conventional abrasives may be used for the mechanical polishing. Diamond particles having a particle size of about 1 μιη are preferably used. A bainitic steel base having a polished surface is thus obtained.

[0033] (Step a: Electro-polishing process)

The bainitic steel base is then subjected to an electro-polishing process in an electrolyte solution. It will be appreciated that any suitable electro-polishing process known in the art may be used herein. The electrolyte solution consists essentially of a corrosive acid and an organic solvent. The term "a corrosive acid" as used herein refers to any acids which can corrode the surface of the bainitic steel base, and preferably contains chlorine atoms. The corrosive acid may be an oxidizing acid. The term "an oxidizing acid" as used herein refers to an acid contains an anion with an oxidation potential higher than the potential of H ⁺ ion. Preferably, the oxidizing acid is at least one selected from a group consisting of, but not limited to, perchloric acid, sulfuric acid and phosphoric acid. The corrosive acid oxides and dissolves the surface of the bainitic steel base to form a native oxide layer having a generally uniform thickness throughout the surface of the bainitic steel base. The concentration of the corrosive acid in the electrolyte solution is preferably 0.30 mol/L or more, and more preferably 0.45 mol/L or more, and most preferably about 0.62 mol/L. When the concentration of the corrosive acid is lower than 0.3 mol/L, a polishing time significantly increases or the surface of the bainitic steel base is not sufficiently corroded due to a low reaction rate of the acid with the bainitic steel base, which may deteriorate the surface smoothness of the bainitic steel base. The upper limit of the concentration of the corrosive acid depends on a type of the corrosive acid used. For example, in case of perchloric acid, the concentration in the electrolyte solution is preferably 0.90 mol/L or less, more preferably 0.75 mol/L or less. Since perchloric acid is unstable and can be explosive at room temperature, the electrolyte solution including prechloric acid higher than 0.90 mol/L is difficult to be handled. Also, the surface of the bainitic steel base may be over-etched, which diminishes the smoothness of the surface. If the surface of the bainitic steel base is under-etched or over-etched, a distinctive contrast cannot be obtained in the subsequent process.

[0034] The organic solvent is used for adjusting a viscosity of the electrolyte solution, which may affect the thickness of the native oxide layer. Examples of the organic solvent include, but not limited to, lower alcohols having 1 to 10 carbon atoms such as methanol, glycerol, and butyl glycol, ethers such as 2-n-butoxyethanol, acetic acid, and mixture thereof. The concentration of the organic solvent in the electrolyte solution may be determined based on the kind and the surface smoothness of the bainitic steel base to be etched. If the viscosity is too low, the electrolyte solution does not sufficiently stay on the surface of the bainitic steel base, and the surface is not uniformly polished.

[0035] The temperature of the electrolyte solution affects the reaction rate of the corrosion. The lower temperature will need more processing time, which is practically not favorable. The higher temperature will accelerate the corrosion but may cause an over-etching. Also, the higher temperature renders the corrosive acid more unstable. In view of these, the temperature of the electrolyte solution is preferably from 283 to 308 K, and more preferably from 293 to 298 K.

[0036] The voltage applied to the electrolyte solution and the duration of the electro polishing process also affect the formation of the native oxide layer. The voltage is preferably in a range of 15 to 45 V, and more preferably 35 to 40 V. When the voltage is less than 15 V, the surface will be etched rather than polished. On the other hand, when the voltage is more than 45 the surface will be destroyed due to high current density. The duration can be determined based on process conditions such as temperature, voltage, concentration of corrosive acid and the like. Typical duration is, but not limited to, 2 seconds.

[0037] The thickness of the native oxide layer is preferably 2 to 4 nm. The thickness of the native oxide layer can be measured from the cross-section STEM (HAADF) images.

[0038] (Step b: Retaining process)

After the completion of the electro-polishing process, the bias is switched off and the bainitic steel base is continuously retained in the electrolyte solution for a given time period. Then, the bainitic steel base is removed from the electrolyte solution and cleaned in methanol and subsequently in ethanol. The surface of the bainitic steel base is corroded and becomes matte. Various masks having different sizes may be used to cover the surface of the bainitic steel base, but a smaller mask such as one providing an electro-polished surface area of 0.1 -0.3 cm ² is preferably used. An ultra-sonic cleaner may be used for stirring the electrolyte solution.

[0039] The duration of this retaining process may depend on the compositions of the bainitic steel, the concentration of the corrosive acid in the electrolyte solution, the processing temperature, and the like. In some cases, the duration can be as short as one second, but typically it is 2- 10 seconds.

[0040] During this retaining process, first and second corrosion product layers are formed selectively on the martensite and austenite phases of the surface of the specimen, respectively. It is believed that both of the martensite and austenite phase are carbon rich (i.e, the carbon contents thereof are higher than the remaining surface), which facilitates an additional corrosion during the retaining process. The corrosion product layer is porous and therefore thicker than the native oxide layer. In principle, the thickness of the corrosion product layer depends on the carbon content of the corroded phase. The austenite phase which has the highest carbon content in the bainitic steel is most intensively corroded to form a thickest corrosion product layer thereon. The martensite phase which has a smaller carbon content than that of the austenite phase is moderately corroded to form a thinner corrosion product layer thereon. The bainitic-ferrite matrix contains least carbon and has high corrosion resistivity in the electrolyte solution, so that no corrosion product layer is formed thereon.

[0041] The thicknesses of the corrosion product layers greatly affect the energy of the backscattered electron, and influences the below-discussed contrast in SEM micrographs. The contrast in the SEM images increases as the thickness of the corrosion product layer increases, and thus a thicker corrosion product layer will give a better result. In order to secure clear contrasts in the SEM images, a mean thickness of each of the first and second corrosion product layers is preferably at least 1 nm. An excessively thick corrosion product layers such as one having a mean thickness greater than 10 nm, however, may involve intensive surface degradation, which makes it difficult to obtain the SEM images having reasonable quality. In view of this, a mean thickness of each of the first and second corrosion product layers is preferably at most 10 nm. Further, in order to clearly distinguish the first corrosion product layer from the second corrosion product layer, the difference in thickness between the first and second corrosion product layers is at least 1 nm. In view of foregoing, a mean thickness of the first corrosion product layer is preferably between 1 nm and 9 nm, and a mean thickness of the second corrosion product layer is preferably between 2 nm and 10 nm. The mean thickness of the corrosion product layer can be measured from the cross-section STEM (HAADF) image. There is a difference in the internal structures between the native oxide layer and the corrosion product layers. The native oxide layer is compact whereas the corrosion product layers have porous structures. In addition, these layers have different chemical compositions; the native oxide layer essentially consists of iron oxides and each of the corrosion product layer is a mixture of oxides and hydroxides. The interface between the layers in the STEM (HAADF) images can be determined by a difference of the signal electrons yield from the oxide and corrosive layer.

[0042] (Step c: Visualizing process)

In this way, a bainitic steel specimen having a native oxide layer and corrosion product layers is prepared. The specimen is then subjected to scanning electron microscopy (SEM) observation. When an electron beam is irradiated on the surface of the specimen, backscattered electrons passing through a thicker corrosion product layer will have a lower energy. Thus, in the SEM image, the austenite phase which has a thicker corrosion product layer appears as a darkest area; the bainitic-ferrite matrix which has no corrosion product layer appears as a brightest area; and the martensite phase which has a thinner corrosion product layer appears as a moderate brightness area. In addition, the backscattered electrons are very sensitive to a chemical composition of a specimen. As a result, distinctive contrasts between the austenite phase and martensite phase and between the martensite phase and the remaining surface are generated, and the austenite and martensite phases in the bainitic steel can be individually visualized at the nanoscale.

[0043] Preferable operation conditions securing the contrast highly suited to quantitative and qualitative image analysis without extensive image post processing are reached by detection of the signal electrons carrying information about the specimen surface composition. It is also advantageous to operate the SEM at low landing energies securing high surface sensitivity. Combination of these operation conditions enables to obtain clear contrast between the austenite phase and the other phases in the bainitic steel.

[0044] In principle, the higher contrast will facilitate the identification of the austenite phase in the SEM micrographs. Here, a contrast ratio C (%) is defined as

C = (S _F - S _y ) / (S _F + S _y ) * 100

where S _Y is average luminance of the corrosion product layer, and Sp is average luminance of the remaining surface in the SEM micrograph. The contrast ratio is preferably at least 1 .00%, more preferably at least 1.25%, and most preferably at least 1.50%.

[0045] The above-described preparation technique ensures that the austenite phase and the martensite phase are covered by the corrosion product layer formed on the native oxide layer. Thus, the landing energy should be so low that the dominant part of the interaction volume of the primary electrons and the specimen is situated within the native oxide layer. To this end, the landing energy of the primary electrons is 3 keV or less, and preferably 1 keV or less. There is an energy dependence of the penetration depth of the primary electrons in solids. At the landing energy higher than 3 keV, the specimen bulk becomes a dominant source of the signal electrons and surface information (i.e., presence of the corrosion product layer) is lost, which deteriorates the contrasts in the SEM image. On the other hand, a sufficient penetration depth of the primary electrons in solids may not be obtained if the landing energy is too small. In this regard, the landing energy of the primary electrons is preferably 0.5 keV or more.

[0046] The bainitic steel may be any carbon steel having nano-scale fine secondary phases including an austenite phase (also referred to as a "γ phase"), a martensite phase, and a bainitic-ferrite matrix therein, such as Transformation Induced Plasticity (TRIP) steel. However, the method is not applicable to stainless steel because stainless steel has a passive layer on its surface which prevents the formation of the corrosion product layer. Other phases present in the bainitic steel may be a ferrite phase, a pearlite phase, a carbide phase, and the like.

[0047] Various types of BSE detectors has been known in the art and may be used in this disclosure. In particular, BSE detectors applicable to this disclosure are an in-lens detector and an out-lens detector. The in-lens detector allows for imaging at very small working distances and generally gives higher contrast between the phases. The in-lens detector has been originally developed for collection of the low-angle BSEs, and can be found only inside advanced instruments. Here, in the present disclosure, "low-angle BSEs" refers to back scattered electrons scattered in the direction of about 40 degree or less for incident direction of primary electrons.

[0048] The out-lens detector is also known as an Everhard-Thorney detector and a standard detector for most SEMs. The out-lens detector is typically situated below a pole piece, and its efficiency is strongly affected by a working distance. The longer working distance will generally result in a lower image quality (i.e. noisy image). There are several types of this detector. One type is an advanced type detector having individually operable concentric segments. The advanced type detector can collect the low angle BSEs even at relatively short working distances, and can prevent any parasitic long angle BSEs by the outer segment switch-off. Thus, the detection efficiency is high and the image quality is very good. This detector can also be operated as a simple conventional type (i.e. all segments are active) and detection of the low-angle BSEs can be acquired by the long working distance. The image quality, however, may decrease with an increase of the working distance. Another type is a conventional compact detector which does not have any segments. This type can detect the low angle BSEs when a long working distance such as larger than 10 mm is utilized. This secures significantly suppressed detection of the high angle BSEs. Although the contrast between phases is visible, the efficiency of the detector and the image quality are lower than those obtained by the in-lens detector and the advanced type out-lens detector.

[0049] (Step d: Characterization process)

Optionally, the method may further include the step of evaluating quantitative and/or qualitative characterization of the visualized austenite and martensite phases. For example, the micrographs acquired by the step c may be used as input date for quantitative image analysis. Also, the micrographs acquired by the step c may be transformed to binary image and statistical data for further characterization of the austenite phase. [0050] In this way, the austenite phase can be visualized as the darkest area, and the martensite phase can be visualized as the intermediate brightness area. The SEM technique of this embodiment has a very high spatial resolution with a low landing energy and can map a large surface area on the specimen in a short time. This enables to obtain statistically significant data and to better understand effects of the size, shape and morphology of the austenite phase on the final mechanical properties of bainitic steels such as TRIP steels.

Examples

[0051] The following describes examples of the present disclosure. A steel (0.12C, 0.02Si, 1.5Mn, (in mass%)) containing low fraction of martensite-austenite (MA) constituents was provided for a bainitic steel to be observed. A specimen of the bainitic steel was prepared according to the preparation technique described above. The electrolyte solution used for the examples was prepared by mixing 300ml of CH ₃ OH, 180 ml of 2-n-butoxyethanol (98%), and 30 ml of HCIO ₄ (60%). The electro-polishing process was conducted at the temperature of 298K at a voltage of 40 V for 2 seconds. After the electro-polishing process, the bias was switched off, and the bainitic steel base was continuously retained in the electrolyte solution for another 2-3 seconds. Afterwards, the specimen was withdrawn from the electrolyte and cleaned in methanol and subsequently in ethanol. A mean thickness of the corrosion product layer formed on the austenite phase of each example was 2 nm, and a mean thickness of the corrosion product layer formed on the martensite phase of each example was 1 nm. The micrographs were collected using two types of SEM. The BSE images were obtained by the dual-beam Helios Nano-lab 600i (FEI) using a concentric backscatter (CBS) detector located below the pole piece with a working distance of 7 mm. The landing energy of the primary electrons was 1 keV. By applying a negative bias on the surface of the specimen, it was possible to obtain a high signal at a low landing energy. The SE micrographs were obtained in the SEM LEO 1530 (Carl Zeiss), which was equipped with an in-lens and out-lens type of SE detector. The working distance between the specimen surface and the pole piece was 5 mm, and the landing energy of the primary electrons was 1 keV. [0052] Example 1

Bainitic steel specimens were prepared by a conventional electro-polishing technique and the technique according to the present disclosure. SEM images were obtained for the both specimens at identical parameter, i.e., at 1 keV landing energy of the primary electrons to secure the visibility of the contrast between the phases according to the present disclosure. Fig. 1 (a) shows the SEM image of a bainitic steel specimen prepared by a conventional electro-polishing technique. Fig. 1 (b) show the SEM image of a bainitic steel specimen prepared by the technique according the present disclosure. As can be seen in Fig. 1 (a), the specimen prepared by the conventional electro-polishing technique exhibits strong crystallographic contrast and it is impossible to differentiate the phases. In contrast, as can be seen in Fig. 1 (b), the SEM image of the specimen prepared by the technique according to the present disclosure enables to separately visualize both of the austenite phase and the martensite phase very clearly.

[0053] Figs. 2(a)-(c) show the effect of the various landing energies on the contrasts of the SEM image. Significant suppression of the crystallographic contrast in Fig. 2 (b) is caused by the presence of a thicker corrosion product layer as a consequence of additional etching in the electrolyte solution, as described above.

[0054] Example 2

SEM images were taken from the same specimen with varying magnitudes of the landing energy to evaluate an effect of the magnitudes of the landing energy on the contrast between the austenite phase and bainitic-ferrite matrix. The SEM images thus taken were shown in Figs. 2(a)-(c). It is apparent that the contrast between the austenite phase and bainitic-ferrite matrix is not clear at higher landing energies, and the austenite phase cannot be visually separated from the bainitic-ferrite matrix (Fig. 2 (c)).

[0055] Example 3

SEM images were taken from the same specimen with using different detectors. Figs. 3(a)-(c) show the contrasts between the austenite phases and the bainitic-ferrite matrix in the SEM images acquired by the SE detectors (Figs. 3(a) and (b)) and by the BSE detector (Fig. 3 (c)). It is appreciated from Figs. 3(a)-(c) that the BSE detector is the most optimal for visualization of MA constituents in bainitic-ferrite matrix. Although the ET detector of SE electrons provided some contrasts, the contrasts were weaker and not all of MA constituents were visualized. The in-lens SE detector did not provide clear contrast.

[0056] Example 4

The technique proposed in the disclosure was verified by means of a comparison with a conventionally used technique for phase identification in multi-phase steels, i.e. an EBSD. The EBSD technique enables to separate the austenite phase from martensite and bainite (ferite) due to its specific crystal lattice. The martensite phase is separated from the austenite and bainite (ferrite) by the "image quality" map obtained by the EBSD (martensite phase has the lowest parameter of image quality due to extraordinary high dislocation density). Thus, the EBSD technique enables to separate all phases, but is not optimal for large area mapping and has insufficient spatial resolution, as mentioned above. Figs. 4(a)-(c) show a result of a comparison between the EBSD and the technique according to the present disclosure. It is appreciated that observation of the bainitic steel surface prepared according to the disclosure by the SEM instrument operated according to the present disclosure enables to differentiate the austenite phase, the martensite phase and the bainitic-ferrite matrix.

[0057] Fig. 5(a) is an EBSD image quality map, Fig. 5(b) is a phase map, and Fig. 5(c) is a BSE image of the same area according to the present disclosure. It is apparent from these figures that the method according to the present disclosure offers higher spatial resolution and surface sensitivity in comparison with the conventional EBSD technique, which makes it extraordinary effective for characterization of the phases in fine MA constituents.

[0058] Example 5

The SEM images of the bainitic steel specimens were taken according to the present disclosure and were used as input data for image analysis to obtain quantitative information about the austenite and the martensite area fraction. The images were processed by a software capable of segmentation using thresholding. Fig. 6 shows a graph of the dependence of the austenite and martensite area fractions acquired by the technique according to the present disclosure (total investigated area was 2138 μηα ² for each specimen) on the annealing temperature. The obtained quantitative information was compared with measurements from a conventional X-ray diffraction (XRD) technique for austenite volume fraction. The austenite phase area fraction acquired according to the present disclosure is in good agreement with the XRD technique. Therefore, the SEM images having clear contrasts between the micrographs that are acquired according to the present disclosure are suited to quantitative image analysis without any image post-processing.

[0059] Example 6

As mentioned above, the SEM images acquired according to the present disclosure enable to separate the phases accurately and are highly suited to image analysis. The micrographs are easily transformed to binary image and statistical data about the phase shape can be obtained. Fig. 7 shows average size, circularity and Feret's diameter of the retained austenite and martensite phases in the specimens annealed at different temperatures (the total investigated area was 2138 μπι ² for each specimen).

[0060]Example 7

The electrolyte solution was prepared by mixing 10 volume % of acetyl acetone, 1 wt% of tetra-methyl-ammonium chloride, and methanol as solvent. The electro-polishing process was conducted at room temperature at a voltage of 100 mV and electric charge of 3 C for 1.5 minutes. After completion of the electro-polishing process, the specimen continuously retaining in the electrolyte solution for 2 seconds. Then, the specimen surface was observed in the same manner as described above. The obtained SEM BSE image using an in-lens type of the BSE detector is shown in Fig. 8. The darkest parts in the image represent the austenite phase as in the case of the previous examples. Both of the austenite phase and the martensite phase are separately visible in the image. Thus, it is proved that, in addition to corrosive acid, organic acids can be used as the electrolyte solution to selectively corrode the austenite phase.

Industrial Applicability

[0061] According to the present disclosure, it is possible to clearly and visually identify austenite and martensite phases in the bainitic steel individually at the nanoscale.