Influence of Electron Beam Irradiation on Surface Roughness of Commercially AISI 5140 Steel
2017 Volume 58 Issue 11 Pages 1519-1523
Details
2017 Volume 58 Issue 11 Pages 1519-1523
In this study, a commercially AISI 5140 steel with two different levels of initial surface state (initial high-roughness (IHR) and initial low-roughness (ILR)) was surface-treated by large-area pulsed electron beam irradiation (LPEBI). Surface morphology in 2D and 3D of the two types of samples after LPEBI treatment was characterized. The results show that the surface roughness of IHR samples decreases clearly with increasing LPEBI pulse numbers, while an opposite trend is found in the ILR samples. It is considered that the final surface roughness is influenced by surface remelting and formation of crater-like structures (CLSs) in local regions. For the IHR samples with amounts of mechanical scratches, the remelting plays the leading role, owing to the high surface energy which provides extra driving force for remelting during LPEBI. In contrast, such extra driving force is less due to the relatively flat surface of the ILR samples, instead remelting the surface suffers from the formation of CLSs in localized region. Compared to the formation of localized CLSs, the surface remelting is beneficial to surface roughness of the LPEBI processed samples.
Many product properties are affected by their surface structure and roughness, such as fatigue strength, wear and corrosion resistance. Smooth surface is usually beneficial to the improvement of mechanical properties1–3). Thus, traditionally, polishing process is chosen as the final step in manufacture of mechanical parts and molds4). However, for products with complicated shapes, the automation of polishing process remains difficult, while manual operation methods are generally used, which is inefficient and costly. High efficient polishing method should be developed for industrial production and application. Intense-pulsed energetic beams such as electron, ion and laser beams have attracted much attention over the past few decades. Among these intense-pulsed beams, large-area pulsed electron beam irradiation (LPEBI) is considered as a relatively new surface polishing and modification method5–7).
During LPEBI treatment, the surface of the polycrystalline materials could be re-melted under irradiation of the high energy density electron beam even without focusing. Due to the high energy density and localized interaction, both heating and cooling of the surface layer are very fast compared to traditional surface treatment. Such non-equilibrium process can easily change microstructure, chemical composition, and stress state of the surface layer8,9). Usually, a thin re-solidified layer can be produced on the surface after LPEBI treatment which is beneficial to the improvement of surface hardness and wear resistance10,11). Moreover, it has also found that surface roughness of the materials could be smoothed due to the surface tension of the melted material during LPEBI treatment12). Okada et al.7,11,14) analyzed the structure of re-solidified layer formed on the electron-beam-polished mold steel and discussed the effect of beam conditions on the surface characteristics. Kim et al.5,15) improved surface quality of AISI 304 metal plates by the LPEBI polishing method. Gao1,2,16) investigated the surface topography, microstructure and nanohardness distribution of pure titanium and titanium alloys after electron beam irradiation. Murray et al.4,17) pointed out that optimizing acceleration voltages and pulse numbers of the electron beam irradiation could improve surface finish and repair cracks induced by electrostatic discharge machining process. Therefore, LPEBI is considered as a multifunctional surface treatment technology.
However, the effect of LPEBI treatment on surface roughness with different initial surface situation is not still reported in open literatures. Therefore, a commercially AISI 5140 steel with two different levels of initial surface state (initial high-roughness (IHR) and initial low-roughness (ILR)) was surface-treated by LPEBI. The surface morphology before and after LPEBI treatment was characterized in detail. The influence of initial surface state on polishing effect of LPEBI was investigated.
A commercial AISI 5140 steel (Fe-0.40C-0.8Cr-0.30Ni-0.23-Si-0.7Mn in mass%) was chosen as the start material. The start materials was normalized at 850℃ for 120 minutes and then cut into samples with dimensions of 15 × 15 × 5 mm. Then, two groups of samples with different degree of initial surface roughness were prepared by grinding. Initial high-roughness (IHR) samples were prepared by using #80 silicon carbide papers, while initial low-roughness (ILR) samples were done by using #2000 ones. Hand-held grinding was conducted on a small desktop belt machine (Nisen, China) with a speed of 25 m/min. The belt grinding time is about 10 seconds. The average of surface roughness (Ra) of the IHR samples and ILR samples were determined as 1.4 µm and 0.1 µm respectively.
LPEBI treatment was performed on a RITM-2M type LPEBI machine (produced by Institute of High Current Electronics, Siberian Branch of Russian Academy of Sciences, Russia) with a beam diameter of 60 mm in the present work. The treatment parameters were as follows: accelerating voltage = 30 kV; energy density = 9.5 J/cm2; number of pulses = 3–45; pulse duration = 5 μs; pulse frequency = 0.2 Hz. All the samples were cleaned in acetone followed by air-drying before LPEBI.
Surface profiles were measured by a surface roughness meter (TR200, produced by TIME, China). Each type of samples was tested three times with a transverse length of 3.2 mm perpendicular to the grinding scratches. The 2D and 3D surface topographies of the samples were characterized by optical microscope (OM), scanning electron microscope (SEM, JEOL JSM-6460LV and Zeiss Sigma HD) and atomic force microscope (AFM), respectively. Both secondary electron imaging (SEI) and back-scattered electron imaging (BSEI) in SEM were performed for characterizing the steel surface morphology. More detail advantages of BSEI technique can be found in the Refs. 18, 19).
Figure 1 displays the surface morphology of the ILR sample before and after 3- pulse LPEBI treatment. As can be seen, there are many mechanical grinding scratches (MGSs) located on the surface before LPEBI (Fig. 1(a)). After LPEBI, a modified layer with about 3.0 μm (average value) in thickness can be clearly observed from the sectional view (Fig. 1(b)). Under the modified layer, it still shows the hypoeutectoid structure mixed by ferrite nodule and pearlite colony, as shown in Fig. 1(b). Figures 1(c) and 1(d) are SEI and BSEI images respectively, it can be seen that the MGSs is less observed in the sample surface after LPEBI treatment. Instead, there are a lot of crater-like structures (CLSs) which has evident eruption feature with a deep center-dimpling shape and a small bottom hole.
SEM micrographs showing surface morphology of ILR sample before and after 3 pulses LPEBI treatment: (a) before LPEBI; (b) sectional view after LPEBI; (c) CLSs after LPEBI; (d) BSEI image showing CLSs. The white arrows in (b) indicate the location of modified surface after LPEBI.
Figure 2 shows OM images of various samples before and after LPEBI treatment. Figures 2(a)–(d) are untreated, 3-pulse, 13-pulse and 45-pulse irradiation treated samples of IHR, respectively, while Figs. 2(A)–(D) are the corresponding ones of ILR samples. MGSs are clearly observed from both untreated IHR (see white arrows in Fig. 2(a)) and untreated ILR samples (see white arrows in Fig. 2(A)), the difference is that the scratches in the former are much coarser and deeper. From Fig. 2(b), it can be seen that after 3-pulse irradiation, the number of the coarse and deep scratches starts to fall, but some CLSs with diameter of about 9.0 μm are newly formed and increase with pulse number (see Fig. 2(c)). The density of CLSs in the 13-pulse sample is higher than that in the 3-pulse sample. With further increase of pulse numbers, the area density of CLSs tends to decline. But the size of CLSs increases continuously with more pulses. When the LPEBI pulse number increases to 45 pulses, as shown in Fig. 2(d), both the scratches and CLSs decrease which results in increasing of the bright region filed in the OM images, corresponding to a relatively flat surface. However, a few of deeper scratches are still present (Fig. 2(d)). For the ILR samples, after 13 pulses LPEBI treatment, the craters are clearly observed and increase with increasing LPEBI pulse number. Meanwhile, the initial scratches almost disappear after 13-pulse irradiation (see Fig. 2(C)). Comparing Fig. 2(D) to Fig. 2(d), the diameter of CLSs in the ILR sample is only about 6.4 μm which is much smaller than that of IHR sample.
Optical micrographs showing surface morphology evolution of IHR (a) to (d) and ILR (A) to (D) samples under LPEBI treatment. (a) and (A) are untreated samples, (b) and (B), (c) and (C), and (d) and (D) are 3-pulse, 13-pulse and 45-pulse treated samples respectively. White and dark arrows mean scratches and craters respectively.
Surface roughness (Ra) plotted as a function of the number of LPEBI pulses is shown in Fig. 3. As can be seen, the average value of the surface roughness of the IHR sample is measured as 1.4 µm before LPEBI treatment, while such value is only 0.1 µm in the ILR sample. The surface roughness of the IHR sample decreases with increasing LPEBI pulse numbers, and it dramatically decreases into a minimum value of about 0.8 µm after 35 pulses irradiation treatment. For the further LPEBI treatment, the surface roughness seems to reach a saturated state with increasing irradiation pulse numbers. In contrast, the average value of the surface roughness of the ILR samples shows a different evolution trend with increasing irradiation pulse numbers. The original value of the ILR sample is about 0.1 µm before LPEBI treatment, and then it increases dramatically into 0.5 µm after 13 pulses irradiation treatment. For the further LPEBI treatment, the average value of the surface roughness increases slightly with increasing irradiation pulse numbers. The value of the ILR sample after 45 pulses is about 0.6 µm.
Surface roughness (Ra) plotted as a function of LPEBI pulse numbers.
Surface profiles along transverse direction perpendicular to the original MGSs traces were measured by a surface roughness meter. The results of the IHR and ILR samples before and after 45-pulse LPEBI treatment are shown in Fig. 4. Figures 4(a) and (b) are surface profiles of the IHR sample before and after LPEBI respectively, while Figs. 4(c) and (d) are corresponding ones of the ILR sample. The comparative analysis of before and after LPEBI treatment shows that the pulsed electron beam irradiation can change the surface profiles of the steels. More importantly, such change is related to the original surface roughness of the steel. For the IHR sample, the rough surface can be modified into a relatively flat one after LPEBI, while for the ILR sample, the surface become rougher after LPEBI.
Surface profiles of IHR and ILR samples before and after 45-pulse LPEBI treatment: (a) IHR sample before LPEBI; (b) IHR sample after LPEBI; (c) ILR sample before LPEBI; (d) ILR sample after LPEBI.
Figure 5 displays the AFM images of the IHR and ILR samples before and after 45 pulses irradiation. Figures 5(a) and (b) show the 3D surface topographies of the IHR sample before and after LPEBI respectively, while Figs. 5(c) and (d) present the 3D surface topographies of the ILR sample. Comparing Fig. 5(a) with 5(c), the MGSs in the IHR sample are wider and deeper than that of ILR sample. After LPEBI treatment, due to the characterization of MGSs being not obvious (see Fig. 5(d)), it indicates that most MGSs are eliminated after LPEBI in the ILR samples. As shown in Fig. 5(b), there are many noises existing in the IHR samples, which should be the remained MGSs after LPEBI treatment (see Fig. 2(d)) and could be reduced by increasing the LPEBI pulse numbers further.
AFM images of IHR and ILR samples before and after 45-pulse LPEBI treatment: (a) IHR sample before LPEBI; (b) IHR sample after LPEBI; (c) ILR sample before LPEBI; (d) ILR sample after LPEBI.
The microstructural changing layer is related to the interaction between the electron beam and sample surface. Such layer has been investigated and divided into successive regions: (i) a melted and rapidly solidified layer on the top surface (fully melted and partially melted layer), (ii) a heat affected zone where solid state phase transformation, deformation and recrystallization may occur (～10 μm) and (iii) a stress wave affected zone (～100 μm)13,20). Obviously, the final surface morphology is closely related to the first region.
To understand the remelting behavior of the surface layer under LPEBI, temperature change in different depth versus time is shown in Fig. 6. The temperature was calculated by a method of thermal-mechanical coupling finite element analysis (Anlysis10.0 software, for more information see Ref. 21)). As can be seen, the calculated maximum surface temperature is about 1960℃ after one pulse irradiation in this study. Since the melting temperature of iron is only 1535℃, the steel surface is most probably melted by LPEBI in this study. In LPEBI treatment, the pulse duration is only a few microseconds. Therefore, the heat conduction along the depth direction is considered to be very small, and only a small amount of heat energy can penetrate into inner layers of the samples. Moreover, because the pulse frequency is 0.2 Hz and the interval between LPEBI is about 5.0 s, which means that the temperature of the sample surface should cool down in some extent. Therefore, it is considered that the thickness of the remelting layer is almost constant regardless of how many pulse numbers of irradiation. As shown in Fig. 1(b), the metallographic change layer has a thickness of about 3 µm in this study, which indicates that fully remelting and solidifying have taken place on the sample surface during LPEBI treatment. However, when the energy density is sufficient high, not only remelting but also evaporation could occur7).
Temperature change in different depth versus time after a pulse treatment of LPEBI.
Due to the existence of impurities, the melting point may be lower in some local regions, which could cause the formation of CLSs combining with second phase particles and various structural defects1,6). It is well known that CLSs on the metal surface after LPEBI treatment is the result of local sublayer melting and eruption through the solid outer surface. The formation of CLSs has been observed in many metals after LPEBI treatment, such as 304L steel15), D2 steel8,10), 316L steel22), Ti alloys13,20). On one hand, the remelting of the surface layer induced by the subsequent pulses will cure the surface by filling the previously formed CLSs23). On the other hand, the new eruption will enlarge the size of CLSs when there are still some impurities, second phase particles or other defects existing underneath the former CLSs.
Obviously, both remelting behavior and formation of MGSs can vary the surface roughness of the sample. Actually, it has been reported that the final surface roughness is mainly influenced by the two aspects during the LPEBI treatment2). On one hand, rough surface can be modified by surface remelting because the molten metal can flow from peak to billabong and reduce the initial MGSs or the new-born CLSs. On the other hand, the rough surface can become rougher due to the formation of CLSs.
In this study, owing to the same procedure and parameter of LPEBI treatment, the variation in surface morphology is considered to be linked to the initial state of the steel samples. It is known that the localized crystal symmetry is destroyed and the lattice become unstable during heating, which results in melting. It is believed that the crystal symmetry of the surface grain is incomplete, which results in that such surface grains have lost their stability before LPEBI treatment. In the IHR samples, there are many MGSs with deep and wide shape, corresponding to high instability or high surface energy. As mentioned above, the steel surface is mainly subjected to remelting during LPEBI treatment. It is considered that the surface energy will provide extra driving force for remelting behavior in addition to the heating from LPEBI treatment. Unlike IHR samples, the ILR samples have relatively flat surface (low level of localized crystal symmetry instability or low surface energy) before LPEBI treatment, which means that the extra driving force for remelting is less compared to that of the IHR samples. Instead, the formation of CLSs in localized regions plays the key role during LPEBI treatment in the ILR samples. Obviously, compared to the formation of localized CLSs, the surface remelting is beneficial to surface roughness of the LPEBI processed samples, because the low-lying areas could be filled up and ridge regions could be levelled through remelting of the sample surface. It is considered that the surface of products or machinery parts with initial high surface roughness could be polished by LPEBI treatment.
In summary, the surface morphology in 2D and 3D of AISI 5140 steel treated by large-area pulsed electron beam irradiation (LPEBI) has been studied. The results show that the quality of polishing of LPEBI is closely related to the initial state of the sample surface. For steel samples with initial high surface roughness could be well polished by LPEBI treatment due to its high surface energy providing extra driving force for remelting. With increasing LPEBI pulse numbers the final surface roughness decreases clearly. For samples with initial low surface roughness, it is not suitable to recommend LPEBI as polishing process.
This study was supported by the National Natural Science Foundation of China (51575073, 51501158 and 51275548), International Cooperation Special Project in Science and Technology of China (2015DFR70480), and Scientific and Technological Research Program of Chongqing (cstc2014gjhz70003 and cstc2017jcyjBX0031).
