Effect of structural elements on surface segregation in interstitial-free (IF) steels.
The segregation of impurity and alloying elements, formed during thermomechanical treatment, has a strong effect on the mobility of the grain boundaries and the formation of a texture in interstitial-free (IF) steels. The formation of the optimum texture is of critical importance for improving the deformation capacity of the IF-steels.
In our previous work  we investigated the kinetics of surface segregation in two rolled IF-steels microalloyed with titanium and with titanium and niobium. It has been shown that the main processes, taking place on the surface in the temperature range 550-750[degrees]C is the segregation of the atoms of S, C, P and the competition between these atoms for the area on the surface.
The conclusions published in  are based on the results of integral measurements of the segregation kinetics in which the average composition of the surfaces of approximately 1000 grains was measured in relation to the heating time at a constant temperature in the previously mentioned temperature range in situ in a superhigh vacuum chamber of an Auger microprobe. The averaging of the composition of the surface in many grains did not make it possible to estimate the possible contribution to the redistribution of the impurities of grain boundary diffusion, investigate the dependence of segregation on the orientation of grain and evolution of the excess phases, etc.
At the same time, an important element of the strategy of development of the IF-steels is the control of the composition and morphology of the excess phases [4-6]. The evolution of these phases during melting and thermomechanical treatment has a strong effect on the behaviour of impurities and, consequently, on the segregation processes. The aim of the present investigations is the examination of the possible effect of the elements of the structure of polycrystals on the kinetics of surface segregation in IF-steels.
2. Experimental materials and procedure
The investigations were carried out on specimens of two steels whose composition (Table 1) and structural condition (the condition after going) were identical with those investigated in greater detail in . The microstructure of the specimens was investigated by the method of high-resolution scanning electron microscopy (SEM) with energy-dispersing microanalysis (EDM). The kinetics of surface segregation was studied byAuger electron spectroscopy (AES) in the temperature range 550-750[degrees]C. The method of preparation of the specimens and the experimental procedure were on the whole identical with those used in , with one important exception. In , the probing beam was defocused so that it was possible to investigate the variation of the surface composition as a function of heating time of the area of approximately 0.16 [mm.sup.2], i.e., and information on the mean composition of approximately 1000 grains.
In this work, the segregation kinetics was investigated in individual points of the surface. For this purpose, the probe was focused into a beam with a diameter of approximately 30 nm so that it was possible to monitor the composition at a local point on the surface which was selected in advance using the scanning electron microscope, built into the Auger microprobe. Local measurements were directed to investigate the possible contribution to the surface segregation of structural defects, and also individual processes taking place in polycrystalline during heating, such as, for example, anisotropy of segregation , grain boundary diffusion of impurities , formation of two-dimensional phases .
The level of segregation was evaluated by measuring the ratio of the intensity ([r.sub.i] = [I.sub.i]/[I.sub.Fe]) of the peak of the element i([I.sub.i]), to the LMM peak of iron ([I.sub.Fe]), with the energy of 703 eV. Here and in the rest of the article, the element i refers to the elements found on the surface: S (LMM 151 eV), P (LMM 120 pV), Ti (LMM 418 eV), C (KLL 272 eV).
In addition to the quantitative analysis of the composition of the individual points on the surface of the specimen, the quantitative distribution of the impurities on the surface was also investigated. For this purpose, two-dimensional charts of the distribution of the elements were recorded during heating.
3. The results
3.1. Nonuniformity of segregation on the surface of individual grains
Local measurements of segregation were carried out on both the individual grains and four different points on the surface of a single grain, positioned at different distances from the area in which the grain boundaries extend to the surface. These measurements make it possible to evaluate the possible contribution of accelerated grain boundary diffusion to the transport of impurity atoms on the surface.
The experimental results show that the nature and rate of segregation for each individual grain in the investigated specimen is do not depend on the distance to the nearest boundary. This is illustrated by Fig. 1 which shows segregation at 560[degrees]C in the centre of the grain (Fig. 1a) and at a point situated on the line of intersection of the grain boundary with the surface (Fig. 1b). The distance between these two points is approximately 10 urn.
In both cases, the degree of segregation of carbon was very high, suppressing sulphur segregation. Approximately after 140-150 min after the start of heating the curve [r.sub.C] [(.sup.t1/2]) deviates from the parabolic law and the increase of the degree of segregation slows down. After heating four 230-240 min, the increase of the carbon peak is almost completely interrupted on the level [r.sub.C] = 1.75 and [r.sub.C] = 1.5 for the boundary and the centre of the grain, respectively. The kinetic curves in Fig. 1 are qualitatively identical with the curves recorded in  in integral measurements. Nevertheless, the level of carbon segregation, recorded in the local measurements, is considerably higher (almost double), and the level of sulphur segregation is lower than that indicated by the results of integral measurements. The slope of the segregation curves in the initial stage of segregation is almost completely independent of the distance to the grain boundary in the entire investigated temperature range.
[FIGURE 1 OMITTED]
In the detailed comparison of the kinetics of segregation to the surface of the individual grains at 550-580[degrees]C, in addition to the standard behaviour (Fig. 1) some grains show anomalous effects associated to some extent with the unexpectedly active behaviour of sulphur. In some cases (Fig. 2a) the anomalously high segregation of sulphur was detected on the surface already at the start of heating. In subsequent stages the sulphur concentration decreased and the carbon concentration increased. In other grains, the initially rapid segregation of carbon was unexpectedly replaced by segregation of sulphur after some heating time (Fig. 2b). These anomalies were detected in both steels.
3.2. Two-dimensional charts of the distribution of segregating elements
The two-dimensional charts of the distribution of sulphur and carbon on the surface of the specimen, recorded during heating at temperatures below 600[degrees]C, illustrate clearly the competition between these elements (Fig. 3). The images of the same section of the surface (Fig. 3a) in Auger electrodes of carbon (Fig. 3a) and sulphur (Fig. 3b) have the form of alternating fragments (spots) with one of the elements being dominant. In most cases, the surface fragment, enriched with carbon, is depleted in sulphur and vice versa. The intensity of the iron peak (Fig. 3b) is almost the same for the entire analysed surface indicating the uniform thickness of the produced segregation layer. The distribution and the dimensions of the spots of the preferential segregation of sulphur carbon usually do not correlate with the grain structure of the investigated section of the surface.
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
As shown previously in , an increase of temperature in the range 600-750[degrees]C results in the interaction of sulphur and phosphorus, and the carbon which appears on the surface in the initial minutes of heating rapidly disappears (the rate of this appearance increases with increasing temperature). The results of the Integra measurements show that at medium temperatures (600-650[degrees]C) the mutual effect of sulphur and phosphorus relatively small, and increase of the heating time is usually accompanied by the simultaneous increase of the concentration of both elements.
At higher temperatures, the interaction between sulphur and phosphorus becomes stronger and is of the competing nature. As the concentration of phosphorus at the given point increases, the carbon content decreases and vice versa. This is clearly indicated by the charts of distribution of sulphur and phosphorus at 700[degrees]C where the surface fragments, enriched with sulphur, alternate with the sections coated with phosphorus (Fig. 4). It should be mentioned that the distribution of the sections of preferential segregation of one of the elements is again almost completely independent of the type of grain structure.
The qualitative results of the distribution charts confirmed the results of measurement of the intensity of the sulphur and phosphorus peaks on the surface of more than 20 grains in the steels A and B after holding for 480 min at 700[degrees]C (Fig. 5). A strong negative correlation (correlation coefficient -904) between the values of [r.sub.S] and [r.sub.P] was detected on the surface of the individual grains.
The anomalies in the segregation of the impurities on the individual grains are also expressed in the strong variation of the segregation rate, primarily, of sulphur in transition from one grain to another. For example, three characteristic types of segregation behaviour, detected in different grains and the steels A and B, were found at 700[degrees]C. In the former case (Fig. 6a) sulphur is the dominant impurity on the surface from the very beginning. The phosphorus atoms also reached the surface. After heating for 80-100 min, the segregation curves of both elements reach saturation at [r.sub.S] [??] 1.4-1.5 and [r.sub.P] [??] 0.8. Subsequently, the concentrations start to decrease slowly.
In the second variant of the development of surface segregation (Fig. 6b), the phosphorus concentration increases preferentially from the very beginning of heating. After heating for approximately 110-120 min, the phosphorus concentration reaches the maximum value [r.sub.P] [??] 1.2-1.3. The phosphorus concentration then stabilises and even slightly drops. The increase of the phosphorus concentration on the surface was accompanied by the increase of the sulphur content. However, the rate of segregation of sulphur in the entire time range investigated is approximately 10 times lower than in the case shown in Fig. 6a.
In the third case (Fig. 6c) in the first 210-230 min the segregation distribution in almost completely identical with that shown in Fig. 6b. However, with further heating the sulphur segregation greatly accelerated. An inflection point is found on the sulphur segregation curve. This is accompanied by a decrease of the phosphorus content.
4.1. Nonuniformity of surface segregation
The local measurements in the individual points in the investigation of the surface segregation kinetics make it possible to estimate the contribution of the individual elements of the microstructure to the redistribution of the impurities.
It has been shown that the rate of segregation in the investigated steels is independent of the distance of the line of intersection of the grain boundary with the surface. In this situation in which the sections in the vicinity of the area in which the grain boundaries and reached the surface were enriched with impurities and a rate higher than the sections away from the boundaries, was detected previously in  for the segregation of Sb in the Fe-Ni-Sb alloys. The acceleration of segregation in the boundary region of the surface was associated with the accelerated diffusion of antimony at the grain boundaries.
[FIGURE 4 OMITTED]
The effect of acceleration of surface segregation was observed in  in the alloys in which the grain boundaries were enriched with antimony as a result of heat treatment. This effect was not found in these alloys if preliminary high-temperature annealing of the grain boundaries resulted in the removal of antimony from the grain boundaries. This result indicates that the grain boundaries in the investigated IF-steels have been purified to remove impurities, especially phosphorus. This conclusion is of considerable practical importance of optimising the rolling conditions, and also the conditions of winding sheets into coils, cold rolling, electroplating because the undesirable effect of the grain boundaries segregation of phosphorus has been detected many times previously in different stages of the thermomechanical treatment of the IF-steels [11-13].
[FIGURE 5 OMITTED]
Without evaluating the contribution of the structural elements to the redistribution of the impurities it is very difficult to explain the observed nonuniformity in the management of the individual sections of the surface (Fig. 3 and 4). The energy of binding of the impurity with the interface depends on the crystallography of the enriched surface  and this is referred to as the anisotropy of segregation and is the main reason for nonuniform enrichment of the individual interfaces in polycrystals. However, in the investigated case (Fig. 3 and 4), the expected correlation between the grain structure, on the one hand, and the dimensions and nature of distribution of the sections of the surface of different composition, on the other hand, has not been detected. This may be explained on the basis of the anisotropy assumptions.
Also, the reason for large variation of the rate of exit of the sulphur to the surface of the individual grains is not clear. Comparing the segregation curves in Fig. 6, in case 1 (Fig. 6a) the rate of segregation of sulphur is very high up to the level [r.sub.S] [??] 1.5. In case 2 (Fig. 6b) the sulphur concentration increases at a rate approximately 10 times lower. In the third case (Fig. 6b) sulphur initially shows the behaviour of type 2 and subsequently its segregation rate increases and approaches that in type 1. In contrast to sulphur, the rate of segregation of phosphorus in the initial in the section is far more stable. It is almost the same for the grains showing different segregation behaviour of sulphur.
4.2. Effect of excess phases
In order to understand the reasons for the observed anomalies in the behaviour of sulphur, it is necessary to consider the microstructure of the investigated materials. Both steels contain equiaxed ferrite grains with a diameter of approximately 15 um with a large number of relatively spherical precipitates. The dimensions and distribution of the inclusions, situated in the vicinity of the surface, were investigated using a high-resolution scanning electron microscope on the total area of approximately 40 000 [micro][m.sup.2] (Table 2). A total of 195 particles was found in steel A and 281 particles in steel B. The mean number of the particles in a single grain in steel B was approximately twice as high as in steel A (0.5 and 0.36, respectively). The mean size of these grains was 0.5 [micro]m and varied from 0.2 to 1 [micro]m. The final particles could not be investigated by this method. Particles larger than 1 [micro]m were not found. The majority of the particles were uniformly distributed over the grain area. Nevertheless, some of these particles formed agglomerates consisting of two or more (no more than six) particles, with the distance between the particles not exceeding 3-5 [micro]m.
[FIGURE 6 OMITTED]
The composition of the particles was investigated by the method of EDA and Auger electron spectroscopy with ion etching. According to the EDA data, the particles are based on titanium and sulphur. The measurements of the intensity of the [K.sub.[alpha]]-lines of sulphur and titanium were carried out for 30 particles (15 particles for each steel). The histograms of the distribution of the number of particles and the function of the ratio of the intensities of the peaks S/Ti in the specimens of both steels in the initial condition are shown in Fig. 7a. Usually, the S/Ti ratio varies in the range 1-2, and this corresponds to the variation of the composition between TiS and Ti[S.sub.2]. The ratio was lower than 0.5 only into particles (both particles were found in steel A).
According to Auger spectroscopy data, some of the particles contained carbon and the composition of these particles was close to [Ti.sub.4][S.sub.2][C.sub.2]. These particles, with a hexagonal structure and referred to as the h-phase, were detected previously in IF-steels [4, 5]. They formed during cooling from the austenite range to the ferrite range, when the titanium sulphide TiS particles, formed initially at a high temperature, transform to carbosulphides [Ti.sub.4][S.sub.2][C.sub.2], by absorption of carbon titanium. In addition, temperatures lower than 930[degrees]C result in the epitaxial growth of the carbide on the surface of the h-phase.
Comparative analysis of the particles in the specimens in the initial (rolled) condition, and also in the specimens after completing segregation measurements at 700[degrees]C and higher shows large changes in the composition of the precipitates in the subsurface layers. The majority of the particles lost sulphur. If in the initial condition the ratio of the intensity of the peaks S/Ti varied between 1 and 2 (Fig. 7a), then in the same specimens after heating at 700[degrees]C in the majority of cases S/Ti [less than or equal to] 0.3-0.4 (Fig. 7b).
It is evident that the measurement temperature results in at least partial dissolution of titanium sulphide and carbosulphides. The releaed sulphur increases the effective concentration of sulphur in the solid solution and greatly increases the rate of segregation of this element. It is highly likely that the formation of titanium on the surface is also associated with the partial dissolution of the titanium sulphides. Thus, the inclusion of the additional source of sulphur atoms in the segregation processes results in a large variation of the segregation rate, detected in the experiments. Evidently, the efficiency of this mechanism depends on the presence or absence of excess sulphides in the subsurface layer.
[FIGURE 7 OMITTED]
Phosphors was often detected on the surface of the particles remaining after partial dissolution at 700[degrees]C, in examination by Auger electron spectroscopy. Evidently, heating is accompanied by segregation of phosphorus to the ferrite-h-phase interface. The phenomenon of this type was detected previously . The presence of an additional sink for the phosphorus atoms weakens the surface segregation of this element.
The possible variants of the effect of dissolution of the subsurface sulphides on the rate of segregation of sulphur and phosphorus are shown schematically in Fig. 8 for 3 cases, identical with those observed in the experiments (Fig. 6). If the particle is in the immediate vicinity of the surface (variant 1 in Fig. 8), sulphur becomes the dominant impurity. In the opposite case (variant 2) when the distance between the particle and the surface is considerably greater than the characteristic diffusion path of sulphur [([D.sub.S]t).sup.1/2]), the presence of such a particle has almost no effect on the rate of segregation. Variant 3 in Fig. 8 is intermediate.
[FIGURE 8 OMITTED]
1. The composition of the surface of the individual grains, formed in the same conditions, may greatly differ. The main reason for this is the observed variation in the rate of segregation of sulphur in transition from grain to grain. The competing interaction between sulphur and other impurities in turn influences the surface segregation of these impurities, for example, carbon and phosphorus.
2. It has been shown that the observed differences in the behaviour of sulphur in the individual grains depend on the presence in the surface layers of fine titanium sulphide (carbosulphide) particles. These precipitates partially dissolved in long-term heating at temperatures above 700[degrees]C. The atoms, released during dissolution, enhance the segregation flows of the sulphur and titanium patterns to the surface and contribute to the displacement of phosphorus from the surface.
[1.] S. Q. Cao, J. X. Zhang, J. S. Wu et al., Mater. Sci. Eng. A, 392, No. 1-2, 203 (2005).
[2.] Z. D. Wang, Y. H. Guo, D. Q. Sun et al., Mater. Charact., 57, No. 4-5, 402 (2006).
[3.] A. V. Krainikov and V. V. Shchigolev, , Metallofiz. Noveishie Tekhnol., 31, No. 2, 1669 (2009).
[4.] H. Matsuta, T. Sato, and M. Oku, ISIJ Intern., 36, S125 (1996).
[5.] M. Hua, C. I. Garcia, and A. J. DeArdo, Metall. Mater. Trans. A, 28, No. 9, 1769 (1997).
[6.] A. Ghosh and S. Chatterjee, Mater. Charact., 55, No. 4-5, 298 (2005).
[7.] A. V. Krajnikov, P. Lejcek, M. Militzer, and J. Adamek, Surf. Sci., 417, 337 (1998).
[8.] M. Militzer, Yu. N. Ivashchenko, A. V. Krajnikov et al., Surf. Sci., 261, 267 (1992).
[9.] A. V. Krajnikov, P. Lejcek, M. Militzer, and J. Adamek, Modern Aspects of Analyt. Chem. (Eds. H.-M. Kuss and U. Telgheder) (Aachen, Mahnheim Verlag, 1997).
[10.] A. V. Krajnikov, M. Militzer, and J. Wieting, Mater. Sci. Tech-nol., 40, No. 11, 877 (1997).
[11.] W. A. Spitzig and R. J. Sober, Metall. Trans. A, 8, 1585 (1977).
[12.] W. A. Spitzig and R. J. Sober, Metall. Trans. A, 8, 651 (1977).
[13.] S. Luo, B. Liu, and D. Zhong, Modern LC and ULC Sheet Steels for Cold Forming, Processing and Properties (Aachen, Verlag Mainz, 1998).
[14.] P. Lejcek and S. Hofmann, Crit. Rev. Sol. State Mater. Sci., 33, 133 (2008).
[15.] J. Rege, M. Hua, C. I. Garcia, and A. J. DeArdo, ISIJ Intern., 40, 191 (2000).
A.V. Krainikov and V.V. Shchigolev *
I.N. Frantsevich Institute of Materials Science, National Academy of Sciences of Ukraine, ul. Krzhizhanovskogo 3, 03680, GSP, Kiev-142, Ukraine
* Donetsk Judicial Institute, ul. Zasyad'ko 13, 83054 Donetsk, Ukraine
Table 1. Chemical composition of the steels (in at.%) Steel C N MN P S Si Ti Nb A 0.002 0.004 0.108 0.01 0.007 0.01 0.069 0.005 B 0.003 0.004 0.194 0.01 0.01 0.01 0.037 0.035 Steel Al Sn A 0.04 0.003 B 0.04 0.003 Table 2. Characteristics of the titanium sulphide particles distributed in the subsurface layers of investigated steels Steel Total number Number of large Number of of particles particles particles in (0.7-1 [micro]m) agglomerates A 195 18 37 B 281 17 57 Steel Mean number of particles per grains A 0.36 B 0.52
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||METALLIC SURFACES AND FILMS|
|Author:||Krainikov, A.V.; Shchigolev, V.V.|
|Publication:||Physics of Metals and Advanced Technologies|
|Date:||Jan 1, 2010|
|Previous Article:||Structure of [Fe.sub.82][Si.sub.2][Fe.sub.6] and [Fe.sub.82][Si.sub.4][B.sub.14] alloys in the amorphous and monocrystalline conditions.|
|Next Article:||Strength, fatigue resistance and the evaluation of cyclic cracking resistance of new deformable aluminium alloys, alloyed with transition and rare...|