The document discusses optimizing the cooling process for high carbon wire rods to favor formation of oxides that can be easily removed by mechanical descaling. Laboratory tests were conducted using a Gleeble simulator to study oxide formation at high temperatures under different cooling rates and laying head temperatures. The results from the simulations will be used to optimize conditions in an industrial wire rod mill to produce wire rods with oxides that are less adherent and easier to remove.
The document discusses optimizing the cooling process for high carbon wire rods to favor formation of oxides that can be easily removed by mechanical descaling. Laboratory tests were conducted using a Gleeble simulator to study oxide formation at high temperatures under different cooling rates and laying head temperatures. The results from the simulations will be used to optimize conditions in an industrial wire rod mill to produce wire rods with oxides that are less adherent and easier to remove.
The document discusses optimizing the cooling process for high carbon wire rods to favor formation of oxides that can be easily removed by mechanical descaling. Laboratory tests were conducted using a Gleeble simulator to study oxide formation at high temperatures under different cooling rates and laying head temperatures. The results from the simulations will be used to optimize conditions in an industrial wire rod mill to produce wire rods with oxides that are less adherent and easier to remove.
The document discusses optimizing the cooling process for high carbon wire rods to favor formation of oxides that can be easily removed by mechanical descaling. Laboratory tests were conducted using a Gleeble simulator to study oxide formation at high temperatures under different cooling rates and laying head temperatures. The results from the simulations will be used to optimize conditions in an industrial wire rod mill to produce wire rods with oxides that are less adherent and easier to remove.
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s u r f c o a t
Study on formation of “easy to remove oxide scale” during mechanical descaling of
high carbon wire rods A. Chattopadhyay a,⁎, P. Kumar b, D. Roy a a Research and Development Tata Steel, Jamshedpur, 831007, India b Indian Institute of Technology, Kharagpur, 721302, India
1. Introduction present at the surface of high carbon wire rod surface. As a result of which of which die life gets affected during subsequent process. Oxide scales, which are deposited on the surface of the steel wire The present paper aims at optimizing the process conditions rods during the process of wire rod rolling need to be removed before during the cooling of wire rods by simulated laboratory tests to favor any further processing like wire drawing. Initially the de-scaling the formation of easily removable oxide scale during mechanical process was used by the pickling process using hydrochloric acid. But, descaling of high carbon wire rods. Although the transformation of nowadays due to the stringency of laws favoring environmental oxide under forced air cooling in simulated lab condition is known [1], protection, the use of HCl is being discouraged. A lot of emphasis is the literature on industrial process is very limited [10,11]. therefore being laid for employment of alternate methods of descaling. One such process used is mechanical descaling. Other 2. Experimental procedure modern methods are also employed today for mechanical descaling purpose in wire rod mills like reverse bending procedure. [1,2]. In order to determine the chemical composition of the adherent It becomes essential for the mechanical descaled wire rods to have oxides after mechanical descaling a sample wire rod (whose scale free surfaces to be able to prevent wear and abrasion of dies composition is given in Table 1) from wire rod mill of Tata Steel was during the wire drawing operations. Subsequently, in order to meet taken. During mechanical descaling most of the oxides get removed. this requirement, engineering the development of desired scale These oxides are collected separately as powder which is analyzed structures becomes a necessity. The laying head temperature (LHT) later on. In order to study the high temperature oxide formation on and cooling rate at the Stelmor conveyor significantly influence the the high carbon wire rods, the samples obtained from the plant were formation of different oxide phases in the scale namely, wüstite (FeO), cut into pieces of 5.5 cm length having 8 mm diameter. Subsequently magnetite (Fe3O4) and hematite (Fe2O3) as shown in Fig. 1 [1–5]. The the samples were pickled in 15% HCl solution with 0.1% inhibitor latter two types of oxide are not desirable for both mechanical (polyamine based) to remove all the scales on the sample. descaling and pickling point of view [6–9]. Though the mechanical The oxide scale formation was simulated using Gleeble-1500. The descaling process is mainly designed for low carbon wire rods, yet for different thermal profile maintained at Gleeble-1500 is given in its application in high carbon wire rod detailed study on the oxide Table 2. Initially the samples were heated to 1200 °C and hold for formation at high temperature for high carbon wire rods is required. austenitising (cooling rate not important) and then cooled to 1000 °C Mechanical descaling process is unable to remove all the oxides in 20 s (rate not important). During heating process the Ar atmosphere was maintained inside the simulator so that oxidation during heating could be prevented. After reaching 1200 °C, air having ⁎ Corresponding author. Tel.: +91 9204058840. 80–85% relative humidity is introduced into the chamber and Ar gas E-mail address: [email protected] (A. Chattopadhyay). was withdrawn simultaneously. After that, cooling from 1000° to 860°
Table 2 Details of thermal profiles maintained in Gleeble.
Laying head temperature Different cooling rates Corresponding Stelmor end
(LHT)(°C) used (°C/s) temperature (°C) 880 7.86 600 8.02 575 8.24 550 860 7.56 600 Fig. 1. Macrophotograph of mechanical descaled wire rod showing adherent oxides. 7.70 575 7.88 550 840 7.07 600 (precise rate 140 °C/s) is carried out. Finally, heat transfer model 7.16 575 7.27 550 predicted cooling rate is maintained and it is kept limited up to the 820 6.92 600 end of Stelmor cooling i.e. up to 600–550 °C, followed by conventional 7.02 575 forced air cooling. The Stelmor end temperature is restricted up to 7.11 550 550 °C because, as per the thermodynamic condition the possibility of hematite formation is higher below 550 °C and the growth rate and phase transformation of oxide scale are negligible. Prior to the thermal Fig. 3 depicts the typical thermal profile predicted by the heat simulation in Gleeble, scale free rods were welded with two transfer model used in wire rod mill following which Gleeble thermocouples made of chromel–alumel. Samples were heated and simulation is carried out. After the Gleeble simulation the oxide cooled with the given thermal profile as predicted by heat transfer composition and their relative amount is measured by semi model used in the wire rod mill. quantitative reitveld method using GAXRD and the result is shown The oxide scales were formed on the surface of the wire rod in Fig. 4 a and b. The result of faster cooling rate is plotted in Fig. 4a. samples are analyzed by using Raman Spectroscopic Analysis [12]. Laser Raman spectrometry was carried out on different samples with a Renishaw Raman System RM 1000B coupled with a Leica DMLM microscope. Excitation of the sample was achieved with a 20 mW diode Laser source of 785 nm wavelengths and filtering of the Rayleigh scattering was achieved through an edge filter at 200 cm− 1 cut off that allows study of Raman scattering only on the Stokes side of the spectrum. Power at incidence on the liquid surface was about 8 mW with laser spot size of about 1.5 µm. For identification of the Raman peaks obtained from Raman Spectroscopy study standard database available is used [13]. The oxide layer is further analyzed by Glancing angle X-Ray Diffraction (GAXRD) using Cu Kα. Phase quantification is done by using semi quantitative reitveld method. The step size [°2Th.] maintained in the X-Ray diffractometer is 0.1000 and scan step time [s] is 1.0000. Scan type is continuous in nature and divergence slit size [°] is 6.2899. In Reitveld method, each measured data point is considered as an independent information parameter and least square fitting is done on the whole pattern by high score plus software. Next, refinement on crystal structure related information i.e. unit cell parameters and atomic co-ordinates is carried out. Beside this refinement due to instrument related parameters like peak profiles, zero shift, sample height etc are also prepared.
3. Results and discussion
Wire rod sample (Fig. 1) taken just after mechanical descaling has shown some adherent oxide scales at the surface. The adherent oxides and oxides collected from below the mechanical descaler are both analyzed under Raman Spectroscope and results are shown in Fig. 2a and b respectively. Fig. 2 a shows that adherent oxide is hematite (αFe2O3 and γ Fe2O3) in nature whereas Fig. 2 b shows that the oxides powders collected from below the mechanical descaler are wustite (FeO) and magnetite (Fe3O4) type of oxides. Later types of oxides are non-adherent in nature. Formation of FeO and Fe3O4 type of oxides are favorable for high carbon wire rods which pass through mechanical descaler route.
Table 1 Chemical composition (wt.%) of the wire rod used for the experiments.
C Mn Si P S Cu Ni Cr Al 0.81 0.8 0.3 0.015 0.01 0.05 0.05 0.05 0.05 Fig. 2. Raman spectroscopy results (a) adherent oxide scale (b) powder scale collected from mechanical descaler. 2914 A. Chattopadhyay et al. / Surface & Coatings Technology 203 (2009) 2912–2915
From Fig. 4a it is evident that at lower LHT, the amount of hematite has increased significantly. As per the oxide stability diagram [9] it is found that, only magnetite and wustite is stable at high temperature. However, since during the cooling process, equilibrium diagram is not maintained properly the hematite formation also takes place. But at higher LHT it is observed that the formation of hematite is significantly low. The variation of magnetite at different cooling rate at different LHT is shown in Fig. 4 b where each data point corresponds to a particular LHT followed by particular cooling rate as described in Table 2. From Fig. 4 b it can be observed that at faster cooling rate the amount of magnetite is almost always higher compare to other two cooling rates. At high temperature as wustite and magnetite scales are stable and dominant. As the cooling process progresses, the tendency of these oxides to decompose to form hematite starts. If cooling rate is maintained high then the time for decomposition of high temperature scales to form hematite is very low and therefore, higher amount of magnetite is observed. Whereas, for slower cooling rate chance of hematite formation is higher and as a result of which lower amount of magnetite is observed. Phase transformation in oxide scale depends on various factors like availability of oxygen, cooling rate etc. As the complete process takes place in normal air condition therefore the availability of oxygen is not a critical factor to be considered. In this
Fig. 4. Quantitative analysis obtained from semi reitveld method (a) relationship between % Fe2O3 with LHT for high cooling rate (b) relationship between % Fe3O4 with LHT for various cooling rates.
type of condition, only cooling rate plays important role in determin-
ing the oxide scale composition.
4. Conclusions
So from the above made discussion following conclusions can be
derived:
1. Fe2O3 is responsible for adherent oxide formation. This type of
oxide causes problem during mechanical descaling process. 2. During Stelmor cooling faster cooling is preferable because at higher cooling rates the amount of adherent oxides formed is less. Fe2O3 forms as a decomposition of higher temperature oxides like FeO and Fe3O4. Because of faster cooling the time needed for decomposition is less as a consequence of which less Fe2O3 forms. 3. Laying head temperature should be kept near 880 °C because at this temperature the oxide scales contain less of Fe2O3. At higher temperature the chance of decomposition of high temperature oxide is less and subsequently less Fe2O3 forms. Based on the above observation one plant trail had been conducted and the result of plant trail confirms the above observations. The Fig. 3. Thermal profile maintained in Gleeble simulation to simulate LHT and stelmor amount of Fe2O3 has come down significantly with higher LHT and condition (a) LHT 860 °C and stelmor 550 °C (b) LHT 820 °C and stelmor 550 °C. faster cooling rate. A. Chattopadhyay et al. / Surface & Coatings Technology 203 (2009) 2912–2915 2915
References [7] W.L. Roberts, Hot Rolling of Steel, Marcel Dekker, Inc., New York, 1983, p. 644. [8] Peter Gillstrom, J. Mater. Process. Technol. 172 (2006) 332. [1] J. Tominaga, K. Wakimoto, T. Mori, M. Murakami, T. Yoshimura, Trans. ISIJ 22 [9] R. Bhattacharya, G. Jha, S. Kundu, R. Shankar, N. Gope, Surf. Coat. Technol. 201 (1982) 646. (2006) 526. [2] M. Krzyzanowski, W. Yang, C.M. Sellars, J.H. Beynon, Mater. Sci.Technol. 19 (2003) [10] R.Y. Chen, W.Y.D. Yuen, Oxid. Met. 53 (2000) 539. 109. [11] R.Y. Chen, W.Y.D. Yuen, Oxid. Met. 56 (2001) 89. [3] D.P. Burke, R.L. Higginson, Scr. Mater. 42 (2000) 277. [12] A. Chattopadhyay, N. Bandyopadhyay, A.K. Das, M.K. Panigrahi, Scr. Mater. 52 [4] S. Birosca, D. Dingley, R.L. Higginson, J. Microsc. 213 (2004) 235. (2005) 211. [5] J. Pickens, Proc. of the 25th Mechanical Working and Steel Processing Conf. 1983, [13] University of Parma, Italy, http://www.fis.unipr.it/phevix/ramandb.html, Date vol. 21, Iron and Steel Society, Philadelphia, 1983, p. 39. Viewed: 15th December, 2008. [6] ASM Metals Reference Book, American Society for Metals, Metal Park, Ohio, (1981), 62.
