Texture Changes in Pass- Rolling of Steel Rods

Neutron-diffraction was used to study the texture development during pass-rolling of 
rods of an austenitic (f.c.c.) steel X8CrNiTi18.10 and a ferritic (b.c.c.) material 
X8CrTi17 in a round→oval→round→oval→round groove sequence. The orientation 
distributions occurring in the core of the pass-rolled rods can be described in a 
good approximation in terms of orthorhombic sample symmetry and qualitatively 
interpreted either as press textures on conditions of anisotropic material flow in the 
plane perpendicular to the press direction or as rolling textures on condition of large 
spreading. However, though they must be characterized by components {hkl}〈uvw〉 the pass-rolling textures are not rolling textures in the usual sense.


INTRODUCTION
Textures occurring in the pass-rolling of metallic materials were firstly studied with wires of circular cross-section by V. Vargha and Wassermann (1933), for instance, more than 50 years ago. From the contributions of these and other authors, which were summarized by Wassermann (1939) und Wassermann and Grewen (1962), one Now: Academy of Sciences of the GDR, Institute of Mechanics, P. O.B. 408,9010. could conclude that after pass-rolling in a ROUND/ROUND groove sequence the orientation changes within the central parts of a wire are similar to those occurring in wire drawing or forging. But already by v. Vargha and Wassermann (1933), it was also shown that the crystallite orientations near the surface cannot be described in terms of fibre textures. During the last two decades particularly the pass-rolling of steel rods and wires on hot-working conditions became more and more important. In this connection diverse types of groove geometry and pass sequences leading to materials with non-circular cross sections are used. The orientation changes associated with such more complicated procedures of pass-rolling have not been investigated systematically until now, but with regard to an exact explanation of the deformation-induced microstructures and changes in the properties of hot-rolled steel rods and wires their knowledge is necessary. The present paper informs on some basic features concerning the texture development in a four-step ROUND/OVAL groove sequence which was studied in an austenitic (f.c.c.) stainless steel X8CrNiTil8.10 and a highly-alloyed ferritic (b.c.c.) material X8CrTil7 after deformation at 973 K (cp. also Schubert et al., 1984;Schubert, 1985). Since the pass-rolling texture formed in a ROUND/OVAL groove sequence cannot be expected to be simple, the main purpose of the work was a proper determination of the sample symmetry and the identification of the most important components of the pass-rolling texture.

EXPERIMENTAL PROCEDURES
The chemical composition of the sample materials is given in Table   1.
For the texture investigations steel bars with initial round-section (diameter 12.3 mm) were used which had been obtained by hotrolling and subsequent annealing at 1273 K/30 min. The deformation in a four-step ROUND-OVAL-ROUND-OVAL-ROUND groove sequence was carried out at 973K with a strain of q9 0.4-0.6 per pass (final diameter 7.5 mm). The strain rate was t 1 s-1. After each rolling step a portion of the resulting wire was immediately quenched in water.
The pass-rolling textures were studied qualitatively by means of  (Matthies 1981(Matthies , 1982, for instance), which is important for a more refined interpretation of the pass-rolling texture, has not been performed until now.

RESULTS AND DISCUSSION
3.1 TEXTURE homogeneity and symmetry In a ROUND/OVAL groove sequence, the deformation geometry of which is illustrated in Figure 1, no rotational symmetry of the orientation distribution of the crystallites is obtained because the various elements of the cross-section are differently deformed and, moreover, the rod (wire) is rotated by :r/2 after each rolling step. However, according to Schubert et al. (1984) a sample-related axis system (1, z, 3) can be introduced, the directions 2 (transverse direction TD) and 3 (normal direction ND) of which are perpendicular to symmetry planes of the deformation process. For this reason the texture within the central parts of the rod (wire) can be supposed to be of orthorhombic sample symmetry. On the other hand, because of the temperature gradient between the core and the surface layers as well as the friction stresses occurring near the surface the pass-rolling texture must also be expected as very inhomogeneous over the cross-section of the rod (wire) especially on hot-working conditions.
In order to study the texture inhomogeneity, specimens were taken from the wires after the first (ROUND/OVAL) and the second (OVAL/ROUND) deformation step of the groove sequence, successively thinned parallel to the rolling plane defined by the directions ,a (rolling direction RD) and 2 (transverse direction TD) of the sample-related axis system ( Figure 1) and studied by means of X-ray diffraction. Results obtained in this way are shown in Figure 2 (Figure 3). This means, that it can be formally described in terms of components {hkl}(uvw) like fiat-rolling textures and quantitatively treated by means of the usual Bunge formalism of texture analysis for sheets (Bunge, 1982; Bunge and Esling, 1982, for instance). If the specimen includes an increasing part of the cross-section around the core of the wire, the symmetry of the observed texture is not significantly influenced, but the orientation distribution becomes weaker and the scattering of the texture components increases. In the measurements presented here the influence of the texture inhomogeneity was minimized by careful preparation of the specimens for neutron diffraction (cp. Schubert, 1985).   {001}(100) component of the initial texture is strongly reduced (Fig. 7).
--During the second (ROUND/OVAL) rolling step the deformation texture of the wire is significantly reduced (Figure 5b). Its mAfter the third and the fourth rolling step no remarkable changes of the deformation texture could be found yet (Figure 7). However, a re-increasing of the cube texture {001}(100) was observed now, which in agreement with the results of hotcompression tests at 973 K (Schubert, 1985) indicates increasing influence of dynamic (and eventually postdynamic) recrystallization on the austenite texture (cp. also Ahlborn et al., 1966).
If the components {hkl}(uvw} of the pass-rolling texture are compared with those of the fiat-rolling textures of hot-worked austenite (Goodman and Hu, 1970;Hu, 1974;Klimanek et al., 1981a, b, for instance;cp. also  the second one of which should be typical for recrystallization and is completely destroyed during the pass-rolling process. wAfter the first rolling pass a well-defined deformation texture is observed (Figure 8). Its dominating part in the Odf (Figure 9a) is an   (Figure 11) but important yet. mBecause of the fact that during hot-working at 973 K only dynamic recovery takes place in the steel X8CrTil7 (Schubert, 1985), the tube texture becomes stronger in the second (OVAL/ROUND) deformation step (Figures 9b, 11). The orientation densities of the fibre component {001} ( uvw ) are not changed in this case.
mDuring the deformation in the third and the fourth rolling pass the tube texture becomes weaker again and the {O01}(uvw) fibre remains practically independent of the rolling procedure. According to Figure 12  , . deformation step Figure 11 Orientation density changes of the main texture components occurring in pass-rolling of X8CrTil7 in a four-step round/oval groove sequence.
12. In the case of the b.c.c, material X8CrTi17 the correspondence between the pass-rolling and the fiat-rolling textures seems to be somewhat better. The dominating orienation tube is close to the so-called , fibre running from {111}(110) to {111}(112), and the so-called c fibre of the flat-rolling texture (Heckler and Granzow, 1970;Inagaki and Suda, 1972;Dirmann et al., 1984;Osterle, 1984, for instance;cp. also  --Although a low hot-working temperature was chosen, the pass-rolling textures of both sample materials are weak and a significant increase of the deformation components takes place only during the first rolling step.
--Although the pass-rolling textures have to be described in terms of orthorhombic sample symmetry they are clearly different from the textures observed in fiat-rolling.
The weak deformation dependence of the texture, which additionally is caused by recrystallization in the case of the austenitic steel X8CrNiTil8.10, can well be explained by the fact that a pass-rolled rod (or wire) is rotated by /2 after each deformation step. Since this procedure leads to an interchange of the transverse and the normal directions , 3 of the sample-related axis system ( Figure   1), it causes redistribution of the crystallite orientations during the subsequent rolling step with respect to the new rolling plane, and prevents a further increase of the orientation densities. In this connection the following geometrical interpretation of the observed orientation canges can be given (cp. Figure 13):  (110) and, particularly, the formation of a fibre {hkl}(llO) during the third and the fourth rolling step, the weakening of the ferrite texture is mainly the consequence of orientation redistribution. Therefore, if there is no influence of dynamical or postdynamic recrystallization, the pass-rolling textures of b.c.c, materials should be more intensive than in f.c.c, alloys. A phenomenological explanation of the meaning of the main texture components {hkl} (uvw) of the pass-rolling textures seems to be possible on the base of the following considerations: According to Figure 1 especially an OVAL/ROUND deformation step can be described as a compression along the normal direction 3, which is connected with very anisotropic material flow (elongation along the rolling direction 1) in the plane perpendicular to the compression axis. It is interesting, that the main texture components {011}(111) of the f.c.c, steel X8CrNiTil8.10 and {111} (110) of the b.c.c, material X8CrTil7 are obtained by combination of the preferred orientations occurring in compression and tension (Table 2).
Figure 1 also shows that, particularly in the ROUND/OVAL rolling pass, the deformation of the central part of a rod can be compared with flat-rolling on condition of large spreading. Indeed, investigating the textures of fiat-rolling iron sheets Schliifer and Bunge (1974) found that increasing spreading favoured the formation of the {lll}(uvw) orientation tube which was discussed above.
Finally it shall be mentioned yet that deformation textures very similar to those of the f.c.c, austenite X8CrNiTil8.10 were also obtained by pass-rolling of A1 and Cu in a groove sequence ROUND/OVAL at room temperature (Klimanek et al., 1985). This indicates that the results concerning the texture components as observed in the present work can be generalized. In order to obtain a physically sufficient explanation of the pass-rolling textures, however, further investigations are necessary.