Journal Information
Vol. 8. Issue 5.
Pages 4151-4162 (September - October 2019)
Download PDF
More article options
Vol. 8. Issue 5.
Pages 4151-4162 (September - October 2019)
Original Article
DOI: 10.1016/j.jmrt.2019.07.024
Open Access
The effect of grain size and initial texture on microstructure, texture, and formability of Nb stabilized ferritic stainless steel manufactured by two-step cold rolling
Daniella Gomes Rodriguesa,
Corresponding author

Corresponding author.
, Cláudio Moreira de Alcântarab, Tarcísio Reis de Oliveirab, Berenice Mendonça Gonzaleza,*
a Department of Metallurgical and Materials Engineering, Universidade Federal de Minas Gerais, Av. Antônio Carlos, 6627, Pampulha, Belo Horizonte, Minas Gerais, Brazil
b Aperam South America Research Center, Praça 1° de Maio, 09, Centro, Timóteo, Minas Gerais, Brazil
This item has received

Under a Creative Commons license
Article information
Full Text
Download PDF
Figures (11)
Show moreShow less
Tables (3)
Table 1. Processing details and the corresponding sample codes.
Table 2. Stored energy after cold rolling.
Table 3. Grain size (GS), texture, and anisotropy coefficient.
Show moreShow less

The present study investigated the influence of grain size and initial texture on the microstructure, texture, and formability of a niobium-stabilized ferritic stainless steel processed by two-step cold rolling. The characterization of the samples was performed by X-ray diffraction, electron backscatter diffraction, and tensile tests in order to evaluate the formability by the average normal anisotropy coefficient. The results showed that the reduction in grain size from 110 ± 21 to 50 ± 7 μm, in the hot rolled band, was favorable to obtain a more homogenous microstructure and smaller grain size after annealing. Coarse grain size and α-fiber (<110>||RD) favored a strong α-fiber after cold rolling. In addition, γ-fiber (<111>||ND) was more intense in the deformation texture of fine grain size samples. The recrystallization texture was constituted by intense γ-fiber (<111>||ND) and weak θ-fiber (<100>||ND) for both conditions. However, the reduction in grain size increased the γ-fiber fraction. Therefore, the r¯=1.92 and γ/θ = 8.12 were obtained to the sample originated from the initial fine grain size. The fine grain size and weak texture in the starting materials were effective in developing a high r¯ value in the samples manufactured by two-step cold rolling.

Ferritic stainless steel
Grain size
Two-step cold rolling
Full Text

Ferritic stainless steels (FSSs) are alloys that mainly consist of Cr and Fe. These alloys have been applied in kitchen cookware, sinks, automobile parts, heaters, and equipment for nitric acid plants due to a combination of corrosion resistance, mechanical properties, and surface quality [1–3]. Niobium is an option to reduce sensitization effects and to stabilize the ferrite phase [4].

Many studies have been conducted to improve the formability of FSSs through the application of several different methods. Accordingly, controlling possible crystallographic orientations enhances the formability. This property can be evaluated by the average normal anisotropy coefficient ( r¯ value). A high r¯ value is associated with an intense fraction of γ-fiber (<111>||ND) and a weak θ-fiber (<100>||ND) after annealing [5,6]. Huh and Engler [7] suggested to perform intermediate annealing during cold rolling. Zhang et al. [8] reported an increase of 25% in the r¯ value through the enhancement of in-grain shear bands formed in the hot-rolled band. Gao et al. [9] reported that warm rolling favored the formation of a nearly uniform γ-fiber recrystallization texture, which significantly improved the steel formability. Yazama et al. [5] pointed out that the refinement of a hot-rolled steel microstructure is one way to control the formation of the γ-fiber.

The initial characteristics influence the formation of microstructure and texture during the thermomechanical process. Schlippenbach [10] reported an increase in α-fiber (<110>||RD) and no change in γ-fiber fraction after 80% deformation, when the grain size was increased from 22 to 150 μm in a low carbon steel. However, Lee et al. [11] reported that an increase from 150 to 500 μm in grain size before cold rolling process led to a slight difference in the intensity of the rolling texture after 87% thickness reduction of a non-oriented electrical steel sheet.

Thus, the present study investigated the effect of grain size and initial texture on the microstructure, texture, and formability of a Nb-stabilized ferritic stainless steel manufactured by the two-step cold rolling process.

2Materials and methods

In this study, fine and coarse grain hot-rolled samples of a ferritic stainless steel containing 0.35% of Nb were supplied by Aperam South America. The chemical composition of the steel under investigation is 0.021% C, 0.024% N, 16.0% Cr, 0.35% Nb, 0.30% Si, 0.22% Ni, and 0.20% Mn. These samples were manufactured in an industrial line until being annealed at 970 °C after hot rolling.

As-received samples, with a thickness of 6.0 mm, were cold rolled in a laboratory rolling mill up to 3.0 mm, corresponding to a 50% reduction in thickness. Following this first reduction, the samples were annealed at 880 °C, with a heating rate of 1.8 °C/s and 24 s of soaking time. Subsequently, the samples were cold rolled up to 0.6 mm, corresponding to an 80% reduction in thickness. Finally, the samples were also annealed at 880 °C, with a heating rate of 23.5 °C/s and 24 s of soaking time. The coarse and initial fine grain size samples, with different processing steps, were recorded as hot-rolled samples (CHR and FHR, respectively). The sample codes are listed in Table 1.

Table 1.

Processing details and the corresponding sample codes.

Sample code  Processing details 
CHR  Coarse initial grain size hot-rolled band 
FHR  Fine initial grain size hot-rolled band 
C-50%CR  Coarse initial grain size after the 50% cold rolling 
F-50%CR  Fine initial grain size after the 50% cold rolling 
C-50%CRA  Coarse initial grain size after the 50% cold rolling and annealing 
F-50%CRA  Fine initial grain size after the 50% cold rolling and annealing 
C-80%CR  Coarse initial grain size after the second cold rolling 
F-80%CR  Fine initial grain size after the second cold rolling 
C-80%CRA  Coarse initial grain size after the second cold rolling and annealing 
F-80%CRA  Fine initial grain size after the second cold rolling and annealing 

The samples were characterized using a Leica DMRM light optical microscopy, X-ray diffraction (XRD) measurement, and electron backscatter diffraction (EBSD). Microstructural characterization and crystallographic texture were carried out in longitudinal sections using EBSD. The samples for microstructural characterization were prepared in a standard manner. For EBSD, the final polishing was performed with a colloidal silica suspension. The samples for X-ray diffraction measurement were mechanically polished and etched with a solution of HNO3 (12%) and HF (3%) at ∼55 °C. The grain size was measured using the linear intercept method, in accordance with the standards established by ASTM E112/13 [12].

The crystallographic textures, determined through X-ray diffraction measurement, were done using a Philips X’Pert PRO MPD texture diffractometer. From the incomplete pole figures of (110), (200), and (211), the orientation distribution functions (ODF) were calculated by the series expansion method, according to Bunge (lmax = 22) [13]. The EBSD scans were performed in a Philips XL-30 SEM with a LaB6 filament and operated at 20 kV. The range of step size varied from 4 to 8 μm. The EBSD data were processed using a TSL/EDAX Analysis™ software, considering a confidence index (CI) of >0.095. The clean up procedure applied in the raw data was performed using neighbor CI correlation.

The stored energy after deformation was calculated according to the hardness results, using Eq. (1)[14], where M = 3.05 is the Taylor factor, a = 0.3 is a constant, G = 80 GPa is the average shear modulus for ferritic stainless steel [15], and b = 2.48 × 10−10  m is the Burgers vector [14]. Vickers microhardness was measured using a Leitz–Wetzlar microhardness tester. The average values were defined from ten measurements with a dwell time of 10 s and a load of 100 gf.

The r¯ and Δr values were determined through tensile tests in order to evaluate the effect of the initial grain size and texture on formability. Specimens were cut from the final annealed sheets at the angles of 0°, 45°, and 90° respecting the rolling direction (RD). The specimens with a gauge length of 50 mm and a width of 20 mm were strained until 15%. The r¯ and Δr values were calculated according to the ASTM E517 standard using the Eq. (2) and Eq.(3)[16].

3Results3.1Microstructures throughout the process

The microstructure developed throughout the thermomechanical process of Nb ferritic stainless steel is shown in Fig. 1, where ND is the normal direction and RD is the rolling direction. From Fig. 1a and b, it was possible to observe the difference in the initial grain size, with CHR and FHR presenting average grain sizes of 110 ± 21 µm and 50 ± 7 µm, respectively. After 50% reduction (Fig. 1c and d), the microstructures were formed by the joining of flattened and elongated grains along the rolling direction. In addition, in-grain shear bands were observed in some grains with an inclination of approximately 45° with respect to RD.

Fig. 1.

Evolution of microstructure throughout the process. a) CHR; b) FHR; c) C-50%CR; d) F-50%CR; e) C-50%CRA; f) F-50%CRA; g) C-80%CR ; h) F-80%CR; i) C-80%CRA; j) F-80%CRA.


After the first annealing process (Fig. 1e and f), the samples were fully recrystallized with a more homogenous grain size distribution in the F-50%CRA sample (Fig. 1f). In the C-50%CRA (Fig. 1e), the microstructure was formed by fine (red arrow) and coarse (blue arrow) grain clusters. The C-50%CRA and F-50%CRA average grain sizes were 28 ± 2 µm and 22 ± 2 µm, respectively. Following the second cold rolling (Fig. 1g and h), the microstructure was formed by the joining of more flattened and elongated grains along the rolling direction. After the final annealing (Fig. 1i and j), both samples were fully recrystallized, presenting a similar microstructure. The cluster formation appeared after the first annealing (Fig. 1e), but it disappeared after the second annealing in the C-80%CRA sample (Fig. 1i). After the second annealing, the average grain sizes of C-80%CRA and F-80%CRA samples were 13 ± 1 µm and 10 ± 1 µm, respectively.

The grain size distributions after annealing are shown in Fig. 2. Both distributions showed similar tendencies. The samples originated from coarse grains showed a larger proportional area with the coarse grains. By contrast, fine grain size was more present in the F-50%CRA and F-80%CRA samples.

Fig. 2.

Grain size distribution of the annealed samples. a) After the first annealing. b) After the second annealing.


The density of dislocations and the stored energy after the first and second cold rolling are shown in Table 2. As can be observed, after the thickness reduction, the density of dislocations and the stored energy of the samples coming from FHR proved to be larger than those from CHR.

Table 2.

Stored energy after cold rolling.

Sample  Hardness (HV)  Ρ (m−2) * E14  Ed (MJ/m3
C-50%CR  248 ± 5  3.19  0.78 
F-50%CR  269 ± 3  4.41  1.07 
C-80%CR  293 ± 7  6.85  1.66 
F-80%CR  299 ± 4  6.97  1.69 
3.2Texture throughout the process3.2.1Starting materials

Fig. 3 shows the inverse pole figures (IPF) through-thickness and the orientation distribution function (ODF) at ϕ2 = 45° section of the CHR and FHR samples. The black boundaries in IPF indicate misorientations of ω > 15° between adjacent grains. The orientation maps, represented in Fig. 3a and b, show that the texture is different from the surface and central layers in both samples. Up to approximately ¼ of the thickness, the texture is characterized by the orientation {110} <uvw> , whereas in the center region, it is characterized by {100} <uvw> and {112} <uvw> , in both samples. It can be observed from </uvw> </uvw> </uvw> ϕ2 = 45° ODF (Fig. 3c and d) that the global texture was characterized by α-fiber with a high intensity between the 001110 and 112110 components and a 110001 Goss component in both samples. However, the α-fiber in the CHR sample was more intense than in the FHR sample up to φ ∼ 50°. The maximum of f(g) ∼ 4.8 in the CHR sample was observed for the orientation 001560 at (5°, 0°, 45°). In the FHR sample, the maximum of f(g) ∼ 2.3 was observed for the shifted orientations of α-fiber ideal.

Fig. 3.

Microstructure and texture of the CHR and FHR samples. a) CHR IPF; b) FHR IPF; c) ϕ2 = 45° ODF CHR; d) ϕ2 = 45° ODF FHR.

3.2.2Sample after 50% cold rolling and annealing

The texture obtained from XRD measurement on the surface and central region after 50% cold rolling is shown in Fig. 4. From Fig. 4a and b, it can be observed that, in the C-50%CR sample, the texture in the central region was stronger than that on the surface. In the central region, the texture was dominated by an intense α-fiber, with a peak at 001110 (f(g) ∼ 17.9), and a weak γ-fiber. On the surface, the α-fiber peak occurred at 112110 (f(g) ∼ 7.7), while the γ-fiber was slightly more intense than in the central region. Similarly, in the F-50%CR sample, the α-fiber in the central region was more intense than on the surface, with a peak at 001110 (f(g) ∼ 9.6). On the other hand, the γ-fiber was more developed in the F-50%CR sample than in the C-50%CR sample, especially in the 111112 components.

Fig. 4.

Texture from XRD measurement at different regions after a 50% reduction. a) C-50%CR surface; b) C-50%CR center; c) F-50%CR surface; d) F-50%CR center.


Fig. 5 shows the texture determined by EBSD after 50% reduction and annealing of C-50%CRA and F-50%CRA. In the inverse pole figure map (Fig. 5a and b), the texture is heterogeneous across the entire thickness, and the 011uvw grains are distributed throughout it. From the C-50%CRA sample (Fig. 5a), various fine grains adjacent to coarse grains with random orientation can be observed, as indicated by the black arrows. By contrast, in the F-50%CRA sample (Fig. 5b), grain size and texture distributions were more uniform throughout the thickness. The global texture showed a remarkable difference in recrystallization texture, as illustrated in Fig. 5c and d. In the C-50%CRA sample (Fig. 5c), it can be noted that the α-fiber and Goss component were developed with a maximum of f(g) ∼ 1.7 in the 011001 Goss component. On the other hand, in the F-50%CRA sample (Fig. 5d), the γ-fiber and Goss component proved to be more developed. A maximum of f(g) ∼ 2.0 was observed for the orientation 554225 at (90°, 60°, 45°).

Fig. 5.

Inverse pole figure map after 50% cold rolling and annealing. a) C-50%CRA; b) F-50%CRA; c) ϕ2 = 45 °C-50%CRA; d) ϕ2 = 45° ODF F-50%CRA.

3.2.3Sample after 80% cold rolling and annealing

Fig. 6 shows the deformation textures determined by XRD measurement on the surface and central regions after 80% reduction. The C-80%CR sample (Fig. 6a and b) presented a stronger α-fiber in the central region with a high intensity of {001}<110> at (0°, 0°, 45°) and {223}<110> at (0°, 40°, 45°). The γ-fiber was slightly more intense in the surface region with a peak of {111}<011> at (60°, 54.7°, 45°). In the F-80%CR sample (Fig. 6c and d), the α-fiber and γ-fiber were well developed in both regions. The peak of α-fiber occurred at {223}<011> at (0°, 40°, 45°), and for the γ-fiber, at {111}<011> at (60°, 54.7°, 45°).

Fig. 6.

Texture from XRD measurement at different regions after second cold rolling. a) C-80%CR surface; b) C-80%CR center; c) F-80%CR surface; d) F-80%CR center.


The recrystallized texture, determined by EBSD, is shown in Fig. 7 after 80% reduction and annealing. Fig. 7a and b shows the inverse pole figure maps of the C-80%CRA and F-80%CRA samples, respectively. The texture is heterogeneous throughout the thickness of both samples, but it is more heterogeneous in the C-80%CRA sample (Fig. 7a) than in the F-80%CRA one (Fig.7b). The texture gradient can be expressed by a uniformity factor (UF) obtained from the EBSD data. The value of UF = 0 describes a perfect homogeneous distribution, while UF = 1 describes a heterogeneous distribution. The C-80%CRA sample (Fig. 7a) displayed a UF = 0.032, whereas the F-80%CRA sample (Fig. 7b) showed a UF = 0.024, both of which obtained through the thickness. At ϕ2 = 45° section (Fig. 7c and d), the γ-fiber is not completely uniform in both samples. However, its intensity in the F-80%CRA sample proved to be stronger, with a peak at 111121 at (30°, 54.7°, 45°). The maximum of f(g) was observed at 554225 at (90°, 60°, 45°) (Fig.7d). Moreover, in the C-80%CRA sample (Fig. 7c), it could be observed that some α-fiber components remained after the annealing process.

Fig. 7.

Inverse pole figure map after the second cold rolling and annealing. a) C-80%CRA; b) F-80%CRA; c) ϕ2 = 45° ODF C-80%CRA; d) ϕ2 = 45° ODF F-80%CRA.


Fig. 8 shows the results of grain boundary distribution (mesotexture) after the final annealing. The mesotexture differentiaties the misorientations existent between grains, and the measurements are sensitive to the step size, especially in the subgrains [17]. Therefore, subgrain structure does not appear when larger step sizes are used. It can be noted from Fig. 8 that the C-80%CRA sample showed 45% of low angle grain boundaries (LAGBs), in which 30% corresponded to an angle comprehended between 2° < ω < 5°. By contrast, the F-80%CRA sample showed a high fraction of high angle grain boundaries (HAGBs).

Fig. 8.

Grain boundary distribution after the second annealing.


An enlarged area is shown in Fig. 9 to closely examine the LAGBs of the C-80%CRA sample. The LAGBs with 2° < ω < 5° are indicated by white lines. This figure shows the IPF with the projections of normal (Fig. 9a) and rolling (Fig. 9b) directions. Both projections were performed to better visualize the misorientations inside the grains. It can be noted that the LAGBs appear mainly in γ-fiber grains. Furthermore, the LAGBs are present between two regions with the same plane, but different direction. The other observations include LAGBs inside grains with the same plane and direction.

Fig. 9.

Enlarged area of the C-80%CRA sample. a) Projection of the normal direction. b) Projection of the rolling direction.


The misorientation inside the grains was evaluated by grain orientation spread (GOS). Fig. 10 exhibits the GOS distribution for the C-80%CRA and F-80%CRA samples. The GOS distribution was very different in both samples. In the F-80%CRA sample, most of the grains showed a GOS < 1°. By contrast, the C-80%CRA sample displayed a GOS up to 4° and few areas with GOS < 1°. The average GOS values then obtained were 1.40° and 0.32° for C-80%CRA and F-80%CRA, respectively.

Fig. 10.

Grain orientation spread distribution after the second annealing.

3.3Texture, average normal and planar anisotropy coefficient

Fig. 11 shows the relationship between r¯ values, Δr values, γ-fiber, and θ-fiber for the final annealed samples. The F-80%CRA exhibited a r¯ value notably higher than the one found in the C-80%CRA sample, while the Δr value was approximately the same in both samples. Furthermore, the F-80%CRA sample showed a predominance of γ-fiber. It is important to note that both samples displayed a similar θ-fiber fraction. Hence, the highest γ/θ ratio was obtained for the F-80%CRA, as indicated in Fig. 11. The grain size (GS), γ and θ-fiber volume fractions, important orientations of γ-fiber, and normal (r¯ value) and planar (Δr) anisotropy coefficients are presented in Table 3.

Fig. 11.

Correlation between r¯ value and texture after the second annealing.

Table 3.

Grain size (GS), texture, and anisotropy coefficient.

Code  GS (µm)  γ-fiber (%)  θ-fiber (%)  {111}<121> (%)  {111}<110> (%)  r¯ value  Δr value 
C-80%CRA  13 ± 1  52.8  8.7  26.7  14.4  1.52  0.57 
F-80%CRA  10 ± 1  59.3  7.3  31.1  16.6  1.92  0.58 

The microstructure formed after the cold rolling process can be significantly affected by the initial features before deformation. From the deformed microstructure (Fig. 1c, d, g, and h), it could be observed that the spacing between the strained grains was lower in the samples from FHR than in those from CHR after cold rolling. In both microstructures, the boundaries tend to arrange themselves almost parallel to the cold rolling axis. As a result, pancake-shaped grains are formed, which are typical after cold rolling, leading to an increase in the boundary area [18]. The in-grain shear bands, presented in some grains, were frequently formed in γ-fiber grains [19,20]. This deformation heterogeneity produces microstructural fragmentation and strain located within the grain [21]. Campos et al. [22] reported evidence of strong strain inhomogeneities in larger grain size (500 μm). According to the authors, some grains showed a larger number of in-grain shear bands, while others had none.

The influence of grain size before cold rolling on the stored energy of deformation can be seen in the results projected in Table 2. The stored energy showed a difference mainly after the first cold rolling. During deformation, the grain boundaries make the slip processes that occur in the grains difficult. The reduction of the previous grain size increases the grain boundary constraints and the uniformity of deformation [23]. According to Humphreys [18], the stored energy tends to increase with a decrease in grain size resulted from low and medium deformations. Furthermore, the stored energy associated with this increase in grain boundary area represents an important part of the stored energy, which will be higher for fine grains and large strains [18].

The recrystallized grain size and uniformity of the microstructure can be linked to the stored energy after cold rolling. The grain size depends on the nucleation and growth rates [23]. The literature demonstrates that high stored energy increases the nucleation rate more than the growth rate [23]. According to the experimental results shown in Table 2, it could be observed that the stored energy after the first cold rolling was larger in the F-50%CR sample than in the C-50%CR one. Thus, the nucleation rate’s probability was higher and more uniform in the matrix for the F-50%CRA sample. This condition is consistent with the microstructural results shown in Fig. 1e and f.

Landgraf et al. [24] reported the formation of grain clusters in silicon steels, which showed the same orientation after recrystallization. In their study, the sample with an initial grain size of 500 μm was cold rolled until reaching a deformation of ε = 2.1. According to the authors, clusters are formed by the nucleation of various nuclei in the same deformed grain. The recrystallization in the large deformed grains occurs independently in each grain.

The difference among the stored energies was not accentuated after the second cold rolling. The recrystallized microstructure showed a difference in the distribution and average grain size (Fig. 1i and j). Furthermore, a difference was also observed in the distribution of grain boundary fractions and in the misorientation within the grains. The C-80%CRA sample revealed a fraction of 45% of LAGBs after the second annealing, while the F-80%CRA exhibited only 17.5% (see Fig. 8). Fig. 9 demonstrated that this high LAGBs fraction occurred due to the nucleation of adjacent grains with very close orientation. This suggests that the distribution of nuclei during the onset of recrystallization was different in the samples. In other words, in the F-80%CRA sample, the distribution of grains during recrystallization was possibly more random than the one that occurred in the C-80%CRA sample. The high fraction of LAGBs led to the enhancement of misorientation in the C-80%CRA sample’s grains.

From the results of deformation texture after cold rolling, it was clear that the development of deformation texture was influenced by the initial features. As shown in Figs. 4 and 6, after both cold rolling processes, the α-fiber was significantly more intense in the CHR sample, reaching a peak at 001110 component. By contrast, the development of γ-fiber was slightly more intense in FHR sample. An important detail was the reduction of α-fiber and the increase of γ-fiber after the second cold rolling process in the CHR sample. It is worth noting that the difference in average grain size before deformation was approximately 60 μm and 6 μm for the first and second cold rolling processes, respectively. Different initial grain size results in different constraints of grain boundaries to the plastic deformation; hence, it can influence to the grains rotation. The rotation during deformation occurs more easily in the coarse-grained sample. According to the literature, the coarse grains have a tendency to deform heterogeneously through deformation banding, while for the fine grains, the deformation is more homogenous, and the constraints of grain boundary are much stronger [18,25]. Therefore, the development of deformation texture shows a close relation to the mode deformation of each grain.

Lee et al. [11] studied the effect of grain size in a non-oriented electrical steel sheet, using two samples with initial grain sizes of 150 μm and 500 μm, with an almost random initial texture. The samples were cold rolled with 87% reduction and annealed at 700 °C, 800 °C, and 900 °C for 5 min. According to the authors, the cold-rolled textures consisted of α-fiber and γ-fiber with similar intensities; therefore, the initial grain size did not influenced the deformation texture. In contrast, Schlippenbach et al. [10] showed an increase of α-fiber and a slight γ-fiber fraction when the initial grain size was enhanced from 22 to 150 μm in a low carbon steel.

According to Inagaki [25], the rolling texture would not be largely influenced by the grain size when it is smaller than 50 μm, due to a few changes in the deformation mode. The influence of grain size consists of the fact that the deformation mode of each grain changes when the grain size is approximately 65 μm. Hence, the effect of grain size is not simple. Moreover, the effect of the initial texture should also be considered in the development of the deformation texture.

From the initial texture presented in Fig. 3, it was observed that, after the first deformation, the {011} <uvw> disappeared, and the α-fiber and γ-fiber increased ( </uvw>Fig. 4). During the cold rolling process, the 011uvw  fiber is unstable under plain strain deformation, and it tends to rotate toward the α-fiber and γ-fiber, favoring the sharpening of both fibers in the cold-rolled sample [7,26–28], as shown in Fig. 4. After the second cold rolling, the intensity of α-fiber in the C-80%CR sample (Fig. 6a and b) was notably stronger than that in the F-80%CR sample (Fig. 6c and d). The presence of α-fiber in the previous sample (Fig. 5c), as well as after the cold rolling process, intensified its intensity in the center region (Fig. 6b). Another contribution is that the grains with random orientation had a tendency to rotate to the α-fiber and γ-fiber, mainly to the {111}<011> component at (60°, 54.7°, 45°) with 80% reduction [7]. On the other hand, the presence of the {111}<112> component in Fig. 5d contributed to the high intensity of the γ-fiber after cold rolling (Fig. 6c and d). The grains {111}<112> in polycrystals are unstable and tend to rotate toward {111}<110> [25].

After the first annealing, the recrystallization texture in the samples was very different. In the C-50%CRA sample (Fig. 5a), the α-fiber and the  011001 Goss component were developed. The α-fiber present after annealing (Fig. 5c) could be retained from the cold-rolled sample (Fig. 4b). During cold rolling, the α-fiber grains stored low energy [7,18,29,30]. Hence, during annealing, the recrystallization rate developed in these grains is low, so this orientation may be retained or consumed by other orientations in the recrystallization process. By contrast, in the F-50%CRA sample (Fig. 5b), the γ-fiber and 554225 component were dominant. The γ-fiber fraction tends to increase with the decreasing in the grain size of the hot-rolled band sample, due to the lower probability of a shear band development during cold rolling [5,29]. Furthermore, the formation of γ-fiber can be attributed to the grain boundary constraints [25]. Additionally, the 111112 recrystallized grains, developed in Fig. 5b, nucleate in the 111011 deformed grains or in boundary grains in ferritic stainless steel [9].

The γ-fiber was strengthened mainly in the F-80%CRA after the second annealing, as shown in Fig. 7. According to the literature, the recrystallization texture depends on the cold-rolled texture and microstructure [31]. The development of the strong γ-fiber, with a peak at 111112 (Fig. 7d), can be attributed to the high intensity of the γ-fiber developed after the second cold rolling process (Fig. 6c and d), mainly in the 111110 component. As mentioned above, the 111112 recrystallized grains are formed by preferential nucleation at 111110 deformed grains [9]. Furthermore, the nucleation of γ-fiber grain occurs adjacent to the previous high angle grain boundaries. The combination of fine grain, γ-fiber, and weak α-fiber (Fig. 5b), before cold rolling, was favorable to the development of the stronger γ-fiber and proved to be more homogenous throughout the thickness (Fig. 7b).

Wang et al. [32] reported that the strong γ-fiber in the deformation texture is favorable in order to obtain an ideal γ-fiber after annealing, and with the combination of a strong α-fiber and a weak γ-fiber, a heterogeneous γ-fiber is thereby obtained. On the other hand, in the C-80%CRA sample (Fig. 6a), the γ-fiber showed a slight displacement of the ideal γ-fiber, and some α-fiber components remained after the second annealing. The α-fiber components between  001011 and 112011 have low stored energy, hence, low recrystallization rate during annealing [7,33]. Therefore, these components tend to recover and prevail in the steel or can be consumed during recrystallization. From Fig. 7c, it can be observed that some α-fiber orientations were retained from the starting materials (Fig. 3b) throughout the thermomechanical process.

Lee et al. [34] reported that the occurrence of plastic anisotropy in steel is correlated to the crystallographic texture. Regarding the r¯ value exhibited in Fig. 11, it was observed that the r¯ value = 1.92 was obtained for the F-80%CRA sample, with 59% of γ-fiber and 7% of θ-fiber. The planar anisotropy was low and positive for both samples. According to Du et al. [35], high r¯ value is obtained for materials which possess a strong γ-fiber and a weak θ-fiber. Hence, the γ/θ ratio is a good parameter to evaluate the deep drawability of the steel. The γ-fiber grains have a high r¯ value and a small planar anisotropy, according to the results calculated using the relaxed constraint method of crystal plasticity [29]. For a good deep drawability, a high r¯ value, a minimum  Δr, and the 111110 and 111112 components are favorable [29,34].

Gao et al. [9] cited that the presence of non-uniform γ-fiber in the recrystallization texture can deteriorate the formability of ferritic stainless steels. They reported that the warm rolling process improves the uniform γ-fiber recrystallization texture, thus enhancing its formability in the final sheet. Therefore, the r¯ value result obtained in the F-80%CRA sample combines a strong γ-fiber, mainly 111112, and a weak θ-fiber.


The grain size and initial texture exert a direct effect on the development of microstructure, texture, and formability of a ferritic stainless steel manufactured by two-step cold rolling. The main results can be summarized as follows:

  • i

    The present study demonstrated that the difference in the initial grain size affected both the final microstructure and the recrystallized grain size after annealing. The influence was more pronounced after the first cold rolling process and annealing. The initial coarse grain size produced a heterogeneous microstructure formed by fine grain cluster adjacent to coarse grain cluster. On the other hand, the initial fine grain size formed a homogenous microstructure. These results point to differences in nucleation rate due to the effect of the previous grain size before the cold rolling process.

  • ii

    The deformation texture was mainly constituted by a strong α-fiber and a weak γ-fiber after both cold rolling process. However, there was a remarkable difference in the intensity of the α-fiber between the samples. The CHR samples showed an α-fiber more intense than FHR ones. This difference was associated with the initial texture present in the hot-rolled band sample.

  • iii

    The γ-fiber was predominant in the recrystallization texture of both samples after the second annealing. However, the coarse grain and the α-fiber present in the starting materials led to the development of slightly shifted components of γ-fiber and α-fiber retained after the second annealing, while the fine grain and weaker α-fiber produced a stronger γ-fiber after annealing. These differences occurred due to the influence of grain size on the development of γ-fiber and the low recrystallization rate of α-fiber.

  • iv

    The fine grain and weak texture in starting materials were advantageous in order to develop a high r¯ value. The r¯ value of 1.92 was achieved for 59% of γ-fiber and 7% of θ-fiber. This finding was mainly attributed to the presence of a strong γ-fiber and a weak θ-fiber.

Conflicts of interest

The authors declare no conflicts of interest.


The authors would like to thank Fundação de Amparo à Pesquisa de Minas Gerais (Fapemig), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, 142380/2013-0), Comissão de Aperfeiçoamento de Pessoal do Nível Superior (Capes-PROEX, 0835/2018), and Aperam South America.

R.P. Siqueira, H.R.Z. Sandim, T.R. Oliveira.
Texture evolution in Nb-containing ferritic stainless steels during secondary recrystallization.
Mater Sci Eng A, 497 (2008), pp. 216-223
F.V. Braga, D.P. Escobar, T.A. Reis, Oliveira NJLD, M.S. Andrade.
Recrystallization of niobium stabilized ferritic stainless steel during hot rolling simulation by torsion tests.
J Mater Res Technol, 5 (2016), pp. 92-99
H. Liu, Z. Liu, G. Wang.
Texture development and formability of strip cast 17% Cr ferritic stainless steel.
ISIJ Int, 49 (2009), pp. 890-896
You Xiang-mi, Jiang Zhou-hua, L.I. Hua-bing.
Ultra-pure ferritic stainless steel – grade, refining operation, and applications.
J Iron Steel Res Int, 14 (2007), pp. 24-30
Y. Yazawa, Y. Ozaki, Y. Kato, O. Furukimi.
Development of ferritic stainless steel sheets with excellent deep drawability by {1 1 1} recrystallization texture control.
JSAE Rev, 24 (2003), pp. 483-488
J. Yang, L. Fan, J. Jia, R. Wu, X. Song, L. Jiang.
Microtexture evolution and grain boundary character distribution of interstitial-free steels with moderate levels of cold rolling reductions.
J Iron Steel Res Int, 20 (2013), pp. 47-52
M.Y. Huh, O. Engler.
Effect of intermediate annealing on texture, formability and ridging of 17%Cr ferritic stainless steel sheet.
Mater Sci Eng A, 308 (2001), pp. 74-87
C. Zhang, Z. Liu, G. Wang.
Effects of hot rolled shear bands on formability and surface ridging of an ultra purified 21%Cr ferritic stainless steel.
J Mater Process Technol, 211 (2011), pp. 1051-1059
F. Gao, Z. Liu, H. Liu, G. Wang.
Texture evolution and formability under different hot rolling conditions in ultra purified 17%Cr ferritic stainless steels.
Mater Charact, 75 (2013), pp. 93-100
U. Von Schlippenbach, F. Emren, K. Lücke.
Investigation of the development of the cold rolling texture in deep drawing steels by odf-analysis.
Acta Metall, 34 (1986), pp. 1289-1301
K.M. Lee, M.Y. Huh, H.J. Lee, J.T. Park, J.S. Kim, E.J. Shin, et al.
Effect of hot band grain size on development of textures and magnetic properties in 2.0% Si non-oriented electrical steel sheet.
J Magn Magn Mater, 396 (2015), pp. 53-64
ASTM E112-13: standard test methods for determining average grain size, ASTM International, (2013),
H.J. Bunge.
Texture analysis in materials science.
Butterworths&Co, (1982),
A.A. Gazder, M. Sánchez-Araiza, J.J. Jonas, E.V. Pereloma.
Evolution of recrystallization texture in a 0.78 wt.% Cr extra-low-carbon steel after warm and cold rolling.
Acta Mater, 59 (2011), pp. 4847-4865
F. Cardarelli.
Materials handbook: a concise desktop reference.
2nd ed., Springer-Verlag London Limited, (2008),
ASTM E517 - Standard test method for plastic strain ratio r for sheet metal, ASTM International, (2012),
O. Engler, V. Randle.
Introduction to texture analysis: macrotexture, microtexture, and orientation mapping.
2nd ed., Taylor& Francis Group, (2008),
F.J. Humphreys, M. Hatherly.
Recrystallization and related annealing phenomena.
2nd ed., Elsevier, (2004),
M. Nave, M.R. Barnett.
Understanding the deformed microstructure of cold rolled IF and LC steel.
Mater Forum, 29 (2005), pp. 562-567
B. Hutchinson, N. Hansen, P. van Houtte, D.J. Jensen.
Deformation microstructures and textures in steels.
Philos Trans Math Phys Eng Sci, 357 (1999), pp. 1471-1485
V.M. Nandedkar, I. Samajdar, K. Narasimhan.
Development of grain interior strain localizations during plane strain deformation of a deep drawing quality sheet steel.
ISIJ Int, 41 (2001), pp. 1517-1523
M.F. Campos, F.J.G. Landgraf, R. Takanohashi, F.C. Chagas, I.G.S. Falleiros, G.C. Fronzaglia, et al.
Effect of the hot band grain size and intermediate annealing on the deformation and recrystallization textures in low silicon electrical steels.
ISIJ Int, 44 (2004), pp. 591-597
R. Abbaschian, L. Abbaschian, R.E. Reed-Hill.
Physical metallurgy principles.
4th ed., Cambridge University Press, (2009),
F.J.G. Landgraf, R. Takanohashi, F.C. Chagas, M.F. de Campos, I.G.S. Falleiros.
The origin of grain size inhomogeneity in semi-processed electrical steels.
J Magn Magn Mater, 215–216 (2000), pp. 92-93
H. Inagaki.
Fundamental aspects of texture formation in low carbon steel.
ISIJ Int, 34 (1994), pp. 313-321
D. Raabe, Z. Zhao, S.J. Park, F. Roters.
Theory of orientation gradients in plastically strained crystals.
Acta Mater, 50 (2002), pp. 421-440
R.P. Siqueira, H.R.Z. Sandim, T.R. Oliveira, D. Raabe.
Composition and orientation effects on the final recrystallization texture of coarse-grained Nb-containing AISI 430 ferritic stainless steels.
Mater Sci Eng A, 528 (2011), pp. 3513-3519
F. Gao, Z.Y. Liu, H.T. Liu, G.D. Wang.
Evolution of through-thickness texture in ultra purified 17% Cr ferritic stainless steels.
J Iron Steel Res Int, 20 (2013), pp. 31-38
R.K. Ray, J.J. Jonas, R.E. Hook.
Cold rolling and annealing textures in low carbon and extra low carbon steels.
Int Mater Rev, 39 (1994), pp. 129-172
J.S.M. Pedrosa, C. Paolinelli S da, A.B. Cota.
Influence of initial annealing on structure evolution and magnetic properties of 3.4% Si non-oriented steel during final annealing.
J Magn Magn Mater, 393 (2015), pp. 146-150
D. Raabe, K. Lücke.
Textures of ferritic stainless steels.
Mater Sci Technol, 9 (1993), pp. 302-312
Z. Wang, Y. Guo, W. Xue, X. Liu, G. Wang.
Effect of coiling temperature on the evolution of texture in ferritic rolled Ti-IF steel.
J Mater Sci Technol, 23 (2007), pp. 337-341
W. Du, J. Lai-zhu, Q.S. Sun, Z.-Y. Liu, X. Zhang.
Effect of hot band annealing processes on microstructure texture and r-value of ferritic stainless steel.
J Iron Steel Res Int, 17 (2010), pp. 58-62
K.M. Lee, M.Y. Huh, S. Park, O. Engler.
Effect of texture components on the lankford parameters in ferritic stainless steel sheets.
ISIJ Int, 52 (2012), pp. 522-529
W. Du, L.Z. Jiang, Q.S. Sun, Z.Y. Liu, X. Zhang.
Microstructure, texture, and formability of Nb+Ti stabilized high purity ferritic stainless Steel.
J Iron Steel Res Int, 17 (2010), pp. 47-52
Copyright © 2019. The Authors
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.