IOP Conference Series: Materials Science and Engineering

PAPER • OPEN ACCESS You may also like

- Investigation of the physical properties of
Effects of Fe, Si, and Cu on Recrystallization Fe/Si for use in a Cu/Fe/Si solar cells
contact
Behavior in High Purity Aluminum Foil M Mebarki and A Layadi

- Mechanism for the Morphological Change

from Trenching to Pitting around
To cite this article: T Yamanoi 2021 IOP Conf. Ser.: Mater. Sci. Eng. 1121 012016 Intermetallic Particles in AA1050
Aluminum
Hiroshi Kakinuma, Izumi Muto, Yoshiyuki
Oya et al.

- Progenitor Mass Distribution of Core-

View the article online for updates and enhancements. collapse Supernova Remnants in Our
Galaxy and Magellanic Clouds Based on
Elemental Abundances
Satoru Katsuda, Tomoya Takiwaki,
Nozomu Tominaga et al.

This content was downloaded from IP address 46.232.209.116 on 29/01/2023 at 17:42

The 19th International Conference on Textures of Materials (ICOTOM19 2021) IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1121 (2021) 012016 doi:10.1088/1757-899X/1121/1/012016

Effects of Fe, Si, and Cu on Recrystallization Behavior in High

Purity Aluminum Foil

T YAMANOI
Aluminum Rolled Products Division, SHOWA DENKO K.K., Sakai, Osaka, Japan

Abstract. The correlation between the recrystallization behavior of high purity aluminum foil,
which is used as the electrode material of electrolytic capacitors, and the influence of Fe, Si, and
Cu contents on the formation of recrystallization texture was systematically investigated. The
following results were obtained: (1) Room-temperature recrystallization, which occurs during
cold rolling of the highest-purity material, was suppressed by adding approximately 10 mass
ppm of Fe, Si, and Cu. The effect was strongest for Cu, followed by that for Si and Fe. (2) The
main component of texture after the final heat treatment changed in the order {001}<100>→
{011}<411>→ {126}<411>→ {123}<412> as the purity of the aluminum foil decreased. In
particular, {126}<411> was an characteristic orientation for extremely coarse grains and was
considered to have originated from the rolling texture (β-fiber). This coarse grain growth was
considered to be related to the decrease in the growth rate of the recrystallized grains depending
on the purity of the matrix.

1. Introduction
In the high purity aluminum foil in electrolytic capacitors, which is one of the main applications of high
purity aluminum foil, the control of grain size and texture in terms of electrostatic capacity and bending
strength is one of the objectives of material design. Fe, Si, and Cu content is known to influence this
phenomenon [1][2][3], but the effects on recrystallization behavior in very small amounts of other elements
have not been identified. In this study, a considerable amount of coarse grains, which are obtained at
temperatures slightly above room temperature, was observed. The study aimed to investigate the
respective effects of Fe, Si, and Cu on recrystallization behavior and to clarify its growth process.

2. Experimental Procedures

2.1. Samples
The chemical composition of the sample used in this study is shown in Table 1. The sample preparation
process is shown in Fig. 1. Samples was casting in book molds for aluminum slabs. The sample was
hot-rolled at a starting temperature of 793 K and air-cooled afterwards. Samples that were heat-treated
after hot rolling were dipped in the salt bath at 723 K, held for 600 s, and water-cooled immediately
afterwards. The rolling reduction ratio between each pass was maintained within the range of 30±10%
during the cold rolling and the temperature of the sample was kept from rising.

2.2. Observation and analysis method

The microstructure was observed using a polarizing microscope, after the sample had been polished
with emery paper, buffed, and anodized through treatment with Barker's solution. The hardening and
softening behaviors of the specimens that exhibited recrystallization were evaluated using Vickers
hardness and tensile tests. Texture was analysed via pole figure measurement using X-ray diffraction
(XRD) and via electron backscattered diffraction (EBSD) measurement using FE-SEM.

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
The 19th International Conference on Textures of Materials (ICOTOM19 2021) IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1121 (2021) 012016 doi:10.1088/1757-899X/1121/1/012016

Table 1. Chemical compositions Book mold

Melting & Casting
of the specimens (mass ppm) casting

Si Fe Cu Al S oaking 883K, 7.2×10４s

Ａ１ 2 1 2 Bal.
S calping Thickness: 35mm ｔ
Ａ２ 2 1 11 Bal.
813K,
Ａ３ 10 1 2 Bal. Heating 1.3～3.6×104 s
Ａ４ 2 12 2 Bal.
Hot rolling thickness:５mmｔ
Ａ５ 11 12 2 Bal.
①No annealig
Ａ６ 1 21 2 Bal. Intermidiate annealing ②723K, 600s
(dipping in salt bath ）
Ａ７ 10 10 10 Bal. Cold rolling Thickness: 0.1mm ｔ
Ａ８ 9 10 15 Bal.
Final annealing ①No annealing
Ａ９ 10 11 24 Bal. ②313～573K,
7.2×103 s
Ａ10 12 11 49 Bal.
Fig. 1. Production and experimental process
Ａ11 2 52 2 Bal. of samples.

3. Experimental Results

3.1. Softening behavior during cold rolling

Some of the samples in this study recrystallized during cold rolling. For example, Fig. 2 shows the
microstructure and {111} pole figure after the cold rolling of the A1 composition sample in Table 1.
Figure 2 shows a microstructure whose recrystallization is almost complete and the pole figure has a
very sharp cube texture.
Figure 3 shows the results obtained upon evaluating the hardening after cold rolling to 0.1 mm thickness
using Vickers hardness test. In the sample shown in Fig. 3, elements other than the composition
concentration plotted on the abscissa are equal to or less than 2 mass ppm, indicating the influence of
Fe, Si, and Cu only. Figure 3 shows that Cu content inhibits softening by only 5 atom ppm (11 mass
ppm) in aluminum in the no-annealing sample after hot rolling, and that the inhibiting effect in the order
of Si and Fe is attenuated. Figure 3(b) shows that Vickers hardness for cold-rolled foils annealed after
hot rolled sheets increased holistically compared to that of the no-annealed foils. This is attributable to
the reduction of the initial dislocation density by the annealing of the hot-rolled sheets, and the
insufficiency of the total strain for recovery and recrystallization during cold rolling. However, no
tendency was observed in the A4, A6, and A11 samples containing Fe only. Further, the order of Vickers
hardness of foils was reversed in no-annealing and annealing samples of the hot rolled sheets containing
a large amount of Fe. Researchers in a previous study on work softening in the system added only Fe to
a high purity aluminum base [4]; therefore, the lattice defects may have disappeared as a result of the
precipitation of Al-Fe intermetallic compounds
on lattice defects such as dislocation cells, A1
subgrain boundaries, and recrystallized grain
Rolling direction

boundaries. Based on these results, Fe is ■ ■

considered to slightly contribute to the increase

in the recrystallization temperature for matrix
■ ■
aluminum, and acceleration of recrystallization
by precipitation during heat treatment is thought 200μm
M ax:28.0
to have an impact on this phenomenon. [5] ■{001}<100>

Traverse direction Level 0.5 1 2 4 8 16 32

Fig. 2. Grain structure and texture after cold rolling.

■{001}<100>

2
The 19th International Conference on Textures of Materials (ICOTOM19 2021) IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1121 (2021) 012016 doi:10.1088/1757-899X/1121/1/012016

45 45
(a) (b) Cu
40 Fe 40 A2
Cu Si
Vickers hardness / Hv

35 A11 35
A2 A3
30 30
Si
25
A3 25 A1 Fe
A4 A6 A11
20 A4 A6 20
A1
15 15
0 10 20 30 0 10 20 30
Concentration / atom ppm
Fig. 3. Vickers hardness of foils after cold rolling, fabricated at (a) no annealing
and annealing at 723 K after hot rolling, as a function of atomic content.

3.2. Effect of Fe, Si, and Cu on microstructure after final heat treatment
Figure 4 shows the microstructure and the {111}pole figure when the cold-rolled foils (0.1 mm
thickness) were heated at 473 K without heating of the hot-rolled sheets.
(a)A4 (d)A5 △

■ ■ ■ ■
△

△
△
■ ■ ■ ■

A5
200μm 200μm ■{001}<100>
■{001}<100> M ax:19.0 △{011}<411> M ax:11.8
(b)A3 △ (e)A7

■ △ ■
▽
▽

△
△
■ ■ ▽

200μm ■{001}<100> 200μm ●{123}<412>

Rolling direction

△{011}<411> M ax:6.3 ▽{126}<411> M ax:59.9

(c)A2 △ (f)A10

■ △ ■
●
●

△
△
■ ■ ●

A2 ●
200μm ■{001}<100> 200μm
△{011}<411> M ax:6.4 ●{123}<412> M ax:22.3
Traverse direction Level 0.5 1 2 4 8 16 32
Fig. 4. Effect of Fe, Si, and Cu on grain structure and texture after final annealing.
No annealing after hot rolling, final annealing conditions: 473 K × 7.2 ks
■{001}<100> ●{123}<412> △{011}<411> ▽{126}<411>

3
The 19th International Conference on Textures of Materials (ICOTOM19 2021) IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1121 (2021) 012016 doi:10.1088/1757-899X/1121/1/012016

Figures 4(a), (b), and (c) show that the Fe, Si, and Cu contents had individual effects on the
microstructure and texture respectively. Figure 4(a)A4 exhibited a remarkably well-developed cube
texture {001}<100>. However, the main orientations of Figs. 4(b)A3 and (c)A2 were not only cube
texture but also the orientation components near {011}<411>, which were intermediate between the
Goss {011}<100> and brass {011}<211> orientations. Before final annealing, Fig. 4(b) represented the
partial recrystallization, Fig. 4(c) represented the working structure, and the result illustrated in Fig. 3
was considered. In the case of Figs. 4(b) and (c), the growth of the cubic texture {001}<100> was
suppressed; thus, the orientation developed from the rolling texture (s-orientation [6]: β-fiber [7]).
Figures 4(d), (e), and (f) show the effect of the amount of Cu at the approximately 10 mass ppm level
of both Si and Fe. In sample Fig. 4(d)A5 (Cu: 2 mass ppm), relatively coarse recrystallized grains with
a crystal grain size of more than 200 μm developed almost uniformly. The pole figure confirmed the
existence of the cube texture {001}<100> and an almost {011}<411> texture was observed in Figs. 4
(b), (c). Next, mixed structures of extremely coarse recrystallized grains (hereafter referred to as huge
grains) and un-recrystallized regions, which extended in the rolling direction of sample (e)A7 (Cu: 10
mass ppm). Based on the texture analysis, this huge grain was considered to be an orientation component
near {126}<411>. This orientation was also identified in Fig. 4(d), so was considered to be not typical
to Fig. 4(e). Further, the rolling texture of the un-recrystallized region was represented by the
{123}<412> orientation. Finally, in Fig. 4(f)A10 (Cu: 49 mass ppm), recrystallized grains were not
observed and huge grains were not also observed. In addition, rolling texture was determined based on
pole figure analysis. As shown in Figs. 4(d), (e), and (f), the grain growth rate in recrystallization was
considered to be related to the growth of the huge grains in Fig. 4(e) owing to a certain amount addition
of Cu content. Additionally, in the samples of A8 (Cu: 15 mass ppm), huge grains were observed as Fig.
4(e), and huge grains were not observed in the sample of A9 (Cu: 24 mass ppm). Therefore, the upper
threshold of Cu contents at which huge grains were expressed in this study was considered to range
between 15 and 24 mass ppm.

4. Discussion

From the pole figure analysis shown in Fig. 4, the influence of Fe, Si, and Cu only on the formation of
cube texture in the final annealing increased in the order of Fe <Si<Cu, and the preferred orientation of
texture changed from {001}<100> to {001}<100> + {011}<411>. Moreover, in the estimated 10 mass
ppm samples of Si and Fe, the preferred orientation changed from {001}<100> to {123}<412>, which
is thought to have stemmed from the rolling texture (β-fiber), and grain growth was remarkably high
in the process.
Figure 5 shows the tension softening curves of samples with different contents of Cu, and approximately
10 mass ppm level of Si and Fe contents. This image shows that softening began to increase in speed at
333 K in the samples of A5 and 433 K in the samples of 160
A7. To investigate the orientation relationship between A5
recrystallized grains and the surrounding matrix, samples 140 A7
A10
of (x)A5 (annealed at 343 K for 4.2 ks) and samples of 120
Tensile strength / MPa

(y)A7 (annealed at 443 K for 7.2 ks) were produced, and

100
EBSD analysis was carried out. The results are shown in (y)
Fig. 5. The IPF MAP, KAM MAP, and ODF were applied 80 (x)
to the respective samples through EBSD analysis. 60
First, recrystallization was considered to be completed in
the blue region where deviation of crystal orientation from 40
adjacent regions was very small from KAM MAP, and the 20
grains and orientations were confirmed from IPF MAP.
0
Then, the region where the recrystallization was 273 373 473 573 673
considered to be proceeding to the surrounding matrix was Annealing temperature / K
selected (inside the square frame in Fig. 6), and finally, Fig. 5. Tension softening curves of high purity
ODF analysis in this region was carried out. aluminum foil with varying Cu content.
(x)A5: after 343 K × 7.2 ks annealing
(y)A7: after 443 K × 7.2 ks annealing

4
The 19th International Conference on Textures of Materials (ICOTOM19 2021) IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1121 (2021) 012016 doi:10.1088/1757-899X/1121/1/012016

(x) A5 (y) A7
IPF MAP
KAM MAP

φ1 (001)[100] φ1

Φ Φ

φ2 = φ2 =

__
(113)[273]
ODF

Fig. 6. Typical inverse pole figure map (IPF MAP), kernel average misorientation map
(KAM MAP), and orientation distribution function (ODF) of coarse-grained specimens.
(x) A5: after 343 K×7.2 ks annealing, (y) A7: after 443 K×7.2 ks annealing

5
The 19th International Conference on Textures of Materials (ICOTOM19 2021) IOP Publishing
IOP Conf. Series: Materials Science and Engineering 1121 (2021) 012016 doi:10.1088/1757-899X/1121/1/012016

Based on the ODF analysis shown in Figure 6, the

Recrystallization
A5-1R
preferred orientation of the recrystallized grain in sample

area
A5-2R
A5-3R
(x) was {001}<100> and {113}<273> in sample (y). A5-4R

Figure 7 shows the angles between the face azimuth A5-1

obtained from the region adjacent to the recrystallized

(x) A5
A5-2

Un-recrystallization area
A5-3
grain and the face azimuth of the recrystallized grain. A5-4
A5-5
Figure 7 shows that the deviation angle of the crystal A5-6
A5-7
orientation with recrystallized grains adjacent to A5-8
A5-9
recrystallized grain {011}<100> of sample (x) was A5-10
A5-11
relatively small from approximately 4° to 27°. And the

recrystallization area
deviation angle of the crystal orientation from the adjacent A7-1
A7-2

(y) A7
un-recrystallized region was approximately 21° to 74°, A7-3

Un-
A7-4
which was large, albeit with some variation. Next the A7-5
A7-6
region adjacent to the {113}<273> grains of sample (y)
was the un-recrystallized region in the observation range, 0 10 20 30 40 50 60 70 80 90
and the deviation angle of the crystal orientation was deviation betweem both plains ⊿θ / °
Fig. 7. Angle between crystal orientation planes between
mainly within the range of 35° to 45°. the un-recrystallized region adjacent to recrystallized
In this analysis, the identification of common axes of grains, and the recrystallized grains.
[8]
rotation showing a coincidence relationship in the region with the deviation angle in Fig. 7 was not
accomplished. However, in sample (y), wherein remarkably grain growth had occurred, the un-
recrystallized region showed a clear anisotropy of crystal orientation and development of the huge grains
in the rolling direction. Therefore, it was inferred that the preferred grain growth had occurred.

5. Conclusion
In this study, high purity aluminum foils with varying Fe, Si, and Cu contents and processes were
analyzed using the XRD/pole figure and the SEM/EBSP method. Based on the results, we discussed the
influence of Fe, Si, and Cu content on the formation of a coarse grain texture. The results are summarized
as follows:
(1) Room-temperature recrystallization, which occurs during the cold rolling of the highest-purity
material, was suppressed by adding several ppm of Fe, Si, and Cu. The effect was strongest for Cu,
followed by Si and Fe.
(2) The main component of texture after the final heat treatment changed in the order
{001}<100>→{011}<411>→{126}<411>→{123}<412> as the purity of aluminum foil decreased.
(3) In this study, extremely coarse recrystallized grains were observed in samples wherein the Si and Fe
contents were both approximately 10 mass ppm and the Cu content was within a certain range. The
preferred orientations of these huge grains were {126}<412>, {113}<273>, etc., which were inferred
to be the preferred growth from the rolling texture (β-fiber).

Acknowledgments
The author would like to thank Dr. M. Sakaguchi for the opportunity to begin this study, and Prof. H.
Inoue for their advice on the measurement and analysis of texture. This study was supported by
SHOWA DENKO K.K. and their members.

References
[1] Miki I and Warlimonto H 1968 Z. Metallkde. 59 408-414
[2] Henmi Z and Nagai T 1967 J. Japan Inst. Metals 31 329-333
[3] Nagahama K and Miki I 1970 J. Japan Inst. Light Metals 20 137-144
[4] Ohno Y and Nakamura H 1977 Aluminium 53 539-542
[5] Sakaguchi M, Yamanoi T and Hasegawa M 1987 Z. Metallkde. 78 80-86
[6] Inoue H 2015 Mater. Trans. 56 61-69
[7] Inoue H 2001 Materia Japan 40 589-591
[8] Kronberg M L and Wilson F H 1949 Metals trans. 185 501-514