Influence of Microstructural Evolution Processed by ECAP on
Corrosion Behavior of Pure Magnesium in RPMI-1640 Medium
Taito Hosaka
1
, Iman Amanina
1
, Naohiro Saruwatari
1
, Shoichiro Yoshihara
1
and Bryan J. MacDonald
2
1
Department of mechanicalEngineering, University of Yamanashi, 4-3-11, Takeda, Kofu-city, Yamanashi, Japan
2
School of Manufacturing and Mechanical Engineering, Dublin City University, Dublin 9, Ireland
Keywords: Microstructural Evolution, ECAP Process, Corrosion Behaviour, Pure Magnesium, RPMI 1640 Medium,
Biomaterial.
Abstract: Influence of microstructure changes caused by Equal-channel angular pressing (ECAP) process on
corrosion behavior of pure magnesium in RPMI-1640 medium was investigated. The grain size of ECAPed
samples (30µm) were greatly reduced compared with the grain size of the annealed sample (200µm). Then,
the immersion test has been carried out for a certain period of time. It was revealed that mass loss of the
ECAPed sample is larger than the as-received sample and the annealed sample. Thus, it could be considered
that many crystal defects yielded by ECAP process reduced the corrosion resistance. However, the
corrosion resistance has been improved to a certain extent according to reduction of crystal defects through
the heat treatment at the recrystallization temperature or lower. In addition, the amount of gas generation of
the ECAP sample after immersion test is larger compared with the as-received sample. Therefore,
correlation between the amount of gas generated and the mass loss was confirmed. Based on qualitative
identification of the elements by Energy Dispersive X-ray Spectrometry (EDS), the corrosion products of
the sample surface after the immersion test has been estimated to be a kind of calcium phosphate. These
above results have indicated the potential for fabrication of magnesium as bioabsorbable materials.
1 INTRODUCTION
Magnesium (Mg) is present in high concentrations in
sea water and the eighth most abundant element on
the earth. It has also excellent specific strength and
low density, only two-thirds that of Aluminum. Then,
Mg and its alloys could be adopted in many
applications including computer parts, mobile
phones, aerospace components and handheld tools
(Wu, 2007).
Mg alloys are also potentially useful for bone
implants and stent applications due to their low
density, inherent biocompatibility, adequate
mechanical properties and fracture toughness higher
than that of ceramics (Mani, 2007). Additionally, the
elastic modulus of Mg alloys (40–45GPa) is closer
to that of human bones (10–40GPa) than other
commonly used implant materials such as Titanium
(106GPa). Eventually, the stress-shielding
phenomena caused by current metallic implants
made of stainless steel or Titanium alloy could be
minimized (Xu, 2007).
Another advantage of Mg in relation to other
metallic implants is the degradability of Mg alloys
which offers the possibility of repair and
reconstruction of vascular compliance with
minimum inflammatory response (Loos, 2007).
In recent years, the application of bioabsorbable
materials for medical implants has drastically
increased (Busch, 2014). Since magnesium is an
essential element in our body, it can be absorbed
into the body without leaving any harm. Therefore,
many studies have been focused on using
magnesium and its alloys for medical devices such
as bioabsorbable stents (Hiromoto, 2012).
Nowadays, the traditional permanent metallic
materials for the stents are modernly manufactured
from stainless steel (316L), nitinol and cobalt-
chromium alloys which have high corrosion
resistance and remain as a permanent implant in the
body. Thus, many limitations such as risk of chronic
irritation caused by released toxic substances which
might appear after long-term placement of the stents.
118
Hosaka T., Amanina I., Saruwatari N., Yoshihara S. and J. MacDonald B.
Influence of Microstructural Evolution Processed by ECAP on Corrosion Behavior of Pure Magnesium in RPMI-1640 Medium.
DOI: 10.5220/0006143201180125
In Proceedings of the 10th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2017), pages 118-125
ISBN: 978-989-758-216-5
Copyright
c
2017 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved
Magnesium is able to overcome many of these
limitations since it could be absorbed into the body.
For the patients with the coronary artery disease, the
magnesium stents could temporarily function as a
scaffold for supporting the vascular stenosis. The
disappearance of the magnesium alloys stent is
necessary after 6-12 (Werkhoven, 2011) months
period. Therefore, the structural integrity of the
implanted stent has to be preserve until the surgical
region has completely heal, hence the corrosion
behaviour of Mg and its alloys have received
attention in recent years. However, study on the
degradation behaviour and mechanisms of
magnesium and its alloys in physiological
environment such as inside the human body have not
been elucidated.
The main factors affecting the corrosion
performance of metallic materials as bioabsorbable
stent include internal factors which are all about the
material itself such as microstructure, distortion,
surface condition and external factors such as
influence of blood flow, body temperature and pH
value. The effect of grain refinement by various
severe plastic deformation methods on the corrosion
behaviour of materials has caught much attention in
recent years. During the last decade, equal-channel
angular pressing (ECAP) has considered as the most
appropriate procedure for the fabrication of
ultrafine-grained (UFG) metals and alloys for
industrial application. It has been widely-known that
ECAP process significantly affects the mechanical
properties of magnesium and its alloys. Although
many studies have been done on the materials
processed by ECAP, their corrosion resistance has
been rarely reported (Song, 2010). The present work
focused on the influence of microstructure change
on corrosion resistance of pure Mg processed by
ECAP.
Therefore, in this study, we attempted to
investigate the immersion test on ECAPed pure Mg
in RPMI 1640 medium. The results would be helpful
for better understanding the corrosion behaviours of
ECAPed pure Mg and its alloys, and explore their
possibility for engineering applications.
2 EXPERIMENTAL
2.1 ECAPed Specimen Preparation
The specimen used for ECAP process was cut from
the extruded pure Mg round bar (99.96%). The
schematic illustration of the ECAP process is shown
in Fig.1. The billets of the specimen for ECAP with
the size of diameter Φ6 mm × length 50 mm were
annealed for 24 hours at 723 K after cutting process.
Then, the billets were repeatedly pressed for 4 times
with a plunger speed of 4 mm/s at 573 K.
Molybdenum disulfide as lubricant was used to
reduce the friction coefficient between the billet and
the die inner wall. During the each pressing process,
the billet was inverted, and then rotated by 90° to the
circumferential direction.
Figure 1: Schematic of ECAP process.
2.2 Experimental Conditions
Table 1 shows the kind of the samples on several
experimental conditions of the different material
types and the subsequent annealing times. Type (a)
shows the ECAPed pure magnesium sample, type
(b) the extruded (as-received) pure Mg sample and
type (c) the annealed at 723 K for 24 hours pure Mg
sample have been studied in comparison with the
several conditions. Furthermore, the subsequent
annealing process below the re-crystallization
temperature was applied to ECAPed samples in
order to verify the effect of crystalline defects on
weakening pure Mg’s corrosion resistance without
increasing grain sizes. It has been processed at 423K
for type (d) 2hours, type (e) 4hours and type (f)
6hours respectively to confirm the influence of the
difference conditions on processing time. Besides,
Park has reported that the subsequent annealing
process could release the strain energy and weaken
deformation texture of ECAPed samples (Park,
2008).
2.3 Microstructure Observation
The samples, used for optical microstructure
observation were cut on perpendicular to the
pressing direction from the ECAPed billet, and
polished with distilled water and etched with acetic
picral (100 ml 6 wt.% picric acid, 28.8 ml 99.7 wt.%
acetic acid and 10 ml distilled water). Optical
microscopy (NICON ECLIPSE MA200, Japan) and
Influence of Microstructural Evolution Processed by ECAP on Corrosion Behavior of Pure Magnesium in RPMI-1640 Medium
119
Table 1: Samples on difference conditions of Mg.
scanning electron microscope (JEOL JSM-6500F,
Japan) equipped with the Electron Back Scatter
Diffraction (EBSD) (EDAX) camera were adopted
to observe the microstructures and more details of
the samples.
The samples for EBSD were also cut perpendicular
to the pressing direction, polished on the emery
papers and buffing with the abrasive diamond
solutions of particle size 1μm and 0.25μm.
Afterward, they were etched with the solution of
10ml HNO
3
, 30ml acetic acid, 40ml H
2
O and 120ml
ethanol for around 10s.
2.4 Microhardness Tests
The Vickers hardness test was carried out to
investigate the mechanical properties of the samples.
Vickers hardness measurements were performed
using micro-hardness testing device (AKASHI
MVK-G3500AT, Japan). The samples were cut from
billets perpendicular to the pressing direction and
polished. Measurements were carried out at least
five times for each sample. The load of Vickers
indenter was 25g.
2.5 Corrosion Tests
The corrosion behaviour of the ECAPed pure Mg
was investigated by the immersion test for certain
period. ECAPed samples used were cut from the
core of the ECAPed billets, cleaned and polished to
avoid the surface contamination of lubricant and so
forth. The surface of all specimens was polished
with #2000 grits emery paper to ensure the same
surface roughness, and then cleaned in acetone by
ultrasonic washing machine. The Size of the test
sample was diameter Φ6mm × thickness 3mm,
respectively. RPMI 1640 medium (Sigma-Aldrich
R8758, Japan) as a corrosion solution was used for
immersion test in all experiments to simulate the
internal environment. The temperature of the
corrosion solution was 310K constantIn this study,
Fetal Bovine Serum (FBS) was not added to the
medium. The initial samples were polished with
emery papers up to #2000 grits and cleaned. The
measuring procedure of the polished samples was
weighed at first, immersed in RPMI1640 medium
(pH=7.4) for various intervals in corrosion test,
cleaned with a corrosion product removal solution
(20g CrO
3
+ 2g Ba (NO
3
)
2
+ 1g AgNO
3
+ 100ml
Distilled water) after corrosion test, washed with
acetone and weighed without corrosion product. The
above steps were repeated and the mass loss in every
interval was measured to evaluate the corrosion rate
of ECAPed samples [g/m
2
]. Mass loss was obtained
by the following equation.
ML = ( W
0
- W ) / S (1)
where, W
0
[g] is mass of the sample before corrosion.
W [g] is mass of after corrosion. S [m
2
] is Initial
surface area of the sample.
Moreover, Changes in the pH value in the
immersion test was recorded for 7 days.
In addition, the surface of the ECAPed sample
after the immersion test for 24 hours was observed to
identify the corrosion products. After the immersion
test, the sample was rinsed with distilled water
quickly, and dried to observe their surface by an
optical microscope. The surface of the sample was
analyzed by the
energy dispersive X-ray
spectroscopy (EDS) with the scanning electron
microscope (Japan electron JSM-7100F)
.
Furthermore, the immersion test for 24 hours of
ECAPed sample and As-received sample was
conducted to measure the amount of generated gas
with the corrosion test of the pure magnesium in
RPMI 1640 medium. Fig.2 shows the method of
immersion test.
BIODEVICES 2017 - 10th International Conference on Biomedical Electronics and Devices
120
Figure 2: Photograph of immersion experiment.
3 RESULTS AND DISCUSSION
3.1 Microstructures of ECAPed Pure
Mg Samples
Fig. 3(a)-(f) indicates the optical microstructures of
the as-received, the annealed, the ECAPed, the
subsequent aging for 2, 4 and 6 hours of the pure
magnesium samples. The average grain size
distribution before ECAP is around 200μm. Thus,
after the ECAP process, the average grain size
distribution of 30μm was reduced. It is noteworthy
that some larger grains are surrounded by some
much smaller grains in the deformed
microstructures. This phenomenon might be resulted
from dynamic recrystallization since the temperature
(at 573K) of ECAP process is higher than the
recrystallization temperature of Mg.
Fig.4(a)–(f) shows the inverse pole figure map of
these samples also. In comparison with the as-
received, the crystal orientation of ECAPed sample
has significantly changed. Thus the effect of the
shear stress applied into the material could be
confirmed.
Fig. 5(a)–(f) shows the image quality map by
EBSD of these samples. The polishing method of the
sample in EBSD was all the same approach. Hence,
it was considered that there is no difference in strain
applied to the surface of each sample by the
polishing. The crystalline of the ECAPed sample
was considered low compared to the untreated
sample and the annealed sample.
On EBSD, The factor affecting the quality of
diffraction patterns in a materials science standpoint
is the perfection of the crystal lattice in the sample.
Thus, any distortions to the crystal lattice of the
sample would be produced to the lower quality
(more diffuse) diffraction patterns. It enables the
Image quality (IQ) parameter to be used to give a
qualitative description of the strain distribution.
As seen in figure (a), (b) and (c), the ECAP
sample is confirmed to be less crystalline in
comparison with the as-received sample and the
annealed sample. It could be considered that ECAP
process increases the crystal defects of the material
by introducing a shear strain to the material at the
bent portion of the die and refining the grains.
In additionfrom figure (a), (d), (e) and (f), the
crystalline of the subsequent aged sample was
slightly increased. Hence, it could be concluded that
crystal defects such as dislocation and strain of
ECAPed sample was recovered by subsequent
annealing from ECAP.
3.2 Microhardness Test
Fig. 6(a)–(f) shows the Vickers hardness results.
Compared to the as-received samples and the
annealed samples, the ECAP processed samples
Figure 3: Microscopic images on surface of Mg.
Influence of Microstructural Evolution Processed by ECAP on Corrosion Behavior of Pure Magnesium in RPMI-1640 Medium
121
Figure 4: Inverse pole figure map of surface of Mg.
have higher value of Vickers hardness. From figure
(a), (d), (e h) and (f), Vickers ardness of the
subsequent aged samples was slightly decreased as
the aging time increases the Vickers hardness
decreases. Based on grain observation in optical
microscope and EBSD, there is no change in grain
size of the ECAP sample and the subsequent aged
sample. Accordingly, It was considered that the
Vickers hardness of the subsequent aged samples
were reduced by the recovery of crystal defects in
the heat treatment below the recrystallization
temperature.
3.3 Corrosion Behaviour in Constant
Immersion Test
Constant immersion test is a more direct and
illustrative approach to detect corrosion behaviour of
a material. Fig.7 shows the mass loss of the
ECAPed, the as-received, the annealed and the
subsequent aging for 2, 4 and 6 hours pure Mg
samples after 168 hours (7 days) immersed in RPMI
1640 medium. From Fig.7, it is obvious result that
the ECAP sample has the largest mass loss in these
conditions. Additionally, it shows the annealed
sample has the most favourable corrosion resistance
among the samples. From the both results of the
grain size observation and the mass loss, during the
corrosion test of the pure Mg in RPMI 1640 medium,
there is no significant effect in grain size, however,
it could be suggested that the other corrosion factor
has remarkable effect. Furthermore, the corrosion
resistance of the subsequent samples was improved
as the heat treatment time increased.
Some researchers have reported that the
corrosion resistance of materials have been
improved by grain refinement which has been
decreasing general corrosion rate and alleviating
corrosion localization (Janecek, 2005). Also, the
better corrosion resistance of ECAPed Cupper (Cu),
Titanium (Ti) and industrial pure Aluminium (Al)
with deformed microstructures than the coarse
grains one have been indicated (Blyanov, 2004).
However, unexpectedly, the ECAP process might
weaken the corrosion resistance of the pure Mg. It
could be considered that many crystal defects
yielded by ECAP process reduced the corrosion
resistance.It was considered that this negative
phenomenon was related to the microstructure
characterization and the corrosion mechanism of
ECAPed pure Mg. It has been verified that the
corrosion has been caused primarily by surface
defects such as grain boundaries and dislocations
(Aust, 1994).
The above results have demonstrated that the
strain-induced crystalline defects are unfavourable
for the corrosion resistance of pure Mg. However the
corrosion behaviour of ECAPed pure Mg could be
improved by subsequent annealing process.
Fig.8(a)–(f) shows the pH value. The pH value in all
conditions was increased for approximately over 0-
12 hours and decreased over 12-24 hours. After 24
hours, the pH value was slightly decreased. This
tendency was particularly remarkable in the pure Mg
ECAPed sample and the subsequent aging for 2, 4
and 6 hours samples. In this result, FBS was not
added to the RPMI 1640 medium. There is a need to
investigate the pH level in the case of enhancing the
buffer function of the medium from now on. The
rapid increase in the pH of the early stage of
immersion test has not been sufficiently examined in
the present study.
Fig.9 shows the amount of gas generated the
ECAPed and the as-received sample after immersion
for 24 hours. The amount of gas generated of the
ECAPed sample was 1.72 ml/cm
2
, the as-received
BIODEVICES 2017 - 10th International Conference on Biomedical Electronics and Devices
122
Figure 5: Image quality map on surface of Mg.
sample was 1.28 ml/cm
2
. The ratio on gas generation
volume of the as-received sample and the ECAPed
sample were similar to the ratio on the mass loss of
the as-received sample and the ECAPed sample.
Therefore, it was suggested that there is a correlation
between the amount of mass loss and the amount of
gas generation.
Fig.10 shows the corroded surface of the
ECAPed sample after immersion test for 24 hours by
SEM. The Localized corrosion products could be
observed on the sample surface. Fig.11 shows the
result of the chemical element analysis in the same
microscopic field by EDS. From the result, the main
component elements of the corrosion products
present in the sample surface were oxygen O,
phosphorus P and calcium Ca. It could be assumed
that the chemical compound produced on sample
surface during the corrosion of pure magnesium in
RPMI 1640 medium was a type of calcium
phosphate. The peak of hydrogen of the light metal
could not be detected by EDS.
Figure 6: Vikers hardness test results.
Figure 7: Relationship between mass loss and immersion
time.
Figure 8: Relationship between pH and Immersion time.
Influence of Microstructural Evolution Processed by ECAP on Corrosion Behavior of Pure Magnesium in RPMI-1640 Medium
123
Figure 9: Volume of gas generation on ECAPed and As-
received Mg.
Figure 10: SEM image of ECAPed pure Mg sample after
immersional test.
4 CONCLUSIONS
Fine grained pure Mg could be manufactured
through the ECAP process at 573K. Also, the
subsequent annealing process below the re-
crystallization temperature was applied by using the
ECAPed samples in order to verify the effect of
crystalline defects on weakening corrosion
resistance of pure Mg without increasing grain sizes.
The effect of microstructure on the corrosion
behavior of the ECAPed pure Mg was investigated
by optical microscopy and SEM observation,
microhardness test and certain period immersion
tests in RPMI 1640 medium.
(1) The difference in the microstructure by ECAP
processing was observed by optical microscope and
SEM with EBSD. The average grain size of ECAPed
samples (30µm) were greatly reduced compared
with the annealed sample (200µm). From the result
of EBSD, the crystal orientation of ECAPed sample
in comparison with the as-received and the annealed
sample has significantly changed. The ECAPed
sample was found to be less crystalline alongside of
the as-received and the annealed sample. It could be
considered that the crystal defects of the material by
introducing a shear strain to the material have been
increased by ECAP process. In addition, it has been
confirmed that crystal defects such as dislocation
and strain of ECAPed sample was recovered by
subsequent aging.
(2) The corrosion behaviour of the ECAPed pure
Mg was investigated by the immersion test for
certain period. From the results of the immersion test,
it has been revealed that the corrosion resistance of
pure Mg in RPMI 1640 medium has been reduced
ECAP process. The strain-induced crystalline
defects were unfavourable for the corrosion
resistance of pure Mg. However, the corrosion
resistance of ECAPed pure Mg could be improved
by subsequent aging process. On the other hand, the
ratio on gas generation volume of the as-received
sample and the ECAPed sample were similar to the
ratio on the mass loss of the as-received sample and
the ECAPed sample. Therefore, it was suggested
Figure 11: EDS images of ECAPed pure Mg sample after immersion test.
BIODEVICES 2017 - 10th International Conference on Biomedical Electronics and Devices
124
that there is a correlation between the amount of
mass loss and the amount of gas generation. (3) The
surface of the ECAPed sample after the immersion
test for 24 hours was observed to identify the
corrosion products by EDS. The localized corrosion
products were observed on the sample surface. Then,
from the result, it could be assumed that the
chemical compound which has been produced on
sample surface during the corrosion of pure
magnesium in RPMI 1640 medium was a type of
calcium phosphate.
ACKNOWLEDGEMENTS
This work was supported by JSPS KAKENHI Grant
Numbers JP16K05974, JP25420010.
REFERENCES
K. T. Aust, U. Erb, G. Palumbo, 1994. Interface control
for resistance to intergranular cracking, Mater. Sci.
Eng. A, 176, pp. 329–334.
A. Blyanov, J. Kutnyakova, N.A. Amirkhanova, V.V.
Stolyarov, R.Z. Valiev, X.Z. Liao, Y.H. Zhao, Y.B.
Jiang, H.F. Xu, T.C. Lowe, Y.T. Zhu, 2004. Corrosion
resistance of ultra fine-grained Ti, Scr. Mater., 51, pp.
225–229.
Hiromoto, S., 2014. Corrosion behavior of magnesium
alloys for biomaterial use, Corrosion engineering of
Japan, 63, pp. 371–377.
Jae Yeol Park, Dong Nyung Lee, 2008. Deformation and
annealing textures of equal-channel angular pressed
1050 Al alloys strips, Mater. Sci. Eng. A, 497, pp.
395–407.
M. Janecek, B. Hadzima, R.J. Hellmig, Y. Estrin, 2005.
The influence of microstructure on the corrosion
properties of cu polycrystals prepared by ECAP,
Metall. Mater., 43, pp. 258–271.
A. Loos, R. Rohde, A. Haverich, S. Barlach, 2007. In vitro
and in vivo biocompatibility testing of absorbable
metal stents, Macromol. Symp., 253, pp. 103–108.
G. Mani, M.D. Feldman, D. Patel, C.M. Agrawal, 2007.
Coronary stents: a materials perspective,
Biomaterials, 28, pp. 1689–1710.
Raila, B., Anne, S., Stefanie R., Simon W., Mathias B.,
Svea P., Stephan F., Katrin S., 2014. New stent surface
materials: The impact of polymer-dependent
interactions of human endothelial cells, smooth muscle
cells, and platelets, Acta Biomaterialia, 10, pp. 688–
700.
Dan Song, AiBin Ma, , Jinghua Jiang, Pinghua Lin,
Donghui Yang, Junfeng Fan, Jung-Gu Kim, 2010.
Corrosion behavior of equal-channel-angular-pressed
pure magnesium in NaCl aqueous solution, Corrosion
Science, 52, pp. 481–490.
Werkhoven R. J.,Sillekens W. H., J. B. J. M. van Lieshout,
2011. Effect of surface roughness on the in vitro
degradation behavior of a biodegradable magnesium-
based alloy, Magnesium Technology, pp. 419–422.
C. S. Wu, Z. Zhang, F.H. Cao, L.J. Zhang, J.Q. Zhang,
C.N. Cao, 2007. Study on the anodizing of AZ31
magnesium alloys in alkaline borate solutions, Appl.
Surf. Sci., 253, pp. 3893–3898.
L. Xu, G. Yu, E. Zhang, F. Pan, K. Yang, 2007. In vivo
corrosion behavior of Mg–Mn–Zn alloy for bone
implant application, J. Biomed. Mater. Res. A., 83, pp.
703–711.
Influence of Microstructural Evolution Processed by ECAP on Corrosion Behavior of Pure Magnesium in RPMI-1640 Medium
125