Добавил:
Sekretar
kiopkiopkiop18@yandex.ru
t.me/Prokururor I Вовсе не секретарь, но почту проверяю
Опубликованный материал нарушает ваши авторские права? Сообщите нам.
Вуз:
Предмет:
Файл:Ординатура / Хирургия / Библиотека им академика М.И. Перельмана / Книга_5370_Библиотеки_им_академика_М_И_Перельмана
.pdf
188 Surface Modification of Magnesium and its Alloys for Biomedical Applications
temperature in deaerated, pH 10, 103MNa2SO4solution. They also reported corrosion pits to be the necessary precursor for hydrogen embrittlement, and cracking. Pits
provided bare, active, film-free sites that permit cathodically generated hydrogen to
enter into the metal matrix. In both commercial and high-purity Mg, fracture surfaces
appeared to be predominantly brittle, with the transgranular regions exhibiting typical
quasi cleavage morphology. In contrast, when tested in air, both materials exhibited
dimpled fracture surfaces as characteristic of ductile failure by microvoid nucleation,
growth, and coalescence. Meletis and Hochman (1984) studied the SCC of 99.9% pure
Mg in 3.3 wt% NaCl þ 2 wt% K
CrO4, and reported failure in solution with 2.3% of
2
strain in comparison with 5.1% strain in air, i.e., 55% reduction in strain in solution as
compared to air (Figure 5.6). Fracture surfaces revealed TGSCC with specific features
of cleavage-like fracture, which initiated at corrosion pits and was accompanied by
H evolution.
Magnesium alloys are also reported to be susceptible to SCC in aqueous environments (Ben-Hamu et al., 2008; Bobby Kannan, Dietzel, Raman, & Lyon, 2007; Bobby
Kannan et al., 2008; Makar, Kruger, & Sieradzki, 1993; Uematsu et al., 2012; Winzer
et al., 2008, 2008a, 2008b). The alloying elements play a crucial role in either bene-
ficial or detrimental manner on SCC of an alloy. The roles of most common alloying
elements (aluminium (Al), zinc (Zn), and manganese (Mn)) and impurity (iron (Fe)) on
the SCC of Mg alloys are discussed below.
5.4.2.1 Influence of aluminium
All Al-containing magnesium alloys (such as AZxx, containing Al and Zn, and AMxx,
containing Al and Mn) have been reported to be susceptible to SCC to some extent in
distilled water and chloride-containing solutions (Winzer et al., 2005). The susceptibility to SCC (represented as the minimum stress to cause SCC) was generally found to
150
Air
100
Stress (MPa)
50
0
01
Figure 5.6 Stress versus strain curves for pure Mg specimens tested in air and corrosive solution
(Meletis & Hochman, 1984).
NaCI - K
2345
CrO
2
Strain (%)
4
6
(s)
Mechanical integrity of magnesium alloys for biomedical applications 189
Данная книга находится в списке для перевода на русский язык сайта https://meduniver.com/
140
120
100
Stress (MPa)
80
60
40
1
AZ61
1
10
10
Time to failure
Figure 5.7 Stress versus time-to-failure curves for various Mg-Al alloys in 40 g L
2
Mg-1% Al
AZ80
10
Mg-3% Al
3
Mg-6% AlFe
4
10
1
NaCl þ 40 g L1Na2CrO4(Fairman & Bray, 1971).
increase with increasing Al concentration, as shown in Figure 5.7 (Fairman & Bray,
1971).
It should be noted that the Al is reported to be a neurotoxicant and to cause neuro-
logical disorders such as Alzheimer’s disease and dementia (Crapper, Krishnan, &
Quittkat, 1976; Forbes, Gentleman, & Maxwell, 1995). However, a few in vivo studies
and in vitro cytocompatibility results have suggested the use of Al-containing Mg alloys (e.g., AZ91) to be appropriate for implant applications (Witte et al., 2005; Wong
et al., 2010; Yuen & Ip, 2010). Nevertheless, the avoidance of Al in the Mg-alloy
biomaterials may be quite useful in not only reducing the SCC susceptibility of the
alloys in chloride-containing environments (such as human body fluid), but also in
addressing the commonly perceived toxicity issues due to Al addition.
5.4.2.2 Influence of zinc
Zn also induces SCC susceptibility in Mg alloys (Baker & Avedesian, 1999). It is also
reported that Mg-Zn alloys when alloyed with zirconium or rare earths (but no
aluminium) have intermediate SCC resistance (Winzer et al., 2005).
5.4.2.3 Influence of manganese
Mg-Mn alloys are among the alloys with high resistance to SCC, and they are generally considered to be immune when loaded up to the yield strength in common test
environments (Baker & Avedesian , 1999).
5.4.2.4 Influence of iron
Fe that is found in commercial Mg alloys as an impurity is known to reduce the corrosion resistance by promoting microgalvanic corrosion. The effect of Fe on SCC of Mg
190 Surface Modification of Magnesium and its Alloys for Biomedical Applications
alloys remains unclear. Perrym an (1951) reported that higher Fe content in Mg alloys
decreased the SCC resistance in distilled water, whereas Timonova (1962) reported
that Fe h ad no effect on SCC of Mg-Al-Zn-Mn alloys.
5.4.2.5 Influence of rare-earth elements
Rare-earth (RE) elements are generally added to Mg alloys for improving the creep
resistance and corrosion resistance. It has been reported that Mg alloys containing
RE may have intermediate SCC resistance (Baker & Avedesian, 1999). Rokhlin has
also reported the addition of neodymium or cadmium to MgeZnezirconium (Zr) alloy
to increase the SCC resistance (Rokhlin, 2003). In a very recent study, Padekar,
Singh Raman, Raja, and Paul (2013) showed that an RE-containing EV31A alloy
possesses improved SCC resistance as compared to an RE-free alloy, AZ91E, particularly during const ant load testing (CLT), which was attributed to the formation of a
robust oxide layer on EV31A.
Influence of the surface corrosion film in SCC susceptibili ty is highly dependent on
the alloy composition and/or the nature of the environment, such as the beneficial role
of the robust films formed due to RE elements (as discussed above Padekar et al.
(2013)), or due to chromate-containing aqueous solution (Ebtehaj et al., 1988).
5.4.3 Characterization of SCC using different techniques
There is no generalized approach that could provide a complete insight into prediction
and mechanistic understanding of SCC. Combinations of a few techniques can qualitatively and quantitatively establish the susceptibility of different met als/alloys to
SCC. The main techniques for characterization of SCC are slow strain rate tensile
(SSRT) testing, U-bend testing, C-bend testing, compact tension (CT) testing, double
cantilever beam (DCB) testing (Jones, 1992), circumferential notch tensile (CNT)
testing (Singh Raman, Rihan, & Ibrahim, 2006), etc. A brief description of some of
these testing methods is provided in this section.
5.4.3.1 Slow strain rate tensile testing
SSRT testing is a uniaxial tensile test conducted at extremely low cross-head speeds in
the presence of corrosive environment (Dietzel, Srinivasan, & Atrens, 2011; Jones,
1992). These tests will invariably produce fracture either by SCC or mechanical failure
or both. Slow straining ensures the opportunity for the environment to interact with
crack tip and also the presence of continuing plastic strain, which encourages the initiation and growth of stress corrosion cracks. The SCC susceptibility is evaluated in
terms of the time taken for failure to occur, the extension at failure, and/or morphology
features of the fracture surface. SSRT is an accelerated laboratory testing and generates
data in a relatively short time. Also, SSRT usually gives conservative results because
of the severity of the tests.
Strain rate is a critical parameter in SSRT testing of Mg alloys. For a specific
combination of alloy and environment, SCC occurs in a narrow window of strain rates
)
Mechanical integrity of magnesium alloys for biomedical applications 191
Данная книга находится в списке для перевода на русский язык сайта https://meduniver.com/
(Singh Raman, 2005). Wearmouth et al. (1973) investigated the role of strain rate for
Mg-7Al alloy in aqueous chlorideechromate solution and attributed the role of strain
to rupture of any surface film to allow localized dissolution or H ingress. Ebtehaj et al.
(1988) also investigated the influence of strain rate on the SCC susceptibility of as-cast
Mg-9Al alloy. They proposed that the film integrity was maintained at low stra in rates,
which prevented H ingress into the matrix, and failure occurred in ductile fashion. As
the stra in rate was increased, the film repassivation decreased, allowing H ingress more
freely, causing embrittlement of the matrix. At relatively higher strain rates, ductile
tearing occurred before embrittlement because of insufficient time available for H
ingress. This explained the maximum SCC susceptibility at intermediate strain rates,
as shown in Figure 5.8 (Ebtehaj et al., 1988).
In contrast to the maximum SCC susceptibility of Mg-Al alloy at intermediate strain
rates in NaCl þ K
CrO4solution (Figure 5.8) reported by Ebtehaj et al. (1988), Winzer
2
et al. (2008a) found the maximum susceptibility of a Mg-Al alloy tested in distilled
water at lowest strain rate (Figure 5.9). This was characterized by the greater decrease
in ultimate tensile strength (UTS) and elongation-to-failure in distilled water as
compared to laboratory air in low strain rate regime (Figure 5.9). The occurrence of
maximum SCC susceptibility at intermediate strain rates, as reported by Ebtehaj
et al. (1988), is attributed to the passivating characteristic of the alloy in the
chromate-containing solution used in their study. The corrosion film developed in
this passivating environment required higher strain rate to sustain SCC, whereas a
350
300
250
Maximum normal stress (MPa)
200
5 g / L K2CrO
–7
10
Dry
air
10
Mg-9AI
35 g / L K
20 g / L K2CrO
4
–6
–5
10
Strain rate (s
CrO
2
Figure 5.8 The effect of strain rate on the
maximum nominal stress to fracture the
specimens exposed to dry air or solutions
containing 5 g L
4
4
–4
10
–1
amounts of K
–3
10
1
NaCI and various
CrO4(Ebtehaj et al., 1988).
2
192 Surface Modification of Magnesium and its Alloys for Biomedical Applications
180
160
140
120
100
80
Stress (MPa)
60
40
20
0
0
H2O, 5 × 10
H2O, 3 × 10
H2O, 10
H2O, 3 × 10
0.01
0.02 0.03 0.04 0.060.05 0.07 0.08 0.09
Apparent strain
–7 s–1
–7 s–1
–8 s–1
Air, 10
–7 s–1
–4 s–1
Figure 5.9 Stress versus apparent strain curves for AZ91 in distilled water and air (Winzer et al.,
2008a).
less passivating solution in Winzer and co-workers’ study (Winzer et al., 2008a) produced corrosion film that could be disrupted at lower strain rate and caused SCC.
In a recent study (Cho udhary & Singh Raman, 2012) on SCC of AZ91D in m-SBF,
SSRT tests were conducted at a strain rate of 2.2 10
7s1
under different conditions: (a) strained in air, (b) strained in m-SBF solution, (c) immediately strained in
air after preimmersion in m-SBF solution for 28 h, and (d) continuously cathodically
charged and simultaneously pulled in m-SBF. The purpose of these tests were to investigate: (1) the occurrence of SCC and its mechanism and (2) whether the loss of the
mechanical property is indeed a result of the synergistic effect of stress and corrosive
environment (i.e., SCC) or it can simply be attributed to the continuously reducing
cross-sectional area of the specimen due to high corrosion rate of Mg alloys. The stress
versus time curves for AZ91D alloy under these conditions are shown in Figure 5.10.
AZ91D tensile specimen, which was strained continuously in m-SBF at strain rate of
2.2 10
7s1
(Figure 5.10(b)), failed in 28 h. Accordingly, a preimmersion time of
28 h was selected for straining the preimmersed AZ91D tensile specimen in air,
assuming the stress-independent corrosion da mage to be similar to that of specimen
strained in m-SBF solution. The specimen pulled in m-SBF solution
(Figure 5.10(b)) showed a considerable reduction in mechanical properties as
compared to the specimen pulled in air (Figure 5.10(a)). The stressetime curve for
the specimen preimmersed in m-SBF for 28 h and then strained in air
(Figure 5.10(c)) was very similar to the specimen pulled in air (Figure 5.10(a)). Therefore, it is fair to infer that the simultaneous effect of stress and corrosive environment
(i.e., SCC) was the primary cause of the loss of mechanical property in the case of
specimen pulled in m-SBF (Figure 5.10(b)), and the stress-independent corrosion
contributed only marginally (as the plots suggest). Also, a considerable loss of the
mechanical strength and a decrease in the time-to-failure were observed in continuous
Mechanical integrity of magnesium alloys for biomedical applications 193
Данная книга находится в списке для перевода на русский язык сайта https://meduniver.com/
180
160
140
120
100
80
Stress (MPa)
60
40
20
0
(a) Air
(b) m-SBF
(c) Pre-immersed
(d) Continuous cathodic charging
0
10 20 30 40 50
acdb
60
Time (h)
Figure 5.10 Stress versus time plots of AZ91D tested at a strain rate of 2.2 107s1: (a) in
air, (b) in m-SBF solution, (c) preimmersed in m-SBF for 28 h and then immediately strained in
air, and (d) continuously charged at 200 mV cathodic to open circuit potential in m-SBF during
the test (Choudhary & Singh Raman, 2012).
cathodic charging conditions (i.e., when anodic dissolution is minimal,
Figure 5.10(d)). Thus, the cathodic charging experiment inferred hydrogen-assisted
stress corrosion cracking (HASCC) as one of the SCC mechanisms. However, the
maximum loss in mechanical property was observed in the case of the specimen
strained in m-SBF solution at open circuit conditions (Figure 5.10(b)), which indicated
possibly of a combined effect of HASCC and anodic dissolution. This combined
mechanism may be plausible for Mg and its alloys that are known to invariably
produce considerable amount of hydrogen during anodic dissolution (Song & Atrens,
1999; Song, Atrens, John, et al., 1997; Song, Atrens, Stjohn, et al., 1997).
The SSRT results of AZ91D tested under different conditions were further examined by fractography. The overall fracture surface of the specimen tested in air is
shown in Figure 5.11(a). The entire surface of the specimen tested in air revealed
dimples at higher magnification (Figure 5.11(b)), confirming the mechanical overload failure. There were also occasional features of brittle fracture, which may be
associated with secondary phase particles of the alloy (predominantly b-phase).
Mix-mode fracture observed in air was consistent with the fractographic features
for the Mg alloy tested in air reported in the literature (Ben-Hamu et al., 2008;
Bobby Kannan et al., 2008). In contrast, the overall fracture surface of th e specimen
tested in m-SBF showed the localized attack at the specimen circumference
(Figure 5.11(c)). At higher magnification, the fractography of this specimen revealed
distinctive feature of SCC such as transgranular cracking and localized cracks
(Figure 5.11(d)).
194 Surface Modification of Magnesium and its Alloys for Biomedical Applications
(a)
(c)
(b)
(d)
(e) (f)
Figure 5.11 Fractograph of the specimen tested in air (a, b), fractograph of the specimen failed
in m-SBF at a strain rate of 2.2 10
m-SBF solution for 28 h and then strained in air, and (f) continuously cathodically charged and
simultaneously pulled in m-SBF (Choudhary & Singh Raman, 2012).
7s1
(c, d), fractographs for the specimens: (e) immersed in
The fractograph for the specimen immersed in m-SBF for 28 h followed by straining
in air showed dimples formation (Figure 5.11(e)) similar to that observed for the specimen tested in air, suggesting the ductile failure. The specimen that was cathodically
charged and simultaneously pulled in m-SBF solution exclusively showed the evidences of the transgranular cracking (arrows, Figure 5.11(f)). Transgranular cracking
observed in the case of cathodically charged conditions was attributed to the mechanism involving hydrogen. In fact, the hydrogen generated during the cathodic charging
diffuses ahead of the crack tip and embrittles the matrix (Winzer et al., 2005).
Mechanical integrity of magnesium alloys for biomedical applications 195
Данная книга находится в списке для перевода на русский язык сайта https://meduniver.com/
5.4.3.2 Fracture mechanics-based approach to SCC
The fracture mechanics-based approach is used for quantitative determination of
important design parameters for the components susceptible to SCC. This type of
study is of particular importance for the medical implant components and materials
for the determination of safe service stress level in physiological conditions. In general,
damage-tolerant design methods are applied for performing these studies, which
assume that structure/component contains initial cracks. The concept of fracture
mechanics is then used to characterize the conditions for propagation of cracks from
these initial flaws. Usually, SCC can occur at stresses below general yield strength
and propagate in elastic body; hence, linear elastic fracture mechanics (LEFM) principles are applied for determining the crack tip stress intensity factor (K
et al., 2011).
According to LEFM, the K
at the crack tip can be given by Eqn (5.1):
I
)(Dietzel
I
¼ YsðpaÞ
K
I
1=2
(5.1)
where Y is a geometrical factor, s is the applied stress, and a is the crack length.
For a given material/environmental condition, the typical schematic of relationship
between stress intensity and crack growth rate is shown in Figure 5.12 (Dietzel et al.,
2011). The value of K
terized by threshold stress intensity for SCC (K
any crack propagation at stress intensities lower than the K
Log crack growth rate da / dt
at which the first measurable crack extension occurs is charac-
I
). Consequently, there will not be
ISCC
. The practical meaning
ISCC
Sub-critical crack growth
12 3
K
K
ISCC
Stress intensity factor K
I
(air)
IC
Figure 5.12 Typical plot of stress intensity factor versus crack growth rate (Dietzel et al., 2011).
196 Surface Modification of Magnesium and its Alloys for Biomedical Applications
of K
become insignificantly low (i.e., <10
lies in the fact that below this stress intensity factor, crack growth rates
ISCC
12ms1
)(Singh Raman & Pal, 2011; Winzer
et al., 2005).
The plot in Figure 5.12 is specifically characterized by three zones: region I (crack
initiation and crack propagation), region II (steady state crack propagation), and region
III (failure by overloading). During region I, the crack growth increases rapidly as the
stress intensity is increased from the K
. The crack veloci ty is reasonably constant
ISCC
in region II and largely independent of the stress intensity factor. As the stress intensity
factor approaches K
(i.e., fracture toughness, the critical stress intensity of the spec-
IC
imen tested in air at which it fails), pure mechanical rupture (ductile) dominates over
subcritical crack extension caused by SCC, and rapid crack growth occurs. Determinations of K
and crack growth rate data for life prediction of components/structure
ISCC
are of prime importance for various applications, such as implants.
K
of AZ91D alloy in m-SBF was determined by the authors’ research group
ISCC
(Choudhary & Singh Raman, 2012), using circumferential notch tensile (CNT) testing.
The fatigue precracked CNT specimens of AZ91D alloy were loaded at different stress
intensities in m-SBF at 37
ing on the applied stress intensities. The K
C. Specimens failed after different time durations depend-
at each applied load was calculated
I
(following the procedure as described in references Choudhary and Singh Raman
(2012), Pal, Ibrahim, and Singh Raman (2011), Rihan, Raman, and Ibrahim (2005))
and plotted against the time-to-failure (Figure 5.13). As evident from Figure 5.13,
time-to-failure increased exponentially with the decreasing stress intensity. For
example, a CNT specimen loaded at a stress intensity of 11.7 MPa m
only 16 h, whereas it took 119 h for a specimen loaded at a stress intensity of
6.8 MPa m
1/2
to fail. The threshold stress intensity was determined by drawing an
1/2
failed in
14
12
)
½
10
(MPa m
8
I
K
6
K
ISCC
4
0 200 600 800
400
(h)
t
f
1000
Figure 5.13 KIversus tfplot of AZ91D alloy tested in m-SBF at 37C(Choudhary and Singh
Raman (2012)).
Mechanical integrity of magnesium alloys for biomedical applications 197
Данная книга находится в списке для перевода на русский язык сайта https://meduniver.com/
asymptote to time-to-failure axis in the plot of stress intensity versus time-to-failure
(Figure 5.13).
The equation of the fitted curve can be given by Eqn (5.2):
t
K
¼ 7:58 exp
I
f
123:58
þ 5:18 (5.2)
According to Eqn (5.2), the value of K
time-to-failure axis in Figure 5.13 (i.e., K
when producing a horizontal asymptote to
I
), was 5.18 MPa m
ISCC
1/2
for AZ91D alloy in
m-SBF. Fracture surfaces of CNT specimens were observed under the scanning electron microscope (SEM) after cleaning of corrosion products. Figure 5.14(a) showed
the overall fractograph of specimen that failed at a stress intensity of 6.8 MPa m
1/2
Fracture surface consists of four regions: machined notch, fatigue precracked region,
SCC zone, and mechanical overload failure zone. At higher magnifications, distinctive
features of each of these regions become clearer. Figure 5.14(b) shows the transition
from the fatigue precracked region to the SCC zone. The fatigue precracked region
(a) (b)
(c)
.
Figure 5.14 SEM fractographs of the specimen tested at KI¼ 6.8 MPa m
fracture surface showing machined notch, fatigue precrack, SCC and mechanical failure zones,
(b) fractograph showing fatigue precracked and SCC zones (arrow indicating the crack
propagation direction), and (c) mechanical failure zone showing the dimples (Choudhary &
Singh Raman, 2012).
1/2
: (a) the overall
Соседние файлы в папке Библиотека им академика М.И. Перельмана