47533 Westinghouse Drive Fremont, California 94539 t 510.683.2000 f 510.683.

2001
We are Nitinol.

www.nitinol.com









Lffect of 1emperature and pn on the Corros|on kes|stance of N|t|no|

ChrlsLlne 1repanler and Alan 8. elLon

roceedlngs of SMS1-2004, ed. M. MerLmann, 8aden-8aden, Cermany, 361-366 (2006).

2006
361






EFFECT OF TEMPERATURE AND pH ON THE CORROSION
RESISTANCE OF PASSIVATED NITINOL AND STAINLESS STEEL

Christine Trpanier and Alan R. Pelton

Nitinol Devices and Components, Fremont, 94539, California, USA











ABSTRACT
The goals of this study were to assess the influence of temperature and pH on the corrosion
resistance of Nitinol. Electropolished NiTi and passivated 316L stainless steel wires were tested
in Hanks physiological solution at temperatures ranging from 10C to 80C and pH varying from
1 to 9. Results show that the ability of Nitinol to repassivate is reduced by an increase in
temperature. No other major effects were observed by a change in temperature. Stainless steel
corrosion rate, on the other hand, was the only parameter affected by an increase in temperature.
Study of the effect of pH revealed that except for a change in the potential for oxygen evolution,
other aspects of the corrosion resistance of NiTi were not affected. In contrast, the results for
stainless steel indicate that both uniform corrosion rate and localized corrosion are affected by a
decrease in pH.


INTRODUCTION
Temperature can affect the corrosion behavior of materials in different ways [1]. If the corrosion
rate is only controlled by the metal oxidation process, the corrosion rate will increase
exponentially with an increase in temperature following an Arrhenius relationship. This is the
underlying principle for exposing implant devices to physiological solutions at temperatures much
higher than normal body temperature. By increasing the temperature (and corrosion rate) the test
is accelerated and can therefore be used in a laboratory setting to obtain information on materials
corrosion resistance in a relatively short period of time. Furthermore, a change in temperature can
also affect the resistance of a material to localized corrosion (pitting). To date, few studies have
investigated the effect of temperature on Nitinol in a physiological environment.

Proceedings of the International Conference on Shape Memory and Superelastic Technologies
October 37, 2004, Kurhaus Baden-Baden, Baden-Baden, Germany
Copyright 2006 ASM International

362
The pH of a solution is an important factor in the corrosion resistance of materials because
hydrogen ions can interact and modify the surface and therefore alter the corrosion resistance of a
material. Moreover, interactions between pH and dissolved species such as chloride ions can
enhance the effect of hydrogen ions. Still, most corrosion studies of implant devices have
investigated the corrosion resistance of Nitinol at a physiological pH of 7.4 without further study
of the effect of other pH. However, the literature reports that physiological pH can range from 1,
such as in gastric secretions with high concentration of chloride ions, up to 9, such as in Brunners
gland secretions [2]. Therefore, the goals of this study are to assess the influence of temperature
and pH on the corrosion resistance of Nitinol and stainless steel.


EXPERIMENTAL METHODS
Electropolished Nitinol (50.8%at. Ni) wire (NDC, Fremont, CA) and passivated (ASTMF86)
316L stainless steel wire (Fort Wayne Metals, Fort Wayne, IN) were used in this study [3]. The
corrosion test method follows ASTM F2129 [4]. To assess the effect of temperature on the
corrosion resistance of Nitinol and stainless steel, Hanks solution (pH of 7.4) was cooled down or
heated up using a temperature controlled water bath to 10C, 20C (room temperature), 37C,
50C, 60C, 70C and 80C. To assess the effect of pH, sufficient quantity of 1N HCl or NaOH
were added to 37C Hanks solution to obtain pH values of 1.0, 7.4 and 9.0. In order to stabilize
the pH, the Hanks solution was modified according to ASTM F2129 by adding 1.45g/L NaHCO
3

and by de-aerating the solution with a mixture of 5% CO
2
in nitrogen.

Unless otherwise specified, three samples were tested for each material and condition. The
corrosion resistance of the samples was characterized in terms of their breakdown potential (E
b
)
and repassivation potential (E
p
). In addition, the corrosion current density (I
corr
) was also
determined. After testing the samples were inspected with an optical microscope (10-60X).


EXPERIMENTAL RESULTS
A summary of the corrosion resistance of Nitinol tested between 10C and 80C is presented in
Table 1 and in Fig. 1. Based on the shape of the polarization curves and inspection of the
specimens after the tests, E
b
coincides with the potential for oxygen evolution for samples tested
between 10C and 70C. Therefore, the small shift in E
b
for samples tested between 10C and
70C appears to be related to a change in the potential at which oxygen is being produced on the
sample rather than to a difference in the corrosion resistance of the material. It is important to note
the small hysterisis between E
b
and E
p
indicating superior ability of the material to repassivate and
low susceptibility to crevice corrosion. Starting at 60C, the samples start to exhibit a different
corrosion behavior. Although E
b
appears to still overlap with oxygen evolution, the ability of the
material to repassivate after breakdown of the oxide layer is progressively reduced. The samples
tested at 80C are characterized by a large hysterisis between E
b
and E
p
suggesting that the
material is less able to repassivate and more susceptible to crevice corrosion at that temperature.

EFFECT OF TEMPERATURE AND pH ON THE CORROSION RESISTANCE 363


Temp.
[C]
E
b

[mV vs SCE]
E
p

[mV vs SCE]
I
corr
[nA/cm
2
]
10 107630 101070 21
20 99852 88920 31
37 90213 8456 54
50 85242 7958 32
60 8366 74814 31
70 77823 54251 81
80 79244 -39309 106
Tab. 1: Corrosion test results for Nitinol at different temperatures (meanSD)




Fig. 1: Variation of E
b
, E
p
and I
corr
as a function of temperature (Nitinol).




The corrosion current density, which is directly related to the corrosion rate, appears to slowly
increase as a function of test temperature (refer to Fig. 1). However, because I
corr
are all within the
same decade, it cannot be concluded that this increase in I
corr
is significant within the temperature
range studied. These results suggest that temperature affects the ability of Nitinol to repassivate.
However, no other effects, such as on the corrosion rate of the material, were observed with a
change in temperature.

The corrosion test results of stainless steel tested between 10C and 80C are presented in Table 2
and in Fig. 2. As the temperature is increased, a shift in the polarization curve towards lower
potentials, similar to the one observed for NiTi, is seen for stainless steel. However, in contrast to
NiTi, the breakdown potential of the oxide layer (E
b
) does not correspond to the potential for
oxygen evolution. Consequently, variation of E
b
indicates a change of the corrosion resistance of
the material to localize corrosion. The distribution E
b
as a function of temperature suggests that E
b

364
is greater for the 10C and 20C groups than for the 37C-80C groups. Statistical analysis
comparing the breakdown potential as a function of temperature using one-way ANOVA and
Tukeys pairwise comparison show only a significant difference in the breakdown potentials
between the 20C and 37C group. Therefore, we cannot conclude that temperature has an effect
on the breakdown potential of stainless steel. Analysis of the variation of I
corr
as a function of
temperature reveals a strong correlation between the two variables. In contrast to NiTi, the
corrosion current density of stainless steel (and consequently the corrosion rate) increases
exponentially as a function of temperature. This result suggests that stainless steel corrosion rate
may be controlled by the metal oxidation process and that it may follow an Arrhenius law. No
other effects were observed by a change in temperature.

Temp.
[C]
E
b

[mV vs
SCE]
E
p

[mV vs SCE]
I
corr
[nA/cm
2
]
10 89018 n/a 42
20 91750 15616* 21
37 59686 n/a 31
50 740128 n/a 41
60 761188 n/a 93
70 635135 n/a 91
80 65067 -447** 296
*2 of 3 samples repassivated **1 of 3 samples repassivated
Tab. 2: Corrosion test results for stainless steel at different temperatures (meanSD)




Fig. 2: Variation of E
b
and I
corr
as a function of temperature (stainless steel).

EFFECT OF TEMPERATURE AND pH ON THE CORROSION RESISTANCE 365

These results suggest that Nitinol and stainless steel have different corrosion mechanisms
(localized corrosion vs. uniform corrosion processes respectively) that are affected by an
increased temperature.

A summary of the corrosion resistance of Nitinol tested at pH of 1.0, 7.4 and 9.0 is presented in
Table 3. Polarization of NiTi in Hanks solution at a pH of 1 resulted in a shift of the curve toward
greater values of potential. Similar to the previous study, E
b
coincides with oxygen evolution.
Therefore, the small shift in E
b
can be related to a change in the potential at which oxygen is
being produced in the solution. This result is in agreement with the Pourbaix diagram for water:
more acidic pH shifts the oxygen evolution reaction to higher potentials [5]. The variation in E
p
is
similar to E
b
so the ability to repassivate of the material is not affected. No major differences
could be found between the samples tested at 7.4 and 9.0. Furthermore, no differences were found
in the corrosion current densities between all groups.

pH E
b

[mV vs SCE]
E
p

[mV vs SCE]
I
corr
[nA/cm
2
]
1.0 13347 12876 113
7.4 107910 9256 61
9.0 10634 94335 61
Tab. 3: Corrosion test results for Nitinol tested at different pH (meanSD)



Therefore, except for a change in the potential for oxygen evolution, the corrosion resistance of
NiTi was not affected by a variation in pH. This result is in agreement with the findings from
another study showing that titanium is passive down to a pH of 0 in HCl solution [6].
Electropolished NiTi is protected by a titanium oxide layer similar to the one found on titanium
which may explain the similar resistance to corrosion in low pH solutions containing chloride ions
[7].

The corrosion test results for stainless steel tested at pH of 1.0, 7.4 and 9.0 are presented in Table
4 and in Figure 8. The polarization curves obtained for stainless steel tested at a pH of 1.0 indicate
that the material was strongly affected by the difference in pH. A decrease in the breakdown
potential and increase in the corrosion current density (which translates to corrosion rate) of the
specimens were observed. It is interesting to note that although the general corrosion resistance of
the material was severely deteriorated by exposure to the acidic solution, the material was able to
repassivate (presence of E
p
). Repassivation may have been possible because the samples in this
group were polarized to low potentials. Repassivation of the material at a pH of 7.4 and 9.0 was
not observed. Increase in the pH from 7.4 to 9.0 did not affect the corrosion resistance of the
material.

These results suggest that Nitinol and stainless steel are affected differently by a decrease in pH.
The titanium oxide layer on Nitinol appears more resistant to a variation in pH than the chromium
oxide covering stainless steel.
366

pH E
b

[mV vs SCE]
E
p

[mV vs SCE]
I
corr
[nA/cm
2
]
1.0 39219 -4275* 11599
7.4 87189 n/a 129
9.0 78694 n/a 132
*3 of 4 samples repassivate
Tab. 4: Corrosion test results for stainless steel at different pH (meanSD)


CONCLUSIONS
This study investigated the effect of temperature and pH on the corrosion resistance of passivated
Nitinol and stainless steel implant materials. The results show that:
An increase in temperature affects the resistance to localized corrosion of Nitinol by
reducing the ability of the material to repassivate. The uniform corrosion rate of Nitinol
was not affected by temperature.
The uniform corrosion resistance (corrosion rate) of stainless steel, however, is affected by
temperature. No significant effects were observed for the resistance of the material to
localized corrosion (pitting).
Except for a change in the potential for oxygen evolution, the corrosion resistance of NiTi
was not affected by a variation in pH.
Both uniform corrosion resistance and localized corrosion (pitting) are affected by a
decrease in pH for stainless steel.


REFERENCES
[1] Corrosion, ASM Handbook, Volume 13, ASM International, 1987, p.13.
[2] A.C. Guyton, J.E. Hall, in Textbook of medical physiology, 9th edition, W.B. Saunders Co.,
1996.
[3] F86-91, in Annual Books of ASTM Standards, Medical Devices and Services, vol. 13.01,
1991.
[4] F2129, in Annual Books of ASTM Standards, Medical Devices and Services, vol. 13.01,
2004.
[5] D.C. Silverman, in Derivation and Application of EMF-pH Diagrams, Electrochemical
Techniques for Corrosion Engineers, Ed. NACE Inter., 1986.
[6] R.J. Chin, K. Nobe, in Electrochem Soc, 119(11), 1972, p.1457.
[7] C. Trepanier, M. Tabrizian, LH. Yahia, L. Bilodeau, D.L. Piron, in J Biomed Mat Res (Appl
Biomater) 43, 1998, p.433.