Materials Transactions, Vol. 43, No. 12 (2002) pp.

3100 to 3104
Special Issue on Biomaterials and Bioengineering

2002
c The Japan Institute of Metals

Corrosion Behavior of Nickel-Free High Nitrogen Austenitic Stainless

Steel in Simulated Biological Environments
Daisuke Kuroda1 , Sachiko Hiromoto1 , Takao Hanawa1 and Yasuyuki Katada2
1
Biomaterials Center, National Institute for Materials Science, Tsukuba 305-0047, Japan
2
Steel Research Center, National Institute for Materials Science, Tsukuba 305-0047, Japan

The corrosion resistance of the nickel-free high nitrogen austenitic stainless steel without manganese, Fe–23Cr–2Mo–1.5N (mass%)
(HNS) as biomaterials, was evaluated by the polarization test in various electrolytes: 0.9%NaCl solution (saline), phosphate buffered saline
(PBS(-)), Hanks’ solution (Hanks) and Eagle’s minimum essential medium (E-MEM). Conventional austenitic stainless steel, 316L, was also
polarized for comparison. The both alloys were spontaneously passivated in all electrolytes. The HNS didn’t show pitting corrosion in the
polarization range in all electrolytes although the 316L showed pitting corrosion. Passive current densities of the HNS in all electrolytes were
lower than those of 316L. Therefore, the HNS shows higher passivity and resistance to pitting corrosion than 316L. The passive current density
in Hanks of HNS was lower than that in saline, indicating that the protectiveness of surface oxide film increased with the existence of inorganic
ions such as phosphate and calcium ions. On the other hand, the passive current density in E-MEM was higher than that in Hanks, but was
lower than that in saline. Consequently, the HNS must show high corrosion resistance in vivo and be a promising biomaterials.

(Received June 27, 2002; Accepted September 20, 2002)

Keywords: nickel-free high nitrogen austenitic stainless steel, biological environment, polarization behavior, biomaterial

1. Introduction denum and nitrogen. Ultimate tensile strength and elonga-

Austenitic stainless steel, 316L, is used for implant mate-
rials such as artificial joint, bone plate and stent because of
good combinations of strength and corrosion resistance. 316L
and conventional austenitic stainless steels contains a large
amount of nickel because nickel is attractive austenite stabi-
lizing element for stainless steels. In addition, 316L have a
large amount of chromium and a small amount of molybde-
num to enhance the strength and corrosion resistance, and to
be non-magnetic materials.
However, the safety, strength and corrosion resistance of
316L are sometimes insufficient for biomedical use. Im-
plant devices such as bone plate and artificial hip joint made
of 316L are sometimes fractured in human body with cor-
rosion, fretting fatigue and so on. The reports of metallic
allergy and toxicity caused by nickel and its salts have re-
cently increased.1–6) The coronary in-stent restenosis caused
by nickel ion released from 316L stent was reported.1) There-
fore, removal of nickel from stainless steel is eagerly de-
manded.
Nickel-free austenitic stainless steels having a large
amount of chromium, manganese, nitrogen and a small
amount of molybdenum, such as Fe–15Cr–(10–15)Mn–
4Mo–0.9N (mass%), Fe–18Cr–18Mn–2Mo–0.9N (mass%),
Fe–(15–18)Cr–(10–12)Mn–(3–6)Mo–0.9N (mass%) and Fe–
(19–23)Cr–(21–24)Mn–(0.5–1.5)Mo–(0.85–1.1)N (mass%)
(BioDur
108), are currently developed to enhance the
strength, corrosion resistance and biocompatibility of stain-
less steels.7–10) In those cases, nitrogen and manganese are
employed instead of nickel to obtain austenitic phase, be-
cause nitrogen and manganese also stabilize austenitic phase
and control mechanical properties of stainless steel as same as
nickel. The strength of nickel-free austenitic stainless steels
depends on nitrogen content, while corrosion resistance is
strongly related to the gross amount of chromium, molyb-
tion to fracture of Fe–(19–23)Cr–(21–24)Mn–(0.5–1.5)Mo–
(0.85–1.1)N (mass%) are over 931 MPa and less than 49%,
respectively, in annealed condition.9) Corrosion resistance of
the alloy is higher than that of nickel containing austenitic
stainless steels.9) Those results indicate that nickel-free stain-
less steel seems to be a mechanically and chemically biocom-
patible material.7–10) However, the reports of rather high tox-
icity of manganese and its salts have recently increased.2, 4, 5)
Thus the nickel and manganese-free stainless steel is neces-
sary for biomedical use.
Nickel-free austenitic stainless steel without manganese,
Fe–23Cr–2Mo–1.5N (HNS) (mass%), is also developed.11, 12)
The nitrogen content of the HNS is higher than that of other
nickel-free austenitic stainless steels leading to the removal
of manganese. Ultimate tensile strength and elongation to
fracture of the HNS are over 1200 MPa and over 40%, re-
spectively, in annealed condition.12) In addition, corrosion re-
sistance of the HNS in artificial seawater is higher than that
of 316L.11–13) Therefore, HNS may be a promising biomate-
rial because of its nickel-free and manganese-free composi-
tion and high strength and corrosion resistance. However, the
corrosion resistance of HNS in biological environment has
never been characterized.
In this study, anodic polarization test of HNS and 316L in
various aqueous solutions was performed to evaluate the cor-
rosion behavior of HNS as biomaterials. Effects of inorganic
ions and amino acids on the corrosion behavior of HNS were
also evaluated.
2. Experimental Methods
2.1 Specimens
A practical size ingot of HNS with a weight of 20 kg was
fabricated by a nitrogen gas pressurized electro-slag remelt-
ing process11) and was hot forged to plates with 1.5 × 10−3 m
 Corrosion Behavior of Nickel-Free High Nitrogen Austenitic Stainless Steel in Simulated Biological Environments 3101

Table 1 Chemical compositions of HNS and 316L (mass%).

C Si Mn P S Ni Cr Mo Al O N Fe
HNS 0.034 0.07 0.13 0.009 <0.001 0.04 23.50 2.04 0.015 0.002 1.34 Bal.
316L 0.019 0.48 1.18 0.038 0.13 12.1 16.72 2.05 — — — Bal.

Table 2 Compositions of PBS(-), Hanks and E-MEM. 3. Results

Concentration/kmol m−3
Composition 3.1 Microstructure
PBS(-) Hanks E-MEM
X-ray diffraction profiles of the HNS and 316L are shown
NaCl 1.37 × 10−1 1.37 × 10−1 1.16 × 10−1 in Figs. 1(a) and (b), respectively. The HNS shows diffraction
KCl 2.68 × 10−3 5.37 × 10−3 5.36 × 10−3 peaks of only austenitic phase. On the other hand, the 316L
CaCl2 1.26 × 10−3 1.80 × 10−3 shows large peaks of austenitic phase and small peaks of fer-
Na2 HPO4 8.18 × 10−3 4.23 × 10−4 8.98 × 10−4 ritic phase. Therefore, the HNS consists with only austenitic
KH2 PO4 1.47 × 10−3 4.41 × 10−4
phase. On the other hand, the 316L consists with austenitic
Mg2 SO4 7.39 × 10−4 8.11 × 10−4
phase containing a small amount of ferritic phase.
NaHCO3 4.17 × 10−3 2.38–2.62 × 10−2
Amino acid 5.5 × 10−3
3.2 Polarization behavior
The polarization curves of HNS and 316L in saline, PBS(-),
Hanks and E-MEM are shown in Figs. 2(a)–(d), respectively.
thickness. The HNS plates with 15 mm×15 mm×1.5 mm was The both alloys showed the same E open values and were spon-
employed as specimens. For comparison, a commercial 316L taneously passivated in all electrolytes. The HNS showed pas-
disk with 2.0 × 10−2 m in diameter and 1.5 × 10−3 m in thick- sive current density (Ips ) up to 1000 mV and the gradual in-
ness was also employed. Chemical compositions of HNS and crease in current density over 1000 mV in all electrolytes. The
316L are listed in Table 1. Microstructures of HNS and 316L 316L also showed Ips up to 250 mV followed by the abrupt
were identified by X-ray diffraction analysis with Cu–Kα ra- increase in current density. The Ips values of the HNS were
diation. Specimen was metallographically polished to mirror lower than those of the 316L in all environments although
finish and rinsed in acetone and de-ionized water. The speci- difference between HNS and 316L in saline, PBS(-) and E-
men for polarization test was coated with epoxy resin except MEM were small. In the case of HNS, the Ips at 500 mV in
for a measurement area of 1.0 × 10−4 m2 and was ultrasoni- PBS(-) and Hanks were constant and slightly lower than that
cally rinsed in ethanol for 1.8 × 102 s. in saline. On the other hand, the Ips at 500 mV in E-MEM
was slightly higher than those in saline, PBS(-) and Hanks.
2.2 Electrolyte However, the current density at 0 mV in E-MEM was lower
0.9%NaCl solution (saline), phosphate-buffered saline than those in saline, PBS(-) and Hanks.
(PBS(-)), Hanks’ solution (Hanks) and Eagle’s minimum es-
sential medium (E-MEM) were employed as electrolytes. The 3.3 Surface morphology
compositions of PBS(-), Hanks and E-MEM are summarized Optical micrographs of the HNS and 316L before and after
in Table 2. Hanks consists with only inorganic ions whose the polarization test in E-MEM are shown in Figs. 3(a)–(d),
composition is similar to that of body fluid. E-MEM consists respectively. Only scratches by polishing were observed on
with inorganic ions and amino acids whose composition is the surface of HNS and 316L before polarization test (Figs.
also similar to those of the body fluid. The electrolytes were 3(a) and (c)). After polarization, the HNS still showed only
deaerated with bubbling N2 gas at a rate of 8.3 × 10−7 m3 s−1 scratches (Fig. 3(b)). However, the 316L showed pits with a
for 3.6 × 103 s and were kept at 310 K. diameter of 100–300 µm and corrosion products covering the
surface (Fig. 3(d)). Similar surface morphology was respec-
2.3 Polarization tively observed on HNS and 316L after the polarization in all
Three electrode type electrolytic cell with a saturated solutions.
calomel electrode (SCE) as a reference electrode and a plat-
inum plate as a counter electrode was used for the polariza- 4. Discussion
tion test. The specimen was immersed in the electrolyte for
1.8 × 103 s to allow for its open-circuit potential (E open ) be- 4.1 Corrosion behavior of HNS and 316L
ing stabilized. Subsequently, potential of the specimen was The HNS and 316L were spontaneously passivated in all
anodically swept from E open at a rate of 3.33 × 10−1 mV s−1 . solutions, indicating that the alloys form passive film in bio-
The sweep was stopped when the current density abruptly in- logical environment. Protectiveness of surface oxide film of
creased over 1.0 A m−2 . Polarization was performed at least the HNS is higher than that of the 316L because the Ips val-
two times under each condition. The surface of the HNS and ues of the HNS were lower than those of 316L in all solutions
316L after the polarization in E-MEM was observed with op- although the differences of Ips between HNS and 316L were
tical microscope. small in saline, PBS(-) and E-MEM. Chromium concentra-
tion in surface oxide film of the HNS was higher than that
3102 D. Kuroda, S. Hiromoto, T. Hanawa and Y. Katada
Fig. 1 X-ray diffraction profiles of (a) HNS and (b) 316L.
of 316L after immersion in Hanks.14) In addition, the surface
oxide on the HNS contained nitrogen that is suggested to en-
hance the resistance to pitting corrosion.15, 16) Therefore, the
high passivity of the film of the HNS is probably caused by
the high concentration of chromium and/or the existence of
nitrogen in the surface oxide film. On the other hand, the
difference between the HNS and 316L in Ips in E-MEM was
small (Fig. 2(d)) probably because the chromium concentra-
tion in the surface oxide film of the HNS immersed in E-MEM
containing fetal bovine serum (FBS) was as same as that of
the 316L.14) The protectiveness of the surface oxide film of
the HNS in biological environment seems to depend on the
chromium concentration in the surface oxide film.
316L showed pitting corrosion with the polarization over
the potential where abrupt increase in current density oc-
curred. Pitting potential (E pit ) on the polarization curves of
the 316L in each electrolyte ranged from 300 to 400 mV, that
was as same as values of E pit in the other report on 316L.10)
The 316L in this study contains a small amount of ferritic
phase (Fig. 1(b)), however the influence of ferritic phase on
resistance to pitting corrosion is negligible.
The HNS did not show abrupt increase in current density in
the polarization range in this study (Fig. 2) and no pit was ob-
served on the surface after the polarization (Fig. 3), indicating
that pitting corrosion hardly occurs on the HNS in vivo. The
increase in current density over 1000 mV is probably caused
by oxygen evolution reaction according to the pH-potential
diagram.17) Therefore, the HNS has the advantage over the
316L in high resistance to pitting corrosion in biological en-
vironments. The higher chromium concentration in HNS than
that in 316L is one reason for high resistance to pitting corro-
sion, because resistance to pitting corrosion of stainless steel
increases with the increase in chromium content.18) Moreover,
following effects of nitrogen are suggested to explain the high
resistance of HNS to pitting corrosion: the nitrogen-enriched
layer at oxide/metal interface may stabilize the surface oxide
film against chloride ion,16, 19–23) anodic dissolution inside pit
maybe inhibited by neutralization of Cl− with NH+ 4 at the ini-
tial stage of pitting15) and nitrate ion possibly formed plays as
an inhibitor to pitting corrosion by reacting with iron ion.20)
On the surface of the HNS immersed in Hanks, enrichment
of nitrogen at oxide/metal interface was also observed by an
Auger electron spectroscopy.14) Hence, the nitrogen-enriched
layer probably plays an important role in the high resistance
to pitting corrosion in biological environment as well as in the
sea water and so on.16, 19–23)
4.2 Effects of inorganic ions and amino acids on HNS
The current density at 0 mV, near E open , is influenced by
both the protectiveness of surface oxide film and the mag-
nitude of cathodic current density because both anodic and
cathodic reactions always concurrently occur in corrosion re-
action especially near E open . Thus, the decreased in current
density at 0 mV is explained by the increase in the protec-
tiveness of surface oxide film and the increase in cathodic re-
action. On the other hand, the Ips at 500 mV reflected the
protectiveness of surface oxide film because cathodic current
density was small enough not to influence the constant an-
odic current density. In Hanks, the current densities at 0 mV
and 500 mV decreased, indicating that the protectiveness of
surface oxide film increased with the existence of inorganic
ions. On the other hand, in E-MEM, the current density at
0 mV decreased while that at 500 mV increased, that is possi-
bly explained by the increase in cathodic current density and
the decrease in the protectiveness of surface oxide film by
amino acids. Phosphate and calcium ions are incorporated
into surface oxide film of the HNS,14) indicating that the in-
corporation of those ions enhance the protectiveness of the
surface oxide film. On the other hand, amino acids seem to
suffer the passivity of the HNS. The effect of amino acids on
the surface oxide film of HNS was suggested to be the prefer-
ential dissolution of iron because concentration ratio of iron
to chromium, [Fe]/[Cr], in the surface oxide film decreased
with the immersion in cell culture medium containing amino
acids and proteins.14) Therefore, the continuous preferential
dissolution of iron during anodic polarization may cause the
 Corrosion Behavior of Nickel-Free High Nitrogen Austenitic Stainless Steel in Simulated Biological Environments 3103

10 4 10 4
10 3 (a) 10 3 (b)
Current Density, I / A m-2

Current Density, I / A m-2

10 2 10 2
10 1 10 1
10 0 HNS 10 0 HNS
316L 316L
10-1 10-1
10-2 10-2
10-3 10-3
10-4 10-4
10-5 10-5
10-6 10-6
-500 0 500 1000 1500 -500 0 500 1000 1500
Potential, E / mV vs. SCE Potential, E / mV vs. SCE

10 4 10 4
(c) (d)
Current Density, I / A m-2

Current Density, I / A m-2

10 3 10 3
10 2 10 2
10 1 10 1
10 0 HNS 10 0
316L 316L HNS
10-1 10-1
10-2 10-2
10-3 10-3
10-4 10-4
10-5 10-5
10-6 10-6
-500 0 500 1000 1500 -500 0 500 1000 1500
Potential, E / mV vs. SCE Potential, E / mV vs. SCE
Fig. 2 Polarization curves of HNS and 316L in (a) saline, (b) PBS(-), (c) Hanks and (d) E-MEM.

5. Conclusions

The HNS showed lower current densities than those of the

316L in anodic polarization test in saline, PBS(-), Hanks and
E-MEM. In addition, pitting corrosion did not occur on the
anodic polarization curve. Therefore, the HNS has large ad-
vantages of its passivity and resistance to pitting corrosion
over 316L in biological environments. The corrosion resis-
tance of the HNS probably increase with the inorganic ions
such as phosphate and calcium ions but possibly decreased
by amino acids possibly causing the preferential dissolution
of iron. Consequently, the HNS must be a promising bioma-
terial because of its nickel-free and manganese-free composi-
tion and high corrosion resistance.

Acknowledgement
Fig. 3 Optical micrographs of the surface of (a) HNS and (c) 316L before
the polarization test and (b) HNS and (d) 316L after the polarization test We would like to thank Ms. Yuko Kohyama and Ms.
in E-MEM. Fumiko Mizuno for their valuable support on the experiments.

REFERENCES
apparent decrease in protectiveness of the surface oxide film
by amino acids. 1) R. Köster, D. Vieluf, M. Kiehn, M. Sommeruer, J. Kähler, S. Baldus, T.
Meinrtz and C. W. Hamm: Lancet 356 (2000) 1895–1897.
2) IARC Working Group on the Evaluation of Carcinogenic Risks to
Humans: IARC Monographs on the Evaluation of Carcinogenic Risks
to Humans, 74, (IARC Press, Lyon, 1999) pp. 241–242.
3104 D. Kuroda, S. Hiromoto, T. Hanawa and Y. Katada
3) E. Yano: Toxicology Today, (Kimpodp, Tokyo, 1994) pp. 119–128.
4) A. Yamamoto, Y. Kohyama and T. Hanawa: J. Biomed. Mater. Res. 59
(2001) 176–183.
5) A. Yamamoto, R. Honma and M. Sumita: J. Biomed. Mater. Res. 39
(1998) 331–340.
6) L. W. Chang, L. Magos and T. Suzuki: Toxicology of Metals, (CRC
Press, New York, 1996) pp. 827–828.
7) J. Menzel, W. Kirschner and G. Stein: ISIJ Int. 36 (1996) 893–900.
8) P. J. Uggowitzer, R. Magdowski and M. O. Speidel: ISIJ Int. 36 (1996)
901–908.
9) R. C. Gebeau and R. S. Brown: Adv. Mater. & Proc. 159 (2001) 46–48.
10) K. Endo, Y. Abiko, M. Suzuki, H. Ohno and T. Kazu: Zairyo-to-Kankyo
47 (1998) 570–576.
11) Y. Katada, M. Sagara, S. Iwasaki, K. Sakuraya and T. Kodama: Cur.
Adv. Mater. Proc. 13 (2000) 116.
12) Y. Katada, H. Uno, M. Sagara, M. Ogawa, H. Baba, S. Iwasaki, K.
Sakuraya, K. Hiroka, T. Kodama and C. Shiga: J. Mater. Process. Tech.
117 (2001) CD-ROM Section B1.
13) M. Sagara, Y. Katada and T. Kodama: Cur. Adv. Mater. Proc. 14 (2000)
694.
14) D. Kuroda, T. Hanawa, S. Hiromoto, Y. Katada and K. Asami: Mater.
Trans. (submitted).
15) G. C. Palit, V. Kain and H. S. Gadiyar: Corrosion 49 (1993) 977–991.
16) Y. C. Lu, R. Bandy, C. R. Clayton and R. C.: J. Electrochem. Soc. 130
(1983) 1774–1776.
17) M. Pourbaix: Atlas of Electrochemical Equilibria in Aqueous Solutions,
(National Association of Corrosion Engineers, Texas, 1974) pp. 540–
544.
18) N. Pessall and J. I. Nurminen: Corrosion 30 (1974) 381–392.
19) C.-O. A. Olsson: Corros. Sci. 37 (1995) 467–479.
20) I. Olefjord and L. Wergrelius: Corros. Sci. 38 (1996) 1203–1220.
21) A. S. Vanini, J. P. Audouard and P. Marcus: Corros. Sci. 36 (1994)
1825–1834.
22) H. P. Leckie and H. H. Uhlig: J. Electrchem. Soc. 113 (1966) 1262–
1267.
23) R. C. Newmann and T. Shahrabi: Corros. Sci. 27 (1987) 827–838.