Effect of Surface Treatment on the Surface Characteristics

of AISI 316L Stainless Steel

Steve I II&VVIIii
Electrostatics and Surface Physics Laboratory
NASA Kennedy Space Center, FL 32889

Guna Selvaduray
Department of Chemical and Materials Engineering
San Jose State University
One Washington Square, San Jose, CA 95192

Abstract The performance of the surface layer is in turn dependent on a

number of its characteristics or features. These include its
The ability of 316L stainless steel to maintain detailed chemical composition, thickness, microporosity,
biocompatibility, which is dependent upon the surface surface charge states, surface roughness, total surface area (as
characteristics, is critical to its effectiveness as an implant opposed to geometric area), and critical surface tension,
material. The surfaces of mechanically polished (MP), among others. These characteristics are highly processing
electropolished (EP) and plasma treated 316L stainless steel dependent. The chemical composition, for instance,-can be
coupons were characterized by X-ray Photoelectron significantly different from that of the bulk. The specific
Spectroscopy (XPS) and Auger Electron Spectroscopy (AES) process employed to alter the surface plays a major role in
for chemical composition, Atomic Force Microscopy for determining its properties. Two processes that are widely
surface roughness, and contact angle measurements for critical employed with stainless steels, for surface modification, are
surface tension. All surfaces had a Ni concentration that was electropolishing and passivation.
significantly lower than the bulk concentration of -43%. The
Cr content of the surface was increased significantly by Surface Characteristics, Corrosion and Biocompatibility
electropolishing. The surface roughness was also improved
significantly by electropolishing. Plasma treatment had the The effect of surface characteristics and surface treatments on
reverse effect - the surface Cr content was decreased. It was the corrosion resistance and biocompatibility of AISI 316L
also found that the Cr and Fe in the surface exist in both the stainless steel has been researched by several investigators in
oxide and hydroxide states, with the ratios varying according the past. Bordji and co-workers (l) investigated the effect of
to surface treatment. glow discharge nitrogen implantation, carbon-doped stainless
steel coating sputtering and low temperature plasma nitriding
Introduction on the biocompatibility of 316L stainless steel, as studied with
human osteoblast and fibroblast cultures. They found the first
Stents, fracture fixation plates and screws, spinal implant two treatments resulted in biocompatible surfaces whereas the
devices, aneurysm clips, intramedullary nails and pins, plasma nitrided surface resulted in dramatic cellular
temporary fixation devices, and surgical instruments, among interactions. The effect of different passivation techniques on
others, have been manufactured from AISI 316L stainless steel the in vitro corrosion resistance of 316L wires was studied by
for several years. While the mechanical performance of Shih et al. 2 Only amorphous oxidation was found to improve
implants and devices may be governed by bulk properties, the corrosion resistance of the alloy. The improvement was
interaction with the environment is governed by the attributed to the removal of the plastically deformed native
characteristics of the surface layer. In the case of biomedical oxide layer and the replacement with a newly grown, more
devices two interactions that are of paramount importance are uniform and compact oxide layer composed of nano-scale
corrosion resistance (which is related to leach rates) and oxide particles with higher oxygen and chromium
biocompatibility. concentrations. The authors reported that the properties of the
surface oxide layer, rather than its thickness, seem to be the
predominant factor in the improvement of in vitro corrosion
resistance. Changes to the wettability characteristics of 3 1-6L One of the issues when dealing with materials for biomedical
stainless steels by Nd:YAG laser treatment and its effects on applications is the residual concentration of microbes and-
the cell response of human fibroblast cells was studied by other biological agents. Changing the surface functional
Lawrence et a1. 3 The wettability characteristics of the 316L group s of materials can change the bacterial adhesion,
were found to improve and were attributed to modifications to depending upon the surface hydrophobicity.'° Boyd et al.
the surface roughness, changes in the surface oxygen content found that an increase in the surface roughness of stainless
and the increase in the polar component of the surface energy. steel surfaces increased bacterial adhesion. 1 Therefore there
Cell proliferation and adhesion on the laser treated 316L were is concern about the effectiveness of surface treatments in
found to be consistently less than on the untreated samples. deleting biological contaminants. Plasma treatment can
The authors - state that this effect is due -entirely to the potentially be effective in removing biological contamination.
increased surface roughness affected by the laser treatment.
The interaction of human cardiac artery endothelial cells with There has been a significant amount of work devoted to
316L stainless steel, with varying degrees of surface studying the effects of various surface treatments on
roughness, was evaluated by Rohly and co-workers. 4 They biocompatibility, as measured or evaluated by different means,
fnd that cell growth was promoted by the overall smoother including in vitro studies. Research that has attempted to
surface of the control specimens over that of the specimens characterize in detail the manner in which surface treatments
that had 60 - 240 grit surfaces. affect the surface chemistry, morphology and thermodynamic
stability, and how these in turn may affect biocompatibility
The relationship of the critical surface tension of a solid and corrosion behavior of implants has not been that
surface to its biocompatibility was reported by Baier. 5 Based extensive. There is a need to study these relationships so that a
on Baier's work, a surface is biocompatible when its critical broader understanding of the factors that affect the
surface tension is between 20 x 10 N/rn and 30 x iø N/rn. biocompatibility of implants can be developed.
Selvaduray and Bueno studied the effect of plastic strain in
combination with electropolishing and passivation on the The overall purpose of this study is to investigate the effect of
critical surface tension of 316L. (6 They found that the surface surface treatment on the surface characteristics and surface
treatment consisting of electropolishing, followed by chemistry of Nih, Co-Cr and AISI 316L metallic alloys so
passivation for 30 minutes in 30% nitric acid resulted in a that their corrosion characteristics and biocompatibility
surface which had a critical surface tension between 20 .-. 30 x behavior can be better understood and correlated. The results
10-3 N/rn. It was also reported that surface plastic strains of up reported in this paper represent the first phase of the study,
to 30% did not result in a significant change to the critical which focused on AISI 3161, stainless steel. The effects of
surface tension. mechanical polishing, electropolishing, and each of these with
subsequent plasma treatment, on the surface roughness,
The relationship between surface charge and cellular adhesion surface chemistry, and critical surface tension of 316L
was investigated by measuring the adhesion strength over a stainless steel were studied.
range of charge densities. (7) The cells were found to show
charge and electrical potential-dependent adhesion maxima, Experimental Procedure
suggesting that surface alloying for optimum adherence would
be a possibility. The AISI 3-16L coupons used for this investigation were 19
mm x 19 mm x 0.737 mm thick. The results of the
Trigwell & Selvaduray found that the composition of the independent chemical analysis that was done to verify that the
surface oxide layer, rather than the surface roughness or chemical composition was withing specifications are shown in
surface area, was more important in determining the corrosion Table 1.
rate of NiTi alloys that had been mechanically polished,
electropolished, chemically etched and plasma etched.8 Three surface treatments were employed: mechanical
polishing, electropolishing, and mechanical and
Electropolishing is an electrochemical process that involves electropolishing each followed by plasma treatment. Prior to
removal of material from the specimen being "polished", in an any surface treatment, the specimens were first cleaned with
electrolyte, with the specimen as the anode. The electrolyte is alkanox in deionized water at 70°C followed by immersion in
usually a H3PO4-H2SO4 solution. Direct current is employed to an ultrasonic cleaner for 1 minute, followed by rinsing in 1
effect the material removal. It is reported that electropolishing liter of deionized water for 30 seconds. If a water break
selectively removes material from the high points at a rate occurred, the alkanox cleaning step was repeated. The
faster than material removal from the depressions or "valleys", specimens were rinsed one more time in 1 liter of deionized
resulting in a smoother surface and thus achieving the water, at 80°C, for 30 seconds.
polishing effect. 9 During this process a film also forms on
the surface of the object being electropolished. The mechanically polished specimens had their surfaces
ground with 1000 grit SiC paper and water to produce a
uniform scratch pattern on the surface. The specimens were Surface Roughness
cleaned again following mechanical polishing.
The average surface roughness of the specimens that were
Table 1: Ghenicel composition of3I6L stainless steel mechanically polished (MP), electropolished EP), and before
and after plasma treatment, is shown in Table 2.
Actual Electropolishing was effective in reducing the surface
Element Specification roughness significantly, down to less than half the surface
Wt % At %
C 0.02 0.09 <0.030 roughness after mechanical polishing, with an improvement in
Mn 1.92 1.95 <2.00 the uniformity of the surface as well, as evidenced by the
0.022 0.04 <0.045 decrease in the standard deviation. Plasma treatment had no
P
0.002 0 <0.030 observable effect on the surface roughness, within the
S
0.29 0.58 < 1.00 experimental parameters.
Si
Cr 17.54 18.86 16.00-. 18.00
Table 2: Effect of electropolishing on sui:face roughness
Ni 13.78 13.12 10.00- 14.00
Mo 2.76 j 1.61 2.60-3.00
NS* MP EP EP+ MP+
Cu 0.09 0.08
Plasma Plasma
Fe Balance 63.66 Balance
Roughness Average (A) 32.15 14.66 14.06 29.83
* Not Specified 3.60 5.23 8.82
Standard deviation (A) 8.03

For electropolishing the specimens were first mechanically

X-ray Photoelectron Spectroscopy
polished, as described above: They were then activated in a
50% concentrated H,SO4 solution at 70°C for 1 minute. The
The high resolution XPS data for the C, 0, Cr, and Fe peaks
electropolishing solution used was 63% H 3PO4, 15% H,SO4
are summarized in Table 3. This information was used to
and 22% deionized water, and maintained at 55°C ± 5°C. The
deduce the changes in surface chemistry that occur as a result
current density was 12.9 x 102 A/rn2 and electropolishing of the surface treatments employed.
was done until the scratches from the mechanical polishing
were removed and a mirror finish obtained. After
Electropolishing : When mechanically polished specimens
electropolishing the specimens were rinsed in deionized water,
were electropolished, the characteristics of the oxygen bonds
dried, and the surfaces examined in a SEM to ensure uniform
changed. There was a drop in oxygen bonding in the oxide
polishing. form and a dramatic increase in oxygen bonding as a
hydroxide. The chemical composition of the chromium on the
Mechanically and electropolished specimens were also surface also changed with electropolishing. In the
subjected to atmospheric plasma glow discharge (APGD)
mechanically polished specimen Cr was bound primarily in
treatment. The advantage of APGD is that it is low
the oxide form (Cr203). After electropolishing, the proportion
temperature (minimizing thermal damage) and can be of Cr in the hydroxide form increased significantly, though the
performed in air without the need for a vacuum chamber. The
majority of the chromium was still bound as an oxide.
specimens were exposed to a 98 % He-2 % 0 2 RF plasma at Electropolishing also caused a change in the binding of the
300W for 5 minutes with a Surfx Technologies Atomflo iron on the surface. There was a decrease in metallic iron, and
1500R plasma source. an increase in Fe bound as FeO, which is adherent to steel
substrates. This was offset by a drop in Fe bound as Fe203 and
The surfaces of the mechanically polished and electropolished
the hydroxide. However, more than 50% of the Fe is still
specimens, before and after plasma treatment, were analyzed
bound as Fe203.
by Auger electron spectroscopy (AES) using a Phi-5600
instrument and X-ray photoelectron spectroscopy (XPS)
The Cr:Fe ratio, which is 1:3.63 in the bulk alloy dropped to
which was a Kratos XSAM 800. Contact angles were
1:1.71, leading to an "enrichment" of chromium on the surface
measured on an AST Products VCA Optima instrument
when 316L is electropolished. It must be pointed out that the
equipped with an environmental chamber set at 37 °C. The Cr:Fe ratio for the mechanically polished specimens also
critical surface tension was determined following the method indicates surface enrichment of Cr, as compared to the bulk,
of ZismanP2"3 probably due to preferential atmospheric oxidation of the
chromium. However, the data in Table 3 indicate that
The surface roughness was measured with a Nanosurf E-AFM electropolishing can result in. further enrichment of chromium
atomic force microscope. Ten area scans were taken on each oxide on the surface.
specimen. The scanned area each time was 10 pm by 10 p.m.
Table 3: Relative chemical composition (atomic percent) of are summarized in Table 4, and one depth profile - for the
surfaces by XPS electropolished specimen - is shown in Figure 1.

EP+ MP+ Table 4: Relative chemical composition (atomic Percent) and

ivLr Lt
- Bond Type I Piasma Plasma oxide layer thickness of surfaces by AES
C-C/C-H 72.2 71.6 74.7 77.1
16.2 14.8 12.4 EP+ MP±
C-0 15.5 Element MP EP
Plasma Plasma
C=O 5.9 5.3 3.9 4.5
0 63 66 69 70
2 COO- 5.9 5.1 3.9 3.7 Cr 16 20 10 10
O-C=0 0.5 1.7 2.7 2.4 Fe 18 10 16 18
0= 59.9 11.9 27.2 65.1 Ni 3 3 5 2
0-H 32.1 70.0 59.7 28.1 Thickness 40 A -35A -- 35 A -75A
{
0(H0) 8.0 18.1 13.1 6.8
Cr 13.0 16.3 11.2 7.9
Electropolishing: It can be seen from the data in Table 4 that
I Cr03 67.3 48.4 68.9 80.1
electropolishing is effective in enhancing the Cr content of the
Cr-OH 19.7 35.3 19.9 12.1 surface oxide layer, along with a reduction in the Fe content.
- Fe 9.4 5.6 5.1 0.6 This becomes clear when the Cr and Fe concentrations in the
mechanically polished and electropolished specimens are
FeO 9.6 16.3 4.5 0.7
compared. It is rather interesting to note that the Ni content is
Fe203 59.8 51.5 76.0 81.7 significantly lower than the bulk composition of 13.78 % in all
Fe-OOH 17.4 12.1 10.7 15.7 the specimens; electropolishing. did not affect the Ni
Fe Sat 3.8 14.6 3.7 1.3 concentration. Electropolishing also does not change the
Cr:FeRatio 1:2.10 1:1.71 1:1.82 1:5.88 thickness of the surface oxide layer significantly.

I IMJU

Plasma Treatment: The effect of plasma treatment, regardless

of whether the specimens were mechanically polished or 800

electropolished, was relatively consistent. The proportion of

oxygen bound in the oxide form increased in the plasma 600

treated specimens, with a slight drop in the proportion bound

as a hydroxide. This is consistent with the finding that the 400

proportion of Cr as Cr2O3 and Fe as Fe203 were both found to

increase appreciably, with corresponding drops in metallic Cr 200

and Fe, and chromium and iron hydroxides. It should be noted

that Fe bound in the form of FeO also dropped dramatically,
leading to the conclusion that the plasma treatment used in this 0 25 50 76 100 .125 150

investigation had the effect of oxidizing divalent Fe. [A] (Ta205 equivalent)
Etch depth
Figure 1: AES depth pro/lie of electropolished 316L stainless
The Cr:Fe ratios indicate a dramatic drop in the chromium steel -
concentration of mechanically polished specimens when they
are plasma treated - from 1:2.10 to 1:5.88. Plasma treating
electropolished specimens also reduced the Cr:Fe ratio, though Plasma Treatment: Treating the surface with a plasma has the
to a far lesser extent. Electropolishing appears to significatitly reverse effect of electropolishing. The Cr content decreases
reduce the "chromium-depleting" effect of plasma treatment. and the Fe content increases. Plasma treatment of the
mechanically polished specimens resulted in a noticeable
Significant amounts of P, S, and Ca were detected on the EP increase in the oxide layer thickness, probably due to
surfaces; these are thought to be residues from the oxidation of the FeO to Fe2O3 . Again, as was deduced from
electropolishing process. Most of this residual contamination the XPS data, electropolishing the specimens first, prior to
was removed by the plasma treatment. plasma treatment, prevented this change in the surface oxide
layer thickness. The Ni content does not appear to be affected
Auger Electron Spectroscopy significantly.

The chemical composition and thickness of the surface oxide.

layer as determined by Auger Electron Spectroscopy (AES)
 Critical Surface Tension Discussion of Results

The critical surface tension values measured at 23°C are The results obtained from AES and XPS are consistent with
summarized in Table 5. The Zisman plot for one condition - one another, namely that electropolishing enriches the Cr
t
the mechanically polished specimens at 23°C - is shown in concentra ion in the surface oxide layer. The fact that
Figure 2. The data indicate that plasma treatment increases the electropolishing is done in an aqueous environment is most
thermodynamic stability of the surfaces, which can be seen by probably the cause for the increase in the chromium hydroxide
inspecting the data at 23°C. This is consistent with the XPS content.
data that show an increase in the Cr203 and Fe203
concentrations following plasma treatment. These compounds The surfaces of the 316L stainless steel specimens tested are
are also thermodynamically more stable than the other Fe and not composed entirely of Cr2O3 ; they are really "mixed
Cr compounds. oxides" and "mixed hydroxides", containing primarily both
chromium and iron oxides and hydroxides. This is consistent
The decrease in the critical surface tension following with the findings reported by others that the surface oxide
electropolishing caii be attributed to the decrease in the Cr2O3 passivation layer on ferrous alloys is a complex mix of oxides,
and Fe203 concentrations, coupled with an increase in the Dr, hydroxides and oxyhydroxides. 4 Electropolishing of
Cr-OH and FeO concentrations, all of which are mechanically polished specimens was found to increase the
thermodynamically less stable than Cr 203 and Fe203. chromium hydroxide content on the surface. Further analysis
is necessary to determine the exact composition. It also
resulted in the FeO concentration increasing and the Fe,
Table 5: Critical surface tension as a function of surface Fe203 and Fe-OOH concentrations decreasing.
treatment
Despite there being 13.78 % Ni in the bulk of the alloy itself,
the surface contains only about 3 %, even for the specimens
Temp 23°C that were mechanically polished. For biomedical applications
MP 28.6 this can be a desirable finding as nickel has been known to
EP 21.0 cause allergic reactions in some individuals.
EP+Plasma 27.5
M.P+Plasma 33.0 Plasma treatment was found not to enhance the surface
15.6 passivation, but reduced the Cr:Fe ratio. When the XPS and
PTFE
AES data are interpreted in combination, plasma treatment of
the surfaces increases the concentration of Cr in the oxide
form, as opposed to the hydroxide form. It also resulted in an
increase of Fe in the Fe203 form. This combined effect
resulted in an increase in the critical surface tension of the
Mechanically Polished (23 C) plasma treated specimens, signifying an increase in the
thermodynamic stability of the. surfaces.

While the surface might be constituted of thermodynamically

more stable species, the extent to which this layer is able to
7O6
isolate the bulk from the host environment is also heavily
dependent on its microstructure. Less adherent and porous
surface layers can provide little protection regardless of
thermodynamic stability.

The potential benefit of the plasma treatment is that it

U 0.3 Cnifical surface energy removed residual contamination from the electropolishing
28.6 dyne/Cm
process. Work is in progress to further evaluate the
0.2 effectiveness of plasma treatment in removing contamination
0 20 40 60 80 from 31 6L surfaces, with different gases. This effort will also
Liquid Surface Tension (dyne/cm)
study the effect of plasma treatment in enhancing surface
passivation.

Figure 2: Zisman plot for mechanically polished specimen, at Conclusions

23°C.
Regardless of the surface treatment, it was found thit the
surface oxide layer on 31 6L stainless steel contains
 significantly less Ni as compared to the bulk composition. Proc. 2' Intl. Conf. on Shape Memory and Superelastic
Electropolishing was found to be effective in enriching the Cr Technologies, 2-6 March, 1977, Asilomar, California, p 383-
content of the surface. However, the surface layer is not 388.
entirely Cr203. It contains both Cr and Fe oxides and 9. Hensel, KB., "Electropolishing," Meta! Finishing, v 98, n
hydroxides. The type of surface treatment affects the relative 1. January 2000, p 440-448
ratios of these constituents. Electropolishing results in a 10. Katsikogianni, M. and Y. F. Missirlis, "Concise review of
smoother surface. Plasma treatment, while it can have the mechanisms of bacterial adhesion to biomaterials and of
potential benefit of eliminating biological contamination, was techniques used in estimating bacteria-material interactions,"
found to affect the surface chemistry in a deleterious manner European Cells and Materials, v 8, 2004, p 37-57.
by causing the Cr concentration to decrease. 11. Boyd R.D., J. Verran, M. V. Jones and M. Bhakoo, "Use
of AFM to determine the effect of substratum surface
Acknowledgements topography on bacterial adhesion," Langmuir, y 18, 2002, p
2343-2346.
The authors wish to acknowledge Michelle Michalenko of 12. Fox, H.W., and Zisman, W.A., "The Spreading of Liquids on.
NASA for her help in obtaining the contact angle Low-Energy Surfaces. L FTFE," J. CoiZo(.Sei., (195C) pp. 14-
measurements and Dr. Jim Mantovani of Florida Institute of 531
Technology for the AFM data. Thanks also to Dr. John Turn 13. Zisman, W.A., "Relation of the Equilibrium Contact Angle to
and Robert McClaine of BAE Emerging Technology Systems Liquid and Solid Constitution," in Contact An gle, Wettability and
for doing the electropolishing and obtaining the chemical Adhesion, ACS, Washington, D.C., (1964)
analysis of the AISI 316L stainless steel samples used. 14. Kruger, J., "Passivity" in ASM Handbook Volume 13A
Corrosion: Fundamentals, Testin g and Protection, S. D.
References Cramer & B. S. Covino Editors, ASM International, 2003, p
61-67.
1. Bordji, K., J-Y Jouzeau, D. Mainard, E. Payan, J-P
Delagoutte and P. Netter, "Evaluation of the effect of three
surface treatments on the biocompatibiity of 316L stainless
steel using human differentiated cells," Biomaterials, v 17, n
5, March 1996, p 491-500.
2. Shih, C. C., C. M. Shih, Y. Y Su, L. H. J. Su, M. S Chang
and S. J. Lin, "Effect of surface oxide properties on corrosion
resistance of 316L stainless steel for biomedical applications,"
Corrosion Science, v 46, n 2, February 2004, p 427-441.
3. Lawrence, J., H. R. Chong, C. K. Chong and L. Hao,
"Laser modification of the wettabiity characteristics of a
316L stainless steel biometal and the effects thereof on human
fibroblast cell response," Lasers in Engineering, v 15, n 1-2,
2005, p 75-90.
4. Rohly, K., N. Istephanous, D. Untereker, I. Trausch, A.
Belu and B. Sahli, "Effect of surface properties on interaction
of 316L stainless steel with RCA endothelial cells,"
Transactions - 7th World Biomaterials Congress, 2004, p 790.
5. Baier, R.E., "The Role of Surface Energy in
Thrombogenisis," Bull. NYAcad. Med., vol.48, (1972) p 257-
272
6. Selvaduray, G. and H. Bueno, "The critical surface tension
of 316L stainless steel: Implications for stent
thrombogenecity," Proceedings of the Materials and Processes
for Medical Devices Conference 2003, September 8-10, 2003,
Anaheim, California, p 69-74.
7. Hallab, N. J., K. J. Bundy, K. O'Connor, R. Clark and R. L.
Moses, "Cell adhesion of biomaterials: correlations between
surface charge, surface roughness, adsorbed protein, and cell
morphology," Journal ofLong-Term Effects of Medical
Implants, v 5,113, 1995, p 209-23 1.
8. Trigwell, S and G. Selvaduray, "Effects of Surface Finish
on the Corrosion of NiTi Alloy for Biomedical Applications,"