Corrosion behavior of 2205 duplex stainless steel

Jeffrey A. Platt, DDS, FAGD, a Andres Guzman, DDS, b Arnaldo Zuccari, MD, DDS, b David W. Thornburg, PE, c Barbara F. Rhodes, BGS, a Yoshiki Oshida, PhD, a and B. Keith Moore, PhD a

Indianapolis and LaPorte, Ind.

The corrosion behavior of 2205 duplex stainless steel was compared with that of AISI type 316L stainless steel. The 2205 stainless steel is a potential orthodontic bracket material with low nickel content (4 to 6 wt%), whereas the 316L stainless steel (nickel content: 10 to 14 wt%) is a currently used bracket material. Both stainless steels were subjected to electrochemical and immersion (crevice) corrosion tests in 37 C, 0.9 wt% sodium chloride solution. Electrochemical testing indicates that 2205 has a longer passivation range than 316L. The corrosion rate of 2205 was 0.416 MPY (milli-inch per year), whereas 316L exhibited 0.647 MPY. When 2205 was coupled to 316L with equal surface area ratio, the corrosion rate of 2205 reduced to 0.260 MPY, indicating that 316L stainless steel behaved like a sacrificial anode. When 316L is coupled with NiTi, TMA, or stainless steel arch wire and was subjected to the immersion corrosion test, it was found that 316L suffered from crevice corrosion. On the other hand, 2205 stainless steel did not show any localized crevice corrosion, although the surface of 2205 was covered with corrosion products, formed when coupled to NiTi and stainless steel wires. This study indicates that considering corrosion resistance, 2205 duplex stainless steel is an improved alternative to 316L for orthodontic bracket fabrication when used in conjunction with titanium, its alloys, or stainless steel arch wires. (Am J Orthod Dentofac Orthop 1997;112:69-79.)

A u s t e n i t i c stainless steel (e.g., AISI type 316L stainless steel) is the most commonly used orthodontic bracket material. It typically has a composition of 18 weight percent (wt%) chromium, 8 wt% nickel (Ni), 2 to 3 wt% molybdenum, and a low carbon content. 1 Its mechanical properties, such as ductility and wear resistance, make it attractive for this application. The corrosion resistance and appearance of stainless steel brackets are relatively good. However, this material is challenged by the hostile environment in the mouth, as it is susceptible to localized corrosion in a low pH environment containing chlorine ions. Austenitic stainless steel exists as a face-centered cubic crystalline structure, formed by heating the alloy above 912 C. 2 To maintain this structure when cooled, nickel is added to stabilize the austenitic phase. To minimize the risk of hypersensitivity reactions from nickel, the corrosion resistance of the
aDental Materials, Department of Restorative Dentistry, Indiana University School of Dentistry. bprosthodontics, Department of Restorative Dentistry, Indiana University School of Dentistry. cTP Orthodontics, Inc. Reprint requests to: Dr. Yoshiki Oshida, Dental Materials Laboratory, Indiana University School of Dentistry, 1121 W. Michigan St., Indianapolis, IN 46202-5186. Copyright 1997 by the American Association of Orthodontists. 0889-5406/97/$5.00 + 0 8/1/75384

stainless steel should be maximized to control the nickel ion release from the alloy. It has been reported that up to 21.5% of the population may exhibit allergic reaction to nickel on patch testing? Case studies have indicated hypersensitivity reactions to nickel, stimulated by exposure to orthodontic brackets. 4-8 It has also been shown that the quantity of nickel exposure is critical if hypersensitivity symptoms are seen. 9 The 2205 stainless steel, a duplex (or dual) phase stainless steel, is being investigated as a material for orthodontic bracket fabrication. Microstructure of the duplex stainless steels is a mixture of austenitic and delta-ferritic phases. TM The delta-ferrite is hard and relatively less ductile. Austenite is softer and more ductile. The combination of these phases results in steel harder than the single-phase austenitic stainless steel (316L) and more ductile than single-phase ferritic stainless steel (430). The austenitic structure exhibits corrosion resistance because both chromium and molybdenum are soluble in the matrix. Chromium adds to the overall resistance through a passivation process by forming a complex spinel-type passive film, (Fe,Ni)O(Fe,Cr)203 .11 Molybdenum increases the ability of stainless steel to resist the localized corrosion including pitting and crevice corrosion, particularly in environments containing chloride ion. 12 The duplex stainless steels contain much less
69

70

Platt et aL

American Journal of Orthodontics and Dentofacial Orthopedics July 1997

, T M A wire "l'i Ni wire SS wire

t
/
i Element

block, 3 1 6 L SS ~7~ 2 2 0 5 SS

Fig. 2. Immersion test sample couple for crevice corrosion.

Table I. Chemical composition of a dual-phase 2205 stainless steel and a single-base 316L stainless steel 13 2205 (%) 0.03 maximum 4-6 31-23 3-3 2 0.03 balance 316L(%) 0.03 maximum 10-14 16-18 2-3 2 0.03 balance

~316L $.$. "~ k 2205 [ S.S.

Fig. 1. Sample configuration of coupled test sample.

nickel than austenitic stainless steels, as seen in Table I. 13 If the corrosion resistance of the duplex stainless steel is equal to or better than the austenitic stainless steel, the risk of nickel hypersensitivity should be reduced. The use of duplex stainless steels in other industries in highly corrosive environments would suggest that the corrosion resistance of these materials is good. The purpose of this study is to compare the corrosion behavior of 2205 stainless steel with that of 316L stainless steel through electrochemical polarization studies and immersion crevice corrosion tests.

C Ni Cr Mo Mn S Fe

mens (approximately 5 5 12 mm) were cast in a "Christmas tree" pattern and separated with separating disks and air coolant. Mechanical polishing was performed with wet SiC papers up to grit No. 800. After mechanical polishing, any samples with casting defects evident on the surface were eliminated from the study. D e n s i t y ( g / c m 3) of each polished sample was calculated and, if the calculated specific density was less than 99% of the averaged value, the sample was also eliminated from the study. The averaged value of density of 316L stainless steel was 7.949 -+ 0.076 g/cm3; whereas that for 2205 stainless steel was 7.953 -+ 0.068 g/cm3.

MATERIALS AND METHODS Materials

All specimens of 316L and 2205 stainless steel used in this study were supplied by TP Orthodontics, Inc. Speci-

Microstructural Observation
After additional polishing with a suspension of 0.05 ~m alumina, one surface of each material was chemically etched with "aqua regia" at room temperature for 5

American Journal of Orthodontics and Dentofacial Orthopedics Volume 112, No. 1

Platt et aL

71

2205 in NaCI (0.9%) N 2 saturated

0.2

run 1 run 2

~A

0.0 O > . -0.2

13-

E~or, (run I)

*,

~c

E,o= (run 2)
=

==

-0.4

10-5

104

10-3

1 0 -2

Log current density mA/cm z

Fig. 3. Tafel slopes of polarization curves of Fig. 4. 2205 in NaCl (0.9%) N 2 s a t u r a t e d

2.0
1.8 --

run#1 run #2

1.61.41.2LU

1.0-

_> 0 . 8 c

0.6-

o. 0 . 4 0.20.0-0.2
-0.4
I I I I I I I

10-5

10-4

10-3

10-2

10-1

10 o

101

102

Log current density mA/cm z Fig. 4. Polarization curves of uncoupled 2205 stainless steel.

72 Platt et al.

American Journal of Orthodontics and Dentofacial Orthopedics July 1997

2205 type stainlesssteel

316L type stainlesssteel

Fig. 5. Optical microstructures.

seconds. The microstructure was photographed at a magnification of 250X, with an optical microscope. Corrosion Tests

Electrochemical corrosion tests--uncoupled." Two samples of mechanically polished 2205 stainless steel and two of 316L stainless steel were subjected to potentiodynamic polarization tests in a 0.9 wt% NaC1 solution, nitrogen purged, at 37 C, to simulate the oral environment. 14 A scanning speed of 1 mV/second, over a potential range of -400 to 1500 mV for 2205 and -700 to 1200 mV for 316L stainless steel was used. The resulting data curves were plotted as corrosion potential in V referenced to a SCE (saturated calomel electrode) versus log current density

(mA/cm2).
Electrochemical corrosion tests--coupled: If two dissimilar metals are in contact with each other, the more noble material will behave in a cathodic (noble)) 5 When a 2205 orthodontic bracket is combined with orthodontic arch wire, there is the possibility of a galvanic (dissimilar

metal) reaction when the couple is exposed to the oral environment. To form a corrosion couple, the rear surfaces of 2205 and 316L stainless steel samples were joined by spot-welding a thin sheet of 316L between them. This spot-welded couple was embedded into epoxy resin in such a way that polished surfaces of each material were exposed to the electrolyte, but the spot welded contact was entirely covered by resin (Fig. 1). An important factor in galvanic corrosion is effect of the ratio of the cathodic and anodic areas. An unfavorable area ratio consists of a large cathode and a small anode. For a given current flow in the galvanic cell, a smaller anode results in a greater current density and hence a greater corrosion rate. a5 On the basis of the results of individual polarization curves of uncoupled 2205 and 316L, it was found that 2205 material behaves cathodic with respect to 316L. To study the surface area effect of a 2205/316L couple, three different ratios of surface areas between 2205 and 316L stainless steels were tested: (1) 1:1 (i.e., both materials had equal exposure area), (2) 1:0.5

American Journal of Orthodontics and Dentofacial Orthopedics Volume 112, No. 1

Platt et aL

73

316L in 0.9%NaCI saturated with N 2

,2 1.0 0.8 0.6 0.4 _ run 1 r

> ~

.2

~_ 0.0
-0.2
, .

-0.4
-0.6
I

10-5

10-4

10-3

10-2

10-1

10 0

101

10 2

Log current density mA/cm 2 Fig. 6. Polarization curves of uncoupled 316L stainless steel.

22051316L, No Mask, in NaCI (0.9%) N 2 saturated

1.4 1.2 1.0 0.8 LU (D 0.6 >

.m

run run run run

3 4 5 6

0.4 0.2 0.0 -0.2 -0.4 -0.6

I I I I ~ I I I I I

"6
0.

10 .5

10-4

10-3

10.2

10-1

100

101

102

103

Log current density mA/cm z Fig. 7. Polarization curves of coupled sample with equal surface area.

74

Platt et al.

American Journal of Orthodontics and Dentofacial Orthopedics July 1997

2205/316L, No Mask, in NaCI (0.9%) N2 saturated

-0.2

-0.3

*
LU

run run run

3 L 4 5 6

1
0 0

L~ o9 > -0.4
.1..,
t-

~&

O.

-0.5

"..
A"

-0.6 10 -5

%
J t i illtj J i i ==ll I i i J p ~lll I

.
I

%
i =

10 -4

10 -3

10 -2

Log current density mA/cm 2 Fig. 8. Tafel slopes of polarization curves of Fig. 7.

T a b l e II. C o r r o s i o n r a t e s

Icorr
Maten
2205 Average 316L Average 2205/316L (1:1) run run run run 3 4 5 6 0.45 0.55 0.50 0.75 1.5 0.9 1.0 0.9 1.2 1.5 run 1 run 2 1.5 1.3 run i run 2

(vJl/cm2)
0.8 1.0

Corrosion rate (MPY)

0.370 0.462 0.416 +- 0.065 0.693 0.601 0.647 --- 0.065 0.208 0.254 0.231 0.346 0.260 0.060 0.693 0.416 0.462 0.524 -+ 0.148 0.416

Figure no.
5, 6 5, 6 7 7 10, 10, 10, 10, 12 12 12 13 13 13 11 11 11 11

(i.e., half of 316L stainless steel was masked), and (3) 1:0.25 (i.e., three quarters of 316L stainless steel was masked), as seen in Fig. 2. Masking was accomplished with plastic tape. The corrosion potential was monitored and recorded. Then the coupled samples were subjected to scanning polarization testing in the same manner as the uncoupled samples. Estimation of corrosion rates: Referring to Fig. 3, which is enlarged from the cathodic and anodic polarization portions of Fig. 4, the corrosion current, I ..... is related to the slope of the plot through the following equation16: I .... = [3a[3C/{2"3([3A + [3C)} 8E/8I (1)

Average 2205/316L (1:0.25) Average 2205/316L (1:0.5) Average run 1 run 2 run 3 run 1 run 2 run 3

0.554
0.693 0.554 -- 0.139

where gE/gI is slope of the polarization resistance plot, 13A, 13 c is anodic and cathodic Tafel constants, and Ico,~ is corrosion current 0xA). The corrosion current, I ..... can be obtained graphically as seen in Fig. 2 by finding an intersecting point of [~a and [3c slopes. The corrosion current can be related

American Journal of Orthodontics and Dentofacial Orthopedics Volume 112,No. 1

Platt et al.

75

22051316L,314 MASK, in NaCl (0.9%) Nasaturated

1.4
1.2 --

1.0

--

run 1 run 2 run 3

0.8

--

UJ

o 0.6og > ._~ 0 . 4 C

0.2-

0,0

--

-0.2 z~

-0.4 -0.6

[]

o
10-5 I 10-4

o ~t,~ ~~;~

I 10-3

10-2

10-1

100

101

102

103

Log current density mA/cm =

Fig. 9. Polarization curves of coupled sample when 3/4 of 316L surface was masked.

directly to the corrosion rate through the following equation, corrosion rate (MPY) = 0.13 X Ico~ x E.W./d (2)

where M P Y is milli-inches per year, E.W. is equivalent weight of the corroding species (g), which is equivalent t o the atomic weight of the corroding element divided by the valence of the element, d is density of the corroding element (g/cm3), and I .... equals corrosion current density

(~A/cm2).
Immersion crevice corrosion tests. Twelve blocks of 316L and 2205 were polished on all sides. As seen in Fig. 2, three types of arch wires were sandwiched between pairs of 316L and 2205 blocks: titanium-molybdenum (TMA), stainless steel (SS), and titanium-nickel (NiTi). Two assemblies were prepared for each stainless steel and each type of arch wire. The assemblies were held together by plastic clips and suspended in 0.9 wt% NaC1 solution at 37 C for 5 weeks. Each specimen was examined for the presence of visible general corrosion and examined under a stereoptical microscope for evidence of crevice corrosion, a

process that causes severe localized corrosion readily visible at 40X magnification. This test was repeated twice, resulting in four measurements for each combination of two bracket materials (2205, 316L) and three arch wires (TMA, SS, NiTi). Corrosion product analysis. To identify the crystalline structures, corrosion products were collected and subjected to transmission electron diffraction (TED), because the amount of collected corrosion product was not large enough to conduct x-ray powder diffraction. The acceleration voltage was held constant at 100 kV. The d-spacings of the corrosion product were calculated from the diameter of the diffraction rings. 14
RESULTS Microstructural Observation

Optical microstructures confirmed the mixed structure of the duplex stainless steel (2205) as c o m p a r e d with the austenitic single-phase 316L, as seen in Fig. 5. In 2205, delta-ferrite (marked pools) is precipitated in the primary gamma-austenitic ma-

76 PIatt et aL

American Journal of Orthodontics and Dentofacial Orthopedics July 1 9 9 7

22051316L,1/2 MASK in NaCl (0.9%) N2saturated

1.4
1.2 o 1.0 0.8
UJ

run 1 run 2 run 3

o O9 0.6
> -~ 0.4

0.2
0.0
/x

-0.2
0
O O

oo

Zx

-0.4 -0.6

[]
I t I F I I I I I

10 -6

10-5 10-4

10-3

10-2

10-1

10 0

101

10 2

10 3

Log current density mA/cmz

Fig. 10. Polarization curves of coupled sample when half of 316L surface was m a s k e d .

trix. In 316L, carbide (i.e., Cr23C3)is precipitated in the single gamma-austenitic matrix phase.
Corrosion Product Analysis

Before estimation of corrosion rate, the term E.W. (equivalent weight) in the equation (2) must be known. To do so, the corrosion product must be identified. The corrosion products collected from both steels were reddish brown and had the same electron diffraction patterns. All diffraction lines were identified as belonging to FeO(OH). 17This result is expected for Fe-based alloys subjected to a relatively mild aqueous corrosive environment. Hence the value of E.W. = M/n = 55.85/2 = 27.93 (g), and density of iron, d, = 7.86 g/cm3. Electrochemical corrosion test--uncoupled. Figs. 4 and 6 show the polarization curves of 2205 and 316L, respectively. It was found that (1) cathodic and anodic polarization behaviors of both steels are very similar to each other up to the current density of 0.1 mA/cm 2 (at about 400 mV), and (2)

beyond this point, 316L stainless steel showed earlier transpassivation than 2205 stainless steel, although a clear passivation stage was not identified in either steel. In the transpassivation regime, the oxide (passive) film starts to dissolve, indicating that the metal substrate is no longer protected from the environment by the passive film. To find the corrosion current (Icorr), the cathodic to anodic transition was replotted, as seen in Fig. 2 for 2205 steel. The calculated corrosion rates in milliinches per year (MPY) are listed in Table II. It was found that the corrosion rate of 2205 was 0.416 +_ 0.065 MPY, which is superior to 0.647 +_ 0.065 MPY for 316L. Electrochemical corrosion test--coupled. 1. Equal surface area ratio of 2205/316L couple: Figs. 7 and 8 show polarization curves of a 2205/316L couple having an equal surface area ratio. Referring to Fig. 8, it was noticed that the anodic Tafel slope, [31, is larger than the cathodic slope, [3c, suggesting that the corrosion process is controlled by the

American Journal of Orthodontics and Dentofacial Orthopedics Volume 112, No. 1

P l a t t et al.

77

Table III. R e s u l t s of c o u p l i n g c o r r o s i o n tests

Material Block 316L l~ ire TiNi General corrosion Block 1 1 1 2 2 1 0 2 0 2 2 1 1 1 0 l 0 0 0 0 0 1 0 0 l Block 2 1 1 2 2 1 0 2 0 2 2 l 1 1 0 1 0 0 0 0 0 0 0 1 0 Wire
1

Crevice corrosion Block 1

--

Block 2
--

1 1

---

---

316L

TMA

1
1 1

+
---

+
---

316L

SS

1 2 2 1 1
1 0 l 0 0

+ + + +
------

+ + + +
------

2205

TiNi

2205

TMA

0 0
0 1 0 1 ---------

2205

SS

For general corrosion: 0: no corrosion observed. 1: corrosion evident in isolated areas. 2: corrosion evident in multiple areas. For crevice corrosion: - : no crevice corrosion. +: crevice corrosion present.

anodic reaction. A precipitated film is formed on anodic areas because of the anode control type polarization. 18 As seen in Table II, for the coupled data with equal areas, the calculated values of corrosion rate for runs 3 through 6 agreed well. The corrosion rate of 0.260 __ 0.060 MPY for 2205 in the 2205/316L couple was much less than 0.416 _+ 0.065 MPY for uncoupled 2205. It is suggested that the less noble 316L stainless steel serve as a sacrificial anode to protect the 2205 stainless steel. 2. Unequal surface area ratio of 2205/316L couple: Fig. 9 shows polarization curves of 2205/316L couples when the surface area ratio is 1:0.25; whereas Fig. 10 represents polarization curves of couples when the surface area ratio is 1:0.5. The results of calculated corrosion rates are also listed in Table II. It was found that there is no significant difference in corrosion rates for both unequal surface area ratios (i.e., 0.524 _+ 0.148 MPY for surface area ratio of 1:0.25 and 0.554 +_ 0.139 MPY for 1:0.5). This finding does not agree with the generally accepted concept of

Fig. 11. General view of crevice corrosion tested samples.

the surface area effect.15 One of the possible reasons for this might be the fact that the surface of the 316L stainless steel at the edge of the masking tape was severely attacked by localized crevice corrosion. Consequently, a majority of corrosion current was concentrated on these crevice corrosion sites rather than distributed over the uncovered surface.
Immersion Crevice Corrosion Test

Fig. 11 shows typical general views of crevice corrosion test couples with wires. After dissembling

78 Platt et aL

American Journal of Orthodontics and Dentofacial Orthopedics July 1997

area covered with wire

holes, gasket surfaces, lap joints, surface deposits, and crevices under bolt and rivet headsJ 5 It is believed that crevice corrosion results from differences in metal ion or oxygen concentration between the crevice and its surroundings. Under the deposit area, oxygen depletion takes place. After oxygen is depleted, no further oxygen reduction occurs, although the dissolution of metal continues. This tends to produce an excess of positive charge in the solution that is balanced by the migration of chloride ions into the crevice. This results in an increased concentration of metal chloride with a pH value as low as 3 to 4 within the crevice. As the corrosion within the crevice increases, the rate of oxygen reduction on adjacent surfaces increases. This cathodically protects the external surfaces. Thus during crevice corrosion, the attack is localized within shielded areas, while the remaining surface suffers little or no damage. 15 The right column of Table III shows the results of crevice corrosion with " - " and " + " notations. None of the 2205 samples exhibit crevice corrosion, whereas the 316L specimens did, as seen in Fig. 12. DISCUSSION The use of 2205 duplex stainless steel in place of 316L appears to result in lower corrosion in a simulated oral environment. However, a corrosion couple alloy may dramatically decrease this benefit. In particular, NiTi in contact with 2205 produces a reduction in corrosion resistance. The results of the uncoupled electrochemical corrosion tests are consistent with previous findings. Cigada performed electrochemical corrosion tests on several stainless steel wires containing Cr, Ni, and Mo in physiologic solutions at 40 C. 19 Duplex stainless Steel exhibited a passivation range from -200 to 650 mV, results very similar to those obtained in the current study. As seen clearly in Fig. 7, the 316L stainless steel did not show a clear passivation under the current testing conditions and it is obvious that 2205 stainless steel exhibits much better corrosion resistance against 37 C, 0.9 wt% NaCI aqueous solution. If two dissimilar materials are in electrical contact, the materials should behave differently than when they were exposed to corrosion media separately. It is also generally believed that the magnitude of this galvanic corrosion depends on the ratio of exposed surface area of the two materials. Hence, tests in this study were designed to simulate the exposed surface areas in contact when an arch wire lies in the slot of an orthodontic bracket.

Fig. 12. Crevice corrosion on 316L stainless steel, as

marked on Fig. 11.

the test couples, the extent of general corrosion was evaluated and scored in three ranks, as indicated in Table III. The 2205 stainless steel exhibited better corrosion resistance in the immersion corrosion tests. After sampling the corrosion products, both surfaces and wires were cleaned with alcohol and distilled water. It was found that the surface of 2205 was contaminated with corrosion products formed from the wire materials, with little evidence of corrosion of the 2205. Then the surfaces were examined for crevice corrosion. Fig. 12 was taken from the area marked with an arrow mark on Fig. 11. Fig. 12 shows crevice corrosion of 316L on areas where the arch wire was in contact. Intense localized corrosion frequently occurs within crevices and other shielded areas on metal surfaces exposed to corrosives. This type of attack is usually associated with small volumes of stagnant solution caused by

American Journal of Orthodontics and Dentofacial Orthopedics Volume 112, No. 1

Platt et al.

79

CONCLUSIONS

Our findings indicate that (1) 2205 stainless steel exhibits better corrosion resistance than 316L stainless steel; (2) when 2205 is coupled with a less corrosion resistant material (316L stainless steel), 2205 stainless steel shows decreases corrosion; and (3) this trend is reduced when the exposed surface areas have a anodic/ cathodic ratio less than 1 and that 2205, unlike 316L, is not subject to crevice corrosion. From the standpoint of corrosion resistance, the use of 2205 as an orthodontic bracket material seems to be justified when the arch wire material is stainless steel or titanium. Use of this alloy could decrease the amount of corrosion products to which a patient would be exposed that could minimize nickel allergy problems potentially associated with orthodontic treatment.

REFERENCES

1. Grimsdottir MR, Gjedet NR, Hensten-Pettersen A. Composition and in vitro corrosion of orthodontic appliances. Am J OrthQd Dentofae Orthop 1992;101:525-32. 2. Van Vlack LH. Elements of materials science and engineering. 6th ed. Reading (MA): Addison-Wesley;1989. 3. Schubert H, Prater E. Nickel dermatitis--a follow-up study. Br J Dermatol 1987;117:(Suppl 32)43.

4. Romaguera C, Vilaplana J, Grimait F. Contact stomatitis from a dental prothesis. Contact Dermatitis 1989;21:204. 5. Espana A, Alonso ML, Sofia C, Guimaraens D, Ledo A. Chronic urticaria after implantation of 2 nickel-containing dental prothesis in a nickel-allergic patient. Contact Dermatitis 1989;21:204-6. 6. Wilson AG, Gould DJ. Nickel dermatitis from a dental prothesis without buccal involvement. Contact Dermatitis 1989;21:53-6. 7. Van Joost T, Roesyanto-Mahadi ID. Combined sensitization to palladium and nickel. Contact Dermatitis 1990;22:227-8. 8. Greppi AL, Smith De, Woodside DG. Nickel hypersensitivity reactions in orthodontic patients. Uni Tor Dent J 1989;3:11-4. 9. Velen NK, Hattel T, Justesen O, Norholm A. Dietary treatment of nickel dermatitis. Acta Derm Venerol 1985;65:138-41. 10. Lula RA. Stainless steel. Metals Park (OH): American Society for Metais;1986: 71-2. 11. Nakayama T, Oshida Y. Identification of the initial oxide films on 18-8 stainless steel in high temperature water. Corrosion 1968;21:336-7. 12. Uhlig HH. The corrosion handbook. 9th ed. New York: John Wiley;1966:i5074. 13. Lula RA. Stainless steel. Metals Park (OH): American Society for Metals;1986: 74. 14. Oshida Y, Sachdeva R, Miyazaki S. Microanalyticai characterization and surface modification of TiNi orthodontic archwires. Biomed Mater Eng 1992;2:51-69. 15. Fontana MG, Greene ND. Corrosion engineering. New York: McGraw-Hill;1967: 28-48. 16. Conway BE. Theory and principles of electrode process. New York: Ronald Press (2o.;1965:101-25. 17. American Society for Testing and Materials. X-ray powder data file, No.26-792: 1960. 18. Tomashov ND. Theory of corrosion and protection of metals. New York: MacMillan;1966:228-48. 19. Cigada A, Rondelli G, Vicentini B, Giaasmazzi M, Roos A. Duplex stainless steels for osteosynthesis devices. J Biomed Mater Res 1989;23:1087-95.

AAO MEETING CALENDAR 1998 - - Dallas, Texas, May 16 to 20, Dallas Convention Center 1999 - - San Diego, Calif., May 15 to 19, San Diego Convention Center 2000 - - Chicago, II1., April 29 to May 3, McCormick Place Convention Center (5th IOC and 2nd Meeting of WFO) 2001 - - Toronto, Ontario, Canada, May 5 to 9, Toronto Convention Center 2002 - - Baltimore, Md., April 20 to 24, Baltimore Convention Center 2003 - - Hawaiian Islands, May 2 to 9, Hawaii Convention Center 2004 - - Orlando, Fla., May 1 to 5, Orlando Convention Center