Oxide Films Formed on Alloys at Moderate Temperatures. Electron

G. S. DOUGLAS , J. M. ZANDER. Journal of the American Ceramic Society 1951 34 (2), 52-59 ... Journal of the Royal Microscopical Society 1950,123-141 ...
0 downloads 0 Views 17MB Size
Oxide Films Formed on

Alloys

at Moderate Temperatures

Electron Diffraction a n d Electron Microscope Study EARL A. GULBRANSEN, R.

T. PHELPS, AND J. W. HICKMAN, Westinghouse

films formed at temperatures of the order of 1000" C. existed in the form of three layers. In a study of eight alloys the two outer layers contained only very small amounts of the alloying elements compared with the original steel except in the case of manganese steel. Kith a few exceptions nonferrous elements present in the alloy before oxidation were concentrated in the innermost of the three layers of scale. Pfiel (14) considers the oxidation process as one involving iron atoms diffusing outward through the oxide film. Recently Kornilov and Sidorishin ( 1 1 ) investigated the oxide films formed on iron-chromium-aluminum solid solution alloys by electron diffraction and chemical analysis in the temperature range 400" to 1000" C. At moderate temperatures an isomorphous mixture of oxides of the spinel type was found. Kith increasing temperature the lattice constant of the solid solution of the oxides decreased and approached that of pure ?-alumina. The oxide film which formed on the surface crumbled readily and had no protective properties.

Supplementing previous papers, electron microscope and electron diffraction data are presented concerning the structure of electrochemically and chemically stripped films from a series of 16 alloys consisting principally of iron, cobalt, nickel, and chromium. Both reflection and transmission methods of electron diffraction were used.

T

HE nature of the protective properties of metals and alloys is a question of great importance in modern technology. Many theories have been proposed, some of which are based on the role of the oxide film in preventing further reaction. .4 complete knowledge of the chemical and physical structure of the oxide.crystals in the surface film may aid in determining the necessary conditions for protection, but unfortunately, a complete knowledge is impossible for very thin films with the present state of instrumentation. In previous papers (8, 10, 16) the use and limitations of the electron diffraction and electron microscope techniques were discussed. This paper presents electron microscope and electron diffraction data concerning the structure of electrochemically and chemically stripped films from a series of 16 alloys consisting principally of iron, cobalt, nickel, and chromium.

Table

1.

Analysis and Preparation of A l l o y Specimens

SURVEY OF LITERATURE

The literature ,on the removal of oxide films from iron and other metals by chemical and electrochemical methods developed largely by Evans and co-workers (4) was reviewed in a previous paper (16). The electrochemical method has been applied to the stripping of oxide films from alloys of iron and chromium by Evans and Stockdale ( 5 ) . A very thin film of oxide was stripped from 13% chromium steel by a long anodic treatment. The a m , which carried striae of the original metal surface, also contained opaque flakes. These workers also obtained from 18-8 stainless sbeel a thin skin which contained large amounts of residual metal. The film was opaque to light. The chemical method has been applied with success to the removal of the surface film on 18-8 stainless steel by Vernon, Wormwell, and Nurse (17). The metals present in the film were determined by chemical analyses and the contents of the corresponding oxides computed. The thickness of the oxide flm increased with the degree of polish. The, effect of polishing was to enrich the chromium in the film as compared with the underlying steel. The surface film of brightly polished specimens contained 90% chromic oxide. Nickel, on the other hand, was not found to be enriched in the surface film. The authors have suggested that the enrichment of chromium is associated with surface flow, I n the macroscopic film thickness range considerable information is available on the enrichment and concentration of the several metals making up the oxide film. Pfiel (14) has made chemical analyses on films of the order of 0.25 cm. (0.1 inch) or more in thickness. These

Electric Corp., Pittsburgh, Pa.

Analysis

Alloy Mild steels

0.18 C. 0 , 0 2 8 S, 0.030 P

Heat Treatment, Hours a t 1000° C., Furnace Cooled 1.5 in dry HI

13 CrFe

13 Cr

10 in dry H2

18-8 SS

18 C r 8 Ni

15 in Ammogas

K42B"

41.7 Ni 2 2 . 5 C o 20 OCr 1 2 . 0 Fe, 1 . 8 9 ' T i . ' 0 . 2 i A1 0 . 5 7 M n 1 . 2 1 S i 79.5' Ni. 13 Cr, 6 . 5 Fe. 0 . 2 5 Mn, 0.25 Si, 0.08 c. 0.20 c u

10 in wet H z

80 Ni, 20 Cr

14 in Ammogas

30 CoFe"

30.4 Co,0.22 C

10 in dry HZ

Hipernik

49 Ni, 49 Fe, 2 M n

10 in dry HI

5 CrFeO

4.63 Cr, 0.044 C

10 in dry HI

5 NiFeO

4.89 Ni, 0.015 C

10 in dry HI

Inconel (3) Nichrome

V

15 in rZmmogas

5 SiFe

.......

10 in dry HI

3 VFe

.......

10 in dry HI

Kovar

54 Fe, 18 Co, 28 Ni

10 in dry HI

4 WFea

4.10 W,0.028 C

10 in dry HI

a

Analyced at research laboratories.

640

Polishing Emery

papers

through

2 / 0 , uniwax wheel, 320

abrasive, uniwax wheel 600 aloxite. No. 3 alumina Emery papers through 1/0. wax wheel, 320 abrasive. chrome rouge, Nos. 1 and 3 alumina Emery papers through 2 / 0 , chrome rouge, Nos. 1 and 3 alumina Emery papers through 2 / 0 , chrome rouge, Nos. 1 and 3 alumina Emery papers through 1/0, 320 abrasive wheel, chrome rouge, Nos. 1 and 3 alumina Emery papers through 1/0. 320 abrasive wheel, chrome rouge, Nos. 1 and 3 alumina Emery papers through 3/0,chrome rouge, Nos. 1 and 3 alumina Emery papers through 1/0, wax wheel, 320 abrasive, chrome rouge, Nos. 1 and 3 alumina Emery papers through 1/0. wax wheel, 320 abrasive, chrome rouge, Nos. 1 and 3 alumina Emery papers through 1/0. wax wheel. 320

1 and 3 alumina Emery papers through 3/0. chrome rouze. Nos. l ' a n d 3 aluminaEmery papers through 3/0. chrome rouee. Nos. l'and 3 alumina- ' Emery papers through 3/0,chrome rouge, Nos. 1 and 3 alumina Emery papers through 3/0,chrome rouge, Nos. 1 and 3 alumina

ANALYTICAL EDITION

October, 1946

64 1

A P P A R A T U S AND M E T H O D

The details of the apparatus and methods used in this study have been described (8, 10). The specimens of the alloys are heat-treated and given a metallographic polish according to the procedures tabulated in Table I. The samples are then mounted in the electron diffraction camera furnace and oxidized under carefully controlled conditions of time, temperature, and oxygen pressure. The surface lattice structure is studied in situ by the electron diffraction reflection method. Three photographs are taken: (1) the sample in vacuum and a t the t,emperature of the experiment; (2) after the oxidation and while the specimen is a t the elevated temperature; (3) after the specimen is cooled under vacuum conditions to room temperature. The sample is now removed and cut into two pieces. Light micrographs are made of the surface of one half of the specimen by reflected light a t 100 and 1000 X. The other half is subjected to the electrolytic or chemical stripping techniques (15). After the film is loosened, it is washed and then manipulated carefully onto a small stainless steel specimen screen. The screen and specimen are vacuum-dried before being placed in the microscope for study. Several electron micrographs are taken at 6800 X of typical portions of the stripped oxide film. The film is also studied by the clect.ron diffraction transmission method, using the electron diffraction adapter of the elect,ronmicroscope.

Table

II. Lattice Parameters of Metals and Metallic Oxides a

b

C

a

Structural Type

2.86 3.16 2.88

...

... ... ... ...

..... .....

4.07

.....

Body-centered cubic Body-centered cubic Body-centered cubic Body-centered cubic Face-centered cubic Hexagonal close-packed Cubic Cubic (diamond) Face-centered cubic Face-centered cubic Face-centered cubic Tetragonal Rhombohedral Rhombohedral Rhombohedral Rlonociinic Rhombohedral Orthorhombic

3 03

3.52

2.61

8.89 5.42 4.28 4.2j 4. l i

4.86 5.3.5

5.42

...

7 2'8 5.43 11.48 4.90 4.44 4.44 5.75 8 40 8.11 8.32 8.42 8.3; 8 3-

8.35 8.51 8.34 8.39

...

,.. ... ...

...

...

...

...

... ... ...

...

2.77

... 7.48

3.82

4.36

3.55 5.39

... ... ...

... ...

...

... ... ... ... , . .

...

...

...

... ...

...

...

...

...

...

...

... ... ...

...

...

2.89 9.42

... ... ... ... ... ... ... ... ...

...

..... ..... .....

.....

..... ..... .....

.....

55'17 +'?

..... .....

53053'

..... .....

..... .....

.....

..... ..... ..... ..... ..... .....

.....

..... .....

.....

C H O I C E AND P R E P A R A T I O N OF SPECIMENS

The alloys in this study may be divided into four genera1 classes, although a given alloy may belong to two or more classes-for example, both Inconel and K42B are refractory alloys but they also have good protective qualities. The following classification is used in the discussion of results: 1. Protective. 13 CrFe, 18-8 stainless steel 2. Refractory. K42B, Inconel, Nichrome V 3. Magnetic. Hipernik, 30 CoFe 4, Sealing. Kovar

In addition t o commercial alloys, certain experimental alloys were made to determine whether the lattice type of the alloying element would have any effect on the oxidation products. These ferrous alloys contain alloying metals of the body-centered cubic, face-centered cubic, and other structural types. In each the percentage of the alloying metal does not exceed 5 7 , . The specimens aye machined from bars of the alloys to cylinders of 0.84 em. (0.375-inch) diameter and 0.94 cm. (0.375-inch) length. After cleaning, they are heat-treated at elevated temperatures in dissociated ammonia or wet or dry hydrogen. The specimens are next given a fine metallographic polish. Details of the heat-treatment and polishing procedures are given in Table I. The specimens are stored in a desiccator over anhydrous calcium chloride until used. The alloy specimen is placed in the electron diffraction camera furnace and heated to the desired temperature. Oxygen to a pressure of 0.1 atmosphere is admitted and the specimen is oxidized for a predetermined time. The time and temperature conditions of the oxidation are estimated from rate measurements where these are available. For most of the alloys no rate measurements are available, so that intelligent guesses as to oxidation conditions are necessary in order that the oxide film thickness shall lie within the optimum range for investigation with the electron microscope. INTERPRETATION OF D A T A

*

ELECTROX DIFFRACTION. The electron diffraction reflection method has been discussed in previous papers (8, 10, 16) and the interpretation of data by the transmission technique has been discussed in a recent work on metals (15). The transmission method yields information on the structure of the whole film,while the reflection method may indicate only the structure of the outer surface. This is important, since the surface structure may not be the same as the structure of the body of the film. The oxide layer which forms initially may consist of various oxides present on the surface in the same mole ratio as the metals in the alloy. As the oxidation proceeds a stratification of layers of the several oxides may occur, since the factors influencing their formation may

be different. Thus, in a binary alloy, an oxide of one of the metals may concentrate in the surface layer while an oxide of the other metal may concentrate in contact with the metallic substrate. The interpretation of the information obtained by the electron diffraction method is more difficult in the case of alloys than of metals. Because of the similarity -of the lattice parameters of many of the oxides, there is difficulty in distinguishing cr-Fen08 from CrZO3,Pl'i0.Fen03from Fe304,etc. Table I1 shows the lattice parameters of the various oxides of interest in this study. Other complicating factors are the phenomena of various types of solid solution of the several oxides in each other. For example, on an iron-chromium alloy where both a-Fe203and CrnOa may be present it is difficult by the electron diffraction technique t o determine whether a-FezOs or Cr203is present or a solid solution of one in the other. There may also occur a random replacement of ferric ions by chromic ions in the Fe304(spinel type) lattice. The occurrence of such solid solution phenomena may have effects on the lattice parameters and thus complicate the identification of the oxidation products. An interesting comparison can be made of the reflection and transmission electron diffraction patterns. In general, one should expect to find certain oxidation products by reflection and additional products by transmission. However, the oxides in the outer layer as determined by reflection may comprise such a small fraction of the complete oxide film that one or more of them may not give good diffraction patterns in the transmission investigations. The two methods of studying surELECTRON MICROSCOPE. faces of opaque bodies by means of the electron microscope have been compared in a recent work on metals (15). Since we are interested primarily in the details of crystals making up the body of the oxide film, the stripped film technique is used exclusively in this study. INFORMATION RECORDED.Electron micrographs of the stripped oxide film are taken at 6800X and enlarged optically to 34,OOOX. The following information is recorded from the electron micrographs of the stripped film: (1) particle size, (2) particle size distribution, (3) particle shape, (4)uniformity in film thickness, and ( 5 ) type of micrograph. The particle size is obtained by averaging measurements on a number of crystals, while the particle size distribution indicates the variation in particle size. The particle shape is determined from an examination

642

Vol. 18, No. 10

INDUSTRIAL AND ENGINEERING CHEMISTRY Table 111.

Electron Diffraction Data

(Oxide filnls formed on alloys a t 0.1 atmosphere of 0%a t various times and temperatures' Time Of

Preoxidation by R

O C .

600 600 600

5 5 40

None a-FezOa, D FesOi. 8 45, a-FeaOs*, D

CrzOS* 110 FezOc* 8.44, a-FenOa, S FezOi* 8.45, a-FaOa*, S

CrzOs, 0 FezOi*, 8 44, a-FaOa, S FesO4, 8 45, a-FezOa, 8

CrtOa, Fe:O4*. 8.41, e CrzOs, S CrtOs. 8

Refractory K42B K42B Inconel Inconel Kichrome V Kichrome V

600 600 600 600 600 600

5 30 5 30 5 30

a-FezOa, D a-Feaor. D FesOa. VD FeaOc, VD CrzOsl VD CrzOal 1 D

FesO4, 8.43, S FeaOi, 8.44, a-FerOa, S FesO4, 8.43, S FeaOi, 8 , 4 3 , 8 CrzOa, S CrzOa. S

FzsOi, 8.43, S FesOi, 8.44, a-FezOa, S FeaO4, 8.43, S FeaOd, 8 . 4 3 , 8 crzoa, S CraOa, E

crtO3. FeaOi, 8 36, S CrtOa. FeaOi*, S Spinel, 8:32, CrrOa, Ni, OD CrrOa, N!O, VDO CrzOa, S CrrOa, >I

Magnetic Mild steel Mild steel hfild steel 30CoFe Hipernik

250 250 300 300 300

5 30 5 30 5

None None Fe&. S. 43, a-FeaOa, DO FeaOc, I D Peso&,310

FerO4, 8.43, D O FeaOc 8.42 31 FeaOl: 8.43: a-FezOs, 310 FeaO4, 8.43, h l Feror. 8.42, a-FezOa, S

FeaO4, FeaO4 FeaOi: FegOi FeaOi:

FeaOl. 8.36. hI FeaOl, 8.39, VS FeaOi. 8.41, SO FeaOi, 8.36, 31 Spinel, 8.34, a-FezOi, NiO, SO

Sealing Kovar

400

5

F d i , 8.43, 31

FesOi, 8.43, SO

FeaOi, 8.43, SO

Miscellaneous 5 CrFe 6 XiFe 6 CoFe 6 MnFe 6 SiFe 3 VFe 4 WFe

400 300 300 300 300 300 300

5 5 5 10 5 5 5

FeaOc, DO Sone Fes04, 8 43, 31 a-FezOaI VD

FesO', 8 43, FeaOl, 8.43, FeaOi. 8 43, FeaOi, 8 43, FeaOi, 8 43, FesOi. 8.42, a-FezOa, 8

Temp.

Protective 13 CrFe 18-8 SS 18-8 SS

R, reflection.

T , transmission.

S, sharp.

SI, medium.

Oxidized by R

Diffraction Patterns Oxidized a t 25' C. by R

Oxidation Min.

Alloy

'

8.43, 8.42 8.43: 8.43 8.42:

DO \I &-FezOi, hIO 31 a-FezOa, S

.

Stripped by T

F e b , 8.35, S

CrzOa, *S a-FezOa, 31 hl

S D 0 D

D, diffuse. 0 , oriented.

V , very.

*, trace

rial. These features are of importance in classifying a micrograph but may not be easily interpreted. LIGHTMICROSCOPE. Light micrographs of the oxidized surface are taken at 1OOX and 1OOOX by the use of reflected light. The oxidation process acts like a chemical etching solution in revealing the grain boundaries of the grains in the metal or alloy surface. The light micrographs also reveal the presence of inclusions and the roughness of Table IV. Lattice Parameter Deviations of Oxides the surface layer. Reflection Patterns Transniission Patterns Composition Composition SPECTROSCOPIC AS.\LYSIS. Whenever possible, Time of and and sections of the stripped films were analyzed specTemp. Oxidation parameter Deviation parameter Deviation Min. troscopically. In some cases these analysee c. % % showed the presence of metal's not found in the CrzOa f0.04 5 f0.30 600 CrzOa FeaOi, 8 . 4 1 f 0 . 12 5 600 alloys but present in the oxides used in the polisha-F&O; +.O.'52 .... .... 5 600 .... .... ing procedures. FesO,. 8 . 4 4 f0.48 5 600

of the more typical shapes in the pattern. Uniformity in film thickness refers to the presence of thick and thin portions of the film. The type of pattern refers to a number of features, including: (1) the sharpness of the crystal edges, ( 2 ) the presence of overlapping crystals, and (3) the presence of extraneous mate-

Alloy

Protective 13 CrFe 18-8 SS

600 600 600 600

18-8 88

Refractory K42B

600 600 600 600

K42B

600

Inconel

600 600 600 600

Inconel

600

Nichrome V Magnetic Mild steel hlild steel Mild steei 30 CoFe Hipernik

,

Sealing . Kovar Miscellaneous 5 CrFe 5 NiFe 5 CoFe

5 MnFe 5 SiFe

5 SiFe 3 VFe 4 KFe

600 600 600

....

Cr?On

-0.02

5 40 40 40

a-FezOa FeaO4, 8 . 4 5

+b'.'30 f0.60

.... .... crzo2

5

FesO4, 8 . 4 3

$0.36

Peso,, 8 . 3 6

-0.48

f'd

f0.07

30 30 30 5 5 5 30 30 30 5 30

FeaO;,'8.44 a-FezOa

CrrO?

FeaO;,' 8 . 4 3

f'd.86

FeaO;,'8. 43

+'0 .'3 6

Cr203 CrzOa

+'O. '2 5

+0.30

CrzOE Xi0 CrzOr CrzOa

FesOi, 8 . 4 3 FeaOi, 8.42 FesOi. 8 . 4 3 a-FezOs FesOr. 8 . 4 3 FesOl. 8.42 a-Fez02

$0.36 +0.24 +0.36

FerOr, 8.36 FeaOi, 8.39 FerOi, 8 . 4 1

$0.36 1-0.24 +o. 10

0.48 -0.72

...

FeaO;,' 8.36 Spinel, 8.34 a-FezOa Si0

5

....

....

.... ....

..,

.'48 +0.32

...

...

....

.... ....

00.0

....

....

...

CroOr +0.07 Spinel, 8. 33 .... CrzOa -0.11 NiO Diffuse

....

diffuse Too diffuse -0.04 +0.04

TOO'

-0.48 -0.12

250 260 300 300 300 300 300 300

5 3;

400

5

FesOi, 8.43

f0.36

FejOi, 8 . 3 5

-0.60

5

FeaOi, 8.43 FeaOl, 8 . 4 3 a-Fe?Os FesO4, 8 . 4 3 FesOl, 8 . 4 3 FesOl, 8 . 4 3

f0.36 f0.36 i , n 2.5 +0.36 f0.36 f0.36

FesOi, S 34 FeaO4, 8 34

-0 72 -0.72 .. .. . -0.60 -0.96

FeJO;.. 8 . 42 a-FezOs

+'O.'24 + O . 16

400 300 300 300 300 300 300 300 300

5

30 5

6 5

!

?

10 5

5 5

5

....

10.25

FesOc.8.35 SpineL8.32 a-Fe;O; Spinel, 8.33 a-FezOa

4-0.12

....

-0.03

0.00

....

0.00 -0.84

fO.08

RESULTS

ELECTROX DIFFRACTION. The results of the electron diffraction study are shown in Table 111. Three reflection and one transmission patterns are taken of the oxide film formed on the surface of the alloy, The first reflection pattern of the metal is taken in the vtacuum of the camera before oxidation and at the temperature of the experiment. The second is taken after the ouidation and at the temperature of the experiment The third is taken after cooling the oxidized sample to 25" C. in a vacuum. The transmission pattern of the stripped oxide film is also included in order to compare the body structure of the film with its surface structure. Table 111 shows the conditions of oxidation, the chemical structure of the surface oxide, the unit cell size (Ao) where readily calculable, and the type of diffraction pattern obtained. Let us consider the oxide film formed on mild steel a t 250' C. and 0.1 atmosphere of oxygen with an oxidation time of 5 minutes. A diffuse and oriented pattern of FeaOI is found by reflection. The unit cell size is calculated to be 8.43d. After

.

ANALYTICAL EDITION

October, 1946

643

deviations of the oxides formed in each oxidation experiment are Composition and Parameter Oxidizing shown in Table IV. This table Conditions Spectrographio Temp. Time Alloy shows the conditions of oxidaTransmission Reflection X-ray d a t a (literature) data C. Min. tion, the composition, and unit Protective 13 CrFe 600 5 FeaOa*, 8 . 4 1 , CrzOl CrzOa FeaO4 8 . 4 0 Cr, Fe, Ag* cell size where readily calculable 18-8 SS 600 5 CrzOs FeaOa*, 8.44. FeaO4: 8 . 4 0 ....... *. and the deviation in per cent a-Fez03 18-8 SS 600 40 CrzOs FeaOa 8.45, FerO4, 8 . 4 0 ......... from the accepted x-ray values a-F' ez0a Refractory given in Table 11. Both the reK42B 5 600 Fea01 8 . 3 6 CrzOa FesO4, 8 . 4 3 FeaO4, 8 . 4 0 Cr, Fe* (Mg) flection a n d t r a n s m i s s i o n K42B 600 30 ~ e r ~ aCi r,h a FeaO4, 8 . 4 4 , FeaOa, 8 . 4 0 . . . . a-FezOa patterns are tabulated. The deSpinel, 8 . 3 2 , CrzOa, FeaO4, 8 . 4 3 Inconel 600 FeaOa, 8 . 4 0 , NiO.CrzOa, Cr, Ni, Fe 5 NiO 8.31 (Al, Si, Co) viations are calculated from the 30 Inconel 600 CnOa, NiO FeaO4, 8 4 . 0 FesOa, 8 . 4 3 unit cell size as given in Table Nichrome V 5 600 CrzOa CrzOr .... Cr, Fe* No Ni) 30 Nichrome V 600 CrzOa CrzOa .... Cr, Fe* [No Ni) I11 or from the d / N values for Magnetic those oxides which do not obey Mild steel 260 5 FeaOa 8 36 FesO4, 8 . 4 3 Fea04 8 40 ....... Mild steel 30 250 FeaOa: 8:39 FeaOa, 8 . 4 2 FeaOa: 8:40 the cubic lattice. Mild steel 5 FeaO4, 8 . 4 1 300 FesOa 8 . 4 3 , FesO4, 8 . 4 0 .. .. . . A summary of the electron a-F'erOs Hipernik Spinel, 8.,34, a- Feaoa 8.42, FeaOa, 8 . 4 0 , NiO.FezOs, M n , Ni, Fe* 400 5 diffraction and spectrographic FezOa, N i 0 a-Fez0a 8.34 30 CoFe Spinel, 8 . 3 6 300 30 FeaOa, 8 . 4 0 , CoO.Fez01, Fe, Co, (Cr*) FeaOa, 8 . 4 3 data obtained on the 16 alloys is 8.39 shown in Table V. Both the Sealing Kovar 400 5 FesOd, 8 . 3 5 FeaOa, 8 . 4 3 FeaO4, 8 . 4 0 transmission and reflection data Miscellaneous are given, together with the 5 CrFe 5 Spinel, 8 . 3 4 400 FeaO4, 8 . 4 3 FesOa, 8 . 4 0 , FeO.Cr203, C r , Fe* 8.35 x-ray data obtained from the 5 NiFe 300 J Spinel, 8 . 3 4 FeaO4, 8 . 4 3 FesO4. 8 . 4 0 , NiO.FezOa, . . .. literature. 8.34 5 CoFe 5 Spinel, 8 . 3 5 300 FeaOr, 8 . 4 3 FerO4, 8 . 4 0 , CoO.FezOs, Fe, Cu, (Al*), ELECTRON MICROSCOPE.Fig8.39 No, Co 5 MnFe Spinel, 8 . 3 2 10 300 Fe:O!,-8.40, MnO.FezOa, Fe, M n , (Cr, Al*) FeaOa, 8 . 4 3 ures 1to 11show the light micro8.51 graphs, electron micrographs, 5 SiFe 5 300 a-FezOa FeaOa, 8 . 4 3 Fer04 8 40 ......* . 3 VFe 300 Spinel, 8 . 3 3 5 FesOd, 8 . 4 2 FesOl: 8 : 4 0 ......... electron diffraction transmission, 300 4 WFe 5 a-FezOs a-Fez08 and electron diffraction reflection * Trace. patterns for the stripped oxide ~ m from s the alloy specimens. The lengths of 1, 10, or 100 micooling the sample to room temperature, the oxide film is stripped crons are shown on the photographs. The light micrographs of from the metal. A medium pattern of FesOd is found using the the unstripped oxide film are taken at 100 and lOOOX, while the transmission method on the stripped film. The unit cell size is electron micrographs of the stripped oxide film a r e t a k e n a t 6 m X calculated to be 8.36 A. Figure 10,b and c, shows the electron diffraction photographs for this oxidation experiment. I n a similar and enlarged optically to 34,OOOX. These micrographs are remanner the results of the other experiments on mild steel and the duced subsequently in the printing process. The actual magnififteen other alloys are shown in Table I11 and Figures 1to 11. fications can be readily calculated from the fact that each centimeter of the length of the micron line shown in the micrograph I n general the lattice parameters obtained in this study deviate equals 10,OOOX. from the accepted x-ray diffraction values. The lattice parameter Table

V.

Summary

of Electron Diffraction and Spectrographic Data

.... .

.. ............

......... . . . .. .

Table

Alloy Protective 13 CrFe 18-8 SS

Oxidizing Conditions 0.1 Atmosphere of 01 Min. OC.

VI.

Electron Microscope Analyses

Film Color on Metal

Figure

of Stripped O x i d e Films of Alloys

Particle Size Size, Distribution,

A.

A.

1 2 3

450 300 700

200 to 700 100 to 600 300 to 1500

Irregular Irregular Irregular

Nonuniform Uniform Nonuniform

Medium, overlapping crystala Medium Sharp, overlapping crystals Medium. clusters of crvstals Sharp, clusters of crystds Medium, overlappin crystals, grain boundarie? ow Medium, overlapping crystals Medium, overlapping crystals Medium, overlapping crystals

Shape

Uniformity

5 5 40

600 600 600

Blue Blue and brown Yellow and blue

5 30 5

600 600 600

Yellow and mauve Reddish blue Light blue

..45

350 450 400

200 to 750 300 to 800 300 to 800

Irregular Irregular Irregular

Nonuniform Nonuniform Nonuniform

30 5 30

600 600 600

Red Light blue Yellow

'i

..

450 250 350

300 to 900 100 t o 450 200 to 550

Irregular Irregular Irregular

Nonuniform Fairly uniform Fairly uniform

5 30

250 250

Dark blue Dark blue

..7

700 800

300 to 1600 400 to 1200

Irregular Irregular

Nonuniform Nonuniform

30 CoFe

5 30

300 300

Light blue, hiatus Light mauve

750 700

250 to 900 300 to 1200

Nonuniform Nonuniform

Hipernik

6

400

Blue

Kovar

5

400

Pink and light blue

Refractory K42B Inconel Inconel Nichrome V Magnetic Mild steel

Miscellaneous 5 CrFe

5

400

Light blue

5 NiFe 5 CoFe

5 5

300 300

Yellow and mauve Pink and blue

5 MnFe 5 SiFe 3 VFe

10 5 5

300 300 300

Light,blue, hiatus Reddish-blue Mauve and blue

4 WFe

5

300

Mauve

.8. ..

Type of Micrograph

sf

Medium overlapping crystals Medium, overlapping crystals, striations Medium, overlapping crystals Sharp, overlapping crystals

300

200 to 500

Irregular Irregular, angular Irregular

..

400

250 to 750

Irregular

Uniform, thicker a t grain boundaries Nonuniform

Medium, overlapping crystals. grain boundaries show Medium, clusters of small crystals

9

250

100 to 400

Irregular

Nonuniform

.. ..

600 400

250 to 1000 250 to 750

Irregular Irregular

Nonuniform h-onuniform

Medium, chains of clusters of crystals Medium, overlapping crystals Diffuse, overlapping crystals. clusters of crvstals

350 500 250

250 to 600 250 to 750 200 to 300

Irregular Irregular Irregular

Uniform Uniform Uniform

300

200 to 400

Irregular, indistinct

Fairly uniform

10 11

.. ..

INDUSTRIAL A N D ENGINEERING CHEMISTRY

614 Figure 1.

Oxide Film of 13% ChromeIron, 1 3 Cr 5-600

a

Vol. 18. No. 10 c

D

a. Elechon micmgmpb,rbimed film b . El~clmndilirstion hsnmirsion. stripped film Eloalron diliraction reflection, film on mohl d. e. Light micrographs, film on m e l d

e.

Table VI summarizes the inform& tion recorded from the electron microscope: (1) color of the oxide film on the metal, (2) particle size in AnLngstr6ms, (3) particle size distribution, (4) partide shape, (5) film uniformity, and (6) type of micrograph.

e

DISCUSSION

The factors which determine the chemical and physical structure of the oxide film on the surface of an alloy have been discussed in a previous paper (10): 1. Rates of formation and diffusion

of the various metal ions and electrons through the oxide lattice 2. The rete of diffusion of the

OWPRTI

~ O I P C I I I P ion nr ntom t,hmnsh

H

loop

a

Stainless Steel, SS5-600 pped film nimion itrippod film ti"" A/," 0" metal ... .., ...... 1 on m0t.I

ANALYTICAL EDITION

October, 1946

Figure 3.

h

1

045

Oxide Film of 18-13 Stsinless Steel, SS40-600