UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An electron diffraction and fractographic study of stress corrosion and cracking 1972

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1972_A1 B57_5.pdf [ 26.18MB ]
Metadata
JSON: 1.0078976.json
JSON-LD: 1.0078976+ld.json
RDF/XML (Pretty): 1.0078976.xml
RDF/JSON: 1.0078976+rdf.json
Turtle: 1.0078976+rdf-turtle.txt
N-Triples: 1.0078976+rdf-ntriples.txt
Citation
1.0078976.ris

Full Text

AN ELECTRON DIFFRACTION AND FRACTOGRAPHIC STUDY OF STRESS CORROSION CRACKING by STUART SAMUEL BIRLEY B.Sc. University of Birmingham, 1965 M.A.Sc. University of British Columbia, 1970 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of METALLURGY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA September, 1972. I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r a n a d v a n c e d d e g r e e a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e t h a t t h e L i b r a r y s h a l l m a k e i t f r e e l y a v a i l a b l e f o r r e f e r e n c e a n d s t u d y . I f u r t h e r a g r e e t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d b y t h e H e a d o f my D e p a r t m e n t o r b y h i s r e p r e s e n t a t i v e s , i t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . D e p a r t m e n t o f M e t a l l u r g y T h e U n i v e r s i t y o f B r i t i s h C o l u m b i a V a n c o u v e r 8, C a n a d a D a t e S e p t e m b e r 11, 1972 On Stress Corrosion The image of stress corrosion I see Is that of a huge unwanted tree, Against whose trunk we chop and chop; But which outgrows the chips that drop; And from each gash made in i t s bark A new branch grows to make more dark The shade of ignorance around i t s base, Where Scientists t o i l with puzzled face. Chemists and metallographers, Technicians and philosophers, Though struggling individually, Their common goal: to f e l l the tree. S.P. Rideout Savannah River Laboratory ABSTRACT An expression was developed, using a kinematical approach, for the relative theoretical intensities of Debye rings formed on electron diffraction patterns obtained from randomly oriented powder aggregates. Calculations of relative theoretical intensities of Debye rings for specific crystal structures were compared with relative visual intensities of Debye rings obtained by high energy electron diffraction from these actual crystal structures. Good correlations were obtained and were taken to j u s t i f y the theoretical approach. An electron diffraction and fractographic study was made of the transgranular stress corrosion cracking of 310 and 304L austenitic stainless steels in boiling aqueous magnesium chloride solution and of 0-brasses (Cu-47.76% Zn, Cu-45.54% Zn, Cu-33.5% Zn-4.5% Sn) i n water, and i n ammoniacal copper sulphate solution. High energy electron diffraction was applied to protuberances on mechanically fractured and stress corrosion fracture surfaces to detect and identify surface corrosion product films and surface trans- formation products. Both metallic transformation products and non- metallic phases were detected on the fracture surfaces. Fractographic techniques were employed (1) to characterize the fracture surfaces and (2) to match up the opposite fracture surfaces produced by stress corrosion cracking. The latter technique established whether or not the microstructural features associated with the topo- graphical features were formed before the crack passed. Etching techniques were employed to detect evidence of transformation products adjacent to stress corrosion cracks and the fracture surfaces. - iv - The observations pointed to the conclusion that the transgranular stress corrosion crack path i n the systems under examination is dictated by the presence of a s train induced transformation product. Moreover, cracking along the interface between the strain induced and parent phases can account for the quasi-cleavage stress corrosion fracture surface topographical features; unusual fracture modes are not required. It was suggested that the conductivity of the oxide product films governs, to some degree, the rate of stress corrosion cracking. - v - ACKNOWLEGEMENTS I sincerely wish to thank D. Tromans for a l l his advice, help and encouragement throughout this period of study. I am very grateful to my wife, Louise, whose considerable encouragement, assistance and sacrifices accompanied the realization of this work. The investigation was conducted with financial assistance from the Defence Research Board (Canada) under grant number 9535-50. - v i - TABLE OF CONTENTS Page 1. INTRODUCTION 1 1.1 Electron Diffraction Studies of Stress Corrosion Cracking 3 1.1.1 Detection and Identification of Corrosion Product Films 3 1.1.2 Orientation Studies 7 1.2 Fractographic Studies of Stress Corrosion Cracking. 9 1.3 Objects of Study and Choice of Material ^ 2. EXPERIMENTAL PROCEDURE 1 7 2.1 Materials Preparation and Testing ^ 7 2.1.1 General 1 7 2.1.2 Materials 19 2.1.2.1 Steels 1 9 2.1.2.2 Brasses 1 9 2.1.2.3 Environments 21 2.1.3 Specimens 21 2.1.4 Mechanical and Stress Corrosion Testing .... 22 2.2 Electron Diffraction 2 8 2.2.1 Introduction 2 8 2.2.2 Intensities of Electron Diffraction Patterns 29 31 2.2.3 Demonstration of Intensity Effects 2.2.4 Theoretical Estimates of Relative Intensities -37 for Electrons - v i i - Page 2.2.4.1 Theoretical Aspects 3 7 2.2.4.2 Justification of Theoretical Aspects 41 2.2.5 Procedure for Identifications of Unknown Phases 47 2.2.6 Application of Electron Diffraction Technique to Fracture Surfaces 47 2.3 Fractography and Metallography 49 2.3.1 Fractographic Techniques 49 2.3.2 Metallographic Techniques 50 3. RESULTS, OBSERVATIONS AND DISCUSSIONS 54 3.1 304L Austenitic Stainless Steel 54 3.1.1 Identification of Phases on Fracture Surfaces 54 3.1.2 Stress Corrosion Crack Path 58 3.1.3 Fractography 61 3.1.4 Discussion 65 3.2 310 Austenitic Steel 71 3.2.1 Identification of Phases on Fracture Surfaces 71 3.2.2 Stress Corrosion Crack Path 74 3.2.3 Fractography 78 3.2.4 Discussion _ 83 3.3 Brasses 88 3.3.1 Identification of Phases on Fracture Surfaces 88 3.3.2 Stress Corrosion Crack Path 97 3.3.3 Fractography 108 3.3.4 Discussion 122 - v i i i - Page 3.4 Corrosion Product Films 126 3.4.1 Oxide Formation on g-Brasses 126 3.4.2 Corrosion Product Formation on 304L Steel... 128 3.4.3 Corrosion Product Formation on 310 Steel ... 140 3.4.4 Discussion 144 3.4.4.1 Austenitic Steels 145 3.4.4.2 g-Brasses 147 3.5 General Considerations 150 4. SUMMARY AND CONCLUSIONS 156 BIBLIOGRAPHY 158 APPENDIX 167 - ix - LIST OF FIGURES Figure Page 1 Outline of experimental procedure * 18 2 Schematic variation of intensities of diffracted and direct beams with distance in .a monocrystal of thickness t 30 3 (a) Schematic of diffraction of incident beam by each of three individual crystallites in a monolayer film. (b) Schematic of multilayer film showing crystallites A and B contained in a column of width XY 30 4 Schematic of c r y s t a l l i t e arrangement at a protuberance 32 5 Electron diffraction patterns of gold. (a) and (b) result from film, (c) from f i l i n g s 34 6 Geometrical arrangement of Debye r i n g formed on screen 39 7 E l e c t r o n d i f f r a c t i o n p a t t e r n from Cu£0 powder 39 8 E l e c t r o n d i f f r a c t i o n p a t t e r n from CuCl powder 45 9 Diagram of specimen s e t up i n low angle e l e c t r o n d i f f r a c t i o n . (a) macroscopic f r a c t u r e s urface i s p a r a l l e l to i n c i d e n t e l e c t r o n beam, (b) e l e c t r o n t r a n s m i s s i o n d i s t a n c e through surface f i l m i s much great e r than f i l m t hickness and depends on angle of l o c a l surface p r o j e c t i o n to i n c i d e n t e l e c t r o n beam... 45 10 E l e c t r o n d i f f r a c t i o n p a t t e r n s obtained from 304L f r a c t u r e s u r f a c e s . (a) D u c t i l e f r a c t u r e a t +154°C showing predominantly untransformed f . c . c . y phase ( a u s t e n i t e ) . (b) D u c t i l e f r a c t u r e at -196°C showing who l l y b.c.c. a' phase ( m a r t e n s i t e ) . (c) Stress c o r r o s i o n f r a c t u r e i n b o i l i n g aqueous MgCl2 at 154°C showing predominantly b.c.c. a' phase ( m a r t e n s i t e ) . . . 55 11 General path of s t r e s s c o r r o s i o n cracks i n 304L exposed to b o i l i n g aqueous MgCl2 at +154°C. (a) O p t i c a l micrograph of a l o n g i t u d i n a l s e c t i o n of specimen i l l u s t r a t i n g branching nature of the cracks. 20x. (b) Enlargement of area i n (a) showing preference f o r t r a n s g r a n u l a r c r a c k i n g . G r a i n boundaries l i g h t l y etched i n o x a l i c a c i d . 250x. (c) Scanning e l e c t r o n micrograph absorbed e l e c t r o n image of p o r t i o n of a c t u a l crack surface e x h i b i t i n g mixed i n t e r g r a n u l a r (faceted) and t r a n s g r a n u l a r (roughened) crack paths. 300x 57 Optical micrographs demonstrating presence of a' martensite, as revealed by the chromate etch, in the region adjacent to stress corrosion crack tips in 304L steel. The martensite paths are visible as parallel and intersecting striations in both (a) 1050x and (b) 700x Examples of topographical patterns on stress corrosion fracture surface of 304L and comparison of these with martensite etch patterns formed near ductile fracture region at -196°C. 1200x. Figures (a), (b), (c) are electron micrographs of pre-shadowed direct carbon replicas obtained from stress corrosion fracture surfaces. The arrow in (c) denotes direction of microscopic crack propagation for this particular area (AB). Figures aa, bb, and cc are optical micro- graphs of martensite etch patterns as revealed by the chromate etch for aa and cc and the oxalic acid etch for bb Scanning electron micrographs showing examples of mating surfaces xl400. Figures 14(b) shows mixed mode fracture (smooth and roughened regions) and is only partly mated owing to loss of portions of grain Preshadowed direct replica electron micrographs of ductile fracture surfaces of 304L. xl400 (a) Fracture at -196°C (b) Fracture at +154°C Electron diffraction patterns obtained from 310 fracture surfaces. (a) Ductile fracture at +154°C showing untransformed fee y phase (austenite). (b) Ductile fracture at -196°C showing fee y phase. (c) Stress corrosion fracture in boiling aqueous MgCl2 at 154°C showing austenite plus oxide General S.C. crack path in 310 exposed to boiling aqueous MgCl2 at 154°C. Figure 17(a) shows very severe branching i n i t i a t i n g from notched region (N). x33. Figure 17(b) shows preference for trans- granular cracking. x250. Both etched i n oxalic acid Figure 17(c) Optical micrograph showing branched nature of crack and crystallographic preference. x250. Etched in oxalic acid. Figure 17(d) Scanning electron micrograph absorbed electron image of portion of actual stress corrosion fracture surface. x255 - x i - Figure Page 18(a) Optical micrographs showing etch patterns adjacent and to the stress corrosion crack tips. Oxalic acid 18(b) etch. x400. Figure 18(a) partially fractured specimen. Figure 18(b) f u l l y fractured specimen. Note oxalic acid etch markings approximately parallel and perpendicular to one of the cracks 7 7 19(a) Scanning electron micrographs of 310 S.C.C. fracture and surfaces. Figure 19(a) shows smooth facets. x800. 19(b) Figure 19(b) shows outlines of grains containing annealing twins. x250 79 19(c) Scanning electron micrograph showing mating of the two fracture surfaces of a 310 S.C.C. specimen. x260 8 0 20(a) Preshadowed carbon replicas showing "annealing and twinned" topography. Figure 20(a) shows smooth 20(b) regions containing fine striations. xl550. Figure 20(b). xl500 8 1 21(a) Preshadowed carbon replica showing mechanical and fracture topography of 310 steel. (a) mechanically 21(b) fractured at +154°C. xl500. (b) mechanically fractured at -196°C. xl500 . . . 8 4 22 Electron diffraction patterns from fracture surfaces of beta brass (a) Stress corrosion fracture surface - water/Cu-45.54% Zn (b) Stress corrosion fracture surface - water/Cu-33.5% Zn-4.5% Sh (c) Mechanically fractured surface - Cu-45.54% Zn (d) Mechanically fractured surface - Cu-33.5% Zn V 90 23 Stress corrosion crack path in Cu/47.76% Zn binary alloy fractured in ammoniacal copper sulphate solution. (a) Intergranular crack path. x65. Etched ammonium persulphate. Optical. (b) As in (a) showing greater detail of transgranular cracking. xl200. Scanning electron microscope. (c) Transgranular cracks, showing cross hatch pattern. Unetched. x200 98 - x i i - Figure Page 24 Stress corrosion crack path in water/Cu-47.76% Zn alloy. (a) Steps on major fracture surface. Unetched x250. (b) Cross hatch pattern. x250. Etched i n Swann and Warlimont solution 99 25 Stress corrosion crack path in water/Cu-45.54% Zn system. (a) Intergranular and transgranular failure (x200). Etched i n acid f e r r i c chloride solution. (b) Steps at major fracture surface (xlOO). Unetched 28 Shows s l i p traces adjacent to indentation. Cracks are par a l l e l to s l i p traces. Emphasized by arrows.. 100 26 Stress corrosion crack paths of Cu-33.5% Zn-4.5% Sn in water. (a) Intergranular cracking with transgranular side branches. x200 unetched. (b) Cross hatch crack path x200. Etched in acid f e r r i c chloride solution 101 27 S.C. Cracks propagating at 45° to tensile axis (direction of arrows). (a) Binary Cu-45.54% Zn in water. Unetched x200. (b) Ternary Cu-33.5% Zn-4.5% Sn in water. Unetched x200 102 103 29 Martensite morphology adjacent to mechanical fracture surface of Cu-47.76% Zn specimen. x250. Etched in ammonium persulphate 103 30 Martensite morphology in mechanically fractured Cu-45.54% Zn specimens. A l l x250 and etched in acid f e r r i c chloride 105 31(a) Martensite morphology in mechanically fractured 31(b) ternary Cu-33.5% Zn-4.5% Sn alloy. A l l xlOO. Etched and in acid f e r r i c chloride. (b) shows proximity of 31(c) fracture surface 106 31(d) Martensite morphology in ternary Cu-33.5% Zn-4.5% Sn and alloy. x250. Etched in acid f e r r i c chloride. 31(e) Figure 31(d) shows proximity of fracture surface ... 32 Mating stress corrosion fracture surfaces. Cu-47.76% Zn in water. x225 107 109 - x i i i - Figure Page 110 33(a) Mating s t r e s s c o r r o s i o n f r a c t u r e surfaces of and Cu-45.54% Zn i n water. x225 33(b) 33(c) General t o p o g r a p h i c a l f e a t u r e s of Cu-45.54% Zn a l l o y (cc) f r a c t u r e d i n water. (c) x225, showing steps. (cc) (d) Showing area s i m i l a r to c i r c l e d area on ( c ) . x2000. (dd) (d) x225, i n t e r s e c t i n g steps p e r p e n d i c u l a r to machined notch. (dd) D e t a i l s of area s i m i l a r to c i r c l e d area on (d) x2000 1 1 1 34(a) Mating s t r e s s c o r r o s i o n f r a c t u r e surfaces o f and Cu-33.5% Zn-4.5% Sn a l l o y f r a c t u r e d i n water. 34(b) x225 1 1 2 34(c) Topographical f e a t u r e s on s t r e s s c o r r o s i o n f r a c t u r e (cc) surface of the Cu-33.5% Zn-4.5% Sn a l l o y / w a t e r . and (c) x225. (cc) An area s i m i l a r to c i r c l e d area. (d) x2000 (d) An area s i m i l a r to c i r c l e d area. x2000... 113 35 Matching f r a c t u r e surface t o p o g r a p h i c a l f e a t u r e s of Cu/47.76% Zn a l l o y S.C.C. i n Mattssons s o l u t i o n . x225 114 36 Topographical d e t a i l s of mechanically f r a c t u r e d Surfaces, showing dimples c h a r a c t e r i s t i c of v o i d coalescence. (a) Cu-45.54% x2000 (b) Cu-33.5% Zn-4.5% Sn x2000 (c) Cu-47.76% Zn x225 117 37 Topographical f e a t u r e s i n S.C.C. Cu-33.5% Zn-4.4% Sn alloy in ammoniacal copper sulphate. Features are same as etch markings, and can be traced round the i n t e r s e c t i n g l i n e . Etched p o l i s h e d surface i n Hull/Garwood (75) et c h . Scanning e l e c t r o n micrograph. x300 120 38 M a r t e n s i t e adjacent to crack i n Cu-33.5% Zn-4.4% Sn a l l o y . x225. Scanning e l e c t r o n micrograph. Etched i n Hull/Garwood etch (75) 120 39 Matching t o p o g r a p h i c a l f e a t u r e s i n Cu-33.5% Zn-4.4% Sn a l l o y subjected to S.C.C. i n ammoniacal copper sulphate s o l u t i o n . x225 121 40 Topographical d e t a i l s of mechanically f r a c t u r e d surfaces of Cu-33.5% Zn-4.4% Sn a l l o y . x225. Scanning e l e c t r o n micrograph 121 - xiv - Figure Page 41 X-Ray images of Fe (b), Cr (c), and Ni (d) in corrosion product on 304L steel. (a) shows porous nature of corrosion product x255. Absorbed electron image 132 42 (a) Infrared spectrum of corrosion product from 304L austenitic steel. (b) From y-Ye^O^ and Fe^O^.. 135 43 DTA curves of (a) magnetite, (b) corrosion product on 304L steel, (c) corrosion product on 310 steel, (d) magnesium oxychlorides (114,115) 136 44 Corrosion product formed on 310 steel. (a) absorbed electron image x225, (b), (c), (d) X-ray image of Fe, Cr, Ni respectively 143 - XV - LIST OF TABLES Table Page I Composition of Stainless Steel 20 II Composition of ct-Brass 20 III Specimen Heat Treatment 23 IV Details of Test Conditions 24 V Yield Stresses of Materials Subject to S.C.C 27 VI I and I for Au 36 e v VII I and I for Cuo0 Powder 42 e v 2 VIII I and I for CuCl Powder 43 e v IX I and I for Fe„0, Powder 44 e v 3 4 X Electropolishing Reagents 51 XI Etching Conditions 52 XII Identification of Phases on Fracture Surfaces by • Electron Diffraction 56 XIII Electron Diffraction Observations - 310 Steel 73 XIV Electron Diffraction Observations from S.C.C. Surface of Binary 47.76% Zn Alloy 8 9 XV Electron Diffraction Composite Ring Analysis for Binary Specimen (45.54% Zinc ) 91 XVI Electron Diffraction Composite Ring Analysis for Ternary Specimen (Cu-33.5% Zn-4.5% Sn) 92 XVII X-Ray Powder Diffraction Analysis Filings (Cu-45.4% Zinc) 9 4 XVIII X-Ray Powder Diffraction Analysis - Filings of Ternary Alloy (Cu-33.5% Zn-4.5% Sn) 9 5 XIX Electron Diffraction Observations for Ternary Strip Alloy (Cu-33.5% Zn-4.4% Sn) I 1 9 - xvi - Table Page XX Phases Detected by Electron Diffraction on the S.C.C. Surfaces 1 2 7 XXI Electron Diffraction Observations of Oxides Formed on 304L Steel 1 3 0 XXII Chemical Analysis of Corrosion Product by Weight %.. ^3Q v XXIII Analysis of X-ray Diffraction Patterns of Corrosion Product 1 3 3 XXIV Electron Diffraction Observations of Corrosion Products Formed on 310 Steel 1 4 2 XXV Variation of S.C.C. Life with Environmental Additions 1 4 8 A l Relative Intensities for X-rays - Fe^O^ 173 A2 Relative Intensities for X-rays - HFe c0 o 174 J o A3 Relative Intensities for Electrons - Fe^O^ and HFe^Og 175 A4 Structure Factors of Au, Cu^O, CuCl and Austenite .. 177 A5 Structure Factors for (3-Brasses and a-Steel 178 A6 Structure Factors for HFe c0 o and Fe«0, 179 - 1 - 1. INTRODUCTION Str e s s C o r r o s i o n Cracking (S.C.C.) may be def i n e d as the p r e - mature f a i l u r e of m a t e r i a l s under the c o n j o i n t a c t i o n of t e n s i l e s t r e s s and unique environment. Ceramics, p l a s t i c s , and metals are a l l s u s c e p t i b l e to S.C.C. under appropriate c o n d i t i o n s . The phenomenon, however, i s of p a r t i c u l a r concern to the metals i n d u s t r y s i n c e normally ductile m a t e r i a l c h a r a c t e r i s t i c a l l y f a i l s i n a b r i t t l e manner. Moreover, many of the environments causing S.C.C. are, i n the absence of s t r e s s , u s u a l l y considered q u i t e innocuous. S.C.C. f r a c t u r e occurs by the advance of narrow cracks through the m a t e r i a l , and thus excludes overload f a i l u r e s a r i s i n g from a general r e d u c t i o n of c r o s s - s e c t i o n a l area by a c o r r o s i v e environment. S.C.C. i n v o l v e s a unique i n t e r p l a y of p h y s i c a l m e t a l l u r g y , e l e c t r o c h e m i s t r y , and sur f a c e chemistry. The m a j o r i t y of S.C.C. s t u d i e s have been approached from one of these p o i n t s of view. The current work, an e l e c t r o n d i f f r a c t i o n and f r a c t o g r a p h i c study of S.C.C, emphasizes the p h y s i c a l , r a t h e r than the chemical, aspects of the phenomenon. Examples of previous work concerning the a p p l i c a t i o n of these techniques i n S.C.C. are b r i e f l y d i s c ussed i n the f o l l o w i n g two s e c t i o n s . In p a r t i c u l a r , an attempt i s made to provide an i n s i g h t i n t o the - 2 - potential of electron di f f rac t ion and fractographic techniques to contribute to the understanding of fundamental aspects of S . C . C . 1.1 Electron D i f f r a c t i o n Studies of S.C.C. 1.1.1 Detection and I d e n t i f i c a t i o n of Corrosion Product Films There have been few instances of a p p l i c a t i o n of el e c t r o n d i f f r a c t i o n techniques to str e s s corrosion cracking or corrosion studies i n S.C.C. environments. Electron d i f f r a c t i o n has been mainly used i n the detection and i d e n t i f i c a t i o n of corrosion product films formed on (a) bulk specimens subjected to S.C.C. (1-6), (b) coupons exposed to corrosive environ- ments (7-14), and (c) on stressed and unstressed f o i l s exposed to corrosive environments (15-19). Electron d i f f r a c t i o n patterns have been obtained from (a) protuberances a r i s i n g from the surface of the films (3,4,6,16) or (b) from selected area d i f f r a c t i o n through the body of the thi n f i l m (1,2,7-15,18,19). The l a t t e r technique almost i n v a r i a b l y requires removal from the substrate. The corrosion product films generally consist of oxides or metals, which are e i t h e r deposited on, or a r i s e from enrichment of one p a r t i c u l a r a l l o y component on metal surfaces. For example, i n the MgC^ cracking of a u s t e n i t i c s t a i n l e s s s t e e l , Staehle (2) considers that a n i c k e l enriched layer forms by p r e f e r e n t i a l d i s s o l u t i o n . On the other hand, Baker et a l . (21), and Nielson (22) prefer that an oxide f i l m i s deposited on the surface of the s t e e l . Surface films are considered important i n S.C.C. because they (a) can influence electrode reactions (20,21,23) and (b) a f f e c t the surface mechanical properties of the metal (24,25), e.g., prevent d i s l o c a t i o n egression from the metal. A d d i t i o n a l l y , b u i l d up of - 4 - c o r r o s i o n product i n a crack may produce s t r e s s r a i s e r s f a c i l i t a t i n g f u r t h e r extension of the crack (22,26), or may r e s t r i c t the passage of e l e c t r o l y t e to the crack t i p r e g i o n . D e t e c t i o n and i d e n t i f i c a t i o n of a f i l m or c o r r o s i o n product on a metal surface has s e v e r a l p o t e n t i a l uses. F i r s t l y , i t can help r e s o l v e any controversy regarding the presence of sur f a c e f i l m s i n some S.C.C. systems. For example, s e v e r a l workers (27-31) c l a i m that the transgranular c r a c k i n g of 304L y s t a i n l e s s s t e e l i n MgCl^ s o l u t i o n only occurs i n the absence of a f i l m , w h i l s t others b e l i e v e t h a t f i l m formation i s a p r e r e q u i s i t e of c r a c k i n g (20,21,32). Moreover, i n the ammoniacal s t r e s s c o r r o s i o n c r a c k i n g of alpha b r a s s , Pugh and h i s colleagues (33-35) observed c r a c k i n g i n " n o n - t a r n i s h i n g " and " t a r n i s h i n g " environments, i n e f f e c t suggesting that a f i l m i s not r e q u i r e d f o r c r a c k i n g . Recent evidence (3,4,19), however, suggests that the former s o l u t i o n s do, i n f a c t , promote the formation of a very t h i n t a r n i s h l a y e r . Secondly, because the f i l m rupture mechanisms of S.C.C. (1,3,4,7, 8,36) r e q u i r e the presence of a f i l m , s u c c e s s f u l f i l m d e t e c t i o n cannot r u l e out the p o s s i b i l i t y that f r a c t u r e occurs by t h i s mechanism. T h i r d l y , i d e n t i f i c a t i o n of c o r r o s i o n product f i l m s may r e v e a l i n f o r m a t i o n on the chemistry of the S.C.C. system. I t i s to t h i s end perhaps, that the m a j o r i t y of e l e c t r o n d i f f r a c t i o n s t u d i e s have been c a r r i e d out. The d e t e c t i o n and i d e n t i f i c a t i o n of phases other than c o r r o s i o n product f i l m s i n t h i n f o i l s or i n bulk corroded specimens i s a l s o a n t i c i p a t e d to c o n t r i b u t e to an understanding of c r a c k i n g . For example, - 5 - i t has been suggested that c r a c k i n g has been a s s o c i a t e d w i t h twinning ( 6 ) , martensite (37-39), or p r e c i p i t a t e s such as hydrides (40) or magnesium- aluminum phases (15) depending on the S.C.C. system. Such fe a t u r e s can d i c t a t e the t r a n s g r a n u l a r crack path and hence the t o p o g r a p h i c a l f e a t u r e s on the a c t u a l s t r e s s c o r r o s i o n f r a c t u r e s u r f a c e . Such s t u d i e s , i n p a r t i c u l a r , should help r e s o l v e the long standing controversy regarding the r o l e of s t r a i n induced martensite i n the t r a n s g r a n u l a r S.C.C. of a u s t e n i t i c s t a i n l e s s s t e e l s . Edeleanu (38,39) had suggested that c r a c k i n g of t h i s m a t e r i a l occurred by d i s s o l u t i o n of s t r a i n induced m a r t e n s i t e formed ahead of an advancing crack t i p . G e n e r a l l y , the theory had met con s i d e r a b l e o p p o s i t i o n , d e t a i l e d by Staehle and L a t a n i s i o n (41). This i s p r i n c i p a l l y based on the f a c t s t h a t the S.C.C. t e s t temperature exceeded t h a t of the M g of the a l l o y and that the S.C.C. process occurs i n the more s t a b l e a u s t e n i t i c s t a i n l e s s s t e e l s . Recent evidence (42), however, suggests t h a t m a r t e n s i t e may form i n a u s t e n i t i c s t e e l s a t the t e s t i n g temperature of 154°C. C l e a r l y , i f martensite i s formed, i s i r r e v e r s i b l e , and does not completely d i s s o l v e during c r a c k i n g , i t should be p o s s i b l e to detect evidence of the phase on the a c t u a l f r a c t u r e s u r f a c e s . E l e c t r o n d i f f r a c t i o n s t u d i e s conducted on the a c t u a l S.C.C. surfaces of bulk specimens are more meaningful, i n terms of crack propagation s t u d i e s , than those conducted on the uncracked s u r f a c e s . In f a c t , w i t h respect to the d e t e c t i o n of m a r t e n s i t e , t w i n s , hydrides and p r e c i p i t a t e s , observations from the uncracked surfaces of bulk specimens have l i t t l e relevance. The m a j o r i t y of work, however, has been concerned w i t h c o r r o s i o n product f i l m s present on the uncracked - 6 - surfaces of test specimens. For example, i t has been shown that cuprous oxide forms on the uncracked surface of alpha brass exposed to ammoniacal copper sulphate solution (2,7,8,14) or to citrate and tartrate solutions (18), or that an oxide forms on the uncracked surface of austenitic stainless steel exposed to FeCl^/MgCl^ environments(21), or water at 300°C (10-12). Nielson's (22) observation of a spinel in the MgC^ cracking of stainless steel, the detection of twinning in Mg-Al alloys fractured in salt solution and d i s t i l l e d water by Oryall and Tromans (6), the identification of corrosion product films on the surface of a-brass formed in a number of environments (3,4,43), and on an Al-Zn-Mg alloy (5) appear to be the only observations relevant to actual stress corrosion fracture surfaces. The lack of electron diffraction studies performed on the actual stress corrosion cracked surface of bulk specimens may be attributed to one of two related reasons. F i r s t l y , the film formed on the fracture surface was assumed to have the same composition as that on the uncracked surface. This is not necessarily a valid assumption since the composition of the environment down the crack may d i f f e r from that in the bulk (44). Alternatively, many workers used small diameter wire (2,45) or thin strip (1,21,33,34,46) specimens tested i n such a manner to result in overload failure. In such cases i t was d i f f i c u l t to perceive any difference between films formed on the uncracked surface and that formed on the actual stress corrosion fracture surface. However, Tromans, Dowds and Leja (47) showed that i f large a-brass rod specimens were subjected to stress corrosion cracking in ammoniacal copper sulphate solution, the corrosion product film formed on the actual stress corrosion surface was, in fact, "yellow" whilst that formed on the uncracked surface exhibited a blue-black coloration. This suggests that examination of actual fracture surfaces may reveal different information from studies on uncracked surfaces. 1.1.2 Orientation Studies The other main application of electron diffraction has been the determination of orientation of corrosion product, or of precipitates or planar dislocation arrays susceptible to attack by corrosive environments. These studies have been performed exclusively on stressed and unstressed thin f o i l s immersed in corrosive environments, or on f o i l s thinned from bulk specimens subjected to S.C.C. The nature of data derived from such electron diffractinn studies i s exemplified below: (1) During the electron diffraction detection and identification of corrosion product, Pickering (16) observed the presence of an epitaxial Au-rich film on the surface of an unstressed Cu-10% Au f o i l exposed to 1 M H^SO^. Pickering and Swann (15) also made similar observations on stressed and unstressed Cu-Au f o i l s immersed in aqua regia, FeCl^ and KCN solutions. The application of electron diffraction for extensive and thorough epitaxy studies i s ably illustrated by the work of Foley et a l . (17) concerning formation of oxides on iron f o i l s immersed in NaOH and H„SO.. 2 4 (2) Tromans and Nutting (48,49) carried out direct S.C.C. tests on thin f o i l s of alpha brass stressed in the microscope holder. Subsequent selected area diffraction suggested that cracks i n i t i a t e d on {111} planes. Similar conclusions were reached after examination • f o i l s thinned from bulk specimens of 70/30 brass stressed for 5 mins in an ammoniacal copper sulphate solution (49). Work on thin f o i l s stainless steel has indicated that cracks also occur on the {111} planes in this material (49,50). - 9 - 1.2 Fract o g r a p h l c Studies of Stress Corrosion Cracking One of the c h a r a c t e r i s t i c s of S.C.C. i s that f r a c t u r e occurs i n a b r i t t l e manner. The crack path i s e i t h e r predominantly t r a n s g r a n u l a r or i n t e r g r a n u l a r depending on the S.C.C. system. G e n e r a l l y , the i n t e r g r a n u l a r b r i t t l e f r a c t u r e surface topography e x h i b i t s a smooth angular faceted appearance, w h i l s t a tra n s g r a n u l a r b r i t t l e f r a c t u r e g e n e r a l l y e x h i b i t s a more complex rough s u r f a c e . G e n e r a l l y , a t r a n s - granular S.C.C. f r a c t u r e surface e x h i b i t s an appearance of cleavage. This i s often termed "quasi-cleavage". However, t h i s can be very misleading s i n c e s i m i l a r topography could a l s o a r i s e from d i s s o l u t i o n processes or from d u c t i l e t e a r i n g on a f i n e s c a l e (53). I t i s the i n t e r p r e t a t i o n of these complex t o p o g r a p h i c a l features which has been the concern of many workers. In f a c t , the s i g n i f i c a n c e or i n t e r p r e t a t i o n of many of these surface markings has yet to be e s t a b l i s h e d . This i s q u i t e important because i t may provide an a d d i t i o n a l clue to e s t a b l i s h i n g the reason f o r the tr a n s g r a n u l a r crack path observed i n many a l l o y s . I t i s p e r t i n e n t to note that f o r i n t e r g r a n u l a r c r a c k i n g , d i s s o l u t i o n or c r a c k i n g occurs along a very obvious s u s c e p t i b l e path, i . e . , the g r a i n boundary. Moreover, the f r a c t u r e face topography i s s a t i s f a c t o r i l y accounted f o r . However, i n the case of t r a n s g r a n u l a r c r a c k i n g , there i s no obvious s u s c e p t i b l e anodic crack path. I t has been suggested that cleavage occurs (54), but i n the case of S.C.C of d u c t i l e fee metals i n p a r t i c u l a r , i t s i n c i d e n c e i s p u z z l i n g . A c c o r d i n g l y , from time to time, i t has been proposed that the crack path i s d i c t a t e d by segregation of s o l u t e to s t a c k i n g f a u l t s (55), d i s s o l u t i o n of moving d i s l o c a t i o n s (32), p r e c i p i t a t e s on s l i p planes (15), d e s t r u c t i o n of short range order (56), s t r a i n induced - 10 - phases (38,39), or hydrides (40), depending on the system under examination. These processes should also determine the observed topographical features but unfortunately few attempts have been made to account for this. The few attempts to account for the observed topographical features i n terms of possible mechanisms of transgranular fracture have generally resulted in controversy. For example, McEvily and Bond (1) s u g g e s t e d that par a l l e l striations on the S.C. surface of ammoniacal copper sulphate fractured a-brass provided evidence for discontinuous crack propagation. They claimed that the striations were similar in appearance to those observed on the fracture surfaces of fatigued specimens and defined the position of the crack front at successive stages during propagation. On the other hand other workers (4,34,47,57) suggested that these markings corresponded to sl i p steps produced in the wake of an advancing crack t i p . Striations present on the MgCl^ fractured surfaces of austenitic stainless steel have also been the subject of controversial inter- pretation. They have been taken as corresponding to crack arrest points (22) , cleavage markings (54), or were associated with dissolution along s l i p planes (32,54) or martensite platelets (58). Nielson (22, 54,58,59) has presented several discussions on the fractography of stainless steels. Before the topographical features-can be related to transgranular cracking i t is essential to establish that the observed markings are associated with microstructural features present prior to the passing of the crack and not associated with microstructural features formed - 11 - in i t s wake. If this is not the case, there i s absolutely no basis for supposing that crack paths can be related to observed topographical features. The successful mating of the two fractured surfaces of the S.C.C. specimens is considered to confirm this relationship. For example, in the aforementioned case of ammoniacal stress corrosion cracking of alpha-brass, i f fracture occurs by a two stage process leading to the formation of crack arrest points, matching striations should be observed on the surface. On the other hand, i f s l i p steps emerge after cracking has occurred, i t i s unlikely that the crack surfaces would match up. Matching surface fractography also rules out the p o s s i b i l i t i e s , in the case of the 154°C MgCl^ cracking of stainless steels, that any phase detected on the fracture surface might have formed on cooling from the test temperature, or by cathodic hydrogen charging of the crack surfaces after fracture. Very few instances of matching S.C.C. surfaces have been reported. Pugh (60) showed that two halves of a fractured specimen of AgCl stress corroded in NaCl solution mate. The observed striations were considered not to be s l i p steps produced after crack tip passage, but rather to be indicative of discontinuous crack propagation. McEvily (61) noted essentially the same effect on the creep rupture of an Al-Zn-Mg alloy using a replica technique. Pugh (57) indicated that this technique was being applied to the a-brass/ammoniacal copper sulphate system, but no results have yet appeared in the literature. There appears to be, in fact, only a handful of successful attempts at mating S.C.C. fracture surfaces of metals. F i r s t l y , in an optical study (22) and secondly, in a scanning electron microscopy study (59) Nielson - 12 - considered two halves of a 154°C MgCl 2 fractured 18/8 austenitic steel specimen to match. In addition he mated up oxide replicas removed from a part i a l l y cracked steel specimen (59). Unfortunately he did not attach any significance to his observations. Oryall and Tromans (6), in a scanning electron microscopy study, observed matching topographical features on the fractured surface of a Mg/8A1 alloy subject to S.C.C. in water and salt solution. Finally, complementary fracture surfaces have been noted i n embrittled Zr alloys (43,53). That there have been few successful attempts to match up the fracture surfaces is not surprising in view of the many technical problems which may be encountered during specimen preparation and the application of microscopy. Prior to the advent of scanning electron microscopy, only optical and replica transmission microscopy techniques were available. D i f f i c u l t i e s are associated with: (1) Specimen preparation: grains or portions of grains may drop out and therefore the surfaces do not match. In addition gross oxidation or corrosion of the fracture faces may occur rendering localized features unrecognizable. (2) Replica preparation: when fracture surfaces contain smooth large facets, d i f f i c u l t y i s often encountered in holding the replica together. Replica fold-over can also be a problem. (3) Location of the key features: in transmission electron microscopy, the specimen holder grid may mask the key matching features. Heating may bend the replica, distorting the feature. In optical studies, in addition to the location problem, the surfaces must be relatively f l a t . This i s often not the case. - 13 - Nielson's oxide double r e p l i c a technique (59) overcomes these problems. Scanning e l e c t r o n microscopy can overcome the l a t t e r two problems although care i s s t i l l required i n l o c a t i n g the p a r t i c u l a r features. In conjunction with a p o s i t i v e i d e n t i f i c a t i o n of the surface markings, the matching surface technique i s a p o t e n t i a l l y powerful t o o l as there i s formed some basis for r e l a t i n g the crack path to any m e t a l l u r g i c a l feature present. The i d e n t i f i c a t i o n of the topographical features can be performed by a comparison of patterns with the morphology of phases detected by metallographic techniques. In p r i n c i p l e i t i s a n t i c i p a t e d that the crack path can be r e l a t e d to twins, martensite, grain boundaries, or perhaps to s l i p or cleavage. Although t h i s technique has been applied to f r a c t u r e studies (62), i t does not appear to have been exploited i n the case of stress corrosion cracking, except by O r y a l l and Tromans (6). - 14 - 1.3 Objects of Study and Choice of Materials The overall effort is for a deeper understanding of the physical reasons for the incidence of transgranular S.C.C. of stainless steels and g-brasses, and in particular to determine whether the trans- granular crack path i s dictated by the presence of shear transformations or any other metallurgical feature. The aim of the i n i t i a l stage of the work is to develop and j u s t i f y an electron diffraction technique suitable for detecting and identifying thin powder films or metallic phases. It i s anticipated that applica- tion to actual S.C.C. surfaces w i l l contribute to the debate concerning the possible presence and hence the importance of a corrosion product film and to resolving the martensite controversy. It should be pointed out that i f S.C.C. is related to martensite, repercussions can be anticipated in real situations where materials developed to exploit the incidence of the martensite transformation are u t i l i z e d . If the transgranular crack path is dictated by the martensite transformation, three basic conclusions must be shown to be consistent with the experimental observations: (1) The S.C.C. fracture surface must contain evidence of the martensite product. (2) The crack tip must be adjacent to a martensitically transformed region through which i t passes. In addition, matching fracture surfaces would establish that the topographical features were formed prior to cracking. (3) The topography of the S.C.C. fracture surface should be related to the morphology of martensite. - 15 - In the current work a f r a c t o g r a p h l c , m e t a l l o g r a p h i c and e l e c t r o n d i f f r a c t i o n procedure i s developed w i t h the express o b j e c t i v e of t e s t i n g these three conclusions. S e v e r a l a l l o y s were s e l e c t e d f o r examination. 304L a u s t e n i t i c s t a i n l e s s s t e e l was chosen as t h i s m a t e r i a l i s unstable w i t h respect to the martensite t r a n s f o r m a t i o n and i s s u s c e p t i b l e to t r a n s g r a n u l a r S.C.C. A 310 grade a u s t e n i t i c s t e e l was examined s i n c e t h i s i s s u s c e p t i b l e to t r a n s g r a n u l a r S.C.C. but i s g e n e r a l l y accepted to be s t a b l e w i t h respect to the martensite transformation. However, recent evidence (42) suggested that martensite might form i n t h i s a l l o y at 154°C, the temperature of the S.C.C. t e s t , and thus the p o s s i b i l i t y that martensite might p l a y a r o l e i n the S.C.C. of 310 s t e e l could not be r u l e d out. Work was performed on a ternary g-brass known to e x h i b i t the martensite transformation (63). I t was a l s o a n t i c i p a t e d that t h i s a l l o y would be s u s c e p t i b l e to t r a n s g r a n u l a r S.C.C. (64). F i n a l l y , b i n a r y g-brasses were examined. These a l l o y s (a) f a i l by tra n s g r a n u l a r S.C.C. (3,64) and (b) e x h i b i t a s t r a i n induced phase transformation (15) under c e r t a i n circumstances. In a d d i t i o n t h i s a f f o r d - ed an opportunity to explore a r e l a t i v e l y new S.C.C. system. I t should be noted that very l i t t l e i s known about the S.C.C. of g-brasses. Most work appears to have been performed i n connection w i t h the complex i n d u s t r i a l l y important a l l o y s . While i t i s known that these a l l o y s are s u s c e p t i b l e to S.C.C. i n a number of aqueous e n v i r o n - ments (64) and that the predominant crack path appears to depend on the environment, no accounts of d e t a i l e d s t u d i e s of the fractography or of surface f i l m s are a v a i l a b l e . The S.C.C. behaviour of bin a r y - 16 - (B-brasses is even more obscure. While these alloys are susceptible to S .C .C . i n gaseous environments (64), only one short study has been reported using aqueous environments (3). The binary alloy was found to exhibit poorer tensile properties i n ammoniacal copper sulphate solution, water, and ambient a i r than i n dry a i r . No serious metallographic studies, determination of crack path, fractography, or detection of surface films appear to have been performed on binary 8-brass/aqueous environment S . C . C . systems. - 17 - 2. EXPERIMENTAL PROCEDURE 2.1 Materials Preparation and Testing 2.1.1 General The essence of the experimental procedure was to produce and compare stress corrosion fracture surfaces with mechanical fracture surfaces of both austenitic steels and B-brasses using fractographic and electron diffraction techniques. The austenitic steel stress corrosion fracture surfaces were produced at 154°C and mechanical fractures at 154°C and -196°C. 3-Brass fracture surfaces were prepared at ambient temperatures. The fracture surfaces were studied by electron diffraction techniques to detect the presence of phases. Topographical differences were studied by scanning electron microscopy and replica electron microscopy. The presence of phases such as martensite and twins below the fracture surface and ahead of stress corrosion cracks was determined by optical observation of sectioned specimens using martensite or twinning etching techniques. Morphological features of the martensite or twins as revealed by the etch patterns were compared with the topographical features of stress corrosion fracture surfaces. The overall experimental procedure outlined in Figure 1 was developed. Successive stages, becoming progressively destructive, were applied to the same specimen. This provides a basis for correlating - 18 - Materials Preparation Testing (Mechanical or S.C.C.) Electron Diffraction Fractography Metallography Figure 1. Outline of experimental procedure. - 19 - the electron diffraction, fractographic and metallographic observations. Specifically the transgranular S.C.C. systems examined were: 304L Austenitic steel in MgCl^ solution boiling at 154°C. 310 Austenitic steel in MgCl 2 solution boiling at 154°C. Cu-47.76% Zn in d i s t i l l e d water at ambient temperature. Cu-47.76% Zn i n ammoniacal copper sulphate solution at ambient temperature. Cu-45.54% Zn in d i s t i l l e d water at ambient temperature. Cu-33.5% Zn-4.5% Sn i n d i s t i l l e d water at ambient temperature. 2.1.2 Materials 2.1.2.1 Steels 304L and 310 grade stainless steel were received in the form of 3/8" diameter bar stock. Analyses in terms of weight percentages are shown in Table I . 2.1.2.2 Brasses 3/8" diameter rods of the binary and ternary g-brasses were produced i n the laboratory. Appropriate quantities of alpha brass of composition shown in Table II, 99.999% zinc and 99.99% t i n were melted in quartz capsules. The melts were allowed to mix for 3 hours at 980°C, with occasional agitation, and the products were quenched vertica l l y into ambient temperature water. The rods were encapsulated in quartz and then homogenized at temperatures in the g-phase f i e l d of the Cu-Zn equilibrium diagram; the higher zinc binary alloy was annealed at 500°C for 3 hours and quenched, the lower zinc binary alloy - 20 - TABLE I - Composition of Stainless Steel Type 304 L Type 310 c 0.053 0.12 Mn 1 . 4 9 1.70 Si 0.56 0.68 Cr 17.31 26.3 Ni 9.5 18.5 Mo 0.21 0.15 T i 0.02 0.17 TABLE II- Composition of a-brass Element j wt % Cu 70.47 Zn 29.38 Fe 0.1 Pb 0.01 Si 0.01 Ag 0.005 Sn 0.01 T i 0.01 others < 0.001 - 21 - and ternary alloys were annealed at 800°C for 3 hours and quenched. Microprobe analysis revealed no variation of composition across the rod diameter in any of the alloys. The composition of alloys were chosen such that the temperature of each alloy exceeded room tempera- ture. (The temperature i s that temperature above which martensite cannot be induced by deformation.) 2.1.2.3 Environments The magnesium chloride solution was made up by adding a sufficient amount of the reagent grade solid to d i s t i l l e d water, while heating the solution, u n t i l the temperature attained a steady 154°C. The ammoniacal copper sulphate solution was of the type used in previous studies (3,4,23,46,47), and contained 0.04 M CuSO^«5H20 and 1.5 M (NH^)2S0^ adjusted to pH 6.8 with approximately 2 N NĤ OH. Separate stock solutions of 6 M (NH^SO^ and 0.5 M CuS0^«5H20 were made up using boiled d i s t i l l e d water and mixing of these solutions was carried out before each series of tests. pH measurements were made with a Corning model 10 pH meter. D i s t i l l e d water was used i n the S.C.C. tests. 2.1.3 Specimens The specimens employed in the mechanical and stress corrosion tests were in the form of 7" cut-offs of 3/8" diameter bar, threaded at either end, with a 60° circumferential notch machined at the centre to provide a reduced diameter of 0.2". This localized cracking to the reduced cross section and provided a large area suitable for electron - 22 - diffraction studies, ensuring that the diffracted electron beam originated from the actual fracture surface as opposed to the uncracked surface. A longitudinal mark was scribed on the specimens to f a c i l i t a t e location of matching areas in the fractographic studies. The specimens were subsequently given annealing treatments shown in Table III. The g-brass specimens were encapsulated i n quartz during these heat treatments. Grain sizes were determined by an intercept method. Note that grain diameters larger than the diameter of the electron beam, i.e., > 250 u, were preferred for the electron diffraction studies. This f a c i l i t a t e d focussing the electron beam on an individual grain. Any ring patterns obtained from these single grains are very significant since they may indicate the incidence of a transformation or the presence of a fine powdery corrosion product film. 2.1.4 Mechanical and Stress Corrosion Testing A summary of test conditions is shown in Table IV. The S.C.C. tests and mechanical fracture tests were performed on a Floor Model Instron machine. The S.C.C. tests at elevated tempera- tures were performed by enclosing the specimen in a pyrex-teflon c e l l f i t t e d with two arms, one to allow introduction of the environment and the other to permit f i t t i n g of a reflux condenser for use with boiling solutions. Heat was supplied by an e l e c t r i c a l heating tape. The whole arrangement was enclosed tightly in a sheet of Mullite wool for insulation. - 23 - TABLE I I I - Specimen Heat Treatment A l l o y Heat Treatment G r a i n S i z e 304L S t e e l 1250°C/1 hour argon/water quenched 100--llOy 310 S t e e l 1100°C/1 hour argon/water quenched 100--HOy Cu-47.76% Zn 550°C/1 hour/water quenched 3-4 mm Cu-45.54% Zn 800°C/1 hour/water quenched 3-4 mm Cu-33.5% Zn -4.5% Sn 810°C/1 hour/water quenched 3-4 mm TABLE IV - Details o f Test Conditions Alloy Environment Temp. Type of Test 304L, 310 MgCl 2 (boiling) 154°C S.C.C. ( f u l l and partial) fracture 304L, 310 Liquid nitrogen -196°C Mechanical fracture 304L, 310 Silicone o i l +154°C Mechanical fracture Cu-47.76% Zn Di s t i l l e d water Ambient S.C.C. fracture Cu-47.76% Zn Air Ambient Mechanical fracture Cu-47.76% Zn Ammoniacal copper Sulphate Ambient S.C.C. fracture Cu-45.54% Zn Di s t i l l e d water Ambient S.C.C. ( f u l l and partial) fracture Cu-45.54% Zn Air Ambient Mechanical Cu-33.5% Zn-4. 5% Sn Di s t i l l e d water Ambient S.C.C. ( f u l l and partial) fracture Cu-33.5% Zn-4. 5% Sn Air Ambient Mechanical - 25 - The procedure was as f o l l o w s . The c e l l was allowed to heat up to temperature before i n t r o d u c t i o n of the environment. I t was necessary to apply a s l i g h t e l a s t i c load to counteract the expansion of the rod due to h e a t i n g , m a i n t a i n i n g an approximately zero l o a d reading. Concurrently the environment was being prepared and on reaching the r e q u i r e d temperature was introduced to the c e l l . F u r t h e r time was allowed f o r thermal e q u i l i b r i u m to occur. This g e n e r a l l y took at l e a s t 30 minutes, and was i n d i c a t e d by a steady reading of load on the recorder. A t e n s i l e load was a p p l i e d to the specimen to i n i t i a t e y i e l d i n g w i t h i n the notched r e g i o n . At t h i s p o i n t the crosshead of the I n s t r o n was locked. S.C.C. was l o c a t e d at the apex of the notch and r e s u l t e d i n r e l a x a t i o n of the load to zero as the crack propagated through the specimen. In t h i s manner, a S.C.C. surf a c e f r e e from overload f r a c t u r e was obtained. Some specimens were removed before c r a c k i n g was completed. These were subsequently prepared f o r m e t a l l o g r a p h i c examination and crack t i p s were examined f o r evidence of adjacent phases. The mechanical t e s t s at 154°C i n s i l i c o n e o i l were performed i n a simple pyrex t e f l o n c e l l heated w i t h e l e c t r i c a l tape. The c e l l was allowed to reach temperature before the heated s i l i c o n e o i l was added. A f t e r thermal e q u i l i b r i u m was reached the specimens were mechanically f r a c t u r e d . The mechanical t e s t s at -196°C were performed i n a t e f l o n - r u b b e r c e l l . L i q u i d n i t r o g e n was added u n t i l effervescence ceased. This process took approximately one hour. The specimens were then r a p i d l y m echanically f r a c t u r e d . - 26 - The tests on the g-brasses were performed at ambient temperature in a single pyrex teflon c e l l . The load was applied u n t i l yielding had occurred whilst the specimen was in the environment. Values of yi e l d stresses observed in the stress corrosion tests of the steels and brasses are lis t e d in Table V. In most instances specimens were removed from their environments as quickly as possible after fracturing. Only in the case of the oxide film studies were specimens purposely kept immersed in the environment. The steel S.C.C. specimens were rinsed in boiling water, followed by alcohol, and dried in acetone. The 154°C mechanically fractured specimens were degreased in trichloroethylene, and dried in acetone. The -196°C specimens were allowed to heat up to room tempera- ture, then dried in acetone. g-Brass specimens were also dried in acetone after a water and alcohol rinse. Extreme care was taken when removing the fractured g-brass speci- mens from the ammoniacal environment. Tromans, Dowds and Leja (47) found that the actual S.C.C. fracture surfaces of a-brass immersed in a similar environment appeared bright yellow whilst the uncracked surfaces were covered with a thick bluish tarnish. The yellow surface remained for a long time provided the specimen was kept immersed in the solution. On removal from the solution and exposure to a i r the surface turned dark blue almost instantaneously. Clearly this would hinder fractographic observation and especially the electron diffraction detection of any metallic phases. Preliminary tests (3) indicated that this phenomenon occurred in the case of g-brasses. The effect was minimized by pouring water into the ammoniacal solution thereby diluting and eventually replacing i t . TABLE V - Y i e l d Stresses of M a t e r i a l s Subject to S.C.C. M a t e r i a l Temperature Y i e l d Stress ( p . s . i . ) 30 4L 154° C 63,220 310 154°C 68,710 Cu-47.76% Zn Ambient 32,100 Cu-45.54% Zn Ambient 31,520 Cu-33.5% Zn -4.5% Sn Ambient 28,300 - 28 - 2.2 Electron Diffraction 2.2.1 Introduction In the current work, diffraction patterns were obtained by high energy (100 Kv) transmission through surface protuberances and not by transmission through thin films removed from the specimen substrate. The most widely used high energy technique has been transmission through thin films as evidenced by the comprehensive reviews of Pinsker (66), Vainshtein (67) and Cowley (68). However, transmission through surface protuberances offers specific advantages relative to the study of corrosion products and other phases associated with fracture surfaces. For example: (1) The corrosion product film can be examined without prior removal from the substrate. (2) ' Films too thin for physical removal can be detected and identified. (3) There is less likelyhood of heat induced transformations occurring in the electron beam because the conducting substrate acts as a heat sink. (4) The relative intensities of rings on electron diffraction patterns, i t w i l l be shown, are characteristic of those predicted for a randomly oriented powder aggregate. Diffraction through relatively uniform thin films, the alternative technique, often produces ring patterns that depend on the thickness, for monolayer polycrystalline films or on the c r y s t a l l i t e size, for multilayer polycrystals. It w i l l become evident that the f i r s t technique is a more reliable indicator of the compound under examination. - 29 - 2.2.2 I n t e n s i t i e s of D i f f r a c t i o n P a t t e r n s I t has been shown (69) that the i n t e n s i t y I of a d i f f r a c t e d beam through a p e r f e c t c r y s t a l i n dynamical c o n d i t i o n s v a r i e s s i n u s o i d a l l y w i t h c r y s t a l l i t e s i z e , according to the equation I = s i n 2 TT ( t / t Q ) ( l + x 2 ) 1 / 2 (1 + x 2 ) where x i s a parameter i n d i c a t i n g a s m a l l d e v i a t i o n from the Bragg angle t Q i s the e x t i n c t i o n d i s t a n c e f o r the r e f l e c t i o n being considered t i s the c r y s t a l l i t e t h i c k n e s s . This i s s c h e m a t i c a l l y shown i n Figure 2. In k i n e m a t i c a l c o n d i t i o n s , a p p l i c a b l e to very t h i n c r y s t a l l i t e s and weak d i f f r a c t e d beams, the i n t e n s i t y of the d i f f r a c t e d beam a l s o v a r i e s i n a s i n u s o i d a l manner. The minimum i n t e n s i t y of the transmitted beam i n k i n e m a t i c a l s c a t t e r i n g does not f a l l to zero, whereas i n dynamical c o n d i t i o n s the minimum i n t e n s i t y does reach zero. I t has been p r e d i c t e d (67a) that the , o t r a n s i t i o n from k i n e m a t i c a l to dynamical c o n d i t i o n s w i l l occur at 50 A o o f o r g o l d , 100 A f o r s i l v e r , and 200 A f o r aluminum. The e x t i n c t i o n depends on c r y s t a l t h i c k n e s s f o r a monolayer of randomly o r i e n t e d c r y s t a l s , and the c r y s t a l s i z e f o r m u l t i l a y e r s of p o l y c r y s t a l s . Figure 3(a) represents three c r y s t a l l i t e s i n a monolayer p o l y c r y s t a l l i n e f i l m , o r i e n t e d f o r t h e i r p a r t i c u l a r h k l Bragg r e f l e c t i o n . I t i s c l e a r t h a t the r e l a t i v e i n t e n s i t i e s of the d i f f r a c t e d beam w i l l depend on the c r y s t a l s i z e t , and indeed f o r the appropriate value of t , e i t h e r one, two or three d i f f r a c t e d beams may be e x t i n g u i s h e d . The - 30 - DIRECT WAVE DIFFRACTEO WAVE MAGNITUDE OF INTENSITY t0 = EXTINCTION DISTANCE F i g u r e 2. Schematic v a r i a t i o n of i n t e n s i t i e s of d i f f r a c t e d and d i r e c t beams w i t h d i s t a n c e i n a monocrystal of t h i c k - ness t . Figure 3. (a) Schematic of d i f f r a c t i o n of i n c i d e n t beam by each of three i n d i v i d u a l c r y s t a l l i t e s i n a monolayer f i l m , (b) Schematic of m u l t i l a y e r f i l m showing c r y s t a l l i t e s A and B contained i n a column of widt h XY. - 31 - same i s true f o r the case of m u l t i l a y e r p o l y c r y s t a l l i n e f i l m s , shown sc h e m a t i c a l l y i n Figure 3(b). In the column of f i l m represented by XY, c r y s t a l l i t e A i s o r i e n t e d w i t h respect to the beam f o r r e f l e c t i o n from h j k ^ l ^ planes and c r y s t a l l i t e B f o r r e f l e c t i o n from h^k^l^ planes. I f the length of c r y s t a l l i t e A i s an i n t e g e r m u l t i p l e o f t , , . , \ » the e x t i n c t i o n d i s tance corresponding to the h ^ k ^ l ^ planes, there w i l l be no r e f l e c t i o n corresponding to the h^k-^l^ planes from the whole column. ( I f the length of c r y s t a l l i t e B i s not a m u l t i p l e i n t e g e r of t , . > o U 2 2 2 J there w i l l be a r e f l e c t i o n from the column from the 1 1 ^ 2 ^ planes.) In a t h i n f i l m of uniform t h i c k n e s s , c o n s i s t i n g of a monolayer of randomly o r i e n t e d c r y s t a l l i t e s , i t i s q u i t e p o s s i b l e that the r e l a t i v e i n t e n s i t i e s of r i n g s on an e l e c t r o n d i f f r a c t i o n p a t t e r n w i l l vary w i t h the thickness of the f i l m and t h e r e f o r e cannot be used to c h a r a c t e r i z e the c r y s t a l l o g r a p h i c s t r u c t u r e of the f i l m . The s i t u a t i o n i s q u i t e d i f f e r e n t f o r d i f f r a c t i o n from protuberances. Figure 4 shows a p o s s i b l e arrangement of c r y s t a l l i t e s i n a protuberance. In t h i s case, the taper puts the specimen i n t o the k i n e m a t i c a l c o n d i t i o n and t h e r e f o r e the beam w i l l not completely be e x t i n g u i s h e d . On the co n t r a r y , i t i s q u i t e p o s s i b l e that a d i f f r a c t i o n p a t t e r n from the t i p can be used to c h a r a c t e r i z e the c r y s t a l l o g r a p h i c s t r u c t u r e . (In s e c t i o n 2.2.4.1 k i n e m a t i c a l c o n s i d e r a t i o n s are employed i n the development of an expression f o r the t h e o r e t i c a l r e l a t i v e i n t e n s i t i e s of Debye r i n g s . ) 2.2.3 Demonstration of I n t e n s i t y E f f e c t s The current s e c t i o n describes the techniques used to i l l u s t r a t e the aforementioned e f f e c t s . Gold f i l m s of d i f f e r e n t t h i c k n e s s e s , and f i l i n g s (powder) were employed. - 32 - Figure 4. Schematic of c r y s t a l l i t e arrangement at a protuberance. - 33 - I n i t i a l l y several thin gold films were prepared. Gold was evaporated from a molybdenum boat producing a non-uniform thickness of film on a perspex sheet, arranged to intercept the hemispherical sphere of evaporation. Small sections of identical dimensional cross- sectional area were cut from selected portions on this sheet. The plastic backing was then dissolved in acetone. The gold films were weighed to ensure that their thicknesses were different. Films i n the range of 180 A-1200 A thickness were required. Electron diffraction patterns taken using the selected area diffraction technique at 100 Kv and the same instrument conditions were different, i.e., corresponded to different thicknesses of film. See Figures 5(a) and 5(b). This difference in electron diffraction patterns was not due to preferred orientation effects. Preferred orientation effects were purposely eliminated since both Au films examined were formed simultaneously on the same perspex substrate. Secondly, diffraction patterns of gold f i l i n g s were obtained at 100 Kv using a technique designed to simulate diffraction from protuberances. This was achieved by fracturing a 3/8" diameter rod of amorphous carbon into approximately 1" lengths. The fractured transverse surface, macroscopically smooth, was swabbed with a colloidal sus- pension of graphite in water (trade name "Aquadag") and when s t i l l wet was immersed in the fi l i n g s and removed. Loose f i l i n g s were removed by tapping the rod lig h t l y . Time was permitted for the "Aquadag" to dry before immersion in the reflection stage of the Hitachi HU-11A electron microscope. Figures 5(a), 5(b) and 5(c) compare the results obtained from the  - 35 - films and the f i l i n g s . Details of the interplanar spacings and visual intensities are tabulated in Table VI . Figures 5(a) and 5(b) are presented in such a manner to reveal very obvious differences between the patterns from two films of different thicknesses. In fact, while the electron diffraction pattern represented in Figure 5(b) closely resembles that of an fee pattern, Figure 5(a) resembles a bec ring pattern. This dramatically illustrates how electron diffraction patterns obtained by transmission through films of relatively uniform thickness may provide misleading information regarding the structure of the material. Three rings absent from the pattern represented in Figure 5(a) correspond to reflection from the 111, 311 and 331 o o planes, corresponding to extinction distances of 117 A (70), 292 A (71), o and 406 A (71) respectively. From consideration of these values, i t o was estimated the film was approximately 800 A thick. (Note that the values of Hirsch et a l . (70) are probably over-estimated by about 10% due to the fact that a two-beam and not a multiple beam treatment was used (71)). The ring pattern obtained by diffraction from f i l i n g s shown in Figure 5(c) differs from that shown in Figure 5(b), in that the intensities of the rings are different. The question remains to be answered whether or not the relative visual intensities of the rings in the pattern are characteristic of the crystal structure. This might be achieved by comparison of the visual observed intensities with some standard index, e.g., similar to that of the ASTM index for X-rays. Unfortunately no such standard index exists. Consequently, the procedure adopted was to compare the relative visual intensities - Fig. 5(a) resembles a bec ring pattern in that the reflections corres- ponding to planes with hkl a l l even are present. For a complete bec pattern a l l h+k+1 = even must be present, e.g. 110,310. TABLE V I - I and I for Au e v Electron Diffraction Observati ons Data Gold Film Gold Film Gold Fi l ings Gold Film 0 d A hkl I V d A hkl I V o d A hkl I V O d A hkl I e 2.40 111 vs Abs 2.36 111 vs 2.355 111 100 2.04 200 s/m 2.10 200 s 2.02 200 s 2.039 200 44.5 1.442 220 s/m 1.442 220 m 1.45 220 w/m 1.442 220 22.9 1.225 311 s Abs 1.23 311 m 1.23 311 23.3 1.188 222 w/m 1.17 222 s 1.18 222 w 1.1774 222 6.42 1.03 400 w 1.023 400 m 1.01 400 vw 1.0196 400 2.57 0.943 331 .m Abs 0.935 331 w 0.9358 331 7.15 0.92 420 m 0.92 420 m/s 0.915 420 w 0.9120 420 6.32 0.836 422 w/m 0.836 422 w/s 0.835 422 w 0.8325 422 4.42 hkl = M i l l e r index of reflecting plane vs = very strong, s = strong, I = theoretical relative intensity for electrons s/m = strong-medium, m = medium e I = relative visual intensity w/m = weak/medium, w = weak - 37 - of the rings with theoretical estimates of the relative integrated intensities for electrons based on kinematical considerations. The basis and j u s t i f i c a t i o n for this procedure is presented in the succeeding section, and the method of calculation is shown in the Appendix. Values of the theoretical relative intensities for electrons corresponding to gold powder are presented in Table VI. A good correlation exists between these values and the observed values for gold powder. 2.2.4 Theoretical Estimates of Relative Intensities for Electrons This section develops a simple expression for the calculation of the theoretical estimates of relative intensities. The theoretical considerations are j u s t i f i e d by experimental comparison of visual observations and theoretical estimates of relative integrated intensities of diffraction rings arising from several specific crystal structures. 2.2.4.1 Theoretical Aspects The following approach was suggested by Tromans (72). There w i l l always be some diffraction under kinematical conditions from a protuberance because of the taper. Under these conditions, the total integrated intensity °^ a diffracted ring corresponding to a plane hkl, per unit area of specimen is given by (67b): = Jo X 2 | hkl 2 i hkl 2 P (1) - 38 - where <•>, , , = the s t r u c t u r e amplitude f o r e l e c t r o n s h k l t = specimen thickness Q = volume of u n i t c e l l X = wavelength J = i n t e n s i t y of i n i t i a l beam o J cL^k^ = i n t e r p l a n a r spacing of h k l plane p = m u l t i p l i c i t y f a c t o r The e l e c t r o n s are c o l l e c t e d on a screen normal to the i n c i d e n t beam, and from geometrical c o n s i d e r a t i o n s , shown i n Figure 6 , j- = tan 29 Since X = 2 d ^ ^ s i n e (Bragg Law) and f o r s m a l l angles sin9 = tan9 = LX then R = ^ k l 2rrLX Therefore the t o t a l l e n g t h of the Debye r i n g = ^ k l The i n t e g r a t e d i n t e n s i t y per u n i t l e n g t h of r i n g i s given by: , = -"-hkl d h k l h k l 2TTLX S u b s t i t u t i n g from Equation (1) - 39 - / 1 SPECIMEN R = RADIUS OF D E B Y E RING L = SPECIMEN - S C R E E N D I S T A N C E Figure 6. Geometrical arrangement of debye ring formed on screen. Figure 7. Electron d i f f r a c t i o n pattern from Cu„0 powder. - 40 - But d\ hkl 2 sine Substituting in Equation (2) 3 J Q X t 2 •^kl = 77772 '*hkl' . 2n lorrLft sm 0 When comparing different rings on a given diffraction pattern from a specific crystal, only relative intensities, and not absolute intensities are required. Hence, on a given diffraction pattern from a given crystal, J , X , ft , t, L, are a l l constant, and thus the relative integrated intensity of rings with respect to each other may be written: T * 7 M L J L nkl. n „. x . 2 Q (relative) sin 6 In practice the relative intensities are normalized with respect to the strongest ring and expressed as a percentage of the strongest ring. Note that for X-rays (73), for kinematical diffraction, ignoring absorption effects, , = IFI 2 p ( l + cos 228) Takl . 2. sm 8 cos9 2 When 0 is small, I^ki maY ^e written as 2^ F|—^- sin 0 where |F| i s the structure amplitude for X-rays. The relative integrated intensity w.r. to the strongest ring i s given by: - 41 - !• - M i a ( r e l a t i v e ) s i n 6 Hence the t h e o r e t i c a l r e l a t i v e i n t e n s i t i e s of Debye r i n g s f o r e l e c t r o n s may be obtained from the formula f o r X-rays when c o n s i d e r i n g k i n e m a t i c a l c o n d i t i o n s , I.e., there i s a geometrical s i m i l a r i t y of p a t t e r n formation f o r e l e c t r o n and X-ray powder d i f f r a c t i o n . The procedure f o r e s t i m a t i n g I ^ k l ^ o r a P a r t :'- c u^ a r ( r e l a t i v e ) c r y s t a l i s given i n the Appendix. Note that i n the c a l c u l a t i o n s , the s c a t t e r i n g f a c t o r s for e l e c t r o n s f of metals and of i o n i c s o l i d s were taken as equal to those f o r the appropriate n e u t r a l atoms. However, f i s most s e n s i t i v e to i o n i c charges when • S" 1® < 0.1 x a x -8 0 10 (67c), i . e . , when d ^ 5 A, there i s a marked d e v i a t i o n i n f from that of the n e u t r a l atom. In such cases, the low angle i n t e n s i t i e s w i l l be e i t h e r over or under-estimated depending on which i o n d i c t a t e s the s t r u c t u r e amplitude. In the current work, t h i s was not a s e r i o u s o problem because the l a r g e s t i n t e r p l a n a r spacing considered was 4.8 A, i . e . , d j ^ f o r Fe^Oy I n a d d i t i o n , the Debye-Waller f a c t o r , c o r r e c t i n g f o r v a r i a t i o n of r i n g i n t e n s i t y w i t h temperature was neglected. Howie (70) has shown that the i n t e n s i t y of the (111) f o r A l , Cu, Au i s independent of temperature over the range 150-800°K. 2.2.4.2 J u s t i f i c a t i o n of T h e o r e t i c a l Aspects R e l a t i v e t h e o r e t i c a l i n t e n s i t i e s corresponding to Au, C^O, CuCl and Te^O^ powders were c a l c u l a t e d by the procedure o u t l i n e d i n the Appendix. The values are shown i n Tables VI-IX. Ring patt e r n s at 100 Kv were obtained from reagent grade Cu 00, CuCl, Fe„0, powders using - 42 - TABLE V I I - I and I f o r Cu o0 Powder e v 2 C r y s t a l l o g r a p h i c Data Observations o o d A h k l I % e d A I V 3.020 110 12 3.04 w 2.465 111 100 2.48 s 2.135 200 28 2.10 m 1.743 211 4 1.75 vw 1.510 220 36 1.52 m 1.350 310 1 1.287 311 27 1.29 w/m 1.233 222 5 1.23 W W 1.141 321 1 1.067 400 4 1.007 330/411 0.5 0.9795 331 9 0.99 W W 0.9548 420 5 I = r e l a t i v e v i s u a l i n t e n s i t y I g = t h e o r e t i c a l r e l a t i v e i n t e n s i t i e s f o r e l e c t r o n s s = s t r o n g , m = medium, w/m = weak medium, w = weak, vw = very weak ww = very very weak - 43 - TABLE V I I I - I and I f o r CuCl Powder e v C r y s t a l l o g r a p h i c Data Observations o d A h k l I % e o d A I V 3.127 111 100 3.10 vs 2.710 200 0.4 1.915 220 50 1.91 s 1.633 311 29 1.64 m 1.563 222 0.2 1.354 400 6 1.36 vw 1.243 331 8 1.23 vw 1.212 420 0.5 1.105 422 10 1.11 vw h k l = M i l l e r indeces of r e f l e c t i n g plane * v = r e l a t i v e v i s u a l i n t e n s i t y I = r e l a t i v e t h e o r e t i c a l i n t e n s i t y f o r e l e c t r o n s vs = very s t r o n g , s = s t r o n g , m = medium, vw = very weak - 44 - TABLE IX - I and I for Fe„0, Powder e v 3 4 Crystallographic Data Observations o d A hkl I % e d A I V 4.85 111 8.5 4.8 w/m 2.96 220 32.5 2.95 m 2.53 311 100 2.50 vs 2.419 222 1.3 2.096 400 32 2.09 m 1.93 331 1 1.713 420 8 1.71 w 1.614 333/511 25 1.62 w/m 1.48 440 50 1.49 m 1.41 531 0.6 1.33 620 3 1.33 WW 1.28 533 7 1.29 vw 1.21 444 5 1.23 vw 1.12 642 3 1.13 vw 1.09 553/731 12 1.07 w 1.05 800 5.5 0.99 660/822 1 0.97 555/751 5 0.96 vw 0.94 840 4 0.89 664 0.5 0.88 931 2.5 0.86 844 8.0 hkl = Miller index of reflecting plane I = theoretical relative intensities for electrons e I = relative visual intensities v vs = very strong, m = medium, w/m = weak-medium, w=weak vw = very weak, ww = very very weak - 45 - Figure 8. Electron diffract ion pattern from CuCl powder. Figure 9. Diagram of specimen set up in low angle electron diffract ion, (a) macroscopic fracture surface is para l le l to incident electron beam, (b) electron transmission distance through sur- face film is much greater than film thickness and depends on angle of local surface projection to incident electron beam. - 46 - the aforementioned technique employed for gold f i l i n g s . Ring patterns corresponding to the Cu^O and CuCl powders are exhibited in Figures 7 and 8. A comparison of the visual relative intensities of the patterns corresponding to the four crystal structures tabulated in Table VI-IX with the theoretical estimates of relative intensities tabulated in Table VI-IX reveals that there is quite a close correlation between the two. This i s taken to ju s t i f y the valid i t y of the theoretical considerations, assumptions and approximations. The following were noted during the course of the experimental and theoretical work: (1) The interplanar spacings revealed on the diffraction patterns are estimated to be accurate to within ̂  3%. (2) Comparison of the theoretical relative intensities for electrons with the relative intensities for X-rays given in the ASTM index reveals that the corresponding values of the former are generally lower, and that the values are in the same sequence of decreasing magnitude. In addition, the relative intensities of the weaker lines at high angles are larger for X-rays than for electrons because the Lorentz-polarization factor for X-rays increases at the high angles. This suggests that for complicated structures where i t is very d i f f i c u l t to calculate the values of theoretical integrated intensities for electrons, one can obtain an approximate estimate from the ASTM card index f i l e . (3) It is doubtful that very low values of relative intensities (below 3̂%) for electrons can be observed. It i s for this reason, for - 47 - example, that i t i s d i f f i c u l t , i f not impossible, to distinguish between Fe^O^ and y-Te^O^ powders by this technique. 2.2.5 Procedure for Identification of Unknown Phases From the above considerations, a simple technique for the identification of unknown phases within certain limits of accuracy emerges. Electron diffraction patterns are obtained from protuberances and the interplanar spacings are measured with an accuracy of ̂  3%. These values are compared with those for known materials l i s t e d in the ASTM X-ray card f i l e or other crystallographic literature. Wherever a good correlation between spacings i s found, the identity of each diffraction pattern is further confirmed by comparing the visual electron diffraction ring intensities with those of estimates of relative intensities for electrons. 2.2.6 Application of Electron Diffraction Technique to Fracture Surfaces This section describes experimental details employed in the current work. Corrosion products and phases present on actual undisturbed fracture surfaces of beta brasses and austenitic steels were identified by electron diffraction studies conducted at 100 Kv. A section of specimen containing the fracture surface was placed in the high resolution electron diffraction accessory of the Hitachi HU-11A electron microscope with the macroscopic plane of the fracture surface par a l l e l to the incident beam of electrons. This is shown schematically in Figure 9(a). Such a technique leads to the formation of electron - 48 - diffraction patterns arising from transmission of electrons through protuberances on the fracture surfaces. Under these conditions, the film present on the fracture surface offers an effective penetration path considerably larger than their thickness. The actual magnitude of the penetration path depends upon the angularity of the protuberances. This is illustrated in Figure 9(b) where the penetration path XY i s seen to be larger than the penetration path AB. Consequently, as the intensities of diffracted beams are increased by increased penetration paths, the technique allows reasonably clear electron diffraction patterns to be obtained from thin corrosion films or other phases. The diffraction patterns appeared as rings which were calibrated with respect to standard gold ring patterns obtained under the same instrument conditions. The patterns were then analyzed by the procedure described in Section 2.2.5. - 49 - 2.3 Fractography and Metallography The combination of topographical studies and surface mating experiments were used in conjunction with the electron diffraction studies. Mating determined whether or not the microstructural features associated with the topography were formed before or after crack propagation. The topographical studies revealed features characteristic of the fracture process. Subsequent metallographic examination using etching techniques f a c i l i t a t e d correlation of the topographical features with the etched substructure and the presence or absence of phases associated with the crack t i p . Details of the fractographic and metallographic techniques are given below. 2.3.1 Fractographic Technique A general topographical survey was performed on a l l specimens using the scanning attachment to the JEOL Co. probe microanalyzer and the Cambridge Stereoscan scanning electron microscope. Finer details of surfaces were obtained by standard indirect and direct replica transmission electron microscopy using the Hitachi HU-11A electron microscope. Mating studies on S.C. fracture surfaces of the 304L and brasses were performed using the above mentioned scanning instruments while those on the 310 steel were performed using the ETEC Autoscan scanning electron microscope. The Stereoscan and the Autoscan had the advantage of performing to a lower magnification than the JEOL Co. probe attachment, i.e., x7 compared with x300. This aided in location of features. The - 50 - ETEC Autoscan and the probe microanalyzer attachment had the advantage of being equipped with a specimen holder wide enough to accommodate two rod off-cut specimens side by side. Thus, a simple X or Y translation would bring the matching feature into the f i e l d of view. Specimens for fractographic examination were generally cut par a l l e l to the macroscopic fracture surface to a maximum length of 1/2" allowed by the Stereoscan. 2.3.2 Metallographic Techniques Both par t i a l l y and f u l l y fractured specimens were examined. Specimens were carefully sectioned with a jewellers saw in a direction perpendicular to the macroscopic fracture surfaces and mounted in "Koldmount". It was essential to observe the microstructure close to the fracture surface and therefore in the case of the f u l l y fractured specimens a small off-cut was positioned adjacent to the fracture surface. Specimens were then abraded to 000 grade emery, polished on either alumina or diamond to 1 P, given a light electropolish in the reagents shown in Table X, and etched with the particular reagents (in Table XI) to reveal phases. Some specimens were also examined under polarized light. An alternative mounting procedure was adopted for some f u l l y fractured specimens (62) to aid relating the etch patterns to the topographical markings. 3/8" transverse off-cuts pa r a l l e l to the macroscopic fracture surface were mounted in "Koldmount" without the jewellers saw sectioning. Grinding was effected in such a way to reveal sections vertical to the fracture surfaces, as in the previous case, but much more of the actual fracture surface was preserved. - 51 - TABLE X - Electropolishing Reagents Alloy Electropolish 304L,310 steels 50 cc Perchloric Acid 1000 cc Acetic Acid 5-15 cc Water Brasses 25 c c Chromic Oxide 133 cc Acetic Acid 7 c c Water - 52 - TABLE XI - Etching Conditions Alloy Phase Etch Remarks 304L a' martensite 16 gms Chrome Oxide 80 gms NaOH 145 gms water Boiling Immersion. 5-10 mins. Ref.74 a' + £ martensite Saturated Oxalic Acid Solution Electrolytic. h. volt. Easily susceptable to pitting. Easy to overetch - check by frequent periodic removal of specimen. Ref.74 310 a 1 e Assumed to be same for 304L. No data available on specific martensite etches for 310 grade 0-brasses martensite/ twins 5 gms FeCl- 10 cc HCl 100 cc H20 Ref.63 10 gm Ammonium Per- sulphate 100 cc HO 25 cc NH70H 4 Immersion 5-10 sees Ref.75 10 gms FeCl_ 30 cc cone. HCl 125 cc Ethanol Immersion 5-10 sees Ref.76 - 53 - After polishing and etching this surface, the plastic was removed by immersion in acetone. On subsequent examination in the scanning electron microscope, i t was possible to relate directly the topographical features to the morphology of the etched phases directly at the line of intersection of the two surfaces. Two other procedures were adopted where d i f f i c u l t y was encountered in detecting optical metallographically observable patterns adjacent to the fracture surfaces. (1) Stress corrosion cracked specimens were examined i n the scanning electron microscope to detect any small scale topographical effects produced by etching. Special reference specimens known to possess the suspected phases were observed under polarized li g h t , and subsequently etched and observed. (2) Taper sections were examined by optical and electron metallography. - 54 - 3. RESULTS, OBSERVATIONS AND DISCUSSION 3.1 304L A u s t e n i t i c S t a i n l e s s S t e e l Times to f a i l u r e of the 304L S.C.C. specimens were i n the range 12-17 hours. 3.1.1 I d e n t i f i c a t i o n of Phases on F r a c t u r e Surfaces E l e c t r o n d i f f r a c t i o n s t u d i e s of f r a c t u r e surfaces revealed that r i n g d i f f r a c t i o n p a t t e r n s obtained from s t r e s s c o r r o s i o n f r a c t u r e surfaces produced at +154°C were s i m i l a r to those obtained from d u c t i l e f r a c t u r e s produced at -196°C, but d i s i m i l a r from p a t t e r n s obtained from d u c t i l e f r a c t u r e s o c c u r r i n g at +154°C. This may be seen by v i s u a l comparison of three r e p r e s e n t a t i v e p a t t e r n s presented i n F i g u r e s 10(a) ( D u c t i l e +154°C), 10(b) ( D u c t i l e -196°C) and 10(c) (Stress C o r r o s i o n +154°C). A more d e t a i l e d study of the i n t e r p l a n a r spacings corresponding to the d i f f r a c t i o n r i n g s , and t h e i r r e l a t i v e v i s u a l i n t e n s i t i e s (1^) confirmed that w i t h i n the l i m i t s of d e t e c t i o n , the d u c t i l e f r a c t u r e s u r f a c e at -196°C was w h o l l y b.c.c. a' phase and the d u c t i l e f r a c t u r e s u r f a c e at +154°C was predominantly f . c . c . y phase (together w i t h some weak r e f l e c t i o n s suggesting a p o s s i b l e t r a c e of a' phase). These two observations are i n reasonable accord w i t h normal expectations of the e f f e c t s of temperature and deformation upon the m a r t e n s i t e t r a n s f o r m a t i o n (42). - 55 - Figure 10. Electron diffraction patterns obtained from 304L fracture surfaces. (a) Ductile fracture at + 154°C showing predominantly untransformed f.c.c. y phase (austenite). (b) Ductile fracture at - 196°C showing wholly b.c.c. a' phase (martensite). (c) Stress corrosion fracture in boiling aqueous MgCl- at 154°C showing predominantly b.c.c. a' phase (martensite). - 56 - TABLE XII - Identification of Phases on Fracture Surfaces by Electron Diffraction Electron Diffraction Observations Crystallographic Data of Known Phases Ductile Fracture +154°C Ductile Fracture -196°C Stress Corrosion +154°C y-phase (f .cc.) e- ( -phase cp.h.) a '-phase (b.c.c) d(X) Iv d(X) Iv d (8) Iv hkl d(X) I % e hkl d(A) I % e hkl d(&) I e % — — — 100 2.20 27 - - — 2.08 V.S. I l l 2.07 100 002 2.07 28 - - - 2.02 V.S. 2.04 V.S. - - - - - - 110 2.03 100 - - - 101 1.94 100 - - - 1.80 S. 1.80 V.W. 200 1.80 43 102 1.51 13 : : 1.47 v . w . 1.44 M. 1.44 M. 200 1.43 13 1.28 s . 1.29 V.W. 220 1.27 21 110 1.27 12 - - 1,20 v . w . 1.16 S. 1.18 s . - - - 103 1.17 14 211 1.17 22 : - - - 200 1.10 2 - • - - 1.11 s . 1.08 v . w . 311 1.08 21 112 1.08 12 - - - - - - 201 1.06 8 - - - 1.05 M/W 222 1.04 5 004 1.04 2 - - - 1.02 M/W 1.03 M/W - - - - - - 220 1.01 6 - - - 202 0.972 2 - - - - - - 104 0.938 1.5 - - - 0.88 W. 0.915 M/W 0.914 M/W 400 0.900 2 - - - 310 0.906 7 - - - 203 0.861 3 - - - 0.83 M/W 0.825 V.W. 0.83 V.W 331 0.-826 6 210 0.831 1 222 0.828 1 - - - 211 0.815 5 - - - 0.81 W 402 0.805 5 114 0.803 3 - - - - - - 105 0.776 2 - - - 0.765 M/W 0.78 M/W - - - 212 0.772 1 321 0.776 6 - - - 204 0.775 0.5 - - - 0,73 W 422 0.735 3 300 0.733 1 - - - Mainly Y-phase Wholly a'-phase Mainly ct1 -phase TABLE X I I - C o n t i n u e d 57 hkl d(A°) Ie(%) Miller Indeces of Bragg-reflecting c r y s t a l planes. Interplanar spacing of c r y s t a l planes i n Angstrom Units. Relative t h e o r e t i c a l integrated i n t e n s i t y of electron-beam r e f l e c t i o n s expressed as a percentage of the strongest r e f l e c t i o n in each phase. Relative v i s u a l i n t e n s i t y of observed r e f l e c t i o n s . V.S. - very strong, S. - strong, M. - medium, W. - weak, V.W. - very weak. General path of stress corrosion cracks i n 304L exposed to b o i l i n g aqueous MgC^ at + 154°C. (a) O p t i c a l micro- graph of a lo n g i t u d i n a l section of specimen i l l u s t r a t i n g branching nature of the cracks. 20x. (b) Enlargement of area in(a)showing preference for transgranular cracking. Grain boundaries l i g h t l y etched i n o x a l i c acid. 250x. (c) Scanning electron micrograph absorbed electron image of portion of actual crack surface exhibiting mixed i n t e r - granular (faceted) and transgranular (roughened) crack paths. 300x. - 58 - However, the stress corrosion fracture surface consisted predominantly of a 1 phase, together with some very weak reflections consistent with traces of parent Y - P n a s e ' Thus, stress corrosion cracking processes had promoted formation of the ct' phase product. The details of the phase identification study are presented in Table XII where the experimentally observed interplanar spacings and relative intensities of reflections are compared with crystallographic data for the y> £ and a' phases. The interplanar spacings of the three known phases were obtained from the work of Otte (77) and are o based upon a la t t i c e parameter for a' phase of a Q(a') = 2.866 A, a o parameter for y~phase of a Q(Y) ~ 3.59 A and a parameter for the e phase o o of a 0 ( e ) = 2.54 A and c Q ( e ) = 4.15 A. Experimental work by Reed (74) on a 304L alloy of similar composition to the present studies confirmed o o that a Q(a') = 2.865 A and a Q(y) = 3.595 A. The integrated intensities I of the electron diffraction reflections for the three known phases e were theoretical estimates, determined as shown in the Appendix. A weighted average of the scattering amplitude, based on the composition of the alloy in terms of Fe, Ni, and Cr was employed in each calculation. 3.1.2 Stress Corrosion Crack Path The general crack paths were of the form commonly observed for fracture in boiling solutions of ^gC^ (78). Cracks were branched (Figure 11(a)) and were predominantly transgranular as revealed by the light oxalic acid etch in Figure 11(b). Occasional regions of mixed intergranular and transgranular crack paths were observed as evidenced by the faceted (intergranular) and roughened regions (transgranular) of - 59 - the scanning e l e c t r o n micrograph i n Figure 11(c). Such f a c e t i n g as evidence of an i n t e r g r a n u l a r crack path has a l s o been reporte d i n type 304L s t e e l by Marek and Hochman (79). N i e l s o n (59) has a l s o ebserved the "mixed mode" type of f r a c t u r e . The p r i n c i p l e question regarding crack paths however, was whether or not the a' phase detected by the e l e c t r o n d i f f r a c t i o n s t u d i e s was formed i n the r e g i o n ahead of the propagating crack t i p or a f t e r the crack had passed. A c c o r d i n g l y , the chromate and o x a l i c e t c h i n g techniques of Reed (74) were employed. Both etchants revealed martensite phases ahead of a crack t i p of p a r t i a l l y cracked and s e c t i o n e d specimens, the chromate etc h being most u s e f u l as t h i s r e v e a l e d a' phase only. The presence of a' phase, as revealed by the chromate e t c h , i s seen as a s e r i e s of p a r a l l e l and i n t e r s e c t i n g s t r i a t i o n s i n the v i c i n i t y of crack t i p s i n Figures 12(a) and 12(b). G e n e r a l l y , the etched martensite region ahead of the crack t i p was of the order of a g r a i n diameter, or l e s s , and the extension of the etched m a r t e n s i t e s t r u c t u r e below the f r a c t u r e s u r f a c e was of the same order of magnitude. Thus, the observations were c o n s i s t e n t w i t h a s t r e s s c o r r o s i o n crack propagating through a re g i o n of a 1 martensite formed a t the crack t i p . At this p o i n t , i t i s p e r t i n e n t to note that c o n s i d e r a b l e d i f f i c u l t y was obtained i n s u c c e s s f u l e t c h i n g of the martensite phases adjacent to the s t r e s s c o r r o s i o n cracks. For example, l i g h t e t c h i n g w i t h o x a l i c a c i d d i d not c l e a r l y e s t a b l i s h the presence of any martensite phases, as may be seen from the absence of any such e t c h i n g c h a r a c t e r i s t i c s i n F i g u r e 11(b). In f a c t , i f the e l e c t r o n d i f f r a c t i o n evidence had not been a v a i l a b l e to support the presence of a' phase, i t i s probable - 6 0 - Figure 12. O p t i c a l micrographs demonstrating presence of a' martensite, as revealed by the chromate etch, i n the region adjacent to s t r e s s c o r r o s i o n crack t i p s i n 304L s t e e l . The martensite l a t h s are v i s i b l e as p a r a l l e l and i n t e r s e c t i n g s t r i a t i o n s i n both (a) 1050x and (b) 700x. - 61 - that based on the early unsuccessful experiments with the etchants, i t would have been concluded that no martensite was present. This suggests that the volume percentage of a' phase may be extremely high immediately adjacent to the stress corrosion fracture surface, but f a l l s off rapidly to approximately zero percent within a distance of approximately one grain diameter below the surface. 3.1.3 Fractography The general topographical features of the stress corrosion fracture surfaces were quite complex. Areas containing chevron patterns with an apparent cleavage characteristic similar to those observed by Nielson (54) were obtained as shown in Figure 13(a). Also areas containing intersection ridges, (Figure 13(b)), and parallel striations have been interpreted (54) as a possible consequence of periodic crack arrest and propagation. However, in the case of Figure 13(c) the interpretation appears to be unlikely because the line joining AB represents the apex of the machined notch on the test specimen. Thus, the expected direction of crack propagation would be normal to AB in the direction of the arrow. It is evident that the striated features do not l i e normal to the direction of crack propagation as required by a periodic crack arrest model. The fracture surfaces exhibit matching topographical detail on each surface as evidenced in Figures 14(a) and 14(b). This confirms that the phases present were indeed formed before failure and is consistent with the propagation of the crack through the a' phase. The mixed mode (transgranular and intergranular) failure shown in Figure 14(b) is - 62 - Figure 13. Examples of topographical patterns on s t r e s s c o r r o s i o n f r a c t u r e surface of 304L and comparison of these w i t h martensite etch patterns formed near d u c t i l e f r a c t u r e r e - gion at - 196°C.1200x. Figures ( a ) , ( b ) , ( c ) are e l e c t r o n micrographs of pre-shadowed d i r e c t carbon r e p l i c a s obtained from s t r e s s c o r r o s i o n f r a c t u r e surfaces. The arrow i n (c) denotes d i r e c t i o n of microscopic crack propagation f o r t h i s p a r t i c u l a r area (AB). Figures aa, b b , and cc are o p t i c a l micrographs of martensite etch patterns as revealed b y the chromate etch f o r aa and cc and the o x a l i c a c i d etch f o r b b . - 63 - Figure 14(a) Figure 14(b) Figures 14(a) and 14(b). Scanning electron micrographs showing examples of mating surfaces xl400. Figure 14(b) shows mixed mode fracture (smooth and roughened regions) and i s only p a r t l y mated owing to loss of portions of grain. - 64 - similar to that shorn in Nielson's matching surface scanning electron fractographs (59). It is also observed in Figure 14(b) that the surfaces exhibit only p a r t i a l topographical matching, and this i s attributed to the loss of p a r t i a l or whole metal grains during handling. Clearly i f such loss is severe, i t can be appreciated that successful surface mating may be impossible to achieve. Comparison of topographical patterns obtained on the stress corrosion fracture surfaces with etch patterns obtained from predominantly a' phase regions were reasonably similar. For example, the etched ct'- martensite lath patterns in the deformed region near a ductile fracture conducted at -196°C are shown in Figures 13(aa), 13(bb) and 13(cc). It i s apparent that there is a reasonable similarity in both configura- tions and spacing of the etched ct' laths with the stress corrosion topography of Figures 13(a), 13(b) and 13(c). A few other examples of complex a'-laths are shown in the work of Reed (74) and Benson et a l . (80). Similarities between etched a' martensite and the stress corrosion crack topography may be expected i f such crack surfaces contain a' phase (Figure 10 and Table XII) and the crack propagated through a region of a' phase (Figure 12). However, one important question remains to be answered and that is whether or not the stress corrosion crack topography i s due simply to the normal fracture behaviour of induced a' phase, or is a more complex fracture process associated with the presence of the environment. Consequently fractographs of ductile fractures produced at -196°C (a' martensite surface) were compared with fractographs of ductile fractures produced at +154°C (predominantly y-phase surface). It was established that both types of fracture - 65 - specimen exhibited cusps characteristic of ductile failure by microvoid formation and coalescence, as evidenced in Figures 15(a) (-196°C) and 15(b) (+154°C). This normal fracture behaviour is unable to account for the stress corrosion surface morphology indicating that the stress corrosion process involves other factors in addition to inducing a'-phase formation, e.g., dissolution. 3.1.4 Discussion The experimental observations on the 304L steel are reasonably consistent with a mechanism of stress corrosion cracking involving formation of a' martensite phase ahead of a crack tip and subsequent propagation of the crack through the ct' phase. Electron diffraction studies, etching of ct' phase in the region of the stress corrosion cracks, and fracture surface topography a l l support such a conclusion and may lend credence to the elements of Edeleanu's quasi-martensite model (38,39). The mechanism of formation of the a' phase during stress corrosion cracking in aqueous MgC^ at +154°C is not entirely clear. It cannot be related solely to a deformation induced transformation at the crack tip because electron diffraction studies showed that the untransformed y-phase predominated on the heavily deformed ductile fracture surface produced at +154°C. (Weak reflections observed in Table XII suggest that traces of a' phase may be present, and chromate etching techniques did yield traces of a' martensite near the ductile fracture surface. However, the fact remains that the ductile fracture surface was predominantly y~phase). - 66 - (a) Fracture at - 19f>cC (b) Fracture at + 154°C - 67 - The possibility exists that a' phase formation may be influenced by a hydrostatic tensile component of the strain f i e l d near the crack tip. Calculations based on the la t t i c e parameters of the y and a' phase, and the number of atoms per unit c e l l show that on transforming from y to ct1 phases there is a volume increase of <v 2%. Therefore, thermodynamic considerations would suggest that application of a hydrostatic tensile strain (volume expansion) may promote the y -+• a' transformation. Generally, notches and cracks in bulk specimens lead to conditions of plane strain near the tip of a crack (notch) under conditions of applied tensile stress (81). This is in effect a t r i a x i a l stress state containing a component of hydrostatic tensile stress and tensile strain. Thus, the hydrostatic component may conceivably contribute to the a' phase formation. However, this may be only part of the answer as the notched specimen fractured in o i l at +154°C also contained a t r i a x i a l stress state (though different from that at the tip of a S.C. crack), and the fracture surface was predominantly untransformed yphase. Further i t is possible to promote stress corrosion cracking in sheet specimens of 304 type austenitic steels where plane strain conditions and t r i a x i a l stress states do not exist to any significant degree. The complete answer may lie in the work of Vaughan et a l . (40) , Holzworth and Louthan (82), and Rhodes (83). Vaughan et a l . (40) and Holzworth and Louthan (82) showed that hydrogen charging of 304L y steel expands the f.c.c. Y-phase lat t i c e and promotes formation of martensite phases. Rhodes (83) presented evidence to show that hydrogen i s evolved at the tip of a stress corrosion crack in aqueous chloride solutions. Thus, any hydrogen adsorbed at the crack tip w i l l - 68 - expand the f.c.c. lattice and promote a hydrostatic tensile strain which, together with any other hydrostatic tensile components near the crack tip, may be a sufficient thermodynamic condition to promote formation of a'-martensite. (It is noteworthy that Edeleanu (38) had hypothesized earlier that hydrogen should be expected at the tip of a stress corrosion crack which might lead to hydrogen embrittlement of austenitic steels). Details of the mechanism of crack propagation through the a' phase at the tip of the stress corrosion crack are not completely resolved. Normal fracture behaviour of the a' phase cannot account for the observed fractography. If hydrogen is involved in a' phase formation i t i s possible that presence of atomic hydrogen may lead to hydrogen embrittlement of the phase by promoting cleavage failure. Benson et a l . (80) promoted embrittlement of 304L steel by hydrogen charging. They obtained a martensite transformation and observed that the fracture surface contained b r i t t l e type fracture facets surrounded by a region of more ductile failure. They attempted to demonstrate a similarity between the fracture facets and the appearance of martensite shear markings on a prepolished surface, suggesting that hydrogen promoted quasi-cleavage of the martensite. However, there was not a good correlation between fractographs of their hydrogen charged specimens and the stress corrosion fractographs-" observed in the present work. The reasonably good similarity between stress corrosion fracture surface topography and the etching (a dissolution process) patterns of a' phase observed in Figure 13 suggests that dissolution in the a' phase is the more probable process of crack propagation. The driving force for dissolution could be the electrochemical potential difference between untransformed f.c.c. austenite, either on the original specimen surface or near a p a r t i a l l y transformed crack tip, and the b.c.c. martensite. This process could involve generation of hydrogen which enters the metal at the crack tip (83) and forms more ct' phase, thus making the process autocatalytic. Furthermore, such a process would tend to make stress corrosion crack propagation independent of the presence of prior martensite, as observed by Hoar and Hines (27). Kelly (84) and Reed (74) have studied the crystallography of the a 1 martensite in the 304 type austenitic steels. They found that a' phase occurs in the form of laths with an austenite habit plane close to (112) (84). However, the laths were contained in sheets lying Y para l l e l to (111) . Thus, in a pa r t i a l l y transformed' region at the tip of a stress corrosion crack the interface between the sheets of a' martensite and the untransformed y-phase would be (111). Such an interface could be a possible path of dissolution because of potential differences between the two phases, thus tending to make the leading edge of the propagating crack follow {111}^ planes, as may be the case in some instances (41,85). In a f u l l y transformed region adjacent to a crack tip, crack propagation could s t i l l occur at the interface between parallel sheets of laths (i.e., along prior (111)^) due to the angular misorientation between the laths in one sheet and the laths in the neighbouring sheet forming a continuous higher energy interface somewhat akin to a grain boundary. Under these conditions, crack propagation rates may be different from the situation for a pa r t i a l l y transformed region. - 70 - Furthermore, Kelly (84) has shown that the long direction of a' martensite laths l i e s p a r a l l e l to the [110]^ direction. Therefore, i f dissolution at the leading crack edge occurs along the interface of the laths in a given (111)^ sheet, a series of dissolution tunnels lying on the austenite {111} <110> system may be formed. Such a system of tunnels has been observed by Nielson (58) using oxide extraction replicas. At the time of his observation, Nielson was not certain whether the tunnels were associated with martensite formation or dislocation configurations in the austensite. In summary, there is a good case to be made for the direct association of martensite transformations with the stress corrosion cracking of 304L austenitic steel. It is not necessarily the only process, other mechanisms may be involved. For example, in annealed specimens composed wholly of y-phase, the crack i n i t i a t i o n process may involve mechanisms based upon austenitic dislocation configurations, sli p step characteristics and alloy chemistry as revealed by Latanision and Staehle (41). However, as cracking proceeds i t appears that martensite formation and subsequent propagation of the crack through this phase is a kinetically more favorable process. Certainly deformation substructure of the y-phase becomes of secondary importance once the crack is propagating through the a' phase. It remains to be investigated whether martensite plays a role in more stable austenitic steels, e.g., 310 grade. 3.2 310 A u s t e n i t i c S t a i n l e s s S t e e l Times to f a i l u r e of the s t r e s s c o r r o s i o n f r a c t u r e specimens were c o n s i s t e n t l y i n the range 92-100 hours, c o n s i d e r a b l y x6) g r e a t e r than those of the 304L s t e e l . 3.2.1 I d e n t i f i c a t i o n of Phases on Fr a c t u r e Surfaces E l e c t r o n d i f f r a c t i o n s t u d i e s conducted on the f r a c t u r e surfaces revealed that r i n g d i f f r a c t i o n p a t t e r n s obtained from the S.C.C. f r a c t u r e s u rfaces produced at +154°C were s i m i l a r to those obtained from mechanically f r a c t u r e d surfaces at 154°C and -196°C, as shown i n Fig u r e s 16(a), 16(b) and 16 ( c ) . A d e t a i l e d study of the i n t e r p l a n a r spacings shown i n Table X I I I corresponding to the d i f f r a c t i o n r i n g s suggests that the surfaces of the mechanically f r a c t u r e d specimens contained predominantly f . c . c . y-phase. This o b s e r v a t i o n t h a t severe deformation d i d not appear to produce any a' marte n s i t e phase i s i n accord w i t h the general f e e l i n g that the 310 s t e e l i s s t a b l e w i t h respect to the martensite t r a n s f o r m a t i o n . E l e c t r o n d i f f r a c t i o n p a t t e r n s obtained from the s t r e s s c o r r o s i o n f r a c t u r e surfaces i n v a r i a b l y contained r i n g s c o n s i s t e n t w i t h the presence of the oxide Y-Fe„0 o or Fe_0,, i n a d d i t i o n to a u s t e n i t e r e f l e c t i o n s w i t h i n the 2 3 3 4 l i m i t s of accuracy of the technique. The s t r e s s c o r r o s i o n f r a c t u r e surface e x h i b i t e d a matte grey appearance as compared w i t h the m e t a l l i c l u s t r e of 304L s t e e l . Some 310 S.C. specimens e x h i b i t e d only oxide r e f l e c t i o n s . The e x p e r i m e n t a l l y observed i n t e r p l a n a r spacings and r e l a t i v e v i s u a l i n t e n s i t i e s of r e f l e c t i o n s presented i n Table X I I I were compared Figure 16. Electron d i f f r a c t i o n patterns obtained from 310 fracture surfaces, (a) Duct i l e fracture at + 154°C showing un- transformed fee y phase (austenite). (b) D u c t i l e fracture at - 196°C showing fee Y phase, (c) Stress corrosion fracture i n b o i l i n g aqueous MgCl 2 at 154°C showing austenite plus oxide. - 73 - TABLE X I I I - E l e c t r o n D i f f r a c t i o n Observations - 310 s t e e l D u c t i l e F r a c t u r e + 154°C ! D u c t i l e F r a c t u r e - 196°C Stre s s C o r r o s i o n + 154°C 0 d A I V O d A I V O d A I V 2.08 vs 2.08 s 2.48 2.09 vw s 1.80 s 1.79 s 1.80 1.73 1.47 s W W vw 1.26 s 1.26 m 1.28 m 1.09 ms 1.09 m 1.09 m 1.04 m 1.05 m 1.04 w 0.88 w 0.82 w 0.82 vw y-phas y-phase Y" •phase + oxide I = r e l a t i v e v i s u a l i n t e n s i t y , v J vs = very s t r o n g , s = s t r o n g , ms = medium/strong, m = medium, w = weak, vw = very weak, ww = very very weak. - 74 - w i t h the c r y s t a l l o g r a p h i c data, shown i n Table X I I , f o r the 304L s t e e l . The t h e o r e t i c a l r e l a t i v e i n t e n s i t i e s f o r the 310 s t e e l are i n d i s t i n g u i s h - able from those of the 304L s t e e l s i n c e the l a t t i c e parameters and the mean s c a t t e r i n g f a c t o r f o r e l e c t r o n s of the two s t e e l s are s i m i l a r . 3.2.2 Crack Path The crack path e x h i b i t e d a much gr e a t e r degree of branching than that e x h i b i t e d i n the 304L grade. The degree of crack branching i s shown i n Figures 17(a)-17(c). Figure 17(a) shows a crack formed across the s e c t i o n of a p a r t i a l l y f r a c t u r e d specimen. The higher m a g n i f i c a t i o n photographs (Figures 17(b) and 17(c)) r e v e a l the apparent c r y s t a l l o - graphic nature of c r a c k i n g ; f o r example, the cracks appear to be d e f l e c t e d by regions of annealing twins q u i t e f r e q u e n t l y . Smith and Staehle (86) have noted such c r y s t a l l o g r a p h i c preference i n high a l l o y a u s t e n i t i c s t e e l s . Scanning e l e c t r o n microscopy detected only apparent t r a n s g r a n u l a r areas, as f o r example, shown i n Figure 17(d). Etch markings were revealed by the o x a l i c a c i d etch near to the crack t i p as shown i n Figures 18(a) and 18(b). They g e n e r a l l y took the form of broad p a r a l l e l bands. In Figure 18(b), these bands could be seen both p a r a l l e l to and p e r p e n d i c u l a r to the crack. S i m i l a r markings were a l s o noted adjacent to the mechanically f r a c t u r e d surfaces produced at-196°C and +154°C. A p p l i c a t i o n of the chromate etch to a l l samples d i d not r e v e a l any a' martensite phase, even when a taper s e c t i o n i n g technique was a p p l i e d . I t was, however, assumed that the 310 a' martensite could be detected by the same etch used f o r 304L a' martensite d e t e c t i o n . In f a c t , a 304L grade sample c o n t a i n i n g a' martensite was etched simultaneously w i t h the 310 grade samples, i n order - 75 - N Figure 17(b) Figures 17(a) and 17(b). General S.C. crack path in 310 exposed to b o i l - ing aqueous MgCl^ at 154°C. Figure 17(a) shows very severe branching initi a t i n g from notched region (N). x23. Figure 17(b) shows preference for transgranular cracking. x250. Both etched in oxalic acid. - 76 - Figure 17(c) Figure 17(d) Figures 17(c) and (d). Figure 17(c) Optical micrograph showing branched nature of crack and crystallographic preference, x.250. Etched in oxalic acid. Figure 17(d) Scanning electron micrograph absorbed electron image of portion of actual stress corrosion fracture surface. x255. - 77 - Figures 18(a) and 18(b). Optical micrographs showing etch patterns adjacent to the stress corrosion crack t i p s . Oxalic acid etch. x400. Fipure 18(a) partially fractured specimen - major crack. Figure 18(b) f u l l y fractured specimen - subsidiary crack. - 78 - to check the performance of the chromate etc h . I t was not immediately c l e a r whether the phase i d e n t i f i e d by the o x a l i c a c i d etch was i n f a c t the e martensite phase, because the etchant r e v e a l s presence of g r a i n and twin boundaries. Thus, i t cannot be precluded that the markings are i n f a c t deformation twins. Deforma- t i o n twins have been observed i n a u s t e n i t i c s t a i n l e s s s t e e l s by Swann and N u t t i n g (55) i n t h i n f o i l s t u d i e s . A t h i r d p o s s i b i l i t y i s tha t the markings may correspond to f a u l t i n g . " T w i n - l i k e " f e a t u r e s have been observed by Douglass et a l . (87) i n 310 s t a i n l e s s s t e e l f o i l s . These workers claimed them as evidence f o r m a r t e n s i t e , but o f f e r e d no proof, e.g., e l e c t r o n d i f f r a c t i o n s t u d i e s . I t i s i n f a c t q u i t e p o s s i b l e that the features correspond to deformation twins or s t a c k i n g f a u l t s . The bands whether i n t e r p r e t e d as e m a r t e n s i t e , deformation twins or f a u l t s w i l l form on {111} planes. 3.2.3 Fractography The general t o p o g r a p h i c a l f e a t u r e s of the S.C.C. f r a c t u r e s u r f a c e were q u i t e d i f f e r e n t from those formed on the 304L s t e e l , and were g e n e r a l l y not as complex. Scanning e l e c t r o n microscopy revealed the surface to co n t a i n smooth f a c e t s as shown i n Figures 17(d), 19(a) and 19(b). O u t l i n e s of g r a i n s c o n t a i n i n g annealing twins can be observed i n Figures 17(d) and 19(b) and are q u i t e frequent. D i r e c t r e p l i c a t r a n s m i s s i o n microscopy was made d i f f i c u l t by the presence of the oxide c o a t i n g . Two r e p r e s e n t a t i v e photomicrographs are presented Figure 19(a) Figure 19(b) Figures 19(a) and 19(b). Scanning electron micrographs of 310 S.C.C. fracture surfaces. Figure 19(a) shows smooth facets. x800. Figure 19(b) shows outlines of grains containing annealing twins. x250. - 80 - Figure 19(c). Scanning electron micrograph showing mating of the two fracture surfaces of a 310 S.C.C. specimen. x260. - 81 - Figure 20(b) Figures 20(a) and 20(b). Preshadowed carbon r e p l i c a s showing "annealing twinned" topography. Figure 20(a) shows smooth regions containing fine s t r i a t i o n s . xl550. Figure 20(b). xl500. - 82 - in Figures 20(a) and 20(b) and suggest that the crack is related to annealing twins. The surfaces are roughened, suggesting dissolution has occurred. Careful examination of Figure 20(a) reveals that the " f l a t " area to the right of the "twinned" region contains many fine striations, and i n addition Figure 20(b) reveals coarse striations. The interpretation of the striations i s not clear. On the one hand, the spacing of the striations corresponds roughly with that of the phase detected by the oxalic acid etch. On the other hand, the striations could merely correspond to sl i p steps formed in the wake of the advancing crack tip. It must be pointed out that care must be exercised when interpreting the significance of these striations when there i s an oxide film present on the fracture surface. For example, the striations in Figure 17(d) exhibit a quasi-cleavage appearance, but may indeed be caused by rupture of the corrosion product deposited in the crack on removing the specimen from the stress corrosion c e l l . The two fracture faces of the S.C.C. specimen match up as evidenced by Figure 19(c). This was found to be very d i f f i c u l t possibly due to the dropping out of grains or parts of grains during preparation for fractography, as a consequence of the very severe degree of crack branching in the material. It could therefore be concluded that the microstructural features associated with the observed topographical features were present prior to cracking. It is perhaps noteworthy that the fragmented grains could be readily picked up by a magnet. The iron oxide corrosion product covering the grains i s probably responsible for this observation, a' i s - 83 - magnetic, y and e phases are not magnetic (42). The iron oxides y-Fe^O^ and Fe^O^ are magnetic (111). Comparison of topographical patterns obtained on the S.C.C. fracture surface with etch patterns of the crack branching were reasonably similar. For example, the transgranular deflections by twinning (Figure 17(b) and 17(c)) could readily be identified both for size and shape with the topographical features noted on the fracture surface (Figures 17(d)). Moreover, as suggested above, the spacing and configura- tion of some surface striations could be related to that of the phase etched by oxalic acid. It i s not currently clear whether the metallic phase present near the actual fracture surfaces is y or e . e phase was not detected by electron diffraction on the mechanically fractured surfaces: phases were noted by etching. This suggests that the markings are possibly not e martensite, consistent with the earlier suggestion of correspondence to deformation twins or faulting. However, irrespective of which phase is present, i t s normal fracture behaviour cannot account for the observed features produced by S.C.C. as evidenced by the formation of dimples characteristic of ductile failure in the mechanically fractured samples. See Figures 21(a) and 21(b). Thus, some other process must be involved. 3.2.4 Discussion The observations are not consistent with a mechanism based on cracking associated with the a' martensite. The failure to observe this phase is consistent with the generally accepted view that i t does - 84 - Figures 21(a) and 21(b). Preshadowed carbon replica showing mechanical fracture topography of 310 steel . (a) mechani- cally fractured at + 154°C. xl500. (b) mechani- cally fractured at - 196°C. xl500. - 85 - not form in the 310 steel. It has never been observed metallographically in this grade of steel, although Douglass et a l . (87) claim evidence for i t s incidence in thin f o i l studies. With reference to the current observations, i f the a' does form, i t must certainly disappear after the crack has passed, i.e., the strain induced phase disappears upon removal of the strain as, for example, is known to occur in the g-brass system (65). Etched bands of either e martensite, deformation twins or faults are very easily observed ahead of the crack t i p , and below the stress corrosion fracture surfaces. The experimental observations are reasonably consistent with the view that the crack path is determined by the presence of one of these three strain induced features. The mechanism of formation of the strain induced features i s not clear. The features may be formed by deformation alone at the crack tip since the same features were noted in the p l a s t i c a l l y deformed specimens. It is possible that cathodically produced hydrogen formed at the crack tip may aid in formation of the e martensite phase. It is now well known (89-92) that hydrogen can embrittle f.c.c. metals and in particular 310 steels. The embrittlement of the 310 grade may arise from the formation of a transformation product or from the embrittlement of the austenite l a t t i c e i t s e l f . It has been determined that hydrogen charging produces an expansion of the la t t i c e and sl i p steps on electropolished 0.0025" sections of 310 stainless steel (92,93). e martensite was detected by X-ray diffraction (93) and markings in hydrogen charged samples considered to be e martensite have been observed - 86 - by metallographic techniques (59). It is probable that the e martensite formation is a consequence of the reduction of stacking fault energy by hydrogen absorption. It is also anticipated that the deformation strain due to expansion of the l a t t i c e w i l l aid i t s formation. On the (Other hand, this w i l l also contribute to the formation of deformation twins or random faulting. Details of the mechanism of crack propagation are not clear. The normal fracture behaviour of the y phase, or e phase cannot account for the topographical features. It i s possible that hydrogen adsorption may embrittle these phases and promote cleavage, but the detailed effects of hydrogen adsorption are not known. Benson et a l . (80) have mechanically fractured a 310 grade steel in a hydrogen environment, but the fractography was of ductile character. Since 310 grade steels are subject to embrittlement by hydrogen (80,89-93), i t is clear that a mechanism of hydrogen embrittlement alone cannot explain the observed fractography. There are slight similarities between the topographical features and etch patterns suggesting that cracking occurs by dissolution. Dissolution can occur along the interface of the bands and the parent y-phase. The driving force may arise from the potential difference between the two phases or from the strain energy of the interface between the parent y-phase and the deformation twin. However, because of the similar structure within the bands and in the y-phase, there may not be as much potential difference between the y and the e phase or twin in the 310 steel, as between the y and the a' martensite in the 304L steel. This may in part account for the slower rate of cracking - 87 - of the 310 s t e e l . Moreover, because the a c t i v e d i s s o l u t i o n path can be produced by deformation alone, and i s t h e r e f o r e present at crack i n i t i a t i o n , the crack has a wide choice of p o s s i b l e paths. This may account f o r the g r e a t e r degree of crack branching than observed i n the 304L grade. As a consequence, the d i s s o l u t i o n current d e n s i t y w i l l be very low, and a l s o i n pa r t may c o n t r i b u t e to the observed slower c r a c k i n g r a t e of the 310 s t e e l . The bands of s t r a i n induced product i n the h i g h a l l o y a u s t e n i t e form on {111}^ planes (55,84). I f d i s s o l u t i o n occurs at the i n t e r f a c e of those bands, c r a c k i n g occurs on {111} planes as i s observed i n high a l l o y a u s t e n i t i c s t e e l s (41). Two sheets on {111} planes can Y i n t e r s e c t along a [110]^ d i r e c t i o n . I f the edge of the crack progresses along t h i s d i r e c t i o n , a s e r i e s of d i s s o l u t i o n tunnels may form along the {111} <110> as observed by N i e l s o n (59). Bands on adjacent {111}^ planes can a l s o i n t e r s e c t to form a (100) " i n t e r l o c k i n g " s u r f a c e . Thus Y fche S.C. f r a c t u r e s u r f a c e may c o n t a i n t o p o g r a p h i c a l f e a t u r e s e x h i b i t i n g p a r a l l e l s t r i a t i o n s . This i s i n f a c t observed. S i m i l a r d i s s o l u t i o n paths, i t may be argued, might operate i n the 304L s t e e l . However, the t o p o g r a p h i c a l f e a t u r e s are completely d i f f e r e n t and t h i s i s c o n s i s t e n t w i t h the suggested d i f f e r e n t d i s s o l u t i o n paths i n the two a l l o y s . In summary, i t i s p o s s i b l e that the t r a n s g r a n u l a r crack path can be a s s o c i a t e d w i t h the m a r t e n s i t e , the i n t e r f a c e between deformation twins or f a u l t i n g and the parent phase. Other f a c t o r s e.g., d i s s o l u t i o n along s l i p planes may be important, but appear to be of secondary importance when m a r t e n s i t e , deformation twins or f a u l t i n g i s present. - 88 - 3.3 8-Brasses The S.C.C. failure times of the water fractured 47.76% Zn binary, 45.54% Zn binary and 33.5% Zn-4.5% Sn ternary specimens were in the ranges 17-18 hours, 5-11 mins, 2-10 mins, respectively. Those corresponding to the ammoniacal copper sulphate fractured binary 47.76% Zn rod specimens, and ternary strip specimens were 3-5 mins and 2 mins, respectively. 3.3.1 Phase Identification Study The analysis of electron diffraction patterns of the 47.76% Zn alloy subjected to S.C.C. in water and in ammoniacal copper sulphate is presented in Table XIV. The rings are consistent with the presence of C^O; no metallic phase was detected. Attempts to reveal any underlying metal phase by electropolishing were unsuccessful since this procedure produced a Cu20 film. See Table XIV. Representative diffraction patterns obtained from the 45.54% Zn binary alloy and 33.5% Zn-4.5% Sn ternary alloy are shown in Figures 22(a) and 22(b). The patterns corresponding to the two alloys are similar. The patterns resemble those corresponding to the mechanically fractured surfaces of these alloys, shown in Figures 22(c) and 22(d). Interplanar spacings and intensities corresponding to the patterns are exhibited in Tables XV and XVI and were generally the same for each fracture surface. Not a l l rings were observed on every plate obtained, but the frequency of observation of specific rings are indicated in the tables. Similar ring patterns were noted in the case of the mechanically TABLE XIV - Electron Diffraction Observations from S.C.C. Surface of Binary 47.76% Zn Alloy S.C.C. i n Water S.C.C. in Water + Electropolish S.C.C. in am- moniacal CuoS0. 2 4 S.C.C. in ammon- iacal Cu 2S0, + electropolish Mechanically Fractured o d A I V o d A I V d A I V ' d A I V d A I V 2.98 vw 2.95 w 2.96 vw 3.0 vw 2.08 vs 2.44 s 2.44 vs 2.44 s 2.44 s 1.67 ww 2.09 ms 2.12 s 2.12 s 2.09 m 1.48 w 1.75 W W 1.73 vw 1.29 ww 1.50 m 1.49 m 1.48 s 1.48 mw 1.205 m 1.05 mw 1.28 mw 1.27 m 1.27 w 1.27 mw 0.94 vw 1.22 mw 1.24 vw 1.23 1.09 w w 1.23 vw .0.86 w , 0.78 vw I = relative visual intensity vs = very strong, s = strong, ms = medium strong, m = medium, mw = medium weak, w = weak, vw = very weak, ww = very very weak. - 90 - Figure 22. Electron diffract ion patterns from fracture surfaces of beta brass. (a) Stress corrosion fracture surface - water/Cu-45.54% Zn (b) Stress corrosion fracture surface - water/Cu-33.5% Zn - 4.5% Sn. (c) Mechanically fractured surface - Cu-45.54% Zn (d) Mechanically fractured surface - Cu-33.5% Zn - 4.5% Sn. TABLE XV - El e c t r o n D i f f r a c t i o n Composite Ring A n a l y s i s f o r Binary Specimen (45.54% Zinc) Stress Corrosion Cracking Mechanical Fracture C r y s t a l l o g r a p h i c Data 0 d A I V No. of pl a t e s observed 0 d A I V No. of pl a t e s observed o d A h k l I % e 2.95 W W 11 2.95 W W 5 2.94 100 < 0.1 2.75 W W 4 2.82 W W 1 2.45 W W 15 2.40-2.44 W W 7 2.12-2.04 vs 15 2.12-2.02 vs 7 2.08 110 100 1.82 vw 5 1.84 vw 7 1.65 vw 7 1.67 vw 7 1.70 111 < 0.1 1.60 W W 6 1.59 W W 6 1.48 m 15 1.49 mw 7 1.47 200 13 1.27 vw 15 1.27 vw 7 1.31 210 < 0.1 1.23 m/w 15 1.24-1.21 mw 7 1.20 211 11 1.15 vw 6 1.15 vw 7 1.12 vw 1 1.06 m 15 1.06 m 7 1.04 0.98 220 300/221 6 < 0.1 0.94 mw 15 0.94 mw 7 0.93 0.889 310 311 7 < 0.1 0.86 mw 15 0.86 mw 7 0.848 222 2 0.805 m 15 0.805 mw 7 0.815 320 < 0.1 h k l = M i l l e r indeces of r e f l e c t i n g planes vs = very s t r o n g , m = medium, w/m = weak me I = t h e o r e t i c a l r e l a t i v e i n t e r g r a t e d i n t e n s i t i e s vw = very weak, ww = very very weak e f o r electrons I = r e l a t i v e v i s u a l i n t e n s i t i e s v TABLE XVI - Electron Diffraction Composite Ring Analysis for Ternary Specimen - (Cu-33.5 Zn-4.5 Sn) Stress Corrosion Cracking Mechanical Fracture Crystallographic Data o No. of 0 No. of o d A I V plates d A I plates d A hkl I % P observed V observed 2.94 v w 10 2.94 W W 7 2.94 100 < 0.1 2.82 W W 5 2.82 W W 1 2.40-2.45 W W 15 2.40-2.44 W W 8 2.12-2.04 vs 15 2.12-2.02 vs 8 2.08 110 100 1,84 vw 5 1.84 vw 8 1.67 vw 11 1.67 vw 8 1.70 111 < 0.1 1.59 W W 10 1.59 W W 8 1.49 w/m 15 1.49 m/w 8 1.47 200 13 1.27 vw 12 1.27 vw 8 1.31 210 < 0.1 1.23 m/w 14 1.24-1.22 m/w 8 1.20 211 11 1.15 vw 8 1.15 vw 8 1.06 m 15 1.06 m 8 1.04 0.98 220 221/300 6 < 0.1 0.94 mw 15 0.94 mw 8 0.93 0.89 310 311 7 < 0.1 0.86 mw 15 0.86 mw 8 0.848 222 2 0.805 mw 15 0.805 mw 8 0.815 320 < 0.1 hkl = M i l l e r indeces of reflecting planes vs = very strong, m = medium, I = theoretical relative intergrated intensities w / m = w e a k medium, vw = very weak, for electrons vvw = very very weak. I = relative visual intensities v - 9 3 - fractured Cu-47.76% Zn alloy, and the interplanar spacings corresponding to the rings are tabulated in Table XIV. Only a few patterns were recorded from the fracture surface of the mechanically fractured Cu-47.76% Zn specimen. The interplanar spacings and the relative visual intensities corresponding to the ring patterns are compared, i n i t i a l l y , with the crystallographic data for Cu-Zn in Table XV and XVI. Because of the large grain size relative to that of the electron beam diameter, incidence of Cu-Zn rings may be indicative of deformation twinning i n the material. However, examination of the observations of Jolley and Hull (94) suggests that the rings corresponding to any Cu-Zn would be masked by any rings arising from a martensite phase transformation product, within the limits of accuracy of the electron diffraction technique. Therefore there could be exclusively martensite on the fracture surfaces, i.e., no deformation twins. Data in the literature (63,75,95-101) pertaining to the crystallography of 3-brass phase transformation products i s too varied to assess any sensible trends, since a variety of martensitic products have been observed in alloys of closely similar composition. Consequently some idea of the actual details of interplanar spacings and relative intensities of the transformation products occurring i n the particular alloys under study was obtained using X-ray powder diffraction of f i l i n g s . The patterns for each alloy were almost indistinguishable. The results presented in Tables XVII and XVIII indeed indicate the presence of Cu-Zn plus a phase transformation product. No detailed attempts were made to identify the phase or phases corresponding to the lines additional to those of Cu-Zn, but i t TABLE XVII - X-ray Powder Diffraction Analysis-Binary Fi l ings (Cu-45.4% Zinc) o d A Intensity Lines Additional to Cu-Zn Relative Intensity of Cu-Zn l i n e s ( l 4 2 > Relative Intensities of Martensite(94) 2.85 m/w 6 2.54 m/w X 15 2.29 w X 10 2.25 w X 20 2.13 vs X 100 2.08-2.02 vs (band) X 100 60 1.96-1.91 w (band) X 35 1.86 . vw 1.70 vw 1 1.51-1.47 ms (band) X 15 15 1.43 'VW X 1. 31 VW 2 1.28 vw X 65 1.27 vw X 1.21-1.19 mw (band) X 29 60 1.14 w X 5 1.10 vw X 5 1.09 vw X 5 1.04 vw 5 N.B. Cu radiation was employed. - 95 - TABLE XVIII - X-ray Powder D i f f r a c t i o n A n a l y s i s - F i l i n g s of Ternary A l l o y (Cu - 33.5% Zn - 4.5% Sn) o d A R e l a t i v e I n t e n s i t y L i n e s i n A d d i t i o n to Cu-Zn Li n e s L i n e s Corresponding to J o l l e y and H u l l Orthorhom- b i c M a r t e n s i t e 3.20 W W X X 2.85 m/w 2.54 m/w X X 2.29 w X X 2.25 w X X 2.13 vs X X 2.08-2.02 vs (band) X 2.01 x 1.96-1.91 w (band) X X 1.86 vw X 1.70 vw X 1.57 W W X X 1.53 W W X X 1.51-1.47 ms (band) Cu/Zn=1.47 A x X 1.43 vw X 1.38 W W X 1.36 W W X 1.32 vw 1.31 vw 1.28 vw X X 1.27 vw X X 1.21-1.19 mw (band) Cu/Zn=1.20A x X 1.14 w X X 1.10 vw X X 1.09 vw X x 1.04 vw N.B. Cu r a d i a t i o n was employed. - 96 - was observed that the interplanar spacings and relative v isual intensities f i t t e d , f a i r l y wel l , the experimental data of Jolley and Hull (94). This suggested the transformation product in these alloys possessed an orthorhombic structure. There were a few extra l ines , however, in addition to the Jolley and Hull (94) data. Note that the X-ray powder dif f rac t ion data of the binary alloy f i l i n g s (Table XVII) was different from that obtained by Massalski and Barrett (65) on f i l i n g s of an alloy of composition close to the Cu-47.76% Zn employed i n the current work. It i s possible that the deformation carried out by these workers was more severe than was given i n the current work. o The interplanar spacings of value greater than 1.04 A and the relative visual intensities of rings on the electron dif f rac t ion patterns correspond to those of lines representing martensite on the X-ray diffract ion pattern, within the limits of accuracy of the techniques. Therefore the electron diffract ion ring patterns are consistent with the presence of martensite. As previously pointed out, however, i f twinning occurs, the corresponding rings on the electron diffract ion patterns w i l l not be distinguished from the rings representing martensite. Therefore i t is not clear whether or not twinning is present conjointly with the phase transformation product on the fracture surfaces. However, as indicated in the next section, information on shear transformations in g-brasses points to the incidence of a martensite transformation rather than twinning. Generally, early work (102) referred to twinning, but no diffract ion studies were performed to support this reference. More recent investigators (63,75,94,96-101) would almost certainly interpret the early observations as incidence of a martensite transformation. - 97 - 3.3.2 Stress Corrosion Crack Path Both the ternary and binary alloys exhibited predominantly trans- granular cracking. In general, two types of crack patterns were observed, f i r s t l y , a transgranular criss-cross crack path and secondly, intergranular with transgranular side branching. The transgranular cracking appeared to have crystallographic preference. Representative examples of the observed crack path are shown in Figures 23-27. Crack patterns observed in the 47.76% Zn binary alloy fractured in ammoniacal copper sulphate is exhibited in Figure 23, and in water in Figure 24. Both transgranular and intergranular modes of fracture of the 45.54% Zn binary in water are shown in Figures 25(a) and 25(b), and of the ternary alloy in Figures 26(a) and 26(b). Cracks in the binary and ternary alloys generally appeared to i n i t i a l l y propagate in a direction approximately 45° to the longitudinal tensile axis of the specimen as evidenced in Figures 27(a) and 27(b). This suggests that cracking is related to a shear process. Microhardness indentations were made on polished surfaces of S.C.C. cracked specimens and in every case examined, a l l cracks were pa r a l l e l to at least one of the s l i p plane traces. An example of this i s shown in Figure 28. Slip in these materials occurs on {110} planes (103-106). When etching techniques were applied to the stress corrosion cracked 47.76% Zn alloy to determine whether any martensite or second phase was associated with the crack path,not a trace of any transformation product was found. The fact that no martensite was observed in the stress corroded alloy was the reason for performing studies with the 45.54% Zn binary and the ternary alloys. These alloys are more unstable - 98 - fractured in ammoniacal copper sulphate solution. (a) Intergranular crack path. x65. Etched ammonium persul- phate. Optical . (b) As in (a) showing greater detail of transgranular crack- ing. xl200. Scanning electron microscope. (c) Transgranular cracks, showing cross hatch pattern. Un- etched x200. - 99 - b Figure 24. Stress corrosion crack path in water/Cu - 47.76% Zn alloy. (a) Steps on major fracture surface. Unetched x250. (b) Cross hatch pattern. x250. Etched in Swann and Warlimont^^ solution. - 100 - * >V.',Vs:*-'>. •. • a Figure 25. Stress corrosion crack path in water/Cu-45.54% Zn system. (a) Intergranular and transgranular failure (x200). Etched in acid fer r ic chloride solution. (b) Steps at major fracture surface (xlOO). Unetched. - 101 - b Figure 26. Stress corrosion crack paths of Cu - 33.5% Zn - 4.5% Sn in water. (a) Intergranular cracking with transgranular side branches. x200 unetched. (b) Cross hatch crack path x200. Etched i n acid ferr ic chloride solution. - 102 - 4* Figure 27. S.C. Cracks propagating at 45° to tensile axis (direction of arrows). (a) Binary Cu - 45.54% Zn in w ter. Unetched x200. (b) Ternary Cu - 33.5% Zn - 4.5% Sn in water. Unetched x200. - 103 - Figure 29. Martensite morphology adjacent to mechanical fracture surface of Cu - 47.76% Zinc specimen. x250. Etched i n ammonium persulphate. - 104 - with respect to the martensite transformation (63,65). Nevertheless, not a trace of martensite was detected metallographically on the specimens subjected to stress corrosion cracking. The fact does remain, however, that there is some transformation product and/or twinning on the stress corrosion fracture surfaces as indicated by the electron diffraction observations. "Twin-like" etched markings were observed on the specimens mechanically fractured, consistent with the electron diffraction observations. Representative photographs i l l u s t r a t i n g the various configurations of this etched product are shown in Figures 29-31. It was not immediately clear whether the etched markings corresponded to twins or martensite. The morphology of the etched phase represented in Figure 30(b) and 31(a) closely resembles that morphology interpreted as martensite by Essenwasser (63). That in Figures 30(a), 30(c), 31(c) and 31(e) resembles the martensite observed by other workers (65,75). The only metallographic evidence of twinning in g-brasses is that presented by Elam (102), resembling that morphology presented in Figure 31(e). Barrett (107) claims that twinning does occur in g-brasses but does not disclose his original source. It currently appears that the evidence is weighted in favour of a martensite rather than a twinning interpretation of the etched markings. Because the electron diffraction patterns revealed the same phases to be present on the mechanically and stress corrosion fractured surfaces, i t is clear that any martensite must be confined to a very thin layer on the stress corrosion fracture surface. This was confirmed when a - 105 - • • • • .J^ Figure 30. Martensite morphology in mechanically fractured Cu - 45.54% Zn specimen. A l l x250 and etched in acid fer r ic chloride. - 106 - a b c Figure 31 (a)(b)(c). Martensite morphology in mechanically fractured ternary Cu - 33.5% Zn - 4.5% Sn alloy. A l l xlOO. Etched in acid ferric chloride. (b) shows proxi- mity of fracture surface. - 107 - Figures 31(d) and 31(e). Martensite morphology in ternary Cu - 33.5% Zn - 4.5% Sn alloy. x250. Etched in acid ferric chloride. Figure 31(d) shows proximity of fracture surface. - 108 - s t r e s s c o r r o s i o n f r a c t u r e surface e x h i b i t i n g r i n g p a t t e r n s was given a 10-second e l e c t r o p o l i s h , and the ring patterns were destroyed. However, the mechanically f r a c t u r e d surface e x h i b i t i n g e l e c t r o n d i f f r a c t i o n r i n g p a t t e r n s was given a s i m i l a r treatment w i t h the same r e s u l t , but martensite e t c h markings were s t i l l v i s i b l e below the f r a c t u r e surface. Thus, i t appeared that the e l e c t r o n d i f f r a c t i o n p a t t e r n s observed i n the case of the mechanically f r a c t u r e d specimens d i d not a r i s e from the coarse s c a l e martensite. I t must th e r e f o r e be concluded that the martensite on the mechanically and s t r e s s c o r r o s i o n f r a c t u r e surfaces must be on a very f i n e s c a l e . 3.3.3 Fractography Topographical features on the water f r a c t u r e d b i n a r y and ternary brasses were very s i m i l a r . Examples of such features observed i n the Cu-47.76% Zn, Cu-45.54% Zn and t e r n a r y a l l o y s are shown i n Fi g u r e s 32, 33 and 34, r e s p e c t i v e l y . The features are q u i t e complex and i n general contained bundles of long p a r a l l e l s t e p s , o c c a s i o n a l l y i n t e r - sected by others. Very o f t e n these bundles themselves contained steps y i e l d i n g complex zig-zag arrangements. In the case of the ammoniacal copper sulphate f r a c t u r e d 47.76% Zn b i n a r y b r a s s , the s t r e s s c o r r o s i o n f r a c t u r e surface features a l s o e x h i b i t e d a p a r a l l e l stepped p a t t e r n as shown i n Figure 35. The f i n e s c a l e features on the s t r e s s c o r r o s i o n f r a c t u r e d surfaces of the b i n a r y and t e r n a r y a l l o y s were very s i m i l a r . Representative examples of the f i n e s c a l e features are e x h i b i t e d i n Figures 33(cc), 33(dd), 34(cc) and 34(d). Apparent cleavage c h a r a c t e r i s t i c s i d e n t i c a l to those noted on the l a r g e r s c a l e (Figure 34(cc)) and r i v e r - 109 - - 110 - b Figures33(a) and 33(b). Mating stress corrosion fracture surfaces of Cu-45.54% Zn i n water. x225. - I l l - d Figures 33(c)(cc)(d)(dd). General topographical features of Cu - 45.54% Zn a l l o y fractured i n water. (c) x225, showing steps. (cc) Showing area s i m i l a r to c i r c l e d area on (c). x2000. (d) x225, i n t e r s e c t i n g steps perpendicular to machined notch, (dd) De t a i l s of area s i m i l a r to c i r c l e d area on (d) x2000. - 112 - Figures 34(a) and 34(b). Mating stress corrosion fracture surfaces of Cu — 33.5/fen - 4.5%Sn alloy fractured in water. x225. - 113 - cc d Figure 34(c)(cc)(d). Topographical features on stress corrosion fracture surface of the Cu - 33.5% Zn - 4.5% Sn alloy/water. (c) x225. (cc) An area similar to circled area. x2000 (d) An area similar to circled area. x2000. - 114 - Figure 35. Matching fracture surface topographical features of Cu/47.76% Zn alloy S.C.C. in Mattssons solution. x225. - 115 - type patterns (Figures 33(dd) and 34(d) were exhibited on the surfaces. Moreover, the surfaces were not entirely smooth, as for example noted in Figures 33(cc), 34(dd) and 34(d) suggesting that some dissolution had occurred. The actual stress corrosion fracture surfaces were found to match up in a l l alloys as evidenced by Figures 32(a), 32(b), 33(a), 33(b), 34(a), 34(b) and 35. This suggested that the microstructural features associated with the topographical features observed were present prior to the passage of the crack. The actual interpretation of the features however is not clear. The large scale steps often l i e normal to the machined notch, as shown for example in Figure 33(d). It is therefore unlikely that the steps are related to discontinuous crack arrest points. It is possible that the markings correspond to {110} cleavage steps (102) or {110} dissolution s l i p planes, but i t is unlikely that these processes alone would produce the martensite. It is also possible that the features are related to martensite. The only guideline for interpretation of the fractographic features is that of Ahlers and Pops (99). They fractured a Cu-48 At % Zn single crystal in a i r , observed features similar to those presented in Figure 34(cc), and concluded that the markings were a consequence of the martensite transformation. Features similar to those in Figure 34(cc) were also observed in a mercury embrittled Cu-50wt% Zn single crystal (3). In this work (3), however, no interpretation of the features was made, but the macroscopic crack plane was shown to be close to the (110) plane, which is close to the martensite habit plane in these materials (96-101). - 116 - One question remains to be answered and that i s whether or not the normal fracture mode of the martensite can account for the observed stress corrosion fracture surface topographical features. However, the mechanically fractured ternary and binary alloy surfaces containing the same phases exhibit cusps characteristic of ductile fracture as evidenced by Figure 36. Therefore, some other process must occur in stress corrosion cracking to account for the observed features, e.g., dissolution. It was unfortunate that phase transformation products adjacent to the crack tip or below the stress corrosion fracture surface of the rod specimens were on too fine a scale for detection by metallographic etching techniques because this experimental evidence would have provided additional support to the view that the crack path in g-brass i s determined by the transformation product. However, i t was found that such evidence could be provided by subjecting strip ternary g-brass specimens to U-bend stress corrosion tests in ammoniacal copper sulphate solution. Stress induced martensite reverts to the parent phase except i n severely p l a s t i c a l l y deformed regions. The severe plastic deformation in the U-bend type of specimen affects a larger volume than in the rod specimens, ensuring that the strain induced martensite be retained over a larger volume, and therefore easily detectable by optical metallographic techniques. In the rod specimens, the region of severe deformation was localized to the plane of the notch, and was of insufficient volume to allow optical metallographic detection, but sufficient for detection by electron diffraction. It is also noted - 117 - a . b Figure 36. Topographical details of mechanically fractured surfaces, showing dimples characteristic of void coalescence. (a) Cu - (b) Cu - (c) Cu - 45.54% x2000 33.5% Zn - 4.5% Sn x2000 47.76% x225. - 118 - that the comparison of the d i f f r a c t i o n data of the brass f i l i n g s w i t h that obtained from the f r a c t u r e surfaces i s j u s t i f i e d , because i n each case the brasses were subjected to severe p l a s t i c deformation. The r e s u l t s of the t e s t s performed on a U-bend S.C.C. specimen of Cu-33.5% Zn-4.4% Sn are presented below. D e t a i l s of i n t e r p l a n a r spacings and r e l a t i v e v i s u a l i n t e n s i t i e s corresponding to the e l e c t r o n d i f f r a c t i o n r i n g patterns taken from the s t r e s s c o r r o s i o n and mechanically f r a c t u r e d surfaces are shown i n Table XIX. Comparison of the t a b u l a t e d values suggests t h a t , w i t h i n the experimental accuracy of the technique, the same phase i s present on both s u r f a c e s . Moreover, i t appears that t h i s phase i s the same as that present on the rod specimen s t r e s s c o r r o s i o n f r a c t u r e s u r f a c e s . F r a c t o g r a p h i c examination r e v e a l s that the macroscopic t o p o g r a p h i c a l f e a t u r e s on the s t r e s s c o r r o s i o n f r a c t u r e d s u r f a c e are d i f f e r e n t from those observed on the rod specimens S.C. f r a c t u r e surfaces s i n c e they show a cross-hatch p a t t e r n as shown i n Figure 37. Using the technique of Almond et a l . (62), these f e a t u r e s can be d i r e c t l y r e l a t e d to the etched m a r t e n s i t e phase at the i n t e r s e c t i o n of the etched and f r a c t u r e s u r f a c e s . See Figure 37. Thus the observed e l e c t r o n d i f f r a c t i o n p a t t e r n must be of the martensite phase. S i m i l a r i t i e s would be a n t i c i p a t e d i f c r a c k i n g occurred along the i n t e r f a c e of the two phases. In f a c t , m e t a l l o g r a p h i c examination shows a crack to be propagating along such a path. See Fi g u r e 38. The to p o g r a p h i c a l f e a t u r e s on the two s t r e s s c o r r o s i o n f r a c t u r e surfaces are complementary (Figure 39) i n d i c a t i n g the martensite was formed p r i o r to f r a c t u r e . The normal f r a c t u r e behaviour of the martensite i s d u c t i l e . - 119 - TABLE XIX - E l e c t r o n D i f f r a c t i o n Observations f o r Ternary S t r i p A l l o y (Cu-33. 5£'Zn-4. 4%'Sn) (Two r e p r e s e n t a t i v e patterns from each f r a c t u r e surface) Mechanical F a i l u r e S t r e s s Corrosion F a i l u r e Ammoniacal Copper Sulphate d l I V <> d A ^"V d A d K *v 3.22 vw 3.22 w 2.75 vw 2.75 vw 2.45 vw 2.40 vw 2.02 vs 2.00 vs 2.02-2.04 vs 2.04 m 1.49 vs 1.49 vs 1.49 w/m 1.44 W W 1.41 vvw 1.41 vw 1.27 vvw 1.23 s 1.23 mw 1.13 vw 1.13 W W 1.14 1.05 vw , m 1.15 1.04 mw W W 0.99 vvw 0.99 vw 0.99 m 0.99 vw 0.875 vw 0.83 mw 0.805 W W 0.805 vw 0.79 0.745 0.705 0.69 0.645 0.60 mw vw m m w . w - I = r e l a t i v e v i s u a l i n t e n s i t y v vs = very s t r o n g , s ^ s t r o n g , m = medium, w/m = weak medium w = weak, vw = very weak, ww = very very weak. - 120 - p i S.C.C. FRACTURE SURFACE POLISHED AND ETCHED SURFACE Figure 37. Topographical features in S.C.C. Cu-33.5% Zn-4.4% Sn alloy in ammoniacal copper sulphate. Features are same as etch markings, and can be traced round the intersecting line. Etched polished surface in Hull/Garwood^7-*) etch. Scanning electron micrograph. x300. Figure 38. Martensite adjacent to crack in Cu-33.5% Zn-4.4% Sn alloy. x225. Scanning electron micrograph. Etched in Hull/Garwood etch(75). - 121 - Figure 39. Matching topographical features i n Cu-33.5% Zn - 4.4% Sn a l l o y subjected to S.C.C. i n ammoniacal copper sulphate s o l u t i o n . x225. faces of Cu - 33.5% Zn - 4.4% Sn a l l o y . x225. Scanning e l e c t r o n micrograph. - 122 - See Figure 40. Thus deformation at the crack tip alone cannot account for the observed features. Therefore some other process must be involved. It is possible that dissolution occurs because of the similarity between the topographical features and etch patterns. The observations are consistent with the suggestion that stress corrosion cracking of ternary 3-brass i s dictated by the presence of a phase transformation product and support the studies conducted with the rod specimens of the binary and ternary alloys. The reason for the cross-hatch morphology of the martensite is not obvious - i t might be due to the high plastic strain involved in deforming the U-bend specimens. However, this is not the complete answer since the mechanically deformed rod specimen exhibited a different microscopic martensite morphology. 3.3.4 Discussion A case can be made for suggesting that the transgranular crack path in these alloys is dictated by a phase transformation product. The mechanism of formation of the martensite phase is not clear. It is possible that i t is related solely to the deformation induced phase transformation at the crack tip since the same phases are produced by mechanical fracture. However, the conditions at the tip of a stress corrosion crack are not quite the same as in a ductile mechanically deforming specimen. The formation of martensite may be influenced by a hydrostatic component of strain which is present at the crack tip. Calculations (63) show that the strains associated with the b.c.c. •> orthorhombic transformation in these materials are: - 123 - 5.9% extension along [100] direction p 1.4% contraction along [Oil] direction p 1.1% contraction along [111] direction p Details of the mechanism of crack propagation are not clear. The normal fracture behaviour (ductile) cannot account for the observed fractography. It i s possible that cleavage, or dissolution along s l i p traces of the orthorhombic structure occurs, but l i t t l e is known of such details. Unlike the case of the stainless steels there i s no reason for suggesting that hydrogen is cathodically produced and promotes cleavage. A similarity between the topography and the etch patterns, shown in Figure 37, suggests that dissolution occurs along the interface between the martensite and parent phase. The driving force for the reaction may arise from the potential difference between the two phases. • It has been reported that the habit plane of the martensite in an alloy of Cu-48 - At % Zn, close to the composition of the 45.54 wt % Zn alloy used in the current work, is (110) (99). Other workers (96,97) p have suggested that for a wide range of composition of 3-brasses the habit plane is (2,11,12) , very close to (110) . Cracking along the p P interface of martensite and parent phase is consistent with the observed crack traces. Ahlers and Pops (99), and others (75,96,98) have noted that the martensite i n 3-brasses takes the form of bands or sheets, the traces of which lie approximately parallel to each other. In a p a r t i a l l y transformed region ahead of a crack tip, cracking along, the interface of - 124 - last martensite band and the parent g-phase w i l l lead to an observed cracking along {110}o planes. In a fu l l y transformed region, cracking p can occur at the interface of an intermediate martensite sheet and the g-phase leading to cracking on {110} planes, or at the intersection g of parallel bands on {110} planes. p Parallel bands of martensite on {110} planes may intersect others p at 90° and cracking might occur along these <100> directions on a P microscopic scale. The crack path could then deflect to other {110} p planes to make up a {110} macroscopic plane containing striations or P steps, as i s observed in the current work, and previously on a mercury embrittled g-brass fracture surface (3). If martensite bands possess similar morphologies on a small scale as observed on the large scale, then the bands are not necessarily in p a r a l l e l formation, e.g., see Figures 30(b) and 31(a). Thus dissolution along the ends of martensite plates may account for the apparent river patterns exhibited in Figures 33(dd) and 34(d). The mechanism of intergranular cracking is not clear. Cracking could occur by the tarnish rupture process (1,4,7,8). This requires sl i p to occur at the grain boundary to open up the crack to allow access of fresh environment; dissolution along s l i p planes may account for the crystallographic transgranular side branches observed. It is also possible that fine scale martensite forms at the grain boundary and this may provide a high energy crack path at the grain boundary/ martensite interface. Dissolution along the interface of the martensite plates and the g-phase would account for the crystallographic nature of the side branching. - 125 - The crack paths of the g-brasses examined do not appear to vary greatly with the alloy composition or the environment used. It does appear, however, that the rate of cracking i s sensitive to several other factors. The rate of cracking is faster for 1) the lower zinc alloy content or ternary alloy in a particular environment (water), and 2) a particular alloy in the ammoniacal environment. The former may be due to the kinetics of the martensite transformation, although there may be a conjoint change in any chemical dissolution process which i s occurring. The second may be because the adsorption, dissolution, and transport kinetics are different in the two solutions. The chemistry of dissolution of g-brasses in these solutions has not been discussed in detail. In summary, a case can be made for the association of stress corrosion cracking of g-brasses with the martensite transformation. It is not necessarily the only process occurring, as, for example, dissolution may be associated with dislocation configurations or be influenced by alloy chemistry. However, when the martensite trans- formation does occur i t appears that propagation of the crack through this phase is kinetically more favorable and that dislocation configura- tions become of secondary importance. - 126 - i 3.4 Co r r o s i o n Product Films Oxides were detected on the a c t u a l undisturbed s t r e s s c o r r o s i o n f r a c t u r e surfaces of specimens which f a i l e d i n a r e l a t i v e l y long p e r i o d , and m e t a l l i c phases only on specimens r a p i d l y cracked. See Table XX. (The MgCl^/FeCl^ environment not h i t h e r t o r e f e r r e d was employed i n t e s t s to be discussed below). I r r e s p e c t i v e of the r o l e of the oxid e , i t i s considered that i t s i d e n t i f i c a t i o n would c o n t r i b u t e to the g r a d u a l l y growing bank of data on c o r r o s i o n product f i l m s formed i n s t r e s s c o r r o s i o n c r a c k i n g . I t i s w i t h t h i s j u s t i f i c a t i o n that the f o l l o w i n g r e s u l t s are repo r t e d . 3.4.1 Oxide Formation on 8-brasses Table XIV i n d i c a t e s that a f i l m c o n s i s t i n g of C^O i s formed on the s t r e s s c o r r o s i o n f r a c t u r e surfaces of the Cu-47.76% Zn a l l o y f r a c t u r e d i n ammoniacal copper sulphate s o l u t i o n and i n water. The f r a c t u r e surfaces e x h i b i t e d a "y e l l o w " appearance, and could correspond o to a f i l m of approximately 1000 A (108). The same c o l o u r a t i o n of Cu^O had been noted on the s t r e s s c o r r o s i o n f r a c t u r e s u r f a c e s of g-brass deformed i n ammoniacal copper sulphate s o l u t i o n (47). No Cu^O was detected on the g-brass specimens of lower z i n c content. The p o s s i b i l i t y must not be discounted that the e l e c t r o n d i f f r a c t i o n technique was not s u f f i c i e n t l y s e n s i t i v e . C^O was not detected on a Cu-47.76% Zn specimen mechanically f r a c t u r e d i n water and kept immersed f o r 17 hours, the same time p e r i o d r e q u i r e d f o r S.C.C. of a s i m i l a r specimen. This suggest t h a t the C^O was formed as a consequence of s t r e s s c o r r o s i o n c r a c k i n g , and not as - 127 - TABLE XX - Phases Detected by Electron Diffraction on the S.C.C. Surfaces S.C.C. System Time to F a i l u r e Phase Detected 304L/MgCl 2 310/MgCl 2 12-17 hrs 92-100 hrs a' ( t r a n s f o r m a t i o n product) oxide Cu-47.76/water Cu-45.54/water Ternary/water 17-18 hrs 5-11 mins 2-10 mins oxide t r a n s f o r m a t i o n product t r a n s f o r m a t i o n product Cu-47.76/Ammoniacal Environment Ternary/Ammoniacal Environment ( s t r i p specimen) 3-5 mins 2 mins oxide t r a n s f o r m a t i o n product 304L/MgCl 2/FeCl 3 310/MgCl 2/FeCl 3 h-3 hrs 7-12 hrs oxide oxide - 128 - a result of "free" corrosion. A g-brass specimen mechanically fractured in ammoniacal copper sulphate solution and kept immersed for 5 minutes prior to removal from the electrolyte, revealed a blue-black colouration of Cu^O. This corresponds to a thickness of several microns (1,34). It appeared, therefore, that a thinner layer of Cu^O was formed as a consequence of S.C.C. Similar observations have been made on a-brass (3,4). 3.4.2 Corrosion Product Formation on 304L Steel No corrosion product film was detected by electron diffraction on the actual stress corrosion fracture surface of the 304L specimen removed f rom the boiling MgCl2 environment immediately after completion of cracking. The surface exhibited a metallic lustre. However, electron diffraction rings consistent with the presence of a spinel structure were obtained from a stress corrosion fracture surface subjected to a further 18 hour immersion in the corrosive environment immediately following S.C.C. The interplanar spacings and relative visual inten- si t i e s corresponding t o the ring pattern are shown in Table XXI and may be compared to the estimates of the theoretical relative intensities for Fe o0. and y-Fe„0_ detailed in the Appendix. Both Fe.O, and 3 4 ' 2 3 1 v 34 y-Fe^O^ possess a spinel structure. The observations of a spinel agrees with-the result of Nielson (22). However, insufficient oxide was formed to allow a chemical analysis to be conducted, and i t was not possible to confirm Nielson's analysis (22). Baker at a l . (21) demonstrated by stress corrosion cracking U-bend tests that additions of f e r r i c chloride to the magnesium chloride - 129 - environment r a p i d l y a c c e l e r a t e s f a i l u r e and simultaneously produces a voluminous c o r r o s i o n product f i l m . They v i s u a l l y observed a c o r r o s i o n product f i l m on the a c t u a l s t r e s s c o r r o s i o n f r a c t u r e s u r f a c e but c a r r i e d out an X-ray d i f f r a c t i o n a n a l y s i s only on that product formed on the o r i g i n a l s u rface present p r i o r to S.C.C. The f i l m possessed a s p i n e l s t r u c t u r e but Baker et a l . (21) d i d not perform a chemical a n a l y s i s on the product. The curr e n t work examines the p r o p e r t i e s of the f i l m i n d e t a i l because more product i s a v a i l a b l e to d i f f r a c t i o n and chemical a n a l y s i s than i s formed i n the p l a i n M gCl 2 environment. Notched t e n s i l e l o a d - r e l a x a t i o n S.C.C. t e s t s were performed i n the Baker e t a l . (21) environment which had a b o i l i n g p o i n t of 125°C and composition of 8% F e C l 3 / 6 7 % MgCl 2/25% water. I t was determined that (a) c r a c k i n g was predominantly t r a n s g r a n u l a r , (b) a voluminous dark brown f i l m formed on the f r a c t u r e s u r f a c e and (c) time to f a i l u r e was reduced by an order of magnitude. The t h i r d e f f e c t of the f e r r i c i o n s was remarkable when i t i s pointed out that lowering of the S.C.C. t e s t temperature of a p l a i n M g C l 2 environment from 154°C to 125°C normally i n c r e a s e s time to f a i l u r e by approximately an order of magnitude (27, 109). The f i l m on the S.C.C. f r a c t u r e s u r f a c e obscured any t o p o g r a p h i c a l fe a t u r e s of the und e r l y i n g metal. E l e c t r o n d i f f r a c t i o n s t u d i e s con- ducted on the a c t u a l S.C.C. f r a c t u r e s urface were c o n s i s t e n t w i t h the presence of a s p i n e l of l a t t i c e parameter s i m i l a r to Fe^O^. The corresponding i n t e r p l a n a r spacings are shown i n Table XXI. S u f f i c i e n t oxide was removed from the S.C.C. f r a c t u r e surface to permit s e v e r a l other t e s t s to be made. - 130 - TABLE XXI - E l e c t r o n D i f f r a c t i o n Observations of Oxides Formed on 304L S t e e l Stress Corrosion 154°C MgC^ plus 18 hrs post immersion i n MgC^ Stress Corrosion F e C l 3 / M g C l 2 125°C o d A I V o d A I V 4.7 vw 4.70 vw 3.6 W W 3.73 W W 2.98 w 3.02 m 2.52 vs 2.52 vs 2.10 m 2.40 W W 1.63 m 2.08 m 1.49 m 1.60 m 1.43 W W 1.49 m 1.28 vw 1.28 W 1.23 W W 1.09 w I = r e l a t i v e v i s u a l i n t e n s i t y vs = very s t r o n g , m = medium, w = weak, vw = very weak, ww - very weak TABLE XXII - Chemical A n a l y s i s of Corro s i o n Product by Weight % Element 304L S t e e l 310 S t e e l Fe 52 45.00 N i 3.46 5.00 Cr 11.0 12.80 Mg 5.6 4.8 - 131 - The powder was very porous as evidenced by the absorbed e l e c t r o n image scanning micrograph as shown i n F i g u r e 41(a). The c o r r o s i o n product could e a s i l y be p i c k e d up by a magnet. ^ 3 ^ 4 -*-s s t r o n g l y magnetic and y-Fe^O^ i s l e s s magnetic (111). Some mixed s p i n e l s , i . e . , s p i n e l s c o n t a i n i n g mixed c a t i o n s p e c i e s , e.g., FeCr^O^, are a l s o magnetic (111). To determine whether the c o r r o s i o n product was p u r e l y an i r o n oxide or a mixed s p i n e l , X-ray images were obtained u s i n g the JEOL Co. microprobe a n a l y z e r . The elements Fe, N i , Cr, 0, T i , Mn, Co, Mg, and C l were detected. X-ray images corresponding to the three major elements Fe, Cr and N i are d i s p l a y e d i n F i g u r e s 41(b)-41(d). The d i s t r i b u t i o n of the elements appeared to be uneven. This might be due to the topography of the c o r r o s i o n p r o d u c t A chemical a n a l y s i s confirmed the f i l m to be i r o n r i c h . The r e s u l t s are shown i n Table XXII. X-ray powder d i f f r a c t i o n p a t t e r n s contained l i n e s corresponding o to a s p i n e l s t r u c t u r e of l a t t i c e parameter 8.37 A, midway between the accepted values f o r Fe.j0^ and y-Te^O^ (110,111). ( I t i s now b e l i e v e d that HFe^Og i s a more c o r r e c t r e p r e s e n t a t i o n of -f-Fe^O^ - see the Appendix.) The values of the i n t e r p l a n a r spacings and i n t e n s i t i e s corresponding to the p a t t e r n are t a b u l a t e d i n Table X X I I I . Note that i t i s d i f f i c u l t to d i s t i n g u i s h by X-ray d i f f r a c t i o n between the s p i n e l I 1 1 1 1 Fe^O^ and s p i n e l s based on Fe^O^ w i t h N i and Cr c a t i o n s u b s t i t u t i o n s because of the s i m i l a r i t y of X-ray s c a t t e r i n g f a c t o r s of Fe, N i and Cr and i n t e r p l a n a r spacings of the s p i n e l s . The X-ray d i f f r a c t i o n s t u d i e s were c o n s i s t e n t w i t h the presence of an i r o n r i c h s p i n e l . There were however, a few e x t r a l i n e s i n the X-ray powder p a t t e r n . - 132 - Figure 41. X-ray images of Fe (b), Cr (c), and Ni (d) in corrosion product on 304L steel . (a) shows porous nature of corrosion product x255. Absorbed electron image. - 133 - TABLE XXIII - Analysis of X-ray D i f f r a c t i o n Patterns of Corrosion Product 304L Corrosion Product 310 Corrosion Product d A d A .1 V 7.9-7.44 m (band) 7.84-7.39 w (band) M 4.835 m M 4.842 vw 3.646 vw 3.684 vw 3.456 w 3.337 vw M 2.962 m M 2.958 m 2.679 WW 2.673 w M 2.527 vs M 2.525 vs M 2.417 WW M 2.228 vw 2.173 vw M 2.090 mw M 2.11-2.08 w (band) M 1.710 w M 1.681 w 1.676 vw 1.644 w 1.646 vw M 1.612 ms M 1.611 m M 1.481 s I M 1.481 w Patterns taken with Co tube with Fe f i l t e r I = r e l a t i v e v i s u a l i n t e n s i t y v vs = very strong, s = strong, m = medium, raw = medium weak, w = weak, vw = very weak M = indicates l i n e s corresponding to magnetite. - 134 - The experimental evidence presented above suggests that the c o r r o s i o n product was an i r o n based s p i n e l together w i t h the presence of t r a c e compounds. Confirmation of Iron-based S g i n e l (1) I n f r a - r e d A n a l y s i s An I.R. tr a n s m i s s i o n spectrum of the c o r r o s i o n product i n a KI p e l l e t was obtained u s i n g a P e r k i n Elmer Model 521 Infra-Red Spectro- photometer. The spectrum, shown i n Figure 42(a), was compared w i t h the s p e c t r a of y-Te^O^ and Fe^O^ determined by P o l i n g (112), shown i n Figure 42(b). Note that P o l i n g p l o t t e d the absorbance, r a t h e r than tr a n s m i t t a n c e , on the o r d i n a t e . The spectrum ofthe c o r r o s i o n product d i d not r e a d i l y resemble that of Y -Fe20.j but c l o s e l y resembled that of Fe^O^. The correspondence of the major peak was not p r e c i s e . This might be due to the presence of Cr and N i c a t i o n s i n the magnetite l a t t i c e (113). F u r t h e r peaks observed a t 1620 cm ^ and 3400 cm ^, not shown i n Figure 42(a), suggested that absorbed water was present i n the c o r r o s i o n product. An i n f r a - r e d spectrum obtained by t r a n s m i s s i o n through a suspension of the c o r r o s i o n product i n N u j o l M u l l contained s i m i l a r peaks con f i r m i n g t h a t absorbed water was present i n the c o r r o s i o n product. (2) D i f f e r e n t i a l Thermal A n a l y s i s (D.T.A.) D.T.A. of the c o r r o s i o n product was performed on the Dupont 900 D i f f e r e n t i a l Thermal An a l y s e r . A^O^ was used as the standard. The r e s u l t of the a n a l y s i s i s shown i n Figure 43(b). The curve resembled ure 42 (a) Infra-red spectrum of corrosion product from 304L austenitic steel , (b) Infra-red spectra of y~F e2 u3 a n ^ F e 3 u 4 ^ " ^ ^ ' aoo 400 too loo tooo I I I I I TCMPCRATURC Figure 43. DTA curves of (a) magnetite, (b) c o r r o s i o n product on 304L s t e e l , (c) c o r r o s i o n product on 310 s t e e l , (d) magnesium o x y c h l o r i d e s » H 5 ) m - 137 - that f o r magnetite shown i n Fi g u r e 43(a) which was ex p e r i m e n t a l l y determined i n the current work. The D.T.A. curve f o r the c o r r o s i o n product a l s o resembled p u b l i s h e d curves (143) of the D.T.A. of magnetite. There were, however, d e v i a t i o n s between the D.T.A. of the c o r r o s i o n product and that of magnetite. C3) Thermal G r a v i m e t r i c A n a l y s i s (T.G.A.) T.G.A. of the c o r r o s i o n product was performed over the range 25-900°C on a Dupont950 Thermal G r a v i m e t r i c Analyser. T.G.A. i n d i c a t e d a gradual l o s s i n weight over the temperature range 25°C-600°C, reaching a maximum of 10% at 600°C. There was no f u r t h e r weight l o s s i n the temperature range 600-900°C. Loss of absorbed water might account f o r p a r t of the t o t a l weight l o s s . The decomposition of t r a c e compounds might account f o r the remainder of the weight l o s s . This i s discussed below. This t e s t does not appear to a i d d i s t i n c t i o n between the p o s s i b l e i r o n oxides contained i n the c o r r o s i o n product. Trace Compounds i n Corro s i o n Product The e x t r a l i n e s i n the X-ray powder p a t t e r n and the d e v i a t i o n s i n the D.T.A. suggested the presence of tr a c e compounds i n the c o r r o s i o n product. Chemical c o n s i d e r a t i o n s suggest that magnesium o x y c h l o r i d e s were probable c o r r o s i o n products s i n c e these could form r e a d i l y by the i n t e r a c t i o n of MgCOH)^ and MgCl^ at elevat e d temperatures (111). The OH' ions could a r i s e f r o c v t h e oxygen r e d u c t i o n r e a c t i o n at the cathode, or by h y d r o l y s i s . I t i s considered that the c o r r o s i o n product contained - 138 - a mixture of o x y c h l o r i d e s s i n c e d i f f e r e n t o x y c h l o r i d e s could form at d i f f e r e n t temperatures (114) or on ageing at p a r t i c u l a r temperatures (114). Hence s e v e r a l species of magnesium o x y c h l o r i d e form on c o o l i n g from 125°C and/or on ageing the cooled product at room temperature. These i n s o l u b l e magnesium products are c o n s i s t e n t w i t h the e x t r a l i n e s observed i n the X-ray d i f f r a c t i o n p a t t e r n of the c o r r o s i o n product. D e t a i l e d X-ray data f o r the many forms of magnesium oxy- c h l o r i d e have been determined by Cole and h i s co-workers (114,115). The t o t a l weight l o s s a s s o c i a t e d w i t h h e a t i n g the c o r r o s i o n product i s due to (a) the l o s s of absorbed water i n the c o r r o s i o n product and (b) the decomposition of the o x y c h l o r i d e s . In the case of decomposition of the o x y c h l o r i d e s , adsorbed water i s e x p e l l e d below 200°C (114). F u r t h e r decomposition of the anhydrous product i s considered to occur by the f o l l o w i n g r e a c t i o n s over the range 350—600°C (114). n Mg(OH) 2MgCl 2 I N MgO + (n - 1)H 20 + H 20 + Mg C l 2 MgO + 2HC1 where n i s an i n t e g e r of value 2, 3, 5, or 9 depending on the temperature of formation of the o x y c h l o r i d e . The gradual weight l o s s observed i n the TGA i s probably i n d i c a t i v e of l o s s of absorbed water from both the s p i n e l and o x y c h l o r i d e s below - 139 - 200°C, and of HC1 and water from the decomposition of the anhydrous o x y c h l o r i d e s above 350°C. Decomposition of the ox y c h l o r i d e s i s complete around 600°C, the temperature on a t t a i n i n g maximum weight l o s s from the product. From c o n s i d e r a t i o n s of the chemical a n a l y s i s , the p r o p o r t i o n by weight of o x y c h l o r i d e s i n the c o r r o s i o n product must be q u i t e s m a l l . I f t h i s percentage i s approximately 5%, and the oxy- c h l o r i d e s l o s e 60% of t h e i r weight on decomposition (114), the t o t a l weight l o s t from the c o r r o s i o n product from t h i s source i s around 3%. Thus, the remainder of the weight l o s t i n the T.G.A. corresponds to removal of adsorbed water from the product. Moreover, i f the p r o p o r t i o n by weight of the ox y c h l o r i d e s i n the c o r r o s i o n product i s approximately 5%, the D.T.A. of the c o r r o s i o n product should resemble that of the major c o n s t i t u e n t provided that the minor c o n s t i t u e n t s do not e x h i b i t powerful peaks. Such a resemblance was observed. I t i s p o s s i b l e that the decomposition of the o x y c h l o r i d e s , represented i n Figure 42(d) (114,115), does not exert a strong i n f l u e n c e on the D.T.A. of the c o r r o s i o n product, s i n c e the st r o n g e s t endothermic peak of the o x y c h l o r i d e s (~^250°C) c o i n c i d e s w i t h the strong endothermic region of the magnetite. I t i s a n t i c i p a t e d that the o x y c h l o r i d e s w i l l decompose by he a t i n g e f f e c t s and vacuo generated by the e l e c t r o n beam during the e l e c t r o n d i f f r a c t i o n s t u d i e s and form a very s m a l l amount of residue MgO i n the c o r r o s i o n product. The i n t e r p l a n a r spacing of the stron g e s t MgO l i n e o (d^QQ = 2.106 A) i s very c l o s e to that of a strong s p i n e l l i n e (̂ Q̂Q = o 2.096 A). This may account f o r the observations of the complex X-ray d i f f r a c t i o n p a t t e r n and the r e l a t i v e l y s i m p l e r s p i n e l e l e c t r o n d i f f r a c t i o n - 140 - pattern. These conclusions may very well account for Nielson's (22) observations. He extracted corrosion product from a crack in y stainless steel subjected to S .C.C. i n MgC^ solution. He examined i t by electron d i f f r a c t i o n , and noted that a diffuse ring pattern gave way gradually to well defined Debye rings consistent with the presence of a spinel . The corrosion product he examined certainly contained the s p i n e l , but i t would also almost certainly contain a very large proportion of "entrapped stress corrosion solution (MgCl^)• This would have s o l i d i f i e d at room temperature and would be d i f f i c u l t to remove by washing. In addition, the sol id would contain the oxychlorides precipitated at elevated temperatures and during cooling. Thus, on electron d i f f r a c t i o n examination, his i n i t i a l observation of diffuse rings would correspond to d i f f r a c t i o n primarily from the magnesium compounds. On further heating i n the electron beam the decomposition product MgO would combine with the spinel already present, to form a more complex spinel . This would account for the well defined di f f rac t ion pattern. In summary, the corrosion product formed on the fracture surfaces of the steels subjected to S .C .C . in the FeCl^/MgCl^ environment appears to consist pr inc ipal ly of an iron r ich spinel based on the magnetite structure. 3.4.3 Corrosion Product Formation on 310 Steel A corrosion product f i lm formed very readily on the stress corrosion fracture surface of the 310 steel subjected.to S . C . C . i n 42% - 141 - MgC^ s o l u t i o n . The f r a c t u r e s urface of the s t e e l e x h i b i t e d a matte grey appearance even a f t e r an immediate removal f o l l o w i n g completion of c r a c k i n g . An e l e c t r o n d i f f r a c t i o n p a t t e r n from t h i s s u r f a c e revealed r i n g s c o n s i s t e n t w i t h the presence of an i r o n oxide. A f t e r a f u r t h e r 18 hour immersion i n the environment f o l l o w i n g S.C.C, a s p i n e l phase formed as i n d i c a t e d by e l e c t r o n d i f f r a c t i o n . D e t a i l s of the i n t e r p l a n a r spacings are presented i n Table XXIV. There was, however, an i n s u f f i c i e n t amount of the c o r r o s i o n product to permit f u r t h e r t e s t s to be conducted. No p u b l i s h e d chemical analyses of the c o r r o s i o n product formed on 310 a u s t e n i t i c s t e e l i n t h i s environment are a v a i l a b l e . 310 grade s t e e l s were a l s o t e s t e d i n the Baker et a l . (21) " r a p i d c r a c k i n g " environment. Observations were very s i m i l a r to those made i n the case of 304L s t e e l . Cracking was predominantly t r a n s - g r a n u l a r . Time to f a i l u r e was reduced by an order of magnitude compared w i t h f a i l u r e i n MgC^. A dark brown c o r r o s i o n product was formed on the a c t u a l f r a c t u r e s u r f a c e . This masked the t o p o g r a p h i c a l f e a t u r e s of the u n d e r l y i n g metal. The p r o p e r t i e s of the c o r r o s i o n product were very s i m i l a r to those of the product formed on the 304L s t e e l . The r e s u l t s of t e s t s performed on the c o r r o s i o n product are summarized below: (1) The c o r r o s i o n product was porous. See Figure 44(a). (2) I t could e a s i l y be pi c k e d up by a magnet. (3) The e l e c t r o n d i f f r a c t i o n p a t t e r n s corresponded to those e x h i b i t e d by a s p i n e l s t r u c t u r e . I n t e r p l a n a r spacings are l i s t e d i n Table XXIV. - 142 - TABLE XXIV - E l e c t r o n D i f f r a c t i o n Observations of Corrosion Products Eormed on 310 S t e e l S t r e s s C o r r o s i o n + 154°C St r e s s C o r r o s i o n + 154°C MgCl p l u s 18 hrs f u r t h e r F e C l 3 / M g C l 2 immersion i n M g C l 2 e d A I d A I V V 4.78 w 4.84 mw 3.65 W W 3.73 vw 3.00 m 2.96 mv 2.74 W W 2.66 vw 2.52 vs 2.51 vs 2.08 m 2.17 vw 1.85 W W 2.08 s 1.68 w 1.86 vw 1.59 w 1.70 vw 1.48 w 1.61 m 1.42 vw 1.50 m 1.28 vw 1.44 vw 1.08 W W 1.29 w 1.09 w 1.04 vw 0.96 w vs = r e l a t i v e v i s u a l i n t e n s i t y = very s t r o n g , s = s t r o n g , m = medium, mw - medium weak, weak, vw = very weak, ww = very very weak. - 143 - Figure 44. Corrosion product formed on 310 steel . (a) absorbed electron image x225, (b) (c) (d) X-ray image of Fe, Cr, Ni respectively. - 144 - (4) The spinel was mixed as evidenced by the X-ray images presented in Figures 44(b-d), and by the chemical analysis presented in Table XXII. (5) X-ray powder diffraction analysis was consistent with the view that the product consisted of a spinel of lat t i c e parameter o 8.375 A plus a mixture of magnesium oxychlorides. Interplanar spacings corresponding to the X-ray patterns are exhibited in Table XXIII. (6) T.G.A. indicated a total weight loss of 7%. (7) The D.T.A. curves shown in Figure 43(c) did not readily resemble that of the corrosion product formed on the 304L steel. Careful examination, however, revealed that there were some sim i l a r i t i e s . In summary, the corrosion product formed during stress corrosion cracking of the 310 grade austenitic stainless steel in the FeCl^/MgC^ environment appeared to be similar to that formed on the 304L grade steel. It i s possible that the corrosion products on the 304L and 310 grade steel contained different quantities of magnesium oxychlorides thus exhibiting minor differences in T.G.A. and D.T.A. 3.4.4 Discussion The role of the corrosion product film in stress corrosion cracking i s not clear. Attention has been directed at the possible implications of oxide films in stress corrosion, as detailed in the introduction. Irrespective of i t s importance, the film i s indicative of a corrosion process whether this is the actual failure process or a secondary reaction. For instance, the oxide may deposit from solution whilst the main mechanism of stress corrosion cracking i s stress sorption cracking, or anodic dissolution. - 145 - The mechanism of S.C.C. has been discussed by many workers i n terms of an e l e c t r o c h e m i c a l process (2,4,21,27-32), w i t h d i s s o l u t i o n o c c u r r i n g at an anodic crack t i p and the crack s i d e s a c t i n g as the cathode of the c e l l . I t i s c l e a r that any oxide which dep o s i t s on the crack s i d e s w i l l i n f l u e n c e the p o t e n t i a l and k i n e t i c s at the e l e c t r o d e . For the e l e c t r o c h e m i c a l r e a c t i o n to be s u s t a i n e d , the oxide must be capable of conducting e l e c t r o n s . Therefore the p o s s i b i l i t y suggested by Kruger (116) that v a r i a t i o n s i n f i l m c o n d u c t i v i t y may a f f e c t S.C.C. ra t e s i s not out of the question. However, no support to t h i s suggestion has appeared i n the l i t e r a t u r e . The main purpose of the curr e n t s e c t i o n t h e r e f o r e i s to assess the observations made i n t h i s work and by Baker e t a l . (21) i n the l i g h t of t h i s suggestion. I t i s necessary to emphasize that the assumption i s made that the oxides as i d e n t i f i e d by e l e c t r o n d i f f r a c t i o n a c t u a l l y e x i s t i n s i t u d u r i ng S.C.C. 3.4.4.1 y - S t e e l s In the case of the s t a i n l e s s s t e e l s , the d i s c u s s i o n i s centered around the c o n d u c t i v i t y of the s p i n e l type s t r u c t u r e . Ion p o s i t i o n s i n the normal and in v e r s e s p i n e l s t r u c t u r e s are w e l l documented (117-121). The c o n d u c t i v i t y of the s p i n e l l a t t i c e i s i n f l u e n c e d by c a t i o n type and by the type of s i t e i t occupies i n the l a t t i c e (119,120). E l e c t r o n t r a n s f e r i s e a s i e s t i n the case of s i m i l a r c a t i o n species of d i f f e r e n t valency i n s i m i l a r types of s i t e (120). The c o n d u c t i v i t y of the normal s p i n e l s , i n g e n e r a l , i s low r e l a t i v e to that of the inv e r s e s p i n e l s . In the l a t t e r , e l e c t r o n t r a n s f e r i s r e l a t i v e l y easy - 146 - si n c e the oc t a h e d r a l s i t e s possess mixed v a l e n c i e s of the same c a t i o n s p e c i e s . However, i n the former, e l e c t r o n s must be t r a n s f e r r e d from c a t i o n s of one species and valency i n one type of s i t e to another c a t i o n species and valency i n the other type of s i t e . This i s a more d i f f i c u l t process. The c o n d u c t i v i t y of s p i n e l s can be changed by c a t i o n s u b s t i t u t i o n s . For example, i n the case of normal s p i n e l s i t i s a n t i c i p a t e d that t r i v a l e n t c a t i o n s u b s t i t u t i o n i n the t e t r a h e d r a l s i t e s , or d i v a l e n t c a t i o n s u b s t i t u t i o n i n the o c t a h e d r a l s i t e s , may i n c r e a s e the c o n d u c t i v i t y . The increase i n c o n d u c t i v i t y i s g r e a t e s t i f the c a t i o n a d d i t i o n s b r i n g about a s p i n e l s t r u c t u r e i n which a p a r t i c u l a r type of s i t e i s occupied by a mixed valency of the same c a t i o n s p e c i e s . S u f f i c i e n t s u b s t i t u t i o n b r i n g s about the i n v e r s e s p i n e l s t r u c t u r e . The i n v e r s e s p i n e l , magnetite, i s of p a r t i c u l a r i n t e r e s t because t h i s appeared to be c l o s e i n i d e n t i t y to the c o r r o s i o n product observed i n the FeCl^/MgC^ S.C.C. of the a u s t e n i t i c s t a i n l e s s s t e e l . T h e o r e t i c a l c o n s i d e r a t i o n s suggest that any c a t i o n s u b s t i t u t i o n would impair i t s c o n d u c t i v i t y . From energy c o n s i d e r a t i o n s (121) d i v a l e n t c a t i o n s i n general w i l l p r e f e r to enter the oc t a h e d r a l s i t e s r e p l a c i n g I | the Fe i o n s . T r i v a l e n t c a t i o n s a l s o p r e f e r the o c t a h e d r a l s i t e s , i n preference to the t e t r a h e d r a l s i t e s . The c o n d u c t i v i t y of Fe^O^ can i n f a c t be reduced s u b s t a n t i a l l y by c a t i o n s u b s t i t u t i o n (119,120). For example, the s p e c i f i c c o n d u c t i v i t y of Fe2A10^ was found to be 55 times lower than that of Fe~0. (119). - 147 - Subtle v a r i a t i o n s i n composition of the c o r r o s i o n product f i l m might a r i s e from a l l o y i n g a d d i t i o n s to the metal,thereby i n f l u e n c i n g S.C.C. l i f e . Small d i f f e r e n c e s i n composition of the c o r r o s i o n product f i l m s formed on the 304L and 310 a u s t e n i t i c s t e e l s subjected to S.C.C. i n the F e C l 3 / M g C l 2 environment were observed i n the current work. There were d i f f e r e n c e s between the S.C.C. l i v e s of the two • a l l o y s . This observation might be accounted f o r by d i f f e r e n c e s i n f i l m c o n d u c t i v i t y . However, t h i s i s only p a r t of the answer because the 304L y s t e e l undergoes a martensite tran s f o r m a t i o n . V a r i a t i o n s i n f i l m composition may a l s o a r i s e from s u b t l e changes i n the environment. I t i s expected that some cati o n s from the environment w i l l become incorpo r a t e d i n the Fe^O^ f i l m and reduce i t s c o n d u c t i v i t y , thereby i n c r e a s i n g S.C.C. l i f e of -the un d e r l y i n g metal. An increase of S.C.C. l i f e of 304L s t e e l was, i n f a c t , observed on ad d i t i o n s of small amounts of N i C l 2 , C r C l 3 and A1C1 3 to the FeCl 3/MgCl environment of the same bulk pH (21). A summary of these r e s u l t s i s presented i n Table XXV. 3.4.4.2 B-Brasses The f i l m formed on the f r a c t u r e surfaces of the 3-brasses i s p r i n c i p a l l y Cu^O. T h e o r e t i c a l c o n s i d e r a t i o n s (120,123) of the p-type semiconductor Cu 20 p r e d i c t that monovalent i o n s u b s t i t u t i o n s would not g r e a t l y i n f l u e n c e the c o n d u c t i v i t y of the Cu 20, whereas d i v a l e n t or t r i v a l e n t s u b s t i t u t i o n would decrease the c o n d u c t i v i t y of the Cu 20. There appears to be no published r e s u l t s on the e f f e c t of trac e I | q u a n t i t i e s of d i v a l e n t z i n c i o n s , Zn , i n the Cu„0 l a t t i c e on the - 148 - TABLE XXV - V a r i a t i o n of S.C.C. L i f e w i t h Environmental A d d i t i o n s Corrodent S o l u t i o n (125°C) I n i t i a l pH Cracking Time MgCl 2 + F e C l 3 + H 20 - A 1.2 0.5 h r s S o l u t i o n A + A1C1 3 1.0 1 hr S o l u t i o n A + C r C l 3 1.6 1 hr S o l u t i o n A + N i C l 2 1.4 4 h r s - 1A9 - conductivity of Cu^O (124). Some preliminary studies on the effect of trace additions of Be on the conductivity of Cu^O have been reported (125) but they are not relevant to the present work. It i s possible that the corrosion product films formed on the g-brasses contain traces of zinc oxide, since Cu^O films formed on ct-brasses subjected to S .C .C . i n ammoniacal copper sulphate solutions contain this oxide (14). This i s a n-type semiconductor i n which cation substitutions are expected to have opposite effects on conductivity to those observed i n the p-type semiconductor (14). Ternary cation substitutions i n the corrosion product w i l l therefore probably be negl igible . No systematic investigation of the effects of additions to the environment or alloying of g-brass on the corrosion product composition have been performed. Moreover no unambiguous studies on the influence of environment composition or alloying additions on stress corrosion cracking times of g-brasses have been conducted. Therefore any relationship between f i lm composition and cracking time must remain the subject of future work. In summary, the role of the oxide f i lm is not clear. It has been suggested that i f cracking progresses by electrochemical dissolution, the film may influence the S .C .C . rate by virtue of small changes i n i t s conductivity effected by changes in f i lm composition, resulting from environmental and metal-alloy additions. However, i t is certainly not the only factor influencing stress corrosion l i f e , as indicated e a r l i e r , the martensite transformation is also important. - 150 - 3.5 General C o n s i d e r a t i o n s The observed fe a t u r e s of s t r e s s c o r r o s i o n c r a c k i n g i n the systems examined are summarized below: (1) S.C.C. occurred predominantly by a t r a n s g r a n u l a r path. I n t e r - g r a n u l a r c r a c k i n g was not excluded. (2) The t r a n s g r a n u l a r crack path was determined by a s t r a i n induced t r a n s f o r m a t i o n product, i . e . , m a r t e n s i t e , twinning or f a u l t i n g . (3) The quasi-cleavage t o p o g r a p h i c a l f e a t u r e s observed on the a c t u a l s t r e s s c o r r o s i o n f r a c t u r e s u r f a c e can be accounted f o r i f c r a c k i n g occurred along the i n t e r f a c e between the tra n s f o r m a t i o n product and the parent phase. Unusual cleavage f r a c t u r e modes were not r e q u i r e d . (4) A l l o y s more unstable w i t h respect to the formation of martensite products f a i l e d more r a p i d l y i n a p a r t i c u l a r environment. The c r a c k i n g time decreased w i t h the removal of z i n c from the b i n a r y B-brass i n the range of compositions examined, and w i t h removal of N i from the a u s t e n i t i c s t e e l s i n going from 310 to the 304L grade. (5) C o r r o s i o n product f i l m s were detected on the s t r e s s c o r r o s i o n f r a c t u r e surfaces of a l l o y s which f a i l e d more s l o w l y . I t was suggested that the c o n d u c t i v i t y of these f i l m s , may be important i n determining s t r e s s c o r r o s i o n c r a c k i n g l i f e . S e v e r a l p o i n t s s t i l l remain the su b j e c t of debate. These are b r i e f l y considered below: (1) The r o l e of_the m a r t e n s i t e : i t s primary f u n c t i o n appears to be that of p r o v i d i n g the t r a n s g r a n u l a r crack path. In a d d i t i o n , i f martensite contacts the g r a i n boundaries (42) the i n t e r f a c e between - 151 - the g r a i n boundary and martensite p l a t e s may be a more fa v o r a b l e s i t e f o r d i s s o l u t i o n . This may a l s o account f o r the i n t e r g r a n u l a r crack path i n the m a t e r i a l s s t u d i e d . The martensite/parent i n t e r f a c e may a l s o provide b a r r i e r s f o r d i s l o c a t i o n s . R e s u l t i n g s t r e s s e s at the i n t e r f a c e may i n f l u e n c e c r a c k i n g by a s s i s t i n g s t r e s s - s o r p t i o n processes (46), d i s s o l u t i o n (2,27- 31), hydrogen embrittlement (89-91,126), or t a r n i s h rupture (2-4,7,8, 36) type processes. The s t r a i n a s s o c i a t e d w i t h the i n t e r f a c e may a i d d i s s o l u t i o n (27), or the s o l u b i l i t y or r a t e of d i f f u s i o n of adsorbing s p e c i e s , e.g., H + (127-129). The observed r e l a t i o n s h i p between c r a c k i n g time and the martensite t r a n s f o r m a t i o n suggests that m a t e r i a l s r e s i s t a n t to S.C.C. can be developed by i n c r e a s i n g the s t a b i l i t y of the m a t e r i a l s w i t h respect to the t r a n s f o r m a t i o n . This should be an a d d i t i o n a l c o n s i d e r a t i o n i n a l l o y development. On the other hand there i s co n s i d e r a b l e c u r r e n t i n t e r e s t i n developing a l l o y s to e x p l o i t the marte n s i t e t r a n s f o r m a t i o n , f o r example, TRIP s t e e l s (130). However, i n view of the present work, repercussions can be expected i n p r a c t i c a l s i t u a t i o n s where these s t e e l s come i n t o contact w i t h c o r r o s i v e environments. TRIP s t e e l s have been reported to be prone to S.C.C. i n the l a b o r a t o r y s i t u a t i o n (131). The i n c i d e n c e of the martensite t r a n s f o r m a t i o n may i n par t account f o r the U-shaped dependence of S.C.C. f r a c t u r e time of a Fe-18Cr base a l l o y , w i t h N i a d d i t i o n s (132). The minimum time i s observed i n the N i range of 8-11%. Staehle has accounted f o r t h i s i n terms of the N i enrichment theory (20). An a l t e r n a t i v e e x p l a n a t i o n which may be considered i s given as f o l l o w s : a d d i t i o n of n i c k e l changes the - 152 - s t r u c t u r e of the s t e e l g r a d u a l l y from a -> a + y -> y . I n i t i a l l y , s m a l l a d d i t i o n s of Ni to the a-phase i n f l u e n c e the d i s s o l u t i o n k i n e t i c s (20) and S.C.C. time drops. In the a + y region,the martensite t r a n s f o r m a t i o n occurs i n the y phase p r o v i d i n g an even more k i n e t i c a l l y f a v o r a b l e crack path. S.C.C. f a i l u r e times decrease f u r t h e r . On the other hand, f u r t h e r n i c k e l a d d i t i o n s s t a b i l i z e the y phase, and S.C.C. f r a c t u r e times begin to i n c r e a s e . This i s p a r t i c u l a r l y n o t i c e a b l e i n the pure y r e g i o n . Other f a c t o r s must be taken i n t o account, e.g., d i s s o l u t i o n k i n e t i c s , p o s s i b l e f i l m e f f e c t s . The q u a n t i t y and d i s t r i b u t i o n of martensite may a l s o be important. For example, i f the tr a n s f o r m a t i o n i s ext e n s i v e and on a f i n e s c a l e , general d i s s o l u t i o n may occur. On the other hand, i f there i s a very s m a l l amount of marte n s i t e present, general d i s s o l u t i o n of the parent phase may occur. From current d e n s i t y c o n s i d e r a t i o n s , t h e r e i s probably an optimum p r o p o r t i o n and d i s t r i b u t i o n of martensite f o r most r a p i d S.C.C. (2) The_mechanism_of_cracking: i t i s p o s s i b l e that c r a c k i n g occurs by d i s s o l u t i o n , s t r e s s s o r p t i o n or hydrogen embrittlement. There i s l i t t l e doubt that d i s s o l u t i o n occurs during S.C.C, but t h i s may not be the main cause of c r a c k i n g . For example, i n the e l e c t r o c h e m i c a l process hydrogen may be produced at the crack t i p (83), or c r i t i c a l absorbing species may be formed (46) and thus the p o s s i b i l i t y t h a t c r a c k i n g occurs by hydrogen embrittlement or s t r e s s s o r p t i o n c r a c k i n g cannot be ignored. A l t e r n a t i v e l y i f a f i l m forms, and d i s l o c a t i o n p i l e - u p s are present at the i n t e r f a c e of the ma r t e n s i t e , i t i s p o s s i b l e that f i l m rupture processes may occur ( 4 ) . - 153 - (3) The reason f o r c r a c k i n g along_the m a r t e n s i t e / p a r e n t _ i n t e r f a c e : The two phases possess d i f f e r e n t c r y s t a l s t r u c t u r e s and t h e r e f o r e i t i s a n t i c i p a t e d that a p o t e n t i a l d i f f e r e n c e e x i s t s between the two phases. I t i s p o s s i b l e that t h i s may be s u f f i c i e n t l y l a r g e to provide the d r i v i n g f o r c e f o r a d i s s o l u t i o n r e a c t i o n . A l t e r n a t i v e l y the i n t e r f a c e may be a high energy s i t e f o r a d s o r p t i o n . When martensite does not form, other f a c t o r s could be important. Cracking may occur along d i s l o c a t i o n p i l e - u p s or s t a c k i n g f a u l t s by d i s s o l u t i o n ( " t u n n e l l i n g " ) (15,32) or a d s o r p t i o n processes. Where d i s l o c a t i o n arrays i n t e r s e c t o t h e r s , or i n t e r s e c t an i n t e r f a c e such as a g r a i n boundary, c o n j o i n t l y w i t h f i l m forming c o n d i t i o n s i t i s p o s s i b l e that f a i l u r e occurs by a f i l m rupture mechanism, e.g., s i m i l a r to t h a t suggested by B i r l e y and Tromans (4 ) . However, when the martensite t r a n s f o r m a t i o n occurs, these paths are l e s s s i g n i f i c a n t and the martensite i n t e r f a c e i s a k i n e t i c a l l y more fa v o r a b l e path. (4) F i l m _ e f f e c t s : The p o s s i b l e e f f e c t s of f i l m s have been poi n t e d out, v i z , a c t i n g as a wedge (22,26), r e s t r i c t i n g access of s o l u t i o n to the crack t i p , or t a k i n g p a r t i n f i l m rupture processes (2-4,36,78), or f i l m c o n d u c t i v i t y e f f e c t s ( s e c t i o n 3.4). I f the c o n d u c t i v i t y of the f i l m i s important, one can a n t i c i p a t e that a d d i t i o n s to the metal ( a l l o y i n g ) or to the environment, may become i n c o r p o r a t e d i n t o the f i l m and i n f l u e n c e the c o n d u c t i v i t y of the f i l m and hence S.C.C. l i f e . This approach o f f e r s a new c o n s i d e r a t i o n to the design of metal/environment systems r e s i s t a n t to S.C.C. Other c o n s i d e r a t i o n s are the e f f e c t of a d d i t i o n s to the environment on - 154 - d i s s o l u t i o n k i n e t i c s , t r a n s p o r t of species i n s o l u t i o n and to the metal on the martensite t r a n s f o r m a t i o n , d i s s o l u t i o n k i n e t i c s and d i s l o c a t i o n d i s t r i b u t i o n . In the current work, i t has been pointed out that i n the case of the 304L s t a i n l e s s s t e e l i n a FeCl^/MgCl^ environment, s m a l l a d d i t i o n s of A l , Cr, and N i c a t i o n s have a p r e d i c t e d q u a l i t a t i v e e f f e c t on c r a c k i n g time. E f f e c t s of a d d i t i o n s to the metal on the c o r r o s i o n f i l m composition and t h e r e f o r e on the c o n d u c t i v i t y have not been documented. Moreover, the i n f l u e n c e of a l l o y i n g on the marte n s i t e t r a n s f o r m a t i o n can be q u i t e pronounced and may make f i l m e f f e c t s l e s s important. However, t h i s remains the s u b j e c t f o r f u t u r e r e s e a r c h . The i n f l u e n c e of temperature on S.C.C. cannot be accounted f o r i n terms of temperature e f f e c t s on the c o n d u c t i v i t y of surface f i l m s alone. For example, i n environments where Fe^O^ i s the p r i n c i p a l c o r r o s i o n product, S.C. c r a c k i n g time i s a n t i c i p a t e d to vary w i t h temperature e x p o n e n t i a l l y . However, the c o n d u c t i v i t y of Fe^O^ does not change a p p r e c i a b l y w i t h temperature above room temperature (133). C l e a r l y , other e f f e c t s of temperature are important. For example, temperature may i n f l u e n c e : 1) the incidence of marte n s i t e t r a n s f o r m a t i o n 2) d i s l o c a t i o n motion (creep) 3) t r a n s p o r t i n s o l u t i o n 4) d i s s o l u t i o n k i n e t i c s 5) d i f f u s i o n processes o c c u r r i n g i n the metal In summary, cases have been made for a correlation between the incidence of strain induced phase transformations and transgranular stress corrosion cracking i n the systems examined. In absence of the transformation product, other factors such as dissolution along s l i p planes or variations in alloy chemistry, might be important. However, when the strain induced transformation did occur i t appeared that cracking along the interface between the strain induced phase and the parent phase was a k i n e t i c a l l y more favorable process. - 156 - 4. SUMMARY AND CONCLUSIONS Electron dif f rac t ion and fractographic techniques have been applied to the transgranular stress corrosion cracking of austentic steels in hot chloride solutions and of g-brasses in water and ammoniacal copper sulphate solutions. Electron di f f rac t ion studies were conducted on actual stress corrosion fracture surfaces and both non-metallic and metallic phases were detected. The fractographic techniques were applied to mate up opposite stress corrosion fracture surfaces and to determine the topographical features associated with the fracture surfaces. The observations were consistent with the following conclusions: (1) The transgranular crack path was determined by a s train induced phase transformation product, i . e . , martensite. In the case of the 310 y s teel , deformation twinning or fault ing may possibly determine the crack path. (2) The topographical features were accounted for by postulating that cracking occurs along the interface between the martensite and parent phase. (3) Alloys more unstable with respect to the formation of martensite products were more susceptible to S . C . C . Thus, i n development of S .C .C . resistant a l loys , the influence of alloy composition on the martensite transformation must now be considered. - 157 - (4) The conductivity of corrosion product films influences stress corrosion cracking l i f e to some degree. This i s now an additional consideration in the development of S . C . C . resistant metal/environmental systems because i t i s anticipated that additions to the environment and to the metal w i l l become incorporated into the surface f i lm and change i t s conductivity. - 158 - BIBLIOGRAPHY 1. McEVILY, A.J. and BOND, A.P. J. Electrochemical Soc. V. 112, p. 131, 1965. 2. HOAR, T.P. and BOOKER, C.J.L. Corrosion Science, V.5, p. 821, 1965. 3. BIRLEY, S.S. M.A.Sc. Thesis, Dept. Metallurgy, University of British Columbia, 1970. 4. BIRLEY, S.S. and TROMANS, D. Corrosion, Vol. 27, No. 7, p. 297-331 1971. 5. MIDDLETON, W.R. and PARKINS, R.N. Corrosion, Vol. 28, No. 3, p. 88 1972. 6. ORYALL, G. and TROMANS, D. Corrosion, Vol. 27, No. 8, p. 334-341 1971. 7. FORTY, A.J. and HUMBLE, P. Ph i l . Mag. Vol. 8, p. 247, 1963. 8. FORTY, A.J. and HUMBLE, P. p. 403, "Environmental-Sensitive Mechanical Behaviour", Ed. A.R.C. Westwood and N.S. Stoloff, Gordon & Breach, N.Y., 1966. 9. DAVIS, J.A. and WILDE, B.E. J. Electrochemical Soc, Vol. 118, No. 11, p. 1348, 1971. 10. NAKAYAMA, T. and OSHIDA, Y. Trans. Japan Inst. Metals, V. 11, p. 245, 1970. 11. NAKAYAMA, T. and OSHIDA, Y. Trans. Japan Inst. Metals, V. 12, p. 214, 1971. 12. NAKAYAMA, T. and OSHIDA, Y. Corrosion, Vol. 24, No. 10, p. 336, 1968. 13. MILLER, G.T., Jr. and LAWLESS, K.R. J. Electrochemical Soc, Vol. 106, No. 10, p. 854, 1959. - 159 - 14. JENKINS, L.H. and DURHAM, R.B. J . E l e c t r o c h e m i c a l S o c , V. 117 No. 6, p. 768, 1970. 15. PICKERING, H.W. and SWANN, P.R. C o r r o s i o n , V o l . 19, No. 11, p. 373, 1963. 16. PICKERING , H.W. "Fundamental Aspects of S.C.C", p. 164, Ed. R.W. Staehle, A.J. F o r t y . N.A.C.E., 1969. 17. FOLEY, C.L., KRUGER, J . and BECHTOLDT, C J . J . E l e c t r o c h e m i c a l S o c , V o l . 114, No. 10, p. 994, 1967. 18. GREEN, A.J.S. and PUGH, E.N. Met. Trans., V o l . 2, No. 5, p. 1379, 1971. 19. SHIMODAIRA, S. and TAKANO, M. "Fundamental Aspects of S.C.C", p. 202, Ed. St a e h l e , Forty and van Rooyen, N.A.C.E., 1969. 20. STAEHLE, R.W. "Fundamental Aspects of S.C.C", p. 284, Ed. S t a e h l e , F o r t y and van Rooyen, N.A.C.E., 1969. 21. BAKER, H.R., BLOOM, M.C, BOLSTER, R.N., and SINGLETERRY, C R . C o r r o s i o n , V o l . 26, No. 10, p. 420, Oct. 1970. 22. NIELS0N, N.A. " P h y s i c a l M e t a l l u r g y of Str e s s C o r r o s i o n F r a c t u r e " , p. 121, Ed. T.N. Rhodin, I n t e r s c i e n c e , New York, 1959. 23. MATTSSON, E. E l e c t r o c h i m A c t a , V o l . 3, p. 279, 1961. 24. WESTW00D, A.R.C. "Fra c t u r e of S o l i d s " , p. 553, Ed. D.C. Druckner and J . J . Gilman, Gordon and Breach, N.Y., 1962. 25. KRAMER, I.R. and DEMER, L . J . Progress i n M a t e r i a l s Science, V o l . 9, p. 131, 1961. 26. PICKERING, H.W. , BECK, F.H. and FONTANA, M.G. C o r r o s i o n , V o l . 18, p. 330t, 1962. 27. HOAR, T.P. and HINES, J.G. J . I . S . I . , V o l . 184, p. 166, 1956. 28. BARNARTT, S. C o r r o s i o n , V o l . 18, p. 322, 1962. - 160 - 29. BARNARTT, S. and D. van ROOYEN. J . E l e c t r o c h e m i c a l S o c , V o l . 108, p. 222, 1961. 30. DODD, R.A. "Fundamental Aspects of S.C.C", p. 351, Ed. S t a e h l e , F o r t y and van Rooyen, N.A.C.E., 1969. 31. HOAR, T.P. C o r r o s i o n , V o l . 19, No. 10, p. 331, 1963. 32. SWANN, P.R. and EMBURY, J.D. "High Strength M a t e r i a l s " , p. 327, Ed. Zackay, Wiley and Sons, New York, 1965. 33. PUGH, E.N., CRAIG, J.V. and MONTAGUE, W.G., Trans. ASM, V o l . 61, p. 468, 1968. 34. PUGH, E.N., CRAIG, J.V. and SEDRIKS, A.J. "Fundamental Aspects of S.C.C", p. 118, Ed. St a e h l e , F o r t y and van Rooyen, N.A.C.E., 1969. 35. PUGH, E.N. and WESTWOOD, A.R.C. P h i l . Mag. V o l . 13, p. 167, 1966. 36. CHAMPION, F.A. "Symposium on I n t e r n a l Stresses i n Metals and A l l o y s " , p. 468, I n s t i t u t e of M e t a l s , London 1948. 37. ROCHA, H.J. S t a h l . u. E i s e n , V o l . 62, p. 1094, 1942. 38. EDELEANU, C. "Stress C o r r o s i o n Cracking and Embrittlement", p. 126, Wiley, New York, 1956. 39. EDELEANU, C J . I r o n and S t e e l I n s t i t u t e (London), V o l . 173, p. 140, 1953. 40. VAUGHAN, D.A., PHALEN, D.I., PETERSON, C L . and BOYD, W.K. C o r r o s i o n , V o l . 19, No. 9, p. 315, 1963. 41. LATANISION, R.M. and STAEHLE, R.W. "Fundamental Aspects of S.C.C.", p. 260, Ed. St a e h l e , Forty and van Rooyen, N.A.C.E., 1969. 42. MANGONON, P.L., J r . and THOMAS, G. Met. Trans., V o l . 1, No. 6, p. 1577, 1970. 43. BIRLEY, S.S. and TROMANS, D. Unpublished Work. U n i v e r s i t y of B r i t i s h Columbia, 1971. - 161 - 44. SMITH, J.A., PETERSON, H.M. and BROWN, B.F. Corrosion, Vol. 26, No. 12, p. 539, 1970. 45. STAEHLE, R.W., ROYUELA, J.J., RAREDON, T.L., SERRATE, E . , MORIN, C.R., and FERRAR, R.V. Corrosion, Vol. 26, No. 11, p. 451, 1970. 46. JOHNSON, H.E. and LEJA, J. Corrosion, Vol. 22, No. 178, 1966. 47. TROMANS, D., DOWDS, N.A., LEJA, J. "Fundamental Aspects of S.C.C", p. 154, Ed. Staehle, Forty and van Rooyen, N.A.C.E., 1969. 48. TROMANS, D. and NUTTING, J. "Fracture of Solids", p. 637, Ed. Druckner and Gilman, Interscience, New York, 1963. 49. TROMANS, D. and NUTTING, J. Corrosion, Vol. 21, p. 145, 1965. 50. LEHTINEN, B. "Fundamental Aspects of S.C.C", p. 277, Ed. Staehle, Forty and van Rooyen, N.A.C.E., 1969. 51. "Fundamental Aspects of Stress Corrosion Cracking", Ed. Staehle, Forty and van Rooyen, National Association of Corrosion Engineers, Houston, 1969. 52. "Physical Metallurgy of Stress Corrosion Fracture", Ed. T.N. Rhodin Interscience Publishers, New York, 1959. 53. AITCHISON, I. and COX, B. Corrosion, Vol. 28, No. 3, p. 83, .1972. 54. NIELSON, N.A. Reference 51, p. 308. 55. SWANN, P.R. and NUTTING, J. J. Institute of Metals (London), Vol. 88, p. 478, 1959/60. 56. SWANN, P.R. Corrosion, Vol. 19, p. 102, 1963. 57. PUGH, E.N. Reference 51, p. 138. 58. NIELSON, N.A. Corrosion, Vol. 20, p. 104, 1964. 59. NIELSON, N.A. Corrosion, Vol. 27, No. 5, p. 173, 1971. 60. PUGH, E.N. Reference 51, p. 84. - 162 - 61. McEVILY, A.J., J r . Reference 51, p. 83. 62. ALMOND, E.A., KING, J.T. and EMBURY, J.D. Metallography, V o l . 3, p. 379-382, 1970. 63. EISENWASSER, J.D. M.A.Sc. Thesis, Dept. of M e t a l l u r g y , U n i v e r i t y of B r i t i s h Columbia, 1971. 64. BAILEY, A.R. Met. Reviews, V o l . 6, No. 21, p. 101, 1961. 65. MASSALSKI, T.B. and BARRETT, C S . J . M e t a l s , p. 455, A p r i l 1957. 66. PINSKER, Z.G. " E l e c t r o n D i f f r a c t i o n " , Butterworth, London 1953. 67. VAINSHTEIN, B.K. " S t r u c t u r e A n a l y s i s by E l e c t r o n D i f f r a c t i o n " , Ed. F e i g l and Spink. M c M i l l a n , N.Y. 1964. 67a. Ref. 67, p. 151. 67b. Ref. 67, p. 191. 67c. Ref. 67, p. 138. 68. COWLEY, J.M. " C r y s t a l S t r u c t u r e Determination by E l e c t r o n D i f f r a c t i o n " , Progress i n M a t e r i a l s Science, V o l . 13, No. 6, Pergamon, 1967. 69. WHELAN, M.J. and HIRSCH, P.B. P h i l . Mag. p. 1121, V o l . 2, p. 1301, 1957. 70. HOWIE, A. " I - E l e c t r o n D i f f r a c t i o n . II-The Nature of Defects i n C r y s t a l s " , I n t e r n a t i o n a l Conference, Melbourne, A.A. Science, 1965. 71. HIRSCH, P.B., HOWIE, A., NICHOLSON, R.B., PASHLEY, D.W., and WHELAN, M.J. " E l e c t r o n Microscopy of Thin C r y s t a l s " , p. 496, Butterworth, London, 1965. 72. TROMANS, D. P r i v a t e Communication, U n i v e r s i t y of B r i t i s h Columbia, 1972. 73. CULLITY, B.D., "Elements of X-ray D i f f r a c t i o n " Addison-Wesley, Mass, 1959. - 163 - 74. REED, R.P. Acta Met., V o l . 10, p. 865, 1962. 75. HULL, D. and GARWOOD, R.D., p. 219. "Mechanism of Phase Transformation i n Me t a l s " , I n s t i t u t e of Me t a l s , Lond., 1965. 76. SWANN, P.R. and WARLIMONT, H. Acta Met., p. 511, V o l . 11, June, 1963. 77. OTTE, H.M. Acta Met. V o l . 5, p. 614, 1957. 78. LOGAN, H.L. "The S t r e s s C o r r o s i o n Cracking of M e t a l s " , W i l e y , New York,1966. 79. MAREK, M. and HOCHMAN, R.F. C o r r o s i o n , V o l . 26, No. 1, p. 5, 1970. 80. BENSON, R.B., J r . , DANN, R.K. and ROBERTS, L.W., J r . . Trans. AIME, V o l . 242, No. 10, p. 2199, 1968. 81. DIETER, G.E., J r . "Mechanical M e t a l l u r g y " , McGraw-Hill, 1961. 82. HOLZWORTH, M.L. and LOUTHAN, M.R., J r . C o r r o s i o n , V o l . 24, No. 4, p. 110, 1968. 83. RHODES, P.R. C o r r o s i o n , V o l . 25, No. 11, p. 462, 1969. 84. KELLY, P.M. A c t a Met., V o l . 13, p. 635, 1965. 85. DENHARD, E.E. Reference 52, p. 223. 86. SMITH, T.J. and STAEHLE, R.W. C o r r o s i o n , V o l . 23, p. 117, 1967. 87. DOUGLASS, D.L., THOMAS, G.T. and ROSER, W.R. C o r r o s i o n , V o l . 20, No. 1, p. 156, 1964. 88. COTTERILL, P. Progress i n M a t e r i a l s Science, V o l . 9, p. 201, 1961. 89. WHITEMAN, M.B. and TROIANO, A.R. C o r r o s i o n , V o l . 21, p. 53, 1965. 90. SHIVELY, J.H., HEHEMANN, R.F. and TROIANO, A.R. C o r r o s i o n , V o l . 23, p. 215, 1967. 91. SHIVELY, J.H., HEHEMANN, R.F. and TROIANO, A.R. C o r r o s i o n , V o l . 22, p. 253, 1966. 92. HOLZWORTH, M.L. C o r r o s i o n , V o l . 25, No. 3, p. 107. - 164 - 93. OKADA, H., HOSOI, Y. and ABE, S. C o r r o s i o n , V o l . 26, No. 7, p. 183, 1970. 94. JOLLEY, W. and HULL, D. J . I n s t . M e t a l s . , V o l . 92, p. 129, 1964. 95. PEARSON, W.B. "A Handbook of L a t t i c e Spacings and S t r u c t u r e s of Metals and A l l o y s " , p. 892, Pergamon P r e s s , New York, 1967. 96. POPS, H. and DULAEY, L. Trans. AIME, V o l . 242, No. 9, p. 1849, 1968. 97. POPS, H. and MASSALSKI, T.B. Trans. AIME, V o l . 238, No. 12, p. 1662 1964. 98. REYNOLDS, J.E. and BEVER, M.B. Trans. AIME, V o l . 194, p. 1065, 1952. 99. AHLERS, M. and POPS, H. Trans. AIME, V o l . 242, p. 1267, 1968. 100. SURYANARAYANA, C. and ANANTHARAMAN, T.R. Met. Trans., V o l . 2, No. 11, p. 3237, 1971. 101. HALL, E.O.. p. 145, "Twinning" Butterworth, 1954. 102. ELAM, C.F. Nature, V o l . 133, p. 723, 1934. 103. GRENINGER, A.B. Trans. AIME, V o l . 128, p. 369, 1938. 104. BASSI, G. and HUGO, J.P. J . I n s t . M e t a l s , V o l . 87, p. 155, 1958/59. 105. BARRETT, C S . J . M e t a l s , V o l . 6, p. 1003, 1954. 106. KRAMER, I.R. and MADDIN, R. J . Met a l s , V o l . 4, p. 197, 1952. 107. BARRETT, C S . p. 379, " S t r u c t u r e of M e t a l s " , McGraw-Hill, N.Y., 1952. 108. KUBASHEWSKI, 0. and HOPKINS, B.C. "Oxidation of Metals and A l l o y s " , Butterworth, London, 1962. 109. KOHL, H. C o r r o s i o n , p. 39, V o l . 23, 1967. 110. SIDGWICK, N.V. "Chemical Elements and Th e i r Compounds", p. 1351, 1950. 111. WELLS, A.F. " S t r u c t u r a l Inorganic Chemistry", Oxford U n i v e r s i t y P r e s s , 1962. 112. POLING, G. J . E l e c t r o c h e m i c a l S o c , p. 959, V o l . 116, No. 7, 1969. - 165 - 113. PATHANIA, R. and OGLE, I . P r i v a t e Communication, U n i v e r s i t y of B r i t i s h Columbia, 1972. 114. COLE, W.F. and DEMEDIUK, T. A u s t r a l i a n , J . Chemistry, p. 235, 8, 1955. 115. DEMEDIUK, T., COLE, W.F., and HEUBER, H.V. A u s t r a l i a n J . Chemistry, p. 215, J3, 1955. 116. KRUGER, J . Ref. 51, p. 297. 117. VERWEY, E.J.W. and HEILMANN, E.L. J . Chem. Phys., p. 174, V o l . 15, No. 4, 1947. 118. SHULL, C.G., WOLLAN, E.O. and KOEHLER, W.C. Phys. Rev., V o l . 84, No. 5, 1951. 119. VERWEY, E.J.W., HAAYMAN, P.W. and ROMEIJN, F.C. J . Chem. Phys., p. 181, V o l . 15, No. 4, 1947. 120. KINGERY, W.D. " I n t r o d u c t i o n to Ceramics", p. 115, Wiley, N.Y., 1967. 121. de BOER, F., van SANTEN, J.H. and VERWEY, E.J.W. J . Chem. Phys. p. 1032, V o l . 18, No. 8, 1950. 122. ROMEIJN, F.C. J . Chem. Phys., p. 304, V o l . 21, 1953. 123. SCULLY, J.C. "The Fundamentals of C o r r o s i o n " , p. 26, Pergamon 1966. 124. KROGER, F.A. "The Chemistry of Imperfect C r y s t a l s " , North H o l l a n d , Amsterdam, 1964. 125. O'KEEFFE, M. and MOORE, W.J. J . Chem. Phys., p. 1324, V o l . 35, 1967. 126. JOHNSON, H.H. Ref. 51, p. 439. 127. ORIANI, R.A. Ref. 51, p. 45. 128. BECK, W. C o r r o s i o n , p. 86, V o l . 26, No. 2, 1970. 129. PHALEN, D.I. and VAUGHAN, D.A. C o r r o s i o n , p. 243, V o l . 24, No. 8, - 166 - 130. ANTOLOVICH, S.D. and SINGH, B. Met. Trans, p. 2135, V o l . 2 No. 9, 1971. 131. ANTOLOVICH, S.D. and CHANANI, G.R. Spring Meeting of Met. Soc. of AIME, Boston, Mass, 1972. 132. COPSON, H.R. Ref. 52. 133. TANNHAUSER, D.S. J . Phys. Chem. S o l i d s , V o l . 23, pp. 25-34, Pergamon, 1962. 134. HENRY, J.F., LIPSON, H. and WOOSTER, W.A. "The I n t e r p r e t a t i o n of X-ray D i f f r a c t i o n Photographs ", Macmillan Co., London, 1951. 135. THOMAS, G. "Transmission E l e c t r o n Microscopy of M e t a l s " , W i l e y , N.Y., 1967. 136. EVANS, R.C. "An I n t r o d u c t i o n to C r y s t a l Chemistry", Cambridge U n i v e r s i t y P r e s s , 1964. 137. SMITHELLS, C.J. "Metals Reference Book", 4th E d i t i o n , p. 69. Butterworth, London. 138. DAVID, I . and WELCH, A.J.E. Trans. Faraday, S o c , p. 1642, V o l . 52, 1956. 139. BRAUN p.B. Nature, p. 1123, Dec. 27, 1952. 140. BLOOM, M.C. and GOLDENBERG, L. Corr o s i o n Science, p. 623, V o l . 5, 1965. 141. " I n t e r n a t i o n a l Tables f o r X-ray C r y s t a l l o g r a p h y " , V o l . 1, Symmetry Groups, Ed. Henry and Lonsdale Kynoch, 1952. 142. A.S.T.M. Index f o r X-rays Card Number 2-1231. 143. CHAKLADER, A.CD. and BLAIR, G.R. J . Thermal A n a l y s i s , p. 165, V o l . 2, 1970. - 167 - APPENDIX Theoretical estimates of I ' ^ ^ ( o r I e),the integrated intensity for electrons, were calculated using the expression r h k l = M T f (section 2.2) sin 9 where 0 is the structure amplitude for electrons, p the multipicity factor (related to the number of hkl planes i n a given f a m i l y ^ 3 4 ^ ) y a n d 0 i s the Bragg angle. The Bragg angle 0 was determined for 100 Kv (X. = 100 Kv 0.037 A ( 1 3 5 ) ) . The structure factor |0| for a given reflecting plane hkl i s given by iw I VN n 2Tri(hu + Kv + lw ) where f ^ is the atomic scattering factor (amplitude) of the n atom positioned at u v w in the unit c e l l . The summation is extended over r n n' n the total of N atoms in the unit c e l l . Values of the atomic scattering amplitude for electrons were obtained from tables compiled by Thomas The crystals of particular interest in the current work were C^O, CuCl, Au, CuZn, Te^O^ and y-Fe20.j and the metallic phases in stainless - 168 - steels - a ' martensite, e martensite and y austenite. The atom positions i n the crystals are given below:- 1. Cu^O-PnSm Two molecules per unit c e l l ^ ^ ^ with two oxygen atoms at 000, 1/2 1/2 1/2 and four copper atoms at 3/4 1/4 1/4, 1/4 1/4 3/4, 3/4 3/4 3/4, 1/4 3/4 1/4. (137) 2. CuCl-F43m Four molecules per unit c e l l with a zinc blende structure with four Cl atoms at 000, 1/2 1/2 0, 1/2 0 1/2, 0 1/2 1/2 and four Cu atoms at 1/4 1/4 3/4, 3/4 3/4 3/4, 1/4 3/4 1/4. 3/4 1/4 1/4 3. Gold-Fm3m Four Au atoms per unit c e l l situated at 000, 1/2 1/2 0, 0 1/2 1/2, 1/2 0 1/2 4. y Austenite-Fm3m Four atoms per unit c e l l positioned at 000, 1/2 1/2 0, 0 1/2 1/2, 1/2 0 1/2 5. e Martensite-P63/mmc Two atoms per unit c e l l situated at 000,1/3 2/3 1/2. 6. a 1 Martensite-Im3m Two atoms per unit c e l l situated at 000, 1/2 1/2 1/2. 7. Binary &-brass-Pm3m Two atoms per unit c e l l situated at 000, 1/2 1/2 1/2. A composition of Cu + 46 At % Zn is assumed and the 000 sites are f u l l y occupied by copper and the 1/2 1/2 1/2 sites are occupied randomly by the Zinc and the remainder of the copper. 8. Ternary B-brass-Pm3m Two atoms per unit c e l l positioned at 000, 1/2 1/2 1/2. - 169 - A composition of Cu +.33.At % Zn + 4 % T i n i s assumed and the 000 s i t e s f u l l y occupied by Copper and the 1/2 1/2 1/2 s i t e s randomly occupied by Z i n c , T i n and Copper r e s p e c t i v e l y . 9. Fe^O^,—Fd3m Inverse s p i n e l l a t t i c e c o n t aining 56 atoms p o s i t i o n e d as , „ (118, 119, 121) fo l l o w s ' ' 32 oxygen atoms 000, 1/2 0 0, 0 1/2 0, 0 0 1/2 0 1/4 1/4, 0 1/2 1/2, 0 3/4 3/4, 0 1/4 3/4, 0 3/4 1/4 1/4 0 1/4, 1/2 0 1/2, 3/4 0 3/4, 1/4 0 3/4, 3/4 0 1/4 1/4 1/4 0, 1/2 1/2 0, 3/4 3/4 0, 1/4 3/4 0, 3/4 1/4 0 1/2 1/4 3/4, 3/4 1/2 3/4, 1/2 3/4 3/4, 1/4 1/2 3/4 1/4 1/4 1/2, 3/4 1/4 1/2, 3/4 3/4 1/2, 1/4 3/4 1/2,1/2 1/2 1/2 1/2 1/4 1/4, 3/4 1/2 1/4, 1/2 3/4 1/4, 1/4 1/2 1/4 8 Fe i n octahedral s i t e s 0 3/4 0, 1/2 1/4 0, 0 1/4 1/2 3/4 1/4 1/4, 1/4 3/4 1/4, 1/2 3/4 1/2, 1/4 1/4 3/4, 3/4 3/4 3/4 - 8 octahedral s i t e s : 3/4 0 0, 0 0 3/4, 1/4 0 1/2, 1/2 0 1/4 0 1/2 1/4, 1/4 1/2 0, 3/4 1/2 1/2, 1/2 1/2 3/4 - 170 - 8 tetrahedral s i tes : 1/8 1/8 1/8, 5/8 5/8 1/8, 3/8 3/8 3/8, 7/8 7/8 3/8 5/8 1/8 5/8, 1/8 5/8 5/8, 7/8 3/8 7/8, 3/8 7/8 7/8 10. Y-Fe 0 -P4 3 ~ 2-3 3~ There is consideral controversy over the exact atom positions i n this material. It i s now believed that the formula is more correctly w r i t t e n ^ ^ 140) & g HFe c 0 o with atom positions similar to j o LiFe^Og, the H atoms replacing the L i atoms. The atom positions are given below ( 1 2 1 > 1 3 9 ' 1 4 1 ) . 32 Oxygen Atoms 1/2 +A , , 1/2 + A . , 1/2 + A , 6 6 6 A 6 , 1/4 - A 6 , 3/4 - A 6 3/4 - A 6 , A 6 , 1/4 - A 6 1/4 - A 6 , 3/4 - A 6 , + A 6 V A 6 A 6 1/2 - A 6 , 3/4 + A g , 1/4 + A 6 1/4 + A 1/2 - A 3/4 + A 3/4 + A , , 1/4 + A , , 1/2 O D - A A 3 , 1/4 - A 4 , 1/4 + A 5 1/4 - A 3 , 3/4 - A 4 , 1/2 - A 5 1/4 + A 5 , + A 3 , 1/4 - A A A 5 , 1/2 + A 3 , 1/2 + A 4 1/4 - A 4 , 1/4 + A 5 , A 3 A 4 , 3/4 + A 5 , 3/4 - A 3 1/2 - A 3 , 1/4 - A 5 , 1/4 + A 4 3 1/4 + A 4 , 1/2 - A 1 / 4 - A 1/4 - A 5 , 1/4 + A 4 , 1/2 - A 3 A 3 , 1/2 + A 5 , A 4 3/4 + A 4 , 1/4 + A 3 , A 5 3/4 - A 5 , 1/2 - A 4 , 3/4 + A 3 1/2 + A 3 , 1/2 + A 4 , A 5 3/4 - A 3 , A 4 , 3/4 + A 5 3/4 + A 5 , 3/4 - A 3 , A 4 1/2 - A 5 , 1/4 - A 3 , 3/4 - A 4 3/4 - A 4 , 1/2 - A 5 , 1/4 - A 3 1/2 + A 4 , A 5 , 1/2 + A 3 3/4 + A 3 , 3/4 - A 5 , 1/2 - A 4 A 4 , A 3 , 1/2 + A 5 A 5 , 3/4 + A 4 , 1/4 + A 3 1/4 + A 3 , A 5 , 3/4 + A 4 1/2 - A 4 , 3/4 + A 3 , 3/4 - A 5 1/2 + A 5 , A 4 , A 3 - 171 - 8 Fe atoms i n tetrahedral s i t e s 1/8 + A 2, 1/8 + A 2, 1/8 + A 2 5/8 + A 2, 5/8 - Aj, 1/8 - ^ 1/8 - A 2 , 5/8 + A 2 , 5/8 - A2 5/8 - A 2 , 1/8 - A 2 , 5/8 + A 2 3/8 - A 2 , 3/8 - A 2 , 3/8 - A 2 7/8 - A 2 , 3/8 + A 2 , 7/8 - A 2 7/8 + A 2 , 7/8 - A 2 , 3/8 + A 2 3/8 + A 2 , 7/8 + A 2 , 7/8 - A 2 12 Fe atoms i n octahedral s i t e s 1/4, 1/2 + A r 11 3/4, 1/4 - h v 1/4 + A x A x , 1/4, 1/2 + A 1 1/4 + A 1 , 3/4, 1/4 - A x 1/2 + A x , h r 1/4 1/4 - A L , 1/4 + A L , 3/4 1/2, 3/4 - A x , 1/2 - A x 0, 0 + A , 3/4 + A 1/2 - 1/2, 3/4 - A x 3/4 + A 1 , 0, A ^ 3/4 - A 1 , 1/2 - A r 1/2 A l t 3/4 + A.^ 0 4 H atoms on octahedral s i t e s 3/4 3/4 3/4, 1/4 0 1/2 1/2 1/4 0 ,0 1/2 1/4 where A^ - A^ are small deviations assumed to be equal. (139) Braun ascribed a value of A= 0.007 for the case of LiFe^Og, but no- where i n the l i t e r a t u r e has a value been ascribed to A f o r the case of HFej-OQ. Accordingly, small a r b i t r a r y q u a n t i t i e s were ascribed to A on a J o t r i a l and e r r o r b a s i s . The following procedure was adopted i n determination of I f o r HFe c0 o since there are no guidelines a v a i l a b l e . F i r s t l y , e J O t h e o r e t i c a l r e l a t i v e i n t e n i t i e s f o r X-ray f or Fe^O^ were ca l c u l a t e d (138) (Table A l) and compared with the most r e l i a b l e data a v a i l a b l e f o r t h i s - 172 - compound, i n order to assess the accuracy of the c a l c u l a t i o n s i . e . to ob- t a i n an accepted degree of agreement between t h e o r e t i c a l c a l c u l a t i o n and observations. This i s given i n Table A l . Secondly, c a l c u l a t i o n s of t h e o r e t i c a l r e l a t i v e i n t e n s i t i e s f o r X-rays were made f o r HFe^O., a s c r i b i n g D O s m a l l values of A , to achieve the same accepted degree of correspondence w i t h the observed d a t a v . This i s shown i n Table A2, f o r A = 0.003. This i s the f i r s t time a value f o r A f o r HFe^Og has been given. T h i r d l y , the t h e o r e t i c a l r e l a t i v e i n t e n s i t i e s f o r e l e c t r o n s were determined using t h i s same value f o r A. (See Table A3). The e f f e c t of r e f l e c t i n g plane h k l , upon the form of |0| f o r the c r y s t a l s t r u c t u r e s are shown i n Tables A4-A6. The t h e o r e t i c a l r e l a t i v e i n t e n s i t i e s f o r e l e c t r o n s of Fe„0, and HFe^On are presented i n Table A3, 3 4 5 8 r and those corresponding to the other c r y s t a l s t r u c t u r e s are given i n the main t e x t . Note on I r o n Oxides 1. Examination of Table A3 r e v e a l s that i t i s almost impossible to d i s - tingush between y - F e o 0 o (HFe.-0o) and Fe„0. powders by the e l e c t r o n d i f f r a c t i o n technique because of the l i m i t a t i o n s i n accuracy of the technique. I | j | | 2. For s p i n e l s i n which some Fe and Fe are replaced by other t r a n s i t i o n i | j [ | metal ions (e.g. N i or Cr ), the changes i n s c a t t e r i n g f a c t o r f o r e l e c t r o n s and l a t t i c parameters are such that the current e l e c t r o n d i f f r a c t i o n techniques would not adequately d i s t i n g u s h them from Fe 0.. - 173 - TABLE Al - Relative Intensities for F e , ^ (X-rays - Comparison With Experimental Observations) 2 2 2 h +k +1* hkl O d A Theoretical Intensity (%) „ „ . (138) Observations 3 111 4.85 5 10 8 220 2.966 26 34 11 311 2.53 100 100 12 222 2.419 7 6 16 400 2.096 24 34 19 331 1.93 1 - 24 422 1.713 12 11 27 333/511 1.614 34 41 32 440 1.480 60 62 35 531 1.41 1 - 40 620 1.327 5 3 43 533 1.279 14 8 44 1 622 1.264 16 2 - 174 - TABLE A2 - Relative Intensities for HFe,.0o (X-rays)- ( O - o o T Comparison with Experimental Observations 2 2 2 h +k +1 hkl 0 d A I T % Observations 1 100 8.35 Absent Absent 2 110 5.9 11 5 3 111 4.82 5 5 4 200 4.18 Absent Absent 5 210 3.73 14 6 6 211 3.41 16 5 8 220 2.95 36 42 9 221 2.78 4.3 2 10 310 2.40 3 vf 11 311 2.52 100 100 12 222 2.41 2.4 Absent 13 320 2.32 4.5 vf 14 321 2.23 4 vf 16 400 2.08 19 33 17 410/322 2.02 Absent Absent 18 411/330 1.97 2 Absent 19 331 1.91 0.06 Absent 20 420 1.87 Absent Absent 21 421 1.82 2.2 2 22 332 1.77 0.3 Absent 24 422 1.70 16 16 25 500/430 1.67 Absent Absent 26 510/431 1.63 0.08 Absent 27 330/511 1.61 38 45 29 520/432 1.55 0.12 vf 30 521 1.53 1.4 3 32 440 1.48 63 65 33 441/522 1.45 Absent Absent 34 530/433 1.43 0.6 Absent 35 531 1.41 0.12 Absent 36 442/600 1.39 0.54 Absent 37 610/532 1.37 0.06 Absent 38 611 1.35 0.54 vf 40 620 1.32 7.5 6 41 621/540/433 1.30 Absent Absent 42 541 1.29 0.8 Absent 43 533 1.27 13.7 11 vf = very faint - 175 - TABLE A3 - Theoretical Intensities for Electrons - a Comparison of Fe.,0, and y Fe~0_ (HFe,-0Q) 2 2 2 h +kZ+lZ hkl o d A (Fe 30 4) I T (Fe304)% 0 d A ( Y -Fe 2 0 3 ) (Y-Fe 20 3)% 1 100 8.35 Absent 2 110 5.9 15 3 111 4.85 8.5 4.82 4 4 200 4.18 Absent 5 210 3.73 15 6 211 3.41 6 8 220 2.966 32.5 2.95 34 9 221 2.78 3.7 10 310 2.64 2 11 311 2.530 100 2.52 100 12 222 2.419 1.3 2.41 2 13 320 2.32 3.5 14 321 2.23 3 16 400 2.096 32 2.08 24 17 410 2.02 Absent 18 411/330 1.97 1.3 19 331 1.93 1 1.91 0.15 20 420 1. 87 Absent 21 421 1.82 1.3 22 332 1.77 0.10 24 422 1.713 8 1.70 8 25 500/430 1.67 Absent 26 510/431 1.63 0.4 27 333/511 1.614 25 1.61 20 29 520/432 1.55 0.05 30 521 1.53 0.05 32 440 1.48 50 1.48 47 33 441/522 1.45 Absent 34 530 1.43 0.026 35 531 1.41 0.6 1.41 0.01 36 442/600 1.39 0.01 37 610 1.37 0.04 38 523/611 1.35 0.02 40 620 1.327 3 1.32 3 41 621/540/433 1.30 Absent 42 541 1.29 0.04 43 533 1.279 7 1.29 6 44 622 1.264 0.4 1.26 0.7 45 542/630 1.24 0.8 46 631 1.23 0.02 48 444 1.211 5 1.21 3 49 700/632 1.19 Absent 50 500/710/543 1.18 0.001 51 551/711 1.17 0.3 1.17 0.001 52 640 1.16 Absent 53 641/720 1.15 0.003 54 552/633/721 1.13 0.003 Cont inued. . . . - 176 - TABLE A3 - Continued 2 2 2 h +k +1 hkl O d A (Fe 30 4) (Fe304)% o d A ( Y -Fe 2 0 3 ) T T (Y-Fe 20 3)% 56 642 1.12 3 1.12 3.5 57 544/722 1.10 Absent 58 730 1.095 0.003 59 553/731 1.092 12 1.09 7.5 61 643/650 1.07 0.002 62 723/651 1.06 0.002 64 800 1.048 5.5 1.04 4.5 65 810,740,652 1.04 Absent 66 544,741,811 1.03 0.002 67 733 1.025 0.01 1.02 0.002 68 820,644 1.02 Absent 69 742,821 0.995 0.002 70 653 0.990 0.002 72 822/660 0.989 1 0.98 1 73 830/661 0.975 Absent 74 831/750/743 0.97 0.001 75 555/751 0.9692 5 0.96 4 v 76 662 0.96 0.2 0.955 0.001 77 654,832 0.95 0.001 78 752 0.945 0.001 80 840 0.939 4 0.935 4 81 900,841,744,663 0.925 Absent 82 833,910 0.92 0.001 83 911,753 0.920 0.1 0.915 0.001 84 842 0.91 Absent 85 760/920 0.905 0.002 86 921,655,761 0.90 0.001 88 664 0.895 0.05 0.89 0.001 89 850,762,843,922 0.88 Absent 90 851,754 0.88 0.001 91 931 0.879 2.5 0.875 2 92 852 0.865 0.001 94 763 0.86 0.001 95 932 0.855 Absent 96 844 0.856 8 0.850 5 - 177 - TABLE A4 - Structure Factors for Au, Cu 9 0, CuCl and Austenitic Steels 2 2 2 h +k +1 . . . . . . , , hkl 0 2 Au or y Cu20 CuCl 1 100 - - 2 110 4f 2 o - 3 111 1 6 f l u or Y 1 6 f C u 1 6 f ? l + 1 6 f C u 4 200 16f2 Au or Y ( 4 f c r 4 fcu ) 2 5 210 - - 6 211 - 4f 2 o - 8 220 16f2 Au or Y ( 2 f 0 + 4 f C u ) 2 ( 4 f C l + 4 f C u > 2 9 300,221 - - 10 310 - - 11 311 16fA2 Au or Y 16f 2 Cu 12 222 16fA2 Au or Y < 2 f o- 4 f Cu> 2 ( 4 £cr 4 £cu» 2 13 320 - - 14 321 - < - 16 400 16 f 2 Au or Y < 2 f o + 4 W 2 ( 4 f C l + 4 f C u ) 2 17 410,322 - - 18 411,330 - 4 f o - 19 331 16f 2 Au orY 16fSu 1 6 f C l + 1 6 f C u 20 420 16f 2 Au or Y ( 2 f 0 - 4 f C u ) 2 (4f -4f ) 2 ^ C l Cu^ 21 421 - - 22 332 - 4f 2  H I o - 24 422 16f 2 Au or Y ( 2 f 0 + 4 f C u ) 2 (4f +4f ) 2 v Cl Cu' Structure factor M i l l e r indeces of crystal planes Atomic scattering factors for electrons: o - oxygen, Cl - chlorine . Cu - copper Atomic scattering factor for electrons:gold Atomic scattering factor for electrons:weighted mean of Fe, Ni and Cr I 01 " hkl - f C u > f o ' f C l - f A u " TABLE A5 - Structure Factors for 3-brasses and b . c . c . a steel 2 2 2 h/+kZ+l Structure Factors hkl Binary 3 Ternary 3 a steel 1 100 t ' 9 2 < f C u - f Z n ) ] 2 t ' 7 A f C u " 6 6 f Z n " - ° 8 W 2 2 110 t 1 ' 0 8 f Cu + - 9 2 hrf t 1 ' 2 6 f C u + » 6 6 f Z n + ' ° 8 W 2 4f 2 a 3 111 t - 9 2 < f C u " f Z n ) ] 2 t ' 7 4 f S n " - 6 6 f Z n " - ° 8 h f 4 5 200 210 f 1 - 0 8 f C u + - 9 2 W 2 t - 9 2 ^ f C u - f Z n ) ] 2 t 1 ' 2 6 f C u + ' 6 6 f Z n + - ° 8 f S n ] 2 t ' 7 4 f C u " ' 6 6 f Z n " - ° 8 W 2 4f 2 a 6 8 9 211 220 300/221 t 1 ' 0 8 f Cu + ' 9 2 h f t 1 ' 0 8 f C u + « 9 2 hrf [ ' 9 2 ( f C u - f Z n ) ] 2 [1.26 f_ + .66 f_ + .08 f_ ] 2 Cu Zn Sn [1.26 f„ + .66 f_ + .08 f_ ] 2 Lu Zn on [.74 f_ - .66 f_ - .08 f_ ] 2 Cu Zn Sn 4f 2 a 4f 2 a 10 11 310 311 f 1 ' 0 8 f Cu + « 9 2 f z / [• 9 2 < f Cu- f Z n ) ] 2 [1.26 f C u + .66 f Z n + .08 f S n ] 2 I ' 7 4 f C u " ' 6 6 f Z n " - ° 8 f S n ] 2 4f 2 a 12 13 222 320 [ 1 - 0 8 f c u + - 9 2 hf t ' 9 2 ( f C u - f Z n ) ] 2 I 1 ' 2 5 f C u + ' 6 6 f Z n + - ° 8 f S n ] 2 [ ' 7 4 f Cu " ' 6 6 f Zn " ' ° 8 h f 4f 2 a 14 321 [1.08 f_ + .92 f_ ] 2 Cu Zn [1.26 f_ + .66 f_ + .08 f c ] 2 Cu Zn Sn 4f 2 16 400 [ 1 ' 0 8 f C u + ' ° 2 f Z n ] 2 [1.26 f . + .66 f + .08 f_ ] 2 Cu zn Sn 4f 2 a f = atomic scattering factor for copper f = atomic scattering factor for zinc f_ = atomic scattering factor for t in f = atomic scattering factor for a steel b n 0 1 (weighted mean of Fe, N i , Cr) - 179 - TABLE A6 - Structure Factors of HFe c 0 o and Fe.O 2 2 2 h +k +1 h k l Structure Factor (HFe^g) Structure Factor (Fe^O^) 1 100 - - 2 110 4(f - f h ) 2 - 3 111 (6f - 4/2f + 2f, ) 2 h 2(4f 1 - 4 / 2 f 2 ) 2 4 200 - - 5 210 8(f - f h ) 2 - 6 211 4(f - f , ) 2 - 8 220 64f 2 ( 8 f x ) 2 9 221/300 - - 10 310 4(f - f h ) 2 - 11 311 (6f + 4/2 f +2f h ) 2 2(4^ + 4 / 2 f 2 ) 2 12 222 16(8f - 3 f - f, ) 2 0 n ( 16f 2 + 32 f Q ) 13 320 8(f - f h ) 2 - 14 321 4(f - f h ) 2 - 16 400 16(f + f + f , ) 2 0 n (8f x - 16f 2 - 32f Q ) 2 17 410/322 - - 18 411/330 4(f - f h ) 2 - 19 331 (6f - 4/2f + 2f, ) 2 n 2 (4^ - 4 / 2 f 2 ) 2 20 420 - - 21 421 8(f - f h ) 2 - 22 332 4(f - f , ) 2 - 24 422 64f 2 ( 8 f x ) 2 25 500/430 - - 26 510/431 4(f - f h ) 2 - 27 333/511 (6f + 4/2f + 2f, ) 2 n 2(4f x + 4/2 f 2 ) 2 Continued - 180 - TABLE A6 - Continued - 2 2 2 h +k +1 hkl Structure Factor (HFe s0 f t) Structure Factor (Fe^O^) 29 520/432 8(f - f h ) 2 - 30 521 4(f - f h ) 2 - 32 440 16 (8f + 5f + f, ) o h <8f. + 16f- + 32f ) 2 1 2 o 33 441/522 - - 34 530/433 4(f - f h ) 2 - 35 531 (6f - 4/2f + 2 f h ) 2 2(4^ - 4 / 2 f 2 ) 2 36 600/442 16(f - f h ) 2 - 37 610 8(f - f h ) 2 - 38 532/611 4(f - f h ) 2 - 40 620 2 64f 41 621/540/443 - - 42 541 4(f - f h ) 2 - 43 533 (6f + 4/2f + 2 f h ) 2 2(4f 1 + 4 / 2 f 2 ) 2 44 622 16(8f - 3f - f, ) 2 o h (-16f 0 + 32f ) 2 2 o 45 542/630 8(f - f h ) 2 - 46 631 4(f - f h ) 2 - 48 444 16 (8f + f + f. ) o n (813, - 16f_ - 32f ) 2 1 2 o 49 700 - - 50 543/500/710 4(f - f h ) 2 - 51 551/711 (6f - 4/2f + 2 f, ) 2 h 2(4f x - 4 / 2 f 2 ) 2 52 640 - - 53 641/720 8(f - f h ) 2 - 54 552/633/721 4(f - f h ) 2 - 56 642 64f 2 ( S f ^ 2 Continued - 181 - TABLE A6 - Continued 2 2 2 hz+k +1 hkl Structure Factor (HF 50 8) Structure Factor ( F e „ 0 . ) 3 4 57 544/722 - - 58 730 4(f - f h ) 2 - 59 731/533 (6f +/2f + 2h h ) 2 2(4f 1 + 4 » / 2 f 2 ) 2 61 643/651 8(f - f , ) 2 - 62 723/651 4(f - f h ) 2 - 64 800 16(8f + 5f + f. ) 2 o n (8f.. + 16f_ + 32f ) 2 l z o 65 740/652/810 - - 66 544/741/811 4(f - f h ) 2 - 67 733 (6f - 4/2f + 2f, ) 2 n 2(4f x - 4 / 2 f 2 ) 2 68 644/820 - - 69 742 4(f - f h ) 2 - 821 8(f - f h ) 2 - 70 643 4(f - f h ) 2 - 72 660/822 64f 2 ( 8 f x ) 2 73 661/830 - - 74 347/750/831 4(f - f h ) 2 - 75 551/751 (6f - 4/2f + 2 f h ) 2 2(4f x + 4 / 2 f 2 ) 2 76 662 16(8f - 3f - f, ) 2 o h (-16f0 + 32f ) 2 £. O 77 654 4(f - f h ) 2 - 832 8(f - f , ) 2 - 78 752 4(f - f , ) 2 - 80 840 64(4fo - f h ) 2 (8f. - 16f_ - 32f ) 2 1 Z. o 81 663,744,900,841 - - 82 833,910 4(f - f , ) 2 - Continued. - 182 - TABLE A6 - Continued 2 2 2 h +k +1 hkl Structure Factor (HFe50g) Structure Factor ( F e . ^ ) 83 911,753 (6f - 4/2f + 2 f u ) 2 n 2(4f x - 4 / 2 f 2 ) 2 84 842 - - 85 760,920 8(f - f h ) 2 - 86 655,761,921 4(f - f h ) 2 - 88 664 64f 2 (Sf^) 2 89 762,843,850,922 - - 90 754,851,930 4(f - f h ) 2 - 91 931 (6f + 4/2f + 2 f h ) 2 2(4f]_ + 4 / 2 f 2 ) 2 93 852 2(f - f h ) 2 - 94 763 2(f - f h ) 2 - 95 932 - - 96 844 16(8fQ + 5f + f h ) 2 (8f, + 16f 0 + 32f ) 2 1 l o f = atomic scattering factor for oxygen f^ = atomic scattering factor for hydrogen f = atomic scattering factor for iron = atomic scattering factor for iron in tetrahedral sites f_ = atomic scattering factor for iron i n octahedral sites

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
Germany 2 0
China 2 31
United States 2 0
City Views Downloads
Unknown 2 0
Beijing 2 0
Sunnyvale 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items