Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Ductility and chemical reactions at the interface between nickel and magnesium oxide single crystals Hasselman, Didericus Petrus Hermannus 1959

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1960_A7 H25 D8.pdf [ 12.82MB ]
Metadata
JSON: 831-1.0105929.json
JSON-LD: 831-1.0105929-ld.json
RDF/XML (Pretty): 831-1.0105929-rdf.xml
RDF/JSON: 831-1.0105929-rdf.json
Turtle: 831-1.0105929-turtle.txt
N-Triples: 831-1.0105929-rdf-ntriples.txt
Original Record: 831-1.0105929-source.json
Full Text
831-1.0105929-fulltext.txt
Citation
831-1.0105929.ris

Full Text

DUCTILITY AND CHEMICAL REACTIONS AT THE INTERFACE BETWEEN NICKEL AND MAGNESIUM OXIDE SINGLE CRYSTALS by DIDERICUS PETRUS HERMANNUS HASSELMAN A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE i n the Department of MINING AND METALLURGY We accept t h i s t h e s i s as conforming to the standard required from candidates f o r the degree of MASTER OF APPLIED SCIENCE. Members of the Department of Mining and Metallurgy THE UNIVERSITY OF BRITISH COLUMBIA August 1959 In 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 the requirements f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make 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 and study. I f u r t h e r agree t h a t permission f o r e xtensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head of my Department or by h i s r e p r e s e n t a t I v e s . I t i s understood t h a t copying or p u b l i c a t i o n of 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 not be allowed without my w r i t t e n p e r m i s s i o n . Department of ^ Wining and Metallurgy The U n i v e r s i t y of B r i t i s h Columbia, Vancouver Canada. Date December 11, 1959 ABSTRACT An i n v e s t i g a t i o n was conducted on the i n t e r a c t i o n between n i c k e l metal and s i n g l e c r y s t a l s of magnesium oxide. The n i c k e l was cleaned with p u r i f i e d hydrogen gas at 800°C and melted under vacuum (5 x 10"^ mm. Hg) i n contact with the magnesium oxide. The i n t e r f a c e was examined metallographic-a l l y . The formation of compounds at the i n t e r f a c e was examined by X-ray d i f f r a c t i o n techniques. The magnesium oxide was p l a s t i c a l l y deformed by thermal stresses which occurred on cooling. S l i p occurred on four s l i p planes only. The s l i p sources were present i n the surface at a depth not exceeding ten microns. They were i n the form of d i s l o c a t i o n half-loops introduced i n the surface during cleavage. The in t r o d u c t i o n of these half-loops was due to the forma-t i o n of cleavage steps. Rows of d i s l o c a t i o n half-loops were due to the m u l t i p l i c a t i o n of a h a l f - l o o p on a s l i p plane oriented i n the d i r e c t i o n of propagation of the cleavage crack. Bond formation consisted of the formation of a magnesium-nickel compound (Mg 2Ni). The bond strength exceeded the stress f o r b r i t t l e f r a c t u r e of magnesium oxide. Attack of the magnesium oxide occurred p r e f e r e n t i a l l y at the perimeter of the i n t e r f a c e . This led to groove-formation, which res u l t e d i n a large hysteresis of wetting. The d i f f u s i o n of n i c k e l into magnesium oxide seemed to take place mainly by d i f f u s i o n along d i s l o c a t i o n s . ACKNOWLEDGEMENT The author i s indebted to Mr. W.M. Armstrong f o r h i s supervision and encouragement and to Mr. R. Butters f o r his t e c h n i c a l advice and assistance. The author i s also e s p e c i a l l y indebted to Dr. E. Teghtsoonian and Mr. K.G. Davis f o r many h e l p f u l discussions. The work was financed by Research Grant 7510-32 provided by the Defence Research Board of Canada. SUMMARY O F S T A F F C O M M E M S A T T H E S I S PRESENTATION December 9., 1959. The s t a f f was i n general agreement that t h i s was a very acceptable t h e s i s . Considerable ingenuity i n both experimental technique and i n Int e r p r e t a t i o n of r e s u l t s has been demonstrated by the candidate. Furthermore j, he has proven h i s a b i l i t y to design and execute good e x p e r i -ments and;assess the r e s u l t s with a minimum of supervision. His o r a l presentation and defence of his thesis were both of good q u a l i t y . Some d i s s a t i s f a c t i o n was expressed concerning the v a l i d i t y of the X-ray evidence offered to substantiate the presence of. Mg2NI compound at the in t e r f a c e between MgO and Ni drop. The candidate did not appear to understand the complete basis f o r the i d e n t i f i c a t i o n , based on both d spacings and r e l a t i v e l i n e i n t e n s i t i e s . Accordingly, i t i s very doubt-f u l that h i s i n t e r p r e t a t i o n of the X-ray films i s c o r r e c t . S i m i l a r objections apply to the i d e n t i f i c a t i o n of the NiO. I t should be noted that t h i s does not a f f e c t the major part of the the s i s i n any way. It was also suggested that some of the conclusions concerning the d i s l o c a t i o n mechanisms operative during the cleavage and deformation processes were only barely warranted by the experimental evidence put forward. A d d i t i o n a l minor corrections were as follows; page 63, paragraph 2: , ;During evaporation of.the drop the term cos 9 continuously decreases This should read; "During evaporation of the drop the component of the cos 9 term p a r a l l e l to plane surface i n the d i r e c t i o n of yMgO decreases In a d d i t i o n , i t i s f e l t that the i n t e r p r e t a t i o n i n terms of the .groove formation and the ensuing modification through the angle p i s not as s i g n i f i c a n t as the candidate suggests. page '93;f. l a s t l i n e : ". . .which contained l i t t l e or no n i c k e l . " should be; " . . . which contained l i t t l e or no m e t a l l i c n i c k e l . 5 ' W. M. Armstrong, ,Professor of Metallurgy. TABLE OF CONTENTS Page „ I. INTRODUCTION 1 A. General Purpose and Scope 1 B. Previous Work • . 4 C. S p e c i f i c Aims of the Present Investigation . . . . 6 I I . EXPERIMENTAL 8 A. Materials 8 1. Magnesium oxide 8 2 . N i c k e l 9 B. Apparatus 10 1. Furnace 13 2 . Vacuum system 14 3 . O p t i c a l system 15 C. Preparation of Materials . 16 D. Experimental Procedure 17 I I I . EXPERIMENTAL RESULTS 19 A. Observation on Interface 19 B. Heating Rate 25 C. Etching by Liquid N i c k e l 27 D. Polarized Light 28 E. Chemical Etching 29 F. Nature of Deformation 33 G. Strength of Bond . 38 H. Debye-Scherrer X-ray Analysis of I n t e r f a c i a l . . . . M a t e r i a l 39 I . D i f f u s i o n of N i c k e l into Magnesium Oxide 39 J . W e t t a b i l i t y 40 Table of Contents (cont»d.) Page IV. DISCUSSION 43 A. P l a s t i c Deformation 43 B. Nature of the P l a s t i c Deformation 43 C. Cause of the P l a s t i c Deformation , 45 D. Temperature at which P l a s t i c Deformation Occurred . . 46 E. P l a s t i c Deformation During Heating . . . 46 F. Study of the P l a s t i c Deformation under P o l a r i z e d Light 47 G. Mechanism of the P l a s t i c Deformation 47 H. Temperature Dependence of the P l a s t i c Deformation . . 49 I. Nature of S l i p Sources 49 J . Mechanisms of Introduction of S l i p Sources i n t o Surface 50 K. S l i p Source Density 54 L. Minimum Size of Stable D i s l o c a t i o n Loop . 54 M. Depth of S l i p Sources 55 N. Number of Dislocations Produced by S l i p Sources . . . 56 0 . Mechanism of D i s l o c a t i o n M u l t i p l i c a t i o n . . . . . . 58 P. I n t r i n s i c S l i p Sources 59 Q. Chemical Reaction at Interface 60 R. Etching of D i s l o c a t i o n s by L i q u i d N i c k e l 63 S. D i f f u s i o n of N i c k e l i n t o Magnesium Oxide 64 T. Strength of Bond 65 V. CONCLUSIONS 66 A. D u c t i l i t y 66 B. Chemical Reactions and D i f f u s i o n . . . . 68 Table of Contents (cont'd.) Page VI. RECOMMENDATIONS FOR FURTHER WORK 69 VII. APPENDICES 70 VIII. BIBLIOGRAPHY FIGURES No. Page 1. MgO - NiO binary system 5 2. T y p i c a l (100) cleavage face of magnesium oxide (xlOO) 9 3. The apparatus assembly of the induction furnace, vacuum system and o p t i c a l system . 11 4. Assembly drawing of the apparatus 12 5. Molybdenum heating elements (dismantled)of the induction furnace 13 6. Magnesium oxide 'as received' and cleaved c r y s t a l s with and without n i c k e l drops .... . . . . . . ........ 16 7. Tapered s e c t i o n (1:6) of magnesium oxide-nickel i n t e r f a c e (x750)... 19 8. Magnesium oxide-nickel i n t e r f a c e as seen through the magnesium oxide (x250) 20 9. Same as F i g . 8 but d a r k - f i e l d i l l u m i n a t i o n (x250) 20 10. Same as F i g . 8 (x2500) 21 11. Interface of n i c k e l metal and magnesium oxide p l a t e l e t polished before heating. Near center,of platelet.(x250) . 22 12. Same as F i g . 11 but near edge of p l a t e l e t where p o l i s h i n g was more severe (x250) 22 13. Interface between n i c k e l drop and p l a t e l e t of magnesium oxide without cleavage steps (x300) 25 14. Magnesium oxide-nickel i n t e r f a c e a f t e r f i r s t slow heating cycle (x250) 26 15. Same area as i n F i g . 14.after second f a s t e r heating cycle (x250) . 26 16. Interface between n i c k e l drop and rod of magnesium oxide under p o l a r i z e d l i g h t (x25) 28 Figures (cont'd.) No. Page 17. Side face of magnesium oxide rod with n i c k e l drop on top surface (x25) 29 18. Chemically etched cleavage face of magnesium oxide (xlOO) . . 29 19. Chemically etched cleavage face near high cleavage step (xlOO). 30 20. Opposite cleavage face from F i g . 18 (xlOO) . 31 21. Etched cleavage face of magnesium oxide (xlOO) 32 22. Same area as i n F i g . 21 but with n i c k e l drop, as seen through the p l a t e l e t and therefore i t s mirror image (xlOO) . . . . . 32 23. Faces of broken rod of magnesium oxide near f r a c t u r e end. Unetched (xl5) 33 24. Faces of broken rod near f r a c t u r e end. Etched (xl5) 34 25. End of broken rod under p o l a r i z e d l i g h t (x30) . . . . . . . . 35 26. Area on rod from which n i c k e l drop was removed by melting. Unetched. (xl25) 35 27. Same as F i g . 26. Etched t h i r t y minutes (x750) 36 28. Bent rods of magnesium oxide, i l l u s t r a t i n g d u c t i l i t y . . . . . 36 29. S l i p l i n e s on polished and broken rod of magnesium oxide. Etched (xlOO) 37 30. S l i p l i n e s on unpolished magnesium oxide rod (xlOO) 37 31. Fracture surface between n i c k e l drop and magnesium oxide p l a t e l e t (x250) ; 38 32. Relative amounts of n i c k e l as indicated by i n t e n s i t y of n i c k e l Ka l i n e a. Drop melted o f f . (Undeformed rod) b. Drop melted o f f . (Deformed rod). c. Drop sheared off.(Undeformed r o d ) . d. Drop sheared off.(Deformed rod). . . . . . . . . 41 Figures (cont'd.) No. Page 33. S l i p systems i n magnesium oxide a. Screw d i s l o c a t i o n s b. Edge d i s l o c a t i o n s 44 34. Nickel-magnesium binary system 61 35. Surface tensions e q u i l i b r i a . 62 36. X-ray powder patterns Nos. 1502 and 1528. (Top i n t e r f a c i a l m a t e r i a l , bottom pure magnesium oxide) . . . . 70 37. A.S.T.M. Card No. 4.0829 f o r magnesium oxide 72 38. A.S.T.M. Card No. 1,126$ f o r magnesium-nickel compound (Mg 2Ni) . 72 39. A.S.T.M. Card No. 7.239 f o r magnesium hydroxide 73 40. X-ray powder pattern No. 1530 of i n t e r f a c i a l material between magnesium oxide and n i c k e l + 2% n i c k e l oxide . ; 74 41. A.S.T.M. Card No. 4.0850 f o r n i c k e l 76 42. A.S.T.M. Card No. 4.0835 f o r n i c k e l oxide 76 43. Drop with necessary dimensions to cal c u l a t e wetting angle . . . 78 TABLES No. Page I . Analysis of High P u r i t y O p t i c a l Grade MgO 8 I I . Analysis of N i c k e l Materials 10 I I I . W e t t a b i l i t y Results 42 IV. Data from Film No. 1528 71 V. Data from F i l m No. 1530 75 DUCTILITY AND CHEMICAL REACTIONS AT THE INTERFACE BETWEEN NICKEL AND MAGNESIUM OXIDE SINGLE CRYSTALS. I . INTRODUCTION A. General Purpose and Scope Recently attempts have been made to combine the properties of metals and ceramics to produce a material s u i t a b l e f o r use at high temperatures. At these temperatures, metals often f a i l as materials of construction through low-strength, low creep resistance or poor resistance to oxidation or other attack. In contrast, ceramics at these temperatures have i n general good strength, creep resistance and oxidation resistance but as compared with metals they are b r i t t l e . I f we could combine these properties of metals and ceramics, the ma t e r i a l produced would have very d e s i r a b l e properties f o r high temperature a p p l i c a t i o n . In the development of these materials the major emphasis thus f a r has been on the development of u s e f u l materials on a mainly empirical b a s i s . The art of preparing the materials, commonly r e f e r r e d to as 'cermets', i s f a r i n advance of a s c i e n t i f i c understanding of the fundamental f a c t o r s involved, 1=8 These factors include metal-ceramic reactions, surface and i n t e r f a c i a l energies and constituent properties such as strength, d u c t i l i t y , oxidation r e s i s t a n c e and thermal conductivity. In t h i s paper the author reports on f u r t h e r i n v e s t i g a t i o n s of the reactions between metal and ceramics. P h y s i c a l wetting o f the ceramic by the' metal i s a major f a c t o r i n producing an e f f e c t i v e bond. This can be studied by considering the - 2 -i n t e r f a c i a l energy involved. A low i n t e r f a c i a l energy i s equivalent to strong bonding, although other f a c t o r s , such as i n t e r f a c i a l stresses a r i s i n g as cooling, a f f e c t the bond r e s u l t i n g at room temperature. For metals and ceramics-of i n t e r e s t , l i t t l e i s known of the i n d i v i d u a l surface energies, much less of the i n t e r f a c i a l energies and f a c t o r s a f f e c t i n g them. At high temperatures the sessile-drop method^*! 0,!! c a n be employed very e f f e c t i v e l y f o r measurements of t h i s type. The angle of contact or wetting angle i s then an i n d i c a t i o n of wetta*» b i l i t y of the ceramic by the metal. The important reactions between the metal and ceramic are the forma-t i o n of chemical compounds at or near the i n t e r f a c e . Modifications of the materials present at the i n t e r f a c e change the bonding f o r c e s . The chemical reactions which modify the i n t e r f a c e depend on the thermodynamics of the system. The amount of these reactions depend on the atmosphere as w e l l as on the composition. Unfortunately thermodynamic data i s u s u a l l y incomplete and a study of i n t e r f a c i a l compounds must often be accomplished by experimental methods only. Strong bonding between metal and ceramic can also be af f e c t e d by the a b i l i t y of the metal to d i f f u s e into the ceramic.&*12 T ^ i s w a y a strong bond develops between the metal and the metal-enriched ceramic. The properties of i n t e r e s t i n cermet bodies include strength and creep resistance at elevated temperatures, corrosion and oxidation resistance, thermal expansion and thermal shock r e s i s t a n c e . The i n d i v i d u a l components and composition employed have a considerable e f f e c t on these properties. Many prop e r t i e s , however, cannot be evaluated on the basis of the properties of the constituents. Strength and creep at elevated temperatures are af f e c t e d by the - 3 -r e f r a c t o r i n e s s of the metal component employed. It i s therefore expected that such composition as thoria-tungsten 1^ are expected to maintain t h e i r strengths at elevated temperatures, even though t h e i r room temperature strength may be l e s s than other compositions. The f a c t that the strengths of cermet bodies 2 may be higher than the component strengths has not been completely explained. Corrosion and o x i d a t i o n resistance of cermets depends on the resistance of the components and on the nature of the products formed. Oxida-t i o n resistance generally i s due to the formation of an oxidation layer which prevents f u r t h e r oxidation. One of the major l i m i t a t i o n s of ceramic materials f o r many a p p l i c a -t i o n s i s t h e i r poor thermal-shock resistance as compared with metals. Thermal-shock resistance increases with high thermal conductivity, high t e n s i l e strength, low c o e f f i c i e n t of expansion and low modulus of e l a s t i c i t y . I f , however, the ceramic concerned possessed a c e r t a i n degree of d u c t i l i t y so that i t would deform p l a s t i c a l l y , t h e cermet produced would be f a i r l y r e s i s t a n t to thermal shock. In addition, i f the metal and ceramic would have c o e f f i c i e n t s of thermal expansion which are reasonably w e l l matched, the stresses caused by differences i n expansion of the components would be low enough such that b r i t t l e f r a c t u r e can not occur. A survey of the available metals and ceramics show that a combination of magnesium oxide and n i c k e l may ;have some of the desirable properties as discussed above. Magnesium oxide has a high melting point,^4 i s chemically stable and i t has been shown that s i n g l e crystals- e x h i b i t consider-able d u c t i l i t y . ^ 5 - 1 9 N i c k e l has a reasonably high melting point (1455°C)j i t has good resistance to chemical attack and i t s c o e f f i c i e n t of expansion 20 21 almost i d e n t i c a l with the c o e f f i c i e n t of expansion of magnesium oxide. > Consequently, these materials were chosen f o r a study of the in t e r a c t i o n s between metals and ceramics„ B. Previous Work, Nickel-magnesia i n t e r a c t i o n s were i n t e n s i v e l y investigated by McFarlane, J r . , 2 2 also by Kingery,^ I t was found by melting a n i c k e l p e l l e t on a p o l y c r y s t a l l i n e magnesia block that the adherence of n i c k e l t o magnesia was e x c e l l e n t . The wetting angle was approximately 110°, while between the n i c k e l and magnesia a sharp i n t e r f a c e was found to e x i s t . McFarlane^2 noticed a black d i s c o l o r a t i o n around the magnesia, which confirms the r e s u l t obtained by Economos and Kingery^ who obtained a black c o l o r a t i o n with n i c k e l and magnesia although i n t h e i r case i t concentrated i t s e l f around the grain boundary. The high bond strength between magnesia and n i c k e l could not be explained s a t i s f a c t o r i l y . The p o s s i b i l i t y of the formation of a compound such as n i c k e l oxide or the formation of a s o l i d s o l u t i o n such as n i c k e l oxide-magnesium oxide were investigated by X-ray analysis but no evidence f o r the existence of a compound or s o l i d s o l u t i o n could be detected. I t was f e l t that the presence of n i c k e l oxide could be responsible f o r the bond between magnesium oxide and n i c k e l . Magnesium oxide and n i c k e l oxide form a continuous range of 23 s o l i d solutions as shown i n F i g . 1, The bond would then e x i s t of a l a y e r of n i c k e l oxide adhering to the magnesium oxide and the n i c k e l adhering to the n i c k e l oxide. However, the amount of n i c k e l oxide present may be too small to be detected by X-ray a n a l y s i s . I n v e s t i g a t i o n s 2 ^ * 2 ^ of the i n t e r a c t i o n between metal vapours and s i n g l e c r y s t a l s of magnesium oxide have shown that the magnesium'oxide i s not attacked by l i t h i u m , sodium, potassium and calcium. Copper, however, etched the surface of the c r y s t a l s of magnesium oxide. The reactions between poly-c r y s t a l l i n e magnesium oxide and r e f r a c t o r y metals such as molybdenum, s i l i c o n , - 5 --Fig, 1„ MgO=NiO binary system 0 •titanium and zirconium showed that the metal penetrated the magnesium oxide along the grain boundaries and i n a l l cases but molybdenum, reacted with the magnesium oxide by eit h e r forming an I n t e r f a c i a l layer or by corroding the surface of the magnesium oxide„ The d i f f u s i o n of metals into s i n g l e c r y s t a l s of magnesium oxide was 26 investigated by Turnbull, I t was found that ruthenium, copper and barium do not d i f f u s e i n t e r s t i t i a l l y . Iron, n i c k e l and cobalt, however, do d i f f u s e and the d i f f u s i o n c o e f f i c i e n t s were determined. The mechanical properties of magnesium oxide are cur r e n t l y being 15-19 investigated. I t was observed that single c r y s t a l s of magnesium oxide exhibit considerable d u c t i l i t y and i t was f e l t that t h i s could lead to the development of a d u c t i l e c e r a m i c , T o determine the p l a s t i c properties, the single c r y s t a l s of magnesium oxide were subjected to tension, bending and compression. I t was observed2''' that magnesium oxide i s most d u c t i l e / when under compression and least when under tension. This i s at t r i b u t e d to - 6 -the f a c t that under tension i t i s easier f o r crack formation to occur which then leads to b r i t t l e f r a c t u r e . Johnston, Stokes and L i 1 ^ investigated the fract u r e mechanism and found the d u c t i l i t y to be a function of the p l a s t i c " " deformation already present explained by the f a c t that regions of p l a s t i c flow act as barriers to the propagation of cracks. Investigations of the temperature dependence of the d u c t i l i t y showed that the magnesium oxide can be deformed to 27 a much greater extent before b r i t t l e f r a c t u r e occurs at high temperatures (1500°C) than at low temperatures. The e f f e c t of surface condition on the d u c t i l e properties was 18 investigated by Stokes, Johnston and L i , who introduced s l i p sources i n the magnesium oxide by s p r i n k l i n g the surface with 46-niesh carborundum p a r t i c l e s . I t was observed that s l i p occurred at a lower stress l e v e l i n sprinkled c r y s t a l s than i n unsprinkled c r y s t a l s . Atmospheric conditions d i d not seem to a f f e c t the p l a s t i c properties as had been observed i n materials of the same c r y s t a l structure such as potassium c h l o r i d e . Other observations, however, seem to contradict t h i s . The mechanical properties of nickel-magnesia cermets were i n v e s t i -22 gated by McFarlane, J r * He found that p a r t l y oxidized specimens were generally stronger than unoxidized ones. Most s i g n i f i c a n t was that thermal shock d i d not lead to f r a c t u r e . It was also found that the transverse strength of specimens subjected to series of thermal shock had increased rather than decreased. C. S p e c i f i c Aims of the Present Investigation. The i n i t i a l purpose of t h i s i n v e s t i g a t i o n was to study the i n t e r -a c t i o n between n i c k e l metal and s i n g l e c r y s t a l s of magnesium oxide i n order to be able to draw conclusions about the bonding mechanisms between metals and - 7 -ceramics i n general. From the work done on the nickel-magnesium oxide system i t could be concluded that l i t t l e was known of the inte r a c t i o n s between n i c k e l and magnesium oxide. The s p e c i f i c aim of the present i n v e s t i g a t i o n was to study the interactions by a l l possible techniques such as X-ray analysis of reaction products at the i n t e r f a c e , metallographic examination of perpendicular and tapered sections through the i n t e r f a c e and X-ray fluorescence to study the d i f f u s i o n of the n i c k e l into the magnesium oxide. In addition, measurements of the w e t t a b i l i t y of the magnesium oxide by the l i q u i d n i c k e l were planned. - 8 -I I . EXPERIMENTAL A. Materia l s . 1. Magnesium oxide. The single c r y s t a l s of magnesium oxide used throughout the i n v e s t i g a -t i o n were transparent.High P u r i t y O p t i c a l Grade Magnesium Oxide supplied by Norton Company. This m a t e r i a l i s available i n large i r r e g u l a r pieces with dimensions approximately 3x2x2 cms, such as shown i n F i g . 6, Table I shows a t y p i c a l chemical a n a l y s i s . Q u a l i t a t i v e analysis by means of X-ray fluorescence„ showed a trace of iron only. TABLE I M a t e r i a l Percentage % SiO ?, less than ,20 $ Fe^Oa less than- .15 % CaO less than .20 % AI3O3 l e s s than .35 % MgO balance Examination with p o l a r i z e d l i g h t revealed the occasional occurr-ence of i n t e r n a l s t r e s s e s . The seemed to be l o c a l i z e d and were thought to be due to the existence of i n c l u s i o n s . The c r y s t a l structure of magnesium oxide i s the rocksa l t - s t r u c t u r e and i s based on two interpenetrating face-centered cubic l a t t i c e s , the l a t t i c e s i t e s of the one l a t t i c e occupied by magnesium ions and the l a t t i c e s i t e s of the other l a t t i c e occupied by oxygen ions. Because of i t s c r y s t a l structure i t possesses perfect cleavage u s u a l l y along {l00} planes and oc c a s i o n a l l y along - 9 -{ i l l } planes. In t h i s i n v e s t i g a t i o n use was made of t h i s property i n the preparation of t h i n p l a t e l e t s on top of which the n i c k e l could then be melted i n order to study i t s i n t e r a c t i o n with the magnesium oxide (see F i g . 6). A t y p i c a l (100) cleavage face i s shown i n F i g . 2. The l i n e s running diagonally across the picture are so-called 'cleavage s t e p s ' 2 ^ and are produced when cleavage i s started on more than one cleavage plane and by the i n t e r s e c t i o n of screw d i s l o c a t i o n s and suitable subgrain boundaries by the cleavage crack. Their height ranges from a few Angstroms to approximately 0.1 micron. F i g . 2. T y p i c a l (100) cleavage face of magnesium oxide (xlOO). N i c k e l The n i c k e l used i n t h i s i n v e s t i g a t i o n was supplied by S h e r r i t t Gordon Mines Ltd. and Mond Nickel Co. Ltd., and was i n the form of powder, and i n the case of S h e r r i t t Ni No. C525 also i n the form of r o l l e d s t r i p , approximately .05 cm. t h i c k . The chemical analyses of the various nickels are - 10 -given i n Table I I , TABLE I I  Analyses of N i c k e l Materials Analysis (maximum percent) M a t e r i a l Ni Co Cu Fe S C S h e r r i t t Ni No. C525 Bal. .08 0.029 0.022 0.016 0.01 S h e r r i t t high p u r i t y Ni Ba l . .008 0.009 .004 .006 .08 S h e r r i t t high Sulphur N i B a l . .061 .013 .011 .020 .011 Mond carbonyl N i Ba l . .005 .008 0.007 .002 0.091 B„ Apparatus The apparatus was designed with the f o l l o w i n g objectives: 1. to produce temperatures as high as 1800°C by induction heating and -5 vacuums as high as 5 x 10 mm. of Hg. 2. to permit the i n t r o d u c t i o n of reducing atmospheres at pressures less than atmospheric pressure and 3. to allow measurements o f drops at temperatures greater than the melting points of the a l l o y s used. The equipment i s shown i n F i g s , 3 and 4 . The apparatus assembly of the induction furnace vacuum system and o p t i c a l system OPTICAL SYSTEM FURNACE ASSEMBLY (1) around glass or photographic piste. (12) Vycor tube. (2) Vertical adjustment sere*. (U) Induction c o i l . (5) Horizontal adjustment track. W Heating element, radiation shield CO Focussing screw. (15) Thermocouple gauge. (5) Adjustable bellows. (16) Ionization gauge. (6) Ocular lens. (17) Gas inlet control. (7) - Objective lens, shutter and i r i s dlaphram. (18) Viewing window. (6) Vertical adjustasent screw. (19) Brass fittings. (9) Hater-cooled optical f l a t . (20) Optical pyrometer. (10) Water-cooled brass f i t t i n g . (21) Light source Interchangeable with (") Magnetic shutter. I F iff° 4, Assembly drawing of the apparatus. ••- 13 -1<> Furnace The heating elements of the furnace were constructed from 0,005 inch molybdenum sheet (see F i g . 5), The susceptor consisted of a closed loop i n the F i g , 5o Molybdenum heating elements (dismantled) of the induction furnace, form of a cylinder 0,75 inches i n diameter and 3°5 inches long, open at both ends, A r a d i a t i o n s h i e l d surrounded the susceptor to prevent excessive heat l o s s . This s h i e l d consisted of an 'open loop" cylinder 1,5 inches i n diameter and 5 inches long, also open at both ends. The susceptor and r a d i a t i o n s h i e l d assembly was supported inside the induction c o i l by a molybdenum-rod framework. This rod was separated from the vycor tubing containing the heating elements by sintered alumina i n s u l a t o r s . The furnace-tube assembly consisted of a vycor (fused s i l i c a ) t u b e , 2 1/2 inches i n diameter and 18 inches i n length. Rubber 0-ring seals and s i l i c o n e high vacuum grease were used i n the vycor-to-brass j o i n t s at the end of the tube. At the camera end of the furnace, a water-cooled o p t i c a l f l a t was - 14 -used to protect the camera lenses from heat r a d i a t i o n . Because of the high vapour pressures of metals at operating temperatures, a magnetic shutter was placed inside the furnace to protect the o p t i c a l f l a t from metal vapours. At the other end of the furnace, a viewing window was b u i l t into the elbow leading to the vacuum system. Outside t h i s window, a l i g h t source and an o p t i c a l pyro-meter were placed such that e i t h e r could be positioned on the centre l i n e of the furnace. Through a water-cooled, copper induction c o i l (1/4 inch O.D.tubing, 21 turns) around the outside of the vycor tube, power was supplied to the heating element by a high-frequency induction generator. This u n i t , Lepel model T-10-3, produces power from 0 to 23,5 K.V.A. at 400,000 cycles per second. The power could be adjusted by varying e i t h e r the plate current or the g r i d current i n the tubes of the generator. Because of the low thermal capacity the furnace could be heated from room temperature to 1500°C i n approximately three minutes and could be cooled from 1500"C to approximately 600°C i n less than one minute, when the power input was lowered instantaneously. Both the rate of heating and cooling would follow logarithmic laws and at 1500°C the rate of heating would be much lower than the rate of cooling. The temperature i n the furnace was measured to i5°C. by a Hartmann and Braun o p t i c a l pyrometer, model T0-10-e. E m i s s i v i t y corrections were not necessary because of the design of the susceptor. Power input to the furnace was l i m i t e d by the loss of strength of the vycor tube at elevated temperatures. The tube was, therefore, air-cooled by means of a 10 inch diameter fan. 2. Vacuum system. The pumping system was designed to produce a high vacuum or a c o n t r o l l e d low vacuum. A mechanical fore pump and a two stage o i l d i f f u s i o n pump produced vacuums i n the range of 10~^ to 10"^ mm. of Hg. With only the - 15 " fore pump operating, a reducing atmosphere could be continuously flushed through the system to maintain a vacuum of 0,5 nim. of Hg„ The high vacuums were measured with a N.R.C. i o n i z a t i o n gauge, type 507, and the low vacuums were measured by a N.R.C thermocouple gauge, type 501. P u r i f i e d reducing atmospheres were obtained by continuously f l u s h i n g dry hydrogen through the system. The hydrogen was supplied by Canada Liquid A i r Co., Ltd. i n 2,000 p s i tanks. The p u r i f i c a t i o n and drying t r a i n consisted of a hydrogen 'deoxo' cartridge i n se r i e s with columns of anhydrous calcium sulphate, s i l i c a g e l , and phosphorous pentoxide. The 'deoxo' cartridge converted the oxygen impurity to water. The s i l i c a g e l and anhydrous calcium sulphate reduced the water concentration i n the hydrogen to 0.005 milligrams per l i t e r of gas and the phosphorous pentoxide reduced the concentration f u r t h e r to less than 0 o0002 milligrams per l i t e r . 3. O p t i c a l system. At temperatures above 1100*0, objects emit s u f f i c i e n t r a d i a t i o n t o produce an ou t l i n e image on a photographic f i l m or ground glass p l a t e . Sharp images of the incandescent drops i n t h i s i n v e s t i g a t i o n could be r e a d i l y obtained at 1500°C i f a camera were c o r r e c t l y designed f o r t h i s purpose. A camera was constructed to give a t e n f o l d magnification. This was accomplished by the combination of a two-component objective lens ( f o c a l length of 27.5 centimeters) and a sing l e component, ocular lens ( f o c a l length of 4.2 centimeters)„ The f i x e d objective lens was equipped with an adjustable shutter and i r i s diaphragm. The ocular lens was separated from the objective lens by a 10 to 15 inch adjustable bellows, and from the ground glass plate by a 42 to 48 inch adjustable bellows. The camera was positioned so that the o p t i c a l axis was i n the plane of the ceramic surface and p a r a l l e l to the axis of the -16 -furnace by means of v e r t i c a l adjustment screws and h o r i z o n t a l adjustment track s . Because of the length of the camera, s e n s i t i v i t y to v i b r a t i o n s was high and a very sturdy framework was necessary. Focussing adjustments were made by means of a screw-thread adjust-ment on the ocular lens holder which moved p a r a l l e l to the o p t i c a l axis of the camera. Before the specimen was heated, preliminary focussing was accomplished with the a i d of a l i g h t source at the f a r end of the furnace. The l i g h t source produced a silhouette of the specimen on the ground glass. C. Preparation of M a t e r i a l s . The s i n g l e - c r y s t a l magnesium oxide was obtained as large i r r e g u l a r pieces such as shown i n F i g . 6. These were cut into rectangular pieces roughly along [lOCJ planes with a Felker 'Di-Met* diamond saw. These rectangular pieces were annealed at 1000°C i n order to f a c i l i t a t e - ^ cleavage. A f t e r annealing, the c r y s t a l s were cleaved by means of a small c h i s e l and hammer into p l a t e l e t s approximately one centimeter square and one millimeter t h i c k or into approximate ly square rods two to three centimeters long and 0.2 centimeter t h i c k as F i g . 6. Magnesium oxide 'as received' and cleaved c r y s t a l s with and without n i c k e l drops. i l l u s t r a t e d i n F i g . 6. Whenever necessary, as discussed under experimental r e s u l t s , the cleaved p l a t e l e t s or rods were polished by immersion i n t o hot phosphoric a c i d . A p o l i s h i n g time of one minute was s u f f i c i e n t to produce a microsc o p i c a l l y smooth surface. Care was taken i n handling the cleaved p l a t e -l e t s and rods to avoid contamination of the surface and the intr o d u c t i o n of p l a s t i c deformation. Metal tweezers were used whenever possible . To detect evidence of p l a s t i c deformation, as discussed under experimental r e s u l t s , a so l u t i o n composed of 30 percent hydrochloric a c i d and 10 percent ammonium chloride was used. The n i c k e l i n a l l experiments, but where mentioned, consisted of the S h e r r i t t Gordon No. C525 r o l l e d s t r i p . This was cut into a small s t r i p a few centimeters long and couple of millimeters wide. This was r o l l e d up lo o s e l y to permit deoxidation on a l l surfaces. Where n i c k e l powder was used i t was compacted i n a c y l i n d r i c a l d i e , 1/4 inch i n diameter, under a pressure of f o r t y tons per square inch. The compacting die was machined to form a compact of a shape required to insure an advancing contact a n g l e . 1 1 D. Experimental Procedure. In a l l experiments but where mentioned, the following procedure was used; the metal was placed on the magnesium oxide p l a t e l e t or rod insi d e the susceptor. With the a i d of the l i g h t source, the specimen was l e v e l l e d . The furnace was assembled and pumped to a fore pump vacuum of 5 x 10"" 3 m i l l i -meters of mercury. P u r i f i e d hydrogen was then flushed through the furnace at a rate s u f f i c i e n t to maintain a vacuum of 0.5 millimeters of mercury. The temperature was slowly r a i s e d to approximately 800°C. Near t h i s temperature, the hydrogen gas ion i z e d and hydrogen-ion cleansing of the specimen took place. A f t e r a heating period of ten minutes, the power was turned o ff and the system was pumped down to a vacuum of 10"-> millimeters of mercury. The - 18 -power was again slowly increased u n t i l the temperature reached 1500°C as measured by the o p t i c a l pyrometer. The specimen was maintained at t h i s tempera-ture f o r ten minutes. In the experiments t o determine the w e t t a b i l i t y of the magnesium oxide by l i q u i d n i c k e l the image of the n i c k e l drop was c a r e f u l l y focussed on the ground glass. The necessary dimensions t o determine the wetting angle were made by means of a p a i r of c a l i p e r s . For the X-ray d i f f r a c t i o n i n v e s t i g a t i o n to determine the formation of any compound or s o l i d s o l u t i o n the n i c k e l drops were sheared from the magnesium oxide. The layer of material adhering to the drop was scraped o f f by means of a small sapphire rod and prepared f o r i d e n t i f i c a t i o n by the X-ray powder method. The powder patterns were obtained by exposures to copper r a d i a t i o n (40 K.V., 15 m.a,) f o r two to f i v e hours. These patterns were indexed and compared with patterns of standard materials recorded i n the A.S.T.M. card index f o r powder patterns. The preparation of specimens to inve s t i g a t e the d i f f u s i o n of n i c k e l i n t o the magnesium oxide by means of X-ray fluorescence i s described under experimental observations. - 19 -I I I . EXPERIMENTAL RESULTS A. Observations on Interface. Perpendicular and tapered sections of the magnesium oxi d e - n i c k e l metal interface showed that the i n t e r f a c e consists of a very sharp l i n e . No evidence of any reaction product or intermediate phase could be found. F i g . 7 shows a tapered s e c t i o n of the i n t e r f a c e at an apparent magnification of 4500x. The step i n the i n t e r f a c e i s one of the higher cleavage steps, i t s height being approximately one micron. F i g . 7. Tapered section (Is6) of magnesium oxide-n i c k e l metal i n t e r f a c e (x750, apparent magnification i s 4500x). F i g , 8 shows a view obtained by looking through the magnesium oxide perpendicular onto the i n t e r f a c e . In a d d i t i o n to the cleavage steps, two perpendicular sets of p a r a l l e l s t r a i g h t lines,, also f a i n t l y v i s i b l e i n F i g , 7, can be seen. Their o r i e n t a -t i o n corresponds to the ^ 1 0 0 ^ c r y s t a l l o g r a p h i c d i r e c t i o n s of the magnesium oxide, - 20 -A l l the l i n e s of both sets were observed to be continuous from one side of the in t e r f a c e to the other. F i g , 8. Magnesium oxide-nickel i n t e r f a c e as seen through the magnesium oxide (x250). A dark f i e l d photomicrograph (see F i g , 9) shows i n addition to the l i n e s , a multitude of etch p i t s , some of them coincident with the l i n e s . F i g . 9» Same as F i g , 8 but d a r k - f i e l d i l l u m i n a t i o n (x250). - 21 -F i g . 10 shows the magnesium oxide-nickel i n t e r f a c e at a magnification of 2500x. This magnification was achieved by mechanically p o l i s h i n g the p l a t e l e t of magnesium oxide u n t i l a t h i n layer of magnesium oxide remained, such that at high magnifications the magnesium oxide p l a t e l e t would not i n t e r f e r e with the focussing of the microscope. Most of the etch p i t s seem to be near cleavage steps indicated by the f a i n t t races running diagonally across the f i g u r e . F i g . 10. Same as F i g . 8 (x2500) To determine whether these l i n e s were produced during cleavage or during the heating process, a f t e r cleavage a p l a t e l e t of the magnesium oxide was polished by b o i l i n g i n hot phosphoric a c i d u n t i l a l l surface i r r e g u l a r i t i e s , such as cleavage steps, were removed. A p o l i s h i n g time of approximately two minutes was s u f f i c i e n t . F i g . 11 i s a view of the i n t e r f a c e between the polished p l a t e l e t and a n i c k e l metal drop located near the center of the p l a t e l e t . It i s c l e a r that the l i n e s can s t i l l be seen. A view near the edge of the p l a t e l e t where the p o l i s h i n g was more severe, shows the l i n e s to be curved, some even crossing others. - 22 n : j /»• 1 c • * I ! 1 rn i • '\-\ i 1 r f 1 j - f - » J1 F i g . 11. Interface of n i c k e l metal and magnesium oxide p l a t e l e t polished before heating. Near center of p l a t e l e t (x250). 1 i • i riji , . ' i ! : j 'l\pffllfl|»flH .j J i. • f (If i i - f Llilkk .. 11 { * | ^  #•> • #•* 1 • I 1 • *) ' ii ! i i j i • * j \ 1 I j 11V 11 i | f i i i }<{* F i g , 12, Same as F i g . 11. but near edge of p l a t e l e t where p o l i s h i n g was more severe (x250). - 23 -Examination of the i n t e r f a c e between a n i c k e l drop and a p l a t e l e t , which was polished i n hot phosphoric a c i d f o r approximately f i v e minutes, showed that l e s s l i n e s seemed to appear on the int e r f a c e than on the i n t e r f a c e between a n i c k e l drop and an unpolished p l a t e l e t . No l i n e s could be detected on the i n t e r f a c e between a n i c k e l drop and a p l a t e l e t which was polished i n hot phosphoric a c i d f o r approximately ten minutes. Numerous specimens of p l a t e l e t s and rods with n i c k e l metal drops were made to see i f these l i n e s could be made to appear on faces other than the i n t e r f a c e between the magnesium oxide and n i c k e l drop. The l i n e s could be detected only on one specimen. On the side faces they were found to be p a r a l l e l to the length of the rod. On the bottom face they were perpendicular to the length of the rod. Occasionally l i n e s could be seen near the drop on the same cleavage face which supported the metal drop as shown i n F i g . 1 6 , An attempt was made to see i f the l i n e s could be detected more e a s i l y by s i l v e r i n g the cleavage faces of the magnesium oxide, but t h i s proved to be r e l a t i v e l y unsuccessful as the s i l v e r d i d not adhere too w e l l to the magnesium oxide i n contrast to the beaker which contained the s i l v e r i n g s o l u t i o n . I t was found that the s i l v e r d i d b r i n g out the l i n e s , but di d not bring out any a d d i t i o n a l l i n e s which could not be detected before s i l v e r i n g . To see whether other metals would produce the l i n e s , specimens were prepared with s i l v e r , copper, cobalt, aluminum and t i n . The temperature was , ra i s e d to approximately 50°C above.,the melting point of the respective metal. A l l metals, but the cobalt, which sheared o f f during the cooling, adhered to the magnesium oxide. In a l l cases no lin e s could be detected. However, heating the s i l v e r drop to 1 4 0 0 ° C r e s u l t e d i n the formation of the l i n e s . In contrast to the nickel-magnesium oxide i n t e r f a c e the interfaces of s i l v e r , copper, aluminum and t i n contained many gas occlusions. - 2k -A magnesium oxide p l a t e l e t was heated and cooled without any metal drop, A drop of mercury was then placed upon the p l a t e l e t and the i n t e r f a c e examined f o r l i n e s . The same was done using indium and gallium. In a l l three cases no l i n e s could be found. A n i c k e l drop was melted on a rod of magnesium oxide and kept at 1500°C t i l l the drop was evaporated away completely. On cooling no l i n e s could be observed on the previous l o c a t i o n of the drop. A n i c k e l drop was melted on a p l a t e l e t of magnesium oxide and heated to 1800°C. Examination of the i n t e r f a c e on cooling revealed no d i f f e r e n c e i n the l i n e s such as width or density compared with the l i n e structure formed on cooling from 1500°C. An attempt was made to see i f any r e l a t i o n s h i p might e x i s t between l i n e s on the i n t e r f a c e s between p l a t e l e t s on opposite sides of a cleavage crack and n i c k e l drops. One example could be found where a one-to-one correspondence seemed to e x i s t between approximately s i x neighboring l i n e s . To check i f any one-to-one correspondence might e x i s t between l i n e s on the i n t e r f a c e s between d i f f e r e n t drops and the same c r y s t a l , a specimen was made which consisted of a cleaved rod supporting four drops of n i c k e l metal. However, no evidence of a one-to-one correspondence could be detected. To see whether the s i z e of p l a t e l e t has any e f f e c t on the nature of the l i n e s , such as density, a - n i c k e l drop was melted on a p l a t e l e t of magnesium oxide cleaved "to a thickness of approximately 1/k inch. Examina-t i o n of the i n t e r f a c e showed that' the number of l i n e s seems to be approxima-t e l y the same as f o r thinner p l a t e l e t s . - 25 -In order to determine the e f f e c t of the cleavage steps on the nature of the l i n e s a rod was selected which on one face had only two cleavage stepso When the interface was examined a f t e r a n i c k e l drop was melted on i t , no l i n e s , such as shown i n F i g . 8, could be detected. However, as shown i n F i g . 13, f a i n t <100> traces can be seen. F i g . 13. Interface between n i c k e l drop and p l a t e l e t of magnesium oxide without cleavage 3teps (x300). In only two of the many specimens made, evidence of cracking i n the magnesium oxide could be detected. This took the form of ^ i O O ^ and<^110^> cracks extending from the top surface approximately half way into the p l a t e l e t 0 In one case the drop was surrounded by an octagon of cracks composed of four <^  1O0Scracks and four<^"110^> cracks. B 0 Heating Rate. To study the e f f e c t of the rate of heating and cooling, a p l a t e l e t of magnesium oxide with a n i c k e l drop was subjected to a number of heating cycles, i n each cycle heating and cooling f a s t e r than i n the previous one. F i g s . 14 and 15 show the same area of the i n t e r f a c e a f t e r the f i r s t and second - 26 -Fj-g« Ik Magnesium oxide-nickel i n t e r f a c e a f t e r f i r s t slow heating cycle (x250). Fi£_j_15_„ Same area of in t e r f a c e as i n F i g . 1L a f t e r second f a s t e r heating c y c l e . - 27 = heating c y c l e . In the f i r s t cycle the temperature was raised to 1500°C i n approximately ten minutes, while cooling took approximately f i f t e e n minutes. In the second cycle both these times were halved. By comparing both f i g u r e s i t can be noticed that during the second heating cycle a d d i t i o n a l l i n e s appeared. In a t h i r d cycle, i n which the temperature was r a i s e d i n approxima-t e l y three minutes and lowered by instantaneously switching o f f the power, no a d d i t i o n a l l i n e s could be detected. Repetition of these experiments showed that these r e s u l t s were reproducible. To study the e f f e c t of thermal c y c l i n g on the l i n e s , a p l a t e l e t with drop was subjected to f i f t e e n heating cycles each time r a i s i n g and lowering the power instantaneously. The in t e r f a c e was observed a f t e r the f i r s t , f i f t h , tenth and f i f t e e n t h cycle. No change i n the l i n e s such as width or i n t e n s i t y could be detected. To subject the magnesium oxide to high thermal shock a rod without . metal drop was heated to 1850°C and cooled q u i c k l y by lowering the power input instantaneously. Another rod was heated ( i n a i r ) to a cherry red and immediate-l y plunged into l i q u i d nitrogen. On both rods a few l i n e s appeared a l l running along the length of the rods. No evidence of fra c t u r e could be detected. C. Etching by Liquid N i c k e l . F i g s , 16 and 26 show evidence of attack of magnesium oxide by l i q u i d n i c k e l as indicated by the concentric curves along the perimeter of the drop. In an experiment where a n i c k e l drop was evaporated completely from a p l a t e l e t the former l o c a t i o n of the drop showed many concentric r i n g s . Other evidence f o r attack i s the occurrence of etch p i t s as shown i n F i g s . 9 and 12. It was found that l i q u i d n i c k e l also attacks magnesium oxide at subgrain boundaries but t h i s cannot be detected v i s u a l l y unless the magnesium - 28 -oxide and the n i c k e l are held at 1500°C f o r at l e a s t t h i r t y minutes. D 0 Polarized Light. -The use of p o l a r i z e d l i g h t revealed the existence of i n t e r n a l s t r a i n . F i g . 16 shows the i n t e r f a c e between a drop and a rod of the magnesium oxide. A s i m i l a r pattern i s obtained viewing the i n t e r f a c e between a drop and a p l a t e l e t . No such pattern could be detected with a p l a t e l e t 1/4 inch t h i c k . F i g . 16. Interface between n i c k e l drop and rod of magnesium oxide under p o l a r i z e d l i g h t (x25) As shown i n F i g . 17 the side face of a rod i n the v i c i n i t y of the drop revealed i n addition to an o v e r a l l pattern, as indicated by the two curved bands, an array of i n d i v i d u a l narrow bands oriented i n ^ 1 1 0 ^ , some extending nearly to the opposite face. No evidence of s t r a i n could be detected i n the v i c i n i t y of the drop on the marnesium oxide rod which did not possess any cleavage steps. - 2 9 -F i g . 17. Side face of magnesium oxide rod with n i c k e l drop on top surface under.polarized l i g h t (x25) E, Chemical Etching. Magnesium oxide p l a t e l e t s were chemically etched to bring out d i s l o c a t i o n s , subgrain boundaries and any p l a s t i c deformation. F i g . 18 shows F i g . 18. Chemically etched cleavage face of magnesium oxide (xlOO) an etched p l a t e l e t with an array of two perpendicular sets of p a r a l l e l rows - 30 -of etch p i t s oriented in<1.00^. The diagonal l i n e from the lower l e f t to middle r i g h t i s a subgrain boundary. The other l i n e s running approximately i n a ^ 1 1 0 ^ d i r e c t i o n are the remains of cleavage steps. F i g . 19 shows a high density of etch p i t s near one of the higher cleavage steps. I t may be noticed that the etch p i t s seem to be i n a broad band p a r a l l e l to the cleavage step. of arrays of etch p i t s such as shown i n F i g . 19 are r e l a t i v e l y rare, many cr y s t a l s not having any. P l a t e l e t s cleaved from c r y s t a l s , which were not annealed, have more of these -^.00^- rows of etch p i t s than p l a t e l e t s cleaved from annealed c r y s t a l s , A rod which did not have any cleavage steps on one face did not show any arrays of ^ ^ 1 0 0 ^ etch p i t s . of F i g . 18 was continuously etched and observed at i n t e r v a l s . I t was found that the etch p i t s making up the arrays became deeper and l a r g e r , then became f l a t bottomed and f i n a l l y disappeared altogether. Most of the etch p i t s F i g . 19. Chemically etched cleavage face, near high cleavage step (xlOO), In examining many etched c r y s t a l s i t was found that the occurrence To examine the nature of these arrays of etch p i t s the c r y s t a l - 31 -became f l a t bottomed when a few microns deep, the deepest being approximately from f i v e to ten microns dees. The shortest distance between two neighboring etch p i t s was approximately one and a hal f to two microns. It was also observed that these ^100^> arrays of etch p i t s seemed to be etched f a s t e r than grown-in d i s l o c a t i o n s or subgrain boundaries. Etching the c r y s t a l which was opposite the c r y s t a l of F i g . 18 at cleavage showed that the same etch p i t s do not occur (see F i g . 20). This has also been observed i n F i g . 20. Opposite cleavage face from F i g . 18 (xlOO). many other c r y s t a l s . S imilar arrays of etch p i t s are v i s i b l e i n F i g . 9. To see whether the existence of these arrays of etch p i t s i s a condition f o r the formation of l i n e s , a n i c k e l drop was melted on a p l a t e l e t of magnesium oxide which a f t e r a l i g h t etch showed subgrain boundaries but no evidence of etch p i t s . Observing the i n t e r f a c e showed the existence of these l i n e s with a l i n e density comparable with that of F i g . 8. To see whether a r e l a t i o n s h i p e x i s t s between the etchpits produced - 32 -by chemical etching and the l i n e s produced when a n i c k e l drop i s melted on a p l a t e l e t , a p l a t e l e t was given a l i g h t etch u n t i l etch p i t s appeared. A n i c k e l drop was then melted on the p l a t e l e t and the l i n e s under the drop were then compared with the etch p i t s i n the same area. F i g . 21 shows the p l a t e l e t a f t e r chemical etching. F i g . 22 shows the same area, as seen through the p l a t e l e t , F i g . 21. Etched cleavage face of magnesium oxide.(xlOO) r F i g . 22. Same area as i n F i g . 21 but with n i c k e l drop, as seen through the p l a t e l e t , and therefore i t s mirror image (xlOO). - 33 -a f t e r a n i c k e l drop was melted on i t . Two examples can be found where a l i n e continues at the end o f a row of etch p i t s . In a d d i t i o n , a l i n e can be seen to go through two s i n g l e etch p i t s , as indicated by the lower arrow. I t can be seen that the presence of a subgrain boundary does not seem to i n t e r f e r e with the formation of the l i n e s . Chemical etching of a rod with a n i c k e l drop covering the complete width of the top cleavage face revealed evidence of p l a s t i c deformation on a l l the other faces, but i n the v i c i n i t y of the drop only. On the side faces two ^110 ^> traces of s l i p l i n e s could be seen and traces of ^ 1 0 0 ^ s l i p l i n e s p a r a l l e l to the top surface could be seen. On the bottom face Only 0-00^ •'. t r a c e s , oriented perpendicularly to the length of the rod, could be detected. F. Nature of Deformation. In order to compare the l i n e s seen on the i n t e r f a c e between the magnesium oxide and the n i c k e l drop with the deformation of a p l a s t i c a l l y deformed magnesium oxide c r y s t a l , a rod was slowly bent by hand t i l l f r a c t u r e occurred. F i g . 23 i l l u s t r a t e s the four faces of the rod at the fractured end. On the tension and compression sides of the rod the l i n e s , which l i e i n (100), F i g . 23. Faces of broken rod of magnesium oxide near fracture end. Unetched (xl5). - 34 -are more or le s s continuous across the face, whereas on the side faces the l i n e s extend inward from the tension and compression faces gradually fading out towards the neutral a x i s . F i g . 24 shows the same rod a f t e r etching. On the four faces addi t i o n a l <!llO^> l i n e s can be seen. As i n t h e case of the <ClOO> li n e s they are continuous across the tension and compression faces and on the side faces extending inward towards the n e u t r a l a x i s . r t \ F i g . 24, Faces of broken rod near fr a c t u r e end. Etched. (xl5). Examining a side face of the rod near the point of f r a c t u r e with p o l a r i s e d l i g h t reveals arrays of bands oriented i n ^ 110^, as shown i n F i g . 25. In order to etch the i n t e r f a c e between a n i c k e l drop and a rod of magnesium oxide the drop was removed by melting i t and allowing i t to r o l l o f f the rod. The rod was then chemically etched. To check the effectiveness of the etchant, some p l a s t i c deformation was introduced d e l i b e r a t e l y i n one end of the rod. F i g , 26 shows that no ^ 1 1 0 ^ - t r a c e s can be seen on the former l o c a t i o n of the drop. In the p l a s t i c a l l y deformed region of the rod ^ l l O ^ s l i p l i n e s could be found, some of them extending into the former l o c a t i o n of the drop showing that ^110^ s l i p l i n e s could be etched. F i g . 25o End of broken rod under polarized l i g h t (x30)„ F i g . 26, Area on rod from which n i c k e l drop was removed by melting. Unetched (xl25). It was observed that the usual f i f t e e n minute etching time was not s u f f i c i e n t t o bring out the <^100^ rows of etch p i t s , whereas a f t e r f i f t e e n minutes the p l a s t i c deformation which was d e l i b e r a t e l y introduced showed up very clearly,, - 3 6 -F i g . 2 7 also shows that l i t t l e or ro etching occurred near the perimeter of the former l o c a t i o n of the n i c k e l drop. Before etching, the l i n e s were observed to be continuous fcc the perimeter as shown i n F i g , 2 6 . F i g . 27. Same as F i g . 26. Etched t h i r t y minutes ( x 7 5 0 ) „ In bending the rods i t was observed that those rod which possessed a face which was cut by the diamond saw were more d u c t i l e than those rods which possessed four cleavage faces. A l l of the rods which possessed a cut face could be bent t c the degree as shown i n Fig.28, Only a few of the F i g . 28. Bent rods of magnesium oxide, i l l u s t r a t i n g d u c t i l i t y . - 37 -cleaved rods which possessed four cleavage faces were as d u c t i l e as t h i s , most of them f r a c t u r i n g a f t e r a small amount of bending. To examine the nature of f r a c t u r e of a polished rod, a rod was polished i n hot phosphoric acid f o r ten minutes a f t e r which i t was fr a c t u r e d by bending. The rod was then chemically etched. The r e s u l t i s shown i n Fig,29, F i g . 29. S l i p l i n e s on polished and broken rod of magnesium oxide. Etched 15 mins. (xlOO). F i g . 30 shows an area of the rod of F i g . 24. I t can be seen that the traces F i g . 30. S l i p l i n e s on unpolished magnesium oxide rod. (xlOO) - 38 -i n the polished rod are of a d i f f e r e n t nature than the traces i n the unpolished rod. G. Strength of Bond To te s t the strength of the bond formed between the n i c k e l metal and the magnesium oxide cleavage face, a drop was sheared from a p l a t e l e t to see where fracture would occur. F i g . 31 shows the fr a c t u r e surface of the drop. By the two sets of cleavage steps i t can be seen that fracture occurred through the magnesium oxide 0 I V -L S I \\ S i - , V \ % i\ F i g . 31. Fracture surface between n i c k e l drop and magnesium oxide p l a t e l e t . (x250). A perpendicular section through the drop showed that the la y e r of magnesium oxide sheared o ff with the drop was approximately twenty microns t h i c k . A drop of n i c k e l composed of S h e r r i t t Grade C 525 and 2 percent n i c k e l oxide was sheared from the magnesium oxide. Fracture occurred at the interface rather than through the magnesium oxide. A drop of n i c k e l was melted on a p l a t e l e t of magnesium oxide by - 39 -bringing the temperature up to the melting point of n i c k e l and holding i t there approximately ten seconds a f t e r the n i c k e l melted. Shearing the drop from the p l a t e l e t revealed that fracture occurred mainly at the i n t e r f a c e and o c c a s i o n a l l y -through the magnesium oxide. H. Debye-Scherrer X-ray Analysis of I n t e r f a c i a l M a t e r i a l . X-ray analysis of the i n t e r f a c i a l m a t erial was made by the Debye-Scherrer powder method. Three materials were investigated? 1, the i n t e r f a c i a l material between magnesium oxide and S h e r r i t t C 525 n i c k e l held at 1500°C f o r one hour. Here evidence was found f o r the existence of a magnesium n i c k e l compound. No evidence could be found f or the presence of n i c k e l oxide or magnesium, (For powder patterns and ca l c u l a t i o n s see Appendix I ) . 2, the i n t e r f a c i a l m a t e r i a l between magnesium oxide and S h e r r i t t C 525 n i c k e l to which 2% n i c k e l oxide had been added. Here only n i c k e l and n i c k e l oxide could be detected. No evidence f o r the existence of magnesium oxide could be found, supporting the experimental observation that when removing the drop from the magnesium oxide, f r a c t u r e occurred at the in t e r f a c e rather than through the magnesium oxide as happens when removing a drop of pure n i c k e l from magnesium oxide. (For powder pattern and calculations see Appendix I I ) , 3, the top layer of the magnesium oxide from which the n i c k e l drop had been removed by-melting. No evidence f o r the presence of n i c k e l or magnesium-n i c k e l compound was found. I. D i f f u s i o n of N i c k e l into Magnesium Oxide. The r e l a t i v e amounts of n i c k e l which d i f f u s e d into the magnesium oxide while at 1500°C were determined by removing the n i c k e l drop from the - 40 -magnesium oxide and by determining the amount of n i c k e l by measuring the i n t e n s i t y of the n i c k e l K-a l i n e by X-ray fluorescence. The specimens were prepared by taking two rods from opposite sides of a cleavage crack. N i c k e l drops of equal weight were then melted on them. One specimen was brought to 1500°C, then cooled so as to introduce p l a s t i c deformation (see under Discussion) and again brought to 1500°C and held at t h i s temperature f o r f i f t e e n minutes. The other specimen was just brought to 1500°C and held there f o r an equal length of time. Enough specimens were prepared such that the drops could be removed from the rods i n three ways: 1 0 f o r c i b l e removal by means of shearing them o f f . 2. by melting and allowing them to r o l l o f f . 3o by holding the specimens at 1500°C t i l l a l l the n i c k e l had evaporated away. The remaining rods were then placed into the X-ray fluorescence machine and the r e l a t i v e i n t e n s i t i e s were plotted g r a p h i c a l l y . The r e l a t i v e peak heights are an i n d i c a t i o n of the r e l a t i v e q u a n t i t i e s of n i c k e l i n each rod. The r e s u l t s f o r removal of the n i c k e l drops by shearing and melting are i l l u s t r a t e d i n Fig.32. In the rods from which the n i c k e l drops were evaporated only a trace of n i c k e l could be detected, which were i n approximately equal amounts 0 J , W e t t a b i l i t y The r e s u l t s of the determination of wetting angles between various n i c k e l materials and magnesium oxide i n the polished or 'as cleaved condition' are shown i n Table I I I . The wetting angles were calculated from the drop dimensions. Measurements were made approximately 1 minute and 20 minutes a f t e r the n i c k e l melted. In the case of the S h e r r i t t C525 + 2% NiO, i t was observed - 41 -3.2,. Relative amounts of n i c k e l as indicated by i n t e n s i t y of n i c k e l K-a l i n e . (Relative peak heights indicate r e l a t i v e amounts). a. Drop melted o f f . b. Drop melted o f f . c. Drop sheared o f f . d; Drop sheared o f f . (Undeformed rod) (Deformed rod) (Undeformed rod) (Deformed rod). that immediately upon melting, the n i c k e l oxide segregated at the surface of the drop and formed what seemed to be a s o l i d coating around the l i q u i d n i c k e l . A f t e r approximately t h i r t y minutes t h i s coating had disappeared but f o r an occasional piece which could be seen to rotate around the drop. This was thought to be due to the s t i r r i n g e f f e c t by the applied magnetic f i e l d . - 42 -TABLE III W e t t a b i l i t y Results. Wetting angle i n degrees Polished 'As cleaved' time a f t e r melting (min.) time a f t e r melting (min.) M a t e r i a l 1 20 1 20 S h e r r i t t C grade No.525 102.1 104.6 103.6 98.9 High p u r i t y n i c k e l 107.4 103.8 101.6 103.2 Carbonyl 114.^ 5 95.5 108.3 102.2 S h e r r i t t high sulphur 103.6 99.1 102.1 102.0 S h e r r i t t C grade + 2fo NiO 78.0 87.2 102.0 97.0 A large hysteresis of wetting was observed when a n i c k e l drop was evaporated from the magnesium oxide. The estimated wetting angles va r i e d between approximately 105° and 80®. During evaporation the wetting angle would decrease t i l l a sudden contraction occurred at which the wetting angle would change discontinuously from a lower to a higher value. (For c a l c u l a -t i o n of wetting angle see Appendix I I I ) . - 43 -IV. DISCUSSION A. P l a s t i c Deformation. Comparing the r e s u l t s obtained by p l a s t i c a l l y deforming a rod by bending and the l i n e structure, as observed on the i n t e r f a c e , one can conclude that the magneium oxide i s p l a s t i c a l l y deformed during some part of the heating cycle when melting a n i c k e l drop on a c r y s t a l of magnesium oxide (see Section C). 1. I d e n t i c a l o r i e n t a t i o n of the observed l i n e s and the s l i p l i n e s on the p l a s t i c a l l y deformed rod. surface of the p l a s t i c a l l y deformed rod are examined with p o l a r i z e d l i g h t (see F i g s . 17 and 25). B. Nature of the P l a s t i c Deformation. Shown i n F i g . 33 are the s l i p systems i n magnesium oxide. I t can be seen that there are four s l i p planes i n c l i n e d at 45° to the cleavage plane and w i l l henceforward be ref e r r e d to as (110)^5 planes. In addition there are two s l i p planes at 90° to the cleavage plane and w i l l be referred to as (110)QQ planes. The corresponding Burger's vectors, which indicate the s l i p d i r e c t i o n f o r each plane, are indicated by the arrows. The main evidence supporting t h i s -I d e n t i c a l patterns can be seen when the i n t e r f a c e and the - 44 -S C R E W D001 BURGERS VECTOR a/2 PlO] Fig.33. S l i p systems i n magnesium oxide, a. Screw d i s l o c a t i o n s b„ Edge d i s l o c a t i o n s . It i s c l e a r that the p l a s t i c deformation under the n i c k e l drop takes place by s l i p over ( 1 1 0 ) ^ planes only. No s l i p occurs over the two ( 1 1 0 ) ^ planes as indicated by the evidence i n F i g . 27. Here no <^110^ traces of etch p i t s can be seen. These <^110^ traces would be expected i f s l i p had occurred over the ( 1 1 0 ) ^ planes as d i s l o c a t i o n s would intersect.the surface and would show up on etching. The f a c t that s l i p takes place;;; over a l l four ( 1 1 0 ) ^ p l a n e s , i s shown i n F i g . 12 where, due to uneven p o l i s h i n g , the s l i p l i n e s i n t e r s e c t on the surface, which Is an i n d i c a t i o n that these s l i p l i n e s are caused by s l i p on d i f f e r e n t planes. On the bent rod s l i p occurs on two (110) and two (110) 45 planes. No s l i p occurs on the remaining (110)^5 planes as t h e i r Burger's vectors are oriented perpendicular to the st r e s s e s . '90 From F i g s , 16 and 17 and the f a c t that s l i p l i n e s can be detected on the other cleavage faces i n the v i c i n i t y of the drop only, i t can be - 45 -concluded that the presence of the drop i s e s s e n t i a l f o r p l a s t i c deformation to take placeo C o Cause of the P l a s t i c Deformation The stresses responsible f o r the p l a s t i c deformation can be produced i n two possible ways; 1 0 The difference i n contraction on cooling between the magnesium oxide and n i c k e l or any compounds formed at the interface„can cause stresses high enough .to deform the magnesium oxide p l a s t i c a l l y . 2. By thermal str e s s e s . Thermal stresses are due t o diff e r e n c e s i n expansion i n d i f f e r e n t parts of one body caused by temperature gradients set up during heating or cooling. The f i r s t p o s s i b i l i t y i s u n l i k e l y when using n i c k e l drops, A c a l c u l a t i o n (see Appendix IV) shows that the di f f e r e n c e between the c o e f f i c i e n t s of thermal expansion of magnesium oxide and n i c k e l i s too small to cause stresses high enough to exceed the y i e l d s t r e s s . I t i s f e l t also that even when the stresses due to thermal contraction were high enough, the -drop would have sheared o f f completely, as i n the case of cobalt. From the various experiments described under experimental r e s u l t s , i t can be seen that thermal stresses can cause p l a s t i c deformation i n magnesium oxide. The fact that a d d i t i o n a l s l i p l i n e s occur when a p l a t e l e t with drop i s heated and cooled f a s t e r at a second cycle i s not s u f f i c i e n t evidence that thermal stresses are responsible f o r the p l a s t i c deformation. I t may w e l l be possible that the p l a s t i c deformation i n the f i r s t cycle produces a d d i t i o n a l s l i p sources which are then activated during the second c y c l e . However, the f a c t that a s i l v e r drop when heated to 1400°C causes p l a s t i c deformation - 46 -whereas no evidence of p l a s t i c deformation can be found when the s i l v e r i s heated to 1000°C i s conclusive evidence that thermal stresses are responsible as a l i q u i d i s not capable of exerting forces other than g r a v i t a t i o n a l . In addi t i o n , a c a l c u l a t i o n of the possible magnitude of the thermal stresses shows that a small temperature d i f f e r e n c e between the center and the perimeter of the i n t e r f a c e between the n i c k e l drop and the magnesium oxide can give r i s e to stresses exceeding the y i e l d s t r e s s (see Appendix V)„ From the f a c t that no l i n e s could be detected on the area of the magnesium oxide p l a t e l e t from which the drop had been evaporated can be concluded th a t the p l a s t i c deformation takes place during the cooling. This i s also supported by the f a c t that the rate of change of temperature i s sub-s t a n t i a l l y higher on cooling than on heating as discussed under equipment, D, Temperature at which P l a s t i c Deformation Occurs, An a c t u a l measurement of the temperature at which the p l a s t i c defor-mation occurred could not be made, but i t i s f e l t that the p l a s t i c deformation took place immediately when cooling began, therefore at a temperature of p r a c t i c a l l y 1500°C„ This i s reasonable when one r e a l i z e s that the region under the drop can lose heat by conduction only and then when areas other than the area under the drop have cooled down. Furthermore the rate of heat loss at these areas occurred by r a d i a t i o n only and w i l l therefore be high at these temperatures, E, P l a s t i c Deformation During Heating. The only way that p l a s t i c deformation could occur during heating i s i f the n i c k e l drop, but not the magnesium oxide, were heated suddenly. It was observed that the n i c k e l supercools at le a s t 50°C, i n d i c a t e d by the experimental observation that the drop suddenly 'flashed', due to the heating of the drop by the sudden release of the heat of melting. This i n e f f e c t , corresponds to an i n f i n i t e heating rate and could give r i s e to thermal stresses i n the magnesium oxide which exceed the y i e l d s t r e s s . However, from the experimental observation that a drop of s i l v e r melted on a p l a t e l e t of magnesium oxi^e can give r i s e to p l a s t i c deformation at temperatures above the melting point of s i l v e r , one can conclude that the supercooling e f f e c t i s not responsible. F, Study of the P l a s t i c Deformation Under Polarized Light. The use of p o l a r i z e d l i g h t shows that, a f t e r cooling, stresses e x i s t i n both the p l a t e l e t s and the rods i n the v i c i n i t y of the n i c k e l drop. Comparison of the patterns shown i n F i g s . 16 and 17 with patterns 31 published by Jessop and H a r r i s v shows that the pattern i n F i g . 16 i s s i m i l a r to the pattern of a photoelastic model of a v e r t i c a l c i r c u l a r plate loaded by a v e r t i c a l compressive f o r c e . The pattern i n F i g . 17 i s s i m i l a r to the e l a s t i c s t r a i n pattern of a photoelastic model of a beam loaded at the center and supported at the ends. G. . Mechanism of the P l a s t i c Deformation. I t i s f e l t that the rods and p l a t e l e t s were deformed i n the following ways I n i t i a l l y at 1500°C both drop and p l a t e l e t or rod are at the same temperature. When the temperature i s lowered, the region under the drop does not cool as f a s t as the rest of the p l a t e l e t or rod. This gives r i s e to thermal stresses. They w i l l be i n the form of compressive and shear stresses (see also Appendix V). As the y i e l d s t r e s s of the region under the drop, because of the higher temperature, i s lower than the y i e l d stress elsewhere i n - 48 -the p l a t e l e t or rod, i t w i l l p l a s t i c a l l y deform whenever the thermal stresses exceed the y i e l d s t r e s s . Upon further cooling the deformed region w i l l contract more than the rest of the p l a t e l e t and w i l l give r i s e to the o v e r a l l ^ e l a s t i c bending moment as seen i n F i g . 17, Upon cooling no a d d i t i o n a l p l a s t i c deformation w i l l take place as the y i e l d stress i s higher at lower temperatures. As the region under the drop w i l l be compressed when deformed, i t w i l l be under tension once the drop and p l a t e l e t are at room temperature. This i s supported by the fac t that o c c a s i o n a l l y cracks can be seen around the perimeter of the drop. Although the region under the drop i s i n tension and therefore the opposite face under compression, the o v e r a l l s t r a i n i s that of a beam supported at i t s ends and loaded at i t s center, because on cooling the region of the rod or p l a t e l e t under the n i c k e l drop contracts more than elsewhere. The fa c t that a bending moment ex i s t s i n the rods i s also supported by the evidence i n F i g . 16, which shows an e l a s t i c s t r a i n pattern i d e n t i c a l to the pattern obtained i n a v e r t i c a l c i r c u l a r plate under a v e r t i c a l compressive load. The experimental observation that a pattern such as i n F i g . 16 could not be detected on the i n t e r f a c e between a n i c k e l drop and a p l a t e l e t 1/4 inch t h i c k , can be explained by the f a c t that the s t r a i n r e s u l t i n g from the bending moment i n a t h i c k p l a t e l e t i s less than i n a t h i n p l a t e l e t . I t i s f e l t that l i t t l e or no cont r i b u t i o n i s made to the stresses i n the magnesium oxide by the differences i n contraction on cooling as the c o e f f i c i e n t s of thermal expansion are nearly i d e n t i c a l . This i s also supported by the experimental observation that no evidence of s t r a i n could be detected i n the v i c i n i t y of the drop melted on the rod which d i d not possess any cleavage steps. - 49 -H, Temperature Dependence of the P l a s t i c Deformation The f a c t that no p l a s t i c deformation could be detected on the i n t e r -faces between p l a t e l e t s of magnesium oxide and drops of copper, aluminum and s i l v e r heated to approximately 50°C above the melting point of the respective metal can be explained by the f a c t that the y i e l d stress of magnesium oxide i s higher at these temperatures than at 1500°C. In addition, as the thermal stresses induced i n the magnesium oxide w i l l be a function of the rate of cool-ing and as cooling takes place by r a d i a t i o n only and therefore depends on the fourth power of the temperature, the thermal stresses at lower temperatures w i l l be s u b s t a n t i a l l y lower and may be low enough not to exceed the y i e l d s t r e s s of the magnesium oxide. I. Nature of S l i p Source's. As the s l i p l i n e s coincide with rows of <^100^ etch p i t s as shown i n F i g s . 9j, 21 and 22, i t can be concluded that the l a t t i c e imperfections which give r i s e to the etch p i t s are responsible f o r the s l i p sources i n magnesium oxide. This i s substantiated by the f a c t that no s l i p occurs on (100)^Q planes, although the thermal stresses present are high enough f o r s l i p to occur on these planes. The etch p i t s are evidence of d i s l o c a t i o n - l o o p s i n the screw o r i e n t a -t i o n introduced i n the surface during cleavage because? 1 0 They are l i n e imperfections i n the c r y s t a l as shown by the experi-mental observation that they continue into the c r y s t a l on continued etching of the s u r f a c e 0 2 0 Their o r i e n t a t i o n corresponds t o the o r i e n t a t i o n of screw d i s l o c a -t i o n s (see F i g . 33)« - 50 -3o They appear only on one side of the cleavage crack as can be seen i n F i g s . 18 and 20 and are therefore introduced during cleavage. 4. They become flat-bottomed on continued etching showing them to be closed loops.^7 5„ They etch f a s t e r than subgrain boundaries and grown-in dislocations,..' -i a This i n d i c a t e s that these d i s l o c a t i o n s are 'fresh' d i s l o c a t i o n s . ° A d d i t i o n a l evidence f o r these d i s l o c a t i o n s to be 'fresh' are the etch p i t s i n F i g . 9* produced by etching of the magnesium oxide by l i q u i d n i c k e l . These etch p i t s , as i n the case of chemical etching, appeared f a s t e r than subgrain boundaries and grown-in d i s l o c a t i o n s . I t i s f e l t that a l l the etch p i t s shown i n F i g . 9 are evidence of d i s l o c a t i o n s introduced during cleavage, although they do not appear i n arrays. J , Mechanisms of Introduction of S l i p Sources into Surface. The formation! of <^ 100^ > rows of d i s l o c a t i o n loops i s c l o s e l y connected with the formation of cleavage steps as; 1, When chemically etching a cleavage face no rows of ^100^ d i s -locations can be detected when the cleavage face does not d i s p l a y cleavage steps. 2, A high d e n s i t y of <Cl00^ > rows of etch p i t s i s produced near high cleavage steps as shown i n F i g . 19. 3. The etch p i t s produced by the l i q u i d n i c k e l as i n F i g . 10 seem to be near low cleavage steps and 4. Many of the rows of d i s l o c a t i o n s terminate on cleavage steps as shown i n F i g . 21, Various mechanisms can be suggested to explain the simultaneous occurrence of cleavage steps and d i s l o c a t i o n s introduced during cleavage: 1 0 As discussed by G i l m a n ^ cleavage steps are produced during cleavage when the cleavage crack i n t e r s e c t s screw-dislocations. These screw d i s l o c a -t i o n s may already be present i n the c r y s t a l or they may be screw d i s l o c a t i o n s introduced during cleavage ahead of the cleavage crack. An i n d i c a t i o n that the i n t e r s e c t i o n of d i s l o c a t i o n s i s responsible f o r the formation of cleavage steps can be seen i n F i g . 10. These d i s l o c a t i o n s introduced during cleavage take the form of closed l o o p s ^ which are nucleated by the high stresses at the t i p of the cleavage crack, and are then bisected by the crack. If the cleavage crack propagates at less than a c r i t i c a l r a te, these half-loops may expand to such a s i z e as to remain stable i n the c r y s t a l . This would require that i d e n t i c a l patterns of etch p i t s should be obtained when chemically etching both sides of the cleavage crack. The f a c t that t h i s i s u s u a l l y not the case, as shown i n F i g s . 18 and 20, i s that the stresses may not be symmetrical around the t i p depending on the method of cleavage and therefore the stresses at only one side of the cleavage crack may be s u f f i c i e n t to expand the half-loops to such a si z e so that they are stable and do not collapse. This i s also supported by the experimental observation that only one example could be found where a one-to-one correspondence existed between s l i p l i n e s on the i n t e r f a c e s between surfaces on opposite sides of a cleavage crack and n i c k e l drops. I f the d i s l o c a t i o n loops nucleated ahead of the 4 cleavage crack grew large enough so that stable half-loops would remain at both sides of the cleavage crack, then a one-to-one correspondence would occur more often. 2. Cleavage steps are also i n i t i a t e d at the s t a r t of the cleavage crack, - 52 -i t being p r a c t i c a l l y impossible to s t a r t the crack on one atomic plane only. I t has been observed experimentally that most of the higher cleavage steps or i g i n a t e at the point at which cleavage was s t a r t e d . During the propagation of the cleavage crack through the c r y s t a l considerable shear w i l l be produced i n the v i c i n i t y of these cleavage steps. This may slow down the crack to such an extent that i t propagates at a rate which permits the loops, nucleated at the t i p of the cleavage crack, to expand and to remain as stable half-loops i n the c r y s t a l , as discussed previously. This i s i l l u s t r a t e d c l e a r l y i n F i g . 19 where the high cleavage step i s one i n i t i a t e d at .the s t a r t of cleavage. This slowed the crack down such that the d i s l o c a t i o n loops nucleated at the t i p grew large enough to be stable as indicated by the presence of etch p i t s . The crack intersected these d i s l o c a t i o n loops and, as discussed above, gave r i s e to a d d i t i o n a l cleavage steps indicated by the i r r e g u l a r l i n e s roughly perpendicular to the high cleavage step. The boundary of the band of etch p i t s indicates the l o c a t i o n where the rate of propagation of the cleavage crack changed discontinuously. 3. It may also be possible that even when the rate of propagation of the cleavage crack i n the v i c i n i t y of cleavage steps i s too high f o r d i s l o c a -t i o n loops to become st a b l e , the shear stresses at the crack f r o n t , due to the propagation of the cleavage steps i n i t i a t e d at the s t a r t of cleavage, may be s u f f i c i e n t to expand the d i s l o c a t i o n loops nucleated at the t i p of the cleavage crack to remain as stable loops i n the c r y s t a l . This mechanism may explain the occasional row of d i s l o c a t i o n s which seem to s t a r t at a cleavage step as shown i n F i g . 21. 4. Some evidence e x i s t s which indicates that the grown-in d i s l o c a t i o n s are responsible f o r the ^ C l O O ^ arrays of d i s l o c a t i o n s as more of these arrays seem to occur i n p l a t e l e t s cleaved from c r y s t a l s which were not annealed than - 53 -i n p l a t e l e t s cleaved from annealed c r y s t a l s . In ad d i t i o n , as shown i n F i g , 21, a row of dislocations seems to be nucleated where a cleavage step i n t e r s e c t s a subgrain boundary. From many observations on etched cleavage faces, i t was concluded that most half-loops were introduced by the second mechanism. By the evidence mentioned, some half-loops were also introduced by the t h i r d and fourth mechanisms. It was never observed that half-loops were introduced by the f i r s t mechanism. The occurrence of rows of d i s l o c a t i o n half-loops i s due t o the m u l t i p l i c a t i o n of a s i n g l e h a l f - l o o p . Once a single half-loop has been nuclea-ted xt can act as a t r i g g e r - " 3 f o r f u r t h e r half-loops i n the same s l i p plane. The m u l t i p l i c a t i o n of a loop once i t has been nucleated i s not necessary f o r the formation of s l i p sources, as i t has been shown experimentally that s l i p l i n e s occurred on the i n t e r f a c e between a n i c k e l drop and a chemically etched p l a t e l e t of magnesium oxide which did not exhibit any <ClOO^ rows of etch p i t s . The s l i p sources then consist of s i n g l e d i s l o c a t i o n h a l f - l o o p s . An example of t h i s may be seen i n F i g s . 21 and 22, where a s l i p l i n e goes through two etch p i t s only. Washburn, Gorum and Parker1''' also observed these d i s l o c a t i o n s , introduced during cleavage. However, i n t h e i r specimens the rows of etch p i t s which showed up a f t e r etching ran along the length of the specimen only. From t h i s i t was concluded that these d i s l o c a t i o n s could not a f f e c t the d u c t i l e properties as the Burger's vector of these d i s l o c a t i o n s i s oriented perpendi-c u l a r l y to the applied s t r e s s 0 In the present i n v e s t i g a t i o n i t was shown that i n cleavage approxi-mately an equal number of s l i p sources were introduced on a l l s l i p planes 0 - 54 -Therefore i t can be concluded that only those half-loops oriented i n the d i r e c t i o n of crack propagation can t r i g g e r the nucleation of rows of h a l f -loops. The half-loons oriented perpendicularly to the d i r e c t i o n of crack "~ propagation do not m u l t i p l y but can s t i l l act as s l i p sources. The conclusions 17 reached by Washburn, Gorum and Parker ' should therefore be a l t e r e d . K, S l i p Source Density A c a l c u l a t i o n based on the number of s l i p l i n e s produced shows that the number of s l i p sources introduced during cleavage varies from a maximum of approximately f i f t e e n thousand s l i p sources per cm.^ f o r a surface with a high density of cleavage steps to no s l i p sources on a surface without cleavage steps (see Appendix VI),. The experimental observation that rods, when bent by hand, exhibited varying degrees of d u c t i l i t y before f r a c t u r e , can be explained by assuming "that the more d u c t i l e rods had more surface sources than the r e l a t i v e l y b r i t t l e rods. This i s supported by the observation that those rods which possessed a face formed by cutting the c r y s t a l with the diamond saw were i n v a r i a b l y d u c t i l e . That a rough surface treatment leads to introduction of s l i p sources has been 18 observed by Stokes, Johnston and L i , who introduced d i s l o c a t i o n s i n the surface of polished magnesium oxide rods simply by s p r i n k l i n g the surface with 46-mesh carborundum powder from a height of three inches. The r e l a t i v e d u c t i l i t y can also be explained i f s l i p took place on one s l i p system only. This way d i s l o c a t i o n p i l e - u p and subsequent crack formation leading to fracture,1 9 i s avoided, L, Minimum Size of Stable D i s l o c a t i o n Loop. A c a l c u l a t i o n of the minimum size of a closed d i s l o c a t i o n loop which can remain i n the c r y s t a l shows t h i s to be approximately one micron. The - 55 -minimum size of a stable d i s l o c a t i o n h a l f - l o o p w i l l probably exceed t h i s f i g u r e somewhato This compares favourably with the average depth of a few microns and the observed distance between etch p i t s of approximately one and one h a l f microns, assuming that these two etch p i t s belong to the same loop. The minimum s i z e of loop i s probably a function of temperature and would depend on the r e l a t i v e temperature dependence of d i s l o c a t i o n tension and minimum shear stress of d i s l o c a t i o n g l i d e . Most of the loops remain stable at 1500°C as they act as s l i p sources at t h i s temperature. In addition, as shown i n F i g , 9, where l i q u i d n i c k e l etched a row of d i s l o c a t i o n s at 1500°C, they did not have a tendency to,move» (For c a l c u l a t i o n see Appendix V I I ) . The experimental observation that no more s l i p i s produced at the highest cooling rate than at a lower cooling rate could be because of t h i s minimum s i z e of stable d i s l o c a t i o n loop. M. Depth of S l i p Sources. From the f o l l o w i n g experimental observations, i t can be concluded that the s l i p sources are located at or just below the cleavage surfaces 1. The d i s l o c a t i o n half-loops introduced during cleavage were responsible f o r the s l i p . These half-loops were observed to extend only a few microns into the surface. 2. No s l i p l i n e s could be observed on the interface between a n i c k e l drop and a p l a t e l e t of magnesium oxide which was polished i n hot phosphoric a c i d f o r about ten minutes. As hot phosphoric acid attacks the surface at a 19 rate of approximately one micron/min, i t can be concluded that no s l i p sources can be found which are more than ten microns below the surface. 3. A d d i t i o n a l , but not conclusive, evidence i s as follows; - 56 -When chemically etching a rod with a n i c k e l drop one can detect two sets of < C l l 0^slip l i n e s and one set of (100) s l i p l i n e s p a r a l l e l to the length of the rod. In the bottom face, only (100) s l i p l i n e s can be seen which are perpendicular to the length of the rod. R e c a l l i n g that s l i p with respect to the i n t e r f a c e takes place over the four (110)^ planes only, one can conclude that the s l i p l i n e s , as observed on the side and bottom faces of the rod, are l i n e s of i n t e r s e c t i o n of s l i p planes o r i g i n a t i n g at or i n the v i c i n i t y of the top surface. This, of course, can al s o be explained by assuming that the thermal stresses exceed the y i e l d s t r e s s i n the surface only. The presence of ^  1 1 0 s l i p l i n e s on the side face of the chemically etched rod could lead one to conclude that s l i p also occurs over (110)cx) planes, but as a (IIO)^Q plane with respect to one cleavage plane i s a (110)^ plane with respect to a perpendicular cleavage plane, one can s t i l l define the s l i p taking place over (110)^ planes only. The d i f f i c u l t y of d e f i n i t i o n on which plane s l i p takes place Is removed i f one defines the s l i p plane with respect to the surface at which the s l i p source i s located. S l i p then takes place on (110)^5 planes only. The <^ 110^ > s l i p traces i n the tension and compression surface of the bent rod, such as i n F i g s . 18 and 2k, are due to s l i p nucleated at the side surfaces, although with respect to the top surface t h i s would be (110)QQ s l i p , N. Number of Dislo c a t i o n s Produced by S l i p Sources. An estimate can be made of the maximum number of d i s l o c a t i o n s produced by the s l i p sources when activa t e d by the thermal st r e s s e s . Each time a d i s l o c a t i o n on a (110)^ plane emerges on the surface, the height of the s l i p l i n e , as measured perpendicularly to the surface, i s increased by a length equal to h a l f the l a t t i c e parameter or approximately 2.1 x 10 cm. As the height of the step, as measured perpendicularly to the surface, equals i t s apparent width when viewed perpendicularly to the surface, i t s height can be - 57 -estimated by measuring i t s width. As shown i n F i g , 10, the width i s no more than h a l f a millimeter wide at a magnification of 2500x, which makes i t s height approximately 2 x 10"''cm. This corresponds to a t o t a l number of approximately one thousand d i s l o c a t i o n s . I t i s f e l t that the s l i p sources can produce a maximum number of d i s l o c a t i o n s a f t e r which they are rendered i n a c t i v e . This can be concluded from the experimental observation that on thermal c y c l i n g no change i n the appearance of the s l i p l i n e s can be detected. In a d d i t i o n , a p l a t e l e t with a n i c k e l drop heated to 1800°C did not seem to have any higher s l i p l i n e s than 27 one cooled from 1500°C, Observations made by Pask also support t h i s view. S l i p sources can be rendered i n a c t i v e through various mechanisms: 1. This could take place by pile-up of d i s l o c a t i o n s against subgrain boundaries or against other regions of s l i p . The combined back-stress of the piled-up d i s l o c a t i o n s could increase the shear stress required t o operate the source to exceed the thermal st r e s s e s . Pile-up against subgrain boundaries does not occur i n magnesium oxide as shown i n F i g . 22. 2. In the case of magnesium oxide the sources activated by the thermal stresses may be destroyed by the deformation i t produces. This i s because these s l i p sources are located near the surface. Each time the source produces a d i s l o c a t i o n i t w i l l have advanced to the surface a distance as measured along the s l i p plane equal to i t s Burger's vector. I f enough d i s -locations can be produced the s l i p source w i l l come out to the surface and w i l l cease to e x i s t . It i s not necessary f o r the source t o move to the surface t o be rendered i n a c t i v e . Assuming the source takes the form of a double-ended - 58 -Frank-Read source i t w i l l cease to operate when the points at which the d i s l o c a t i o n i s anchored are w i t h i n a distance from the surface equal to h a l f the c r i t i c a l length of the source. The c r i t i c a l length of a double-ended Frank-Read source is" the distance between the anchoring points below which, at a certain' stress l e v e l , the source w i l l not operate. As only half the, stress i s needed to operate a single-ended Frank-Read source of the same length, the same str e s s w i l l operate a single-ended source h a l f the length of the double-ended source. Therefore, the source can approach the surface to wit h i n a distance equal to h a l f i t s length. A c a l c u l a t i o n of the c r i t i c a l length of a double-ended Frank-Read source, based on a y i e l d - s t r e s s of 3.1 x 10^ dynes per cm.2 at 1500°C, showed i t to be approximately one micron. The source w i l l then cease to operate when i t s anchoring points are w i t h i n ha l f a micron from the surface. 0„ Mechanism of D i s l o c a t i o n M u l t i p l i c a t i o n . One mechanism by which the d i s l o c a t i o n half-loops can act as s l i p sources i s by the formation of jogs i n the d i s l o c a t i o n loops by e i t h e r the loo p s ' i n t e r s e c t i n g a grown-in d i s l o c a t i o n or by the loops i n t e r s e c t i n g loops nucleated on other s l i p planes. I f the loop acquires one jog i t can operate as a Frank-Read h a l f -source. I f the loop acquires two jogs, i t can act as a double-ended Frank-Read source. 5 2 If the density of grown-in d i s l o c a t i o n s i s 10 per cm. then a h a l f -loop with a diameter of f i v e microns w i l l , on the average, have intersected 0.01 d i s l o c a t i o n . The p r o b a b i l i t y of jog formation does seem rather low. However, i t i s not necessary f o r the loop to have acquired jogs when formed. - 59 -I t can form jogs when undergoing expansion under applied shear stresses. This way the loop may i n t e r s e c t a considerable amount of d i s l o c a t i o n s which then leads to anchoring. P. I n t r i n s i c S l i p Sources. In addition to s l i p sources i n the surface introduced during cleavage, evidence ex i s t s f o r the presence of ' i n t r i n s i c ' s l i p sources. This can be concluded from the f a c t that a rod which was polished f o r ten minutes i n hot phosphoric acid to remove a l l surface sources, exhibited evidence of p l a s t i c deformation when bent and subsequently etched as shown i n F i g . 23. I t i s c l e a r that the sources responsible for s l i p i n the polished rod are of a d i f f e r e n t nature from the surface sources i n an 'as cleaved rod', by comparing F i g . 29 with F i g . 30 which shows an area of a p l a s t i c a l l y deformed rod, which had not been polished. The s l i p l i n e s i n the polished rod appear to be wider than the s l i p l i n e s i n the rod i n the 'as cleaved' condition. This suggests that the i n t r i n s i c sources, when activated under a shear s t r e s s , generate d i s -l o c a t i o n s by a mechanism d i f f e r e n t from the mechanism by which the surface sources generate d i s l o c a t i o n s . I t i s f e l t that the f a i n t <^110> l i n e s observed i n F i g . 13 may be due to s l i p caused by the generation of d i s l o c a t i o n s by the i n t r i n s i c sources. I t can be concluded, however, that the i n t r i n s i c sources contribute very l i t t l e to the d u c t i l i t y of magnesium oxide cr y s t a l s , activated by thermal stresses, such as occur under the n i c k e l drop. This could be due t o the fact that these thermal stresses may not be high enough to activate the ' i n t r i n s i c sources'. This i s supported by observations by Stokes, Johnston and L i where i t was found that the a c t i v a t i o n s t r e s s f o r s l i p i n polished magnesium oxide rods was much greater than the stress f o r s l i p i n rods which contained surface sources. - 60 -Chemical Reactions at Interface. From the experimental observations a hypothesis may be developed f o r the type of chemical reactions which occurred at the i n t e r f a c e . The reaction according to MgO + Ni ±> Mg + NiO probably d i d not take,place. F i r s t , no evidence f o r the formation of n i c k e l oxide; has been found and second, the n i c k e l oxide i s unstable at-these temperatures and pressures, as calculated i n Appendix V i l l a . A more probable source f o r the magnesium i s the d i s s o c i a t i o n of the magnesium oxide into f r e e magnesium and oxygen according to 2 MgO ^ 2 Mg + 0 ? This d i s s o c i a t i o n i s not impossible, as indicated by a thermodynamic c a l c u l a -t i o n (Appendix V l l l b ) . In addition, i t was experimentally observed that the magnesium oxide i s h i g h l y v o l a t i l e . This magnesium then a l l o y s with the n i c k e l to form the magnesium n i c k e l compound. This r e s u l t s i n the removal of the magnesium, which leads to a f a s t e r decomposition of the magnesium oxide at the i n t e r f a c e . A high enough d r i v i n g force i s present i n the form of the heat of formation and the entropy of mixing so that the compound i s stable at these temperatures. F i g , 34 shows that the compound exi s t s below 768°C only. . .-r- - r -A c a l c u l a t i o n of the minimum thickness of the i n t e r f a c i a l layer such that complete wetting w i l l occur shows t h i s to be 38.4 (See Appendix VI I I c ) , The experimental observation of 'etching' of the magnesium oxide by the l i q u i d n i c k e l at the perimeter of the i n t e r f a c e i s evidence that the decomposition of magnesium oxide occurs p r e f e r e n t i a l l y at the perimeter of the i n t e r f a c e . °C Atomic Percentage Magnesium 10 20 30 40 50 60 70 80 •'• 90 , 1500 1400 - 2600 2400 2200 Boiling Point 2000 - 1800. - 1600 - 1400 1200 1000 - 800 600 - 400 Ni 10 20 30 40 50 60 70 80 Weight Percentage Magnesium 90 Mg F i g . 3k* Nickel-magnesium binary system. - 62 -The d r i v i n g f o r c e responsible f o r the p r e f e r e n t i a l attack at the perimeter of the i n t e r f a c e i s provided by the e q u i l i b r i u m between the surface-and i n t e r f a c i a l tensions. Shown i n F i g s . 35a and 35b are the configurations of a n i c k e l drop on an etched and unetched magnesium oxide c r y s t a l . Replacing b. Etched F i g . 35. Surface tensions e q u i l i b r i a . the i n t e r f a c i a l energies per u n i t area by i n t e r f a c i a l tensions per unit length and r e s o l v i n g these forces i n the plane of the f i g u r e r e s u l t s in the e q u i l i b r i a shown above. B i s the angle through which the i n t e r f a c i a l tension - 63 -between the magnesium oxde and the n i c k e l i s rotated due to the appearance of the groove at the perimeter. The d r i v i n g force f o r the reaction which leads to the groove forma-t i o n r e s u l t s from the f a c t that when a groove i s formed a l l the forces are i n equilibrium. On a f l a t surface t h i s i s the case f o r the h o r i z o n t a l components only. The equilibrium of the forces i n the h o r i z o n t a l plane i s given by y M g 0 - yj j i c o s 6 ^ ^ MgO-Ni c o s D u r i n S evaporation of the drop the term y N j _ cos 0 continuously decreases. This upsets the equilibrium. The drop w i l l contract c a t a s t r o p h i c a l l y when the imbalance of the equilibrium has reached such a value that the drop can r e a t t a i n the equilibrium configuration i t w i l l have on a f l a t surface. The cycle repeats i t s e l f when a new groove i s formed. It i s t h i s groove formation which i s responsible f o r the observed hysteresis of wetting. I t can be concluded that l i t t l e or no value can be attached to the wetting angles reported i n Table I I I . Most of these w i l l represent a non-equilibrium wetting angle. Only those angles which are measured immediately a f t e r a. contraction of the drop would be an i n d i c a t i o n of the w e t t a b i l i t y of the magnesium oxide. I t i s clear that the method of study-ing metal-ceramic i n t e r a c t i o n by means of w e t t a b i l i t y experiments does not lend i t s e l f to the system magnesium oxide-nickel. R, Etching of Dislo c a t i o n s by Liquid N i c k e l Liquid n i c k e l also etches magnesium oxide p r e f e r e n t i a l l y at d i s l o c a -t i o n s and subgrain boundaries. The d r i v i n g force f o r t h i s i s the state of higher energy of the l a t t i c e at d i s l o c a t i o n s and subgrain boundaries. This leads to a f a s t e r decomposition of the magnesium oxide, which r e s u l t s i n etch p i t formation. - 64 -S o D i f f u s i o n of N i c k e l i n t o Magnesium Oxide. The d i f f u s i o n of n i c k e l i n t o magnesium oxide sin g l e c r y s t a l s seems to occur mainly by d i f f u s i o n along d i s l o c a t i o n s . This i s indicated by a much slower rate of etching of old d i s l o c a t i o n s compared with fresh d i s l o c a t i o n s . These f r e s h d i s l o c a t i o n s were introduced at room temperature a f t e r d i f f u s i o n took place. I f the n i c k e l were d i s t r i b u t e d uniformly through the magnesium oxide c r y s t a l , the etching rate f o r a l l d i s l o c a t i o n s would be approximately equalo A d d i t i o n a l evidence that d i f f u s i o n occurs along d i s l o c a t i o n s i s that greater amounts of n i c k e l are present i n those c r y s t a l s which were p l a s t i c a l l y deformed. The a c t u a l r e l a t i v e amounts of d i f f u s e d n i c k e l were not determined. Removal of the drops by shear removes part of the d i f f u s e d n i c k e l . Removal o f the drop by melting, may introduce extra d i s l o c a t i o n s . Also, some of the n i c k e l may d i f f u s e out as indicated by the experimental observation that l i t t l e n i c k e l could be found i n those c r y s t a l s from which the drop was removed by evaporation. Although i n the preparation of the specimen? part of the n i c k e l was removed, i t i s f e l t that s u f f i c i e n t evidence e x i s t s to conclude that b u l k - d i f f u s i o n takes place by d i f f u s i o n along d i s l o c a -t i o n s . Close analysis of the r e s u l t s obtained by Turnbull i n d i c a t e that those d i s l o c a t i o n s introduced i n the surface during cleavage also a f f e c t the d i f f u s i o n of n i c k e l i n t o magnesium oxide. During evaporation of the drop between contractions, equilibrium could be preserved i f i n t e r f a c i a l tension near the perimeter could be decreased. This could take place by extra d i f f u s i o n of the n i c k e l into the magnesium oxide. The i n t e r f a c i a l tension between a n i c k e l r i c h magnesium oxide and n i c k e l w i l l be lower than the i n t e r f a c i a l tension between pure - 65 -magnesium oxide and n i c k e l . I t i s f e l t that t h i s mechanism i s responsible f o r the f a c t that l i t t l e or no chemical etching occurred near the perimeter of the in t e r f a c e as shown i n F i g . 27. T» Strength of Bond. The bond strength between magnesium oxide and n i c k e l exceeds the fra c t u r e stress of the magnesium oxide. On removal of the n i c k e l drop from the magnesium oxide, f r a c t u r e occurs through the magnesium oxide rather than through the interface.„ The formation of the magnesium-nickel compound at the int e r f a c e i s necessary f o r the formation of the strong bond. This i s indicated by the experimental observation that a drop held at a temperature just above i t s melting point f o r approximately ten seconds does not adhere as w e l l as those held at 1500°C f o r t e n minutes. The amount of compound formed i n ten seconds w i l l be less than the amount formed i n ten minutes. The presence of n i c k e l oxide at the int e r f a c e has a detrimental e f f e c t on the bond strength. When removing a n i c k e l drop, which i n i t i a l l y contained 2% n i c k e l oxide, f r a c t u r e occurred at the in t e r f a c e between the drop and magnesium oxide„ Bond strength between magnesium oxide and metals, i n general, w i l l be a f f e c t e d by the formation of gas occlusions at the in t e r f a c e . It was observed that on the interfaces between magnesium oxide and metals such as s i l v e r , copper, aluminum and t i n , many gas occlusions formed. Moore and 8 Thornton made the same observation f o r the system g o l d - s i l i c a . They a t t r i -buted the formation o f gas occlusions to the presence of traces of oxygen i n the vacuum used. - 66 -V. CONCLUSIONS A 0 D u c t i l i t y The magnesium oxide under the n i c k e l drop i s deformed by thermal stresses which are developed during cooling. S l i p takes place on the four •\110>slip planes i n c l i n e d at L5° to the cleavage plane. The s l i p does not seem t o be affe c t e d by the presence of subgrain boundaries. The amount of p l a s t i c deformation i s a function of the rate of . cooling. More deformation occurs at greater cooling r a t e s . At the maximum''-1 cooling rates possible no more s l i p i s produced than at a lower cooling r a t e . At less than the maximum cooling rate a l l the s l i p sources present are act i v a t e d . Because of the p l a s t i c deformation, a bending moment exi s t s upon cooling. As the deformation occurred by compression, the region of the magnesium oxide d i r e c t l y under the n i c k e l drop w i l l be under tension when cooled to room temperature. Heating metal drops, such as aluminum, copper and s i l v e r to a temperature 50°C above t h e i r respective melting points, does not give r i s e to p l a s t i c deformation when cooled, due t o lower thermal stresses and a higher y i e l d point of the magnesium-oxide. When a s i l v e r drop i s heated to 1400°C, thermal stresses on cooling are high enough to give p l a s t i c deformation. The s l i p sources activ a t e d by these thermal stresses are introduced i n the surface of the magnesium oxide during cleavage. The average depth of these sources i s of the order of a few microns. No s l i p sources activated by these thermal stresses e x i s t at a depth exceeding ten microns. The i n t r o d u c t i o n of s l i p sources i n the surface by cleavage i s associated with the formation of cleavage steps. No s l i p sources e x i s t i n surfaces which do not have cleavage steps. The s l i p source density varies from a maximum of f i f t e e n thousand p sources per cm. f o r a surface with a high density of cleavage steps to no s l i p sources f o r a surface without cleavage steps. The s l i p sources introduced i n the surface during cleavage take the form of d i s l o c a t i o n h a l f - l o o p s . The formation of cleavage steps during the propagation of the cleavage crack slows down the rate of propagation of the crack or gives r i s e t o enough shear such that d i s l o c a t i o n loops,nucleated ahead of the cleavage crack, grow to a size such that they can remain as stable half-loops i n the surface when cut by the passing cleavage crack. The calculated minimum s i z e of stable d i s l o c a t i o n loop agrees w e l l with the observed value of approximately one micron. Etching of these cleavage faces r e s u l t s i n the production of etch p i t s at these d i s l o c a t i o n loops. Rows of etch p i t s are due to the m u l t i p l i c a -t i o n of d i s l o c a t i o n loops located on s l i p planes oriented p a r a l l e l to the d i r e c t i o n of propagation of the cleavage crack. Continued etching removes the half-loops from the c r y s t a l . The m u l t i p l i c a t i o n of the half-loops i s probably due to formation of jogs, which then anchor the d i s l o c a t i o n . Under an applied shear stress the half- l o o p s can then m u l t i p l y and do not move out of the c r y s t a l . The number of d i s l o c a t i o n s produced by a s l i p source i s approximate-l y one thousand. The s l i p source stops operating due t o d i s l o c a t i o n pile-ups or by moving to the surface due to the deformation i t causes. - 68 -This method of inducing thermal stresses i n cleavage surfaces lends i t s e l f very w e l l to studying the behaviour of surface s l i p sources i n magnesium oxide and may w e l l be applicable to other materials. - ... B. Chemical Reactions and D i f f u s i o n . The bonding mechanism between n i c k e l and sing l e c r y s t a l s of magnesium oxide consists of the formation of a magnesium n i c k e l (Mg 2Ni) compound as indicated by X-ray analysis of the i n t e r f a c i a l material. The bonding force between the n i c k e l and magnesium oxide exceeds the fracture stress of magnesium oxide, shown by the f a c t that removal of the n i c k e l drop leads to fr a c t u r e through the magnesium oxide. The w e t t a b i l i t y r e s u l t s showed a large hysteresis of wetting. This hysteresis of wetting i s due t o the formation of a' 'groove' at the perimeter of the i n t e r f a c e between the n i c k e l and magnesium oxide. For t h i s reason, the method of studying metal-ceramic i n t e r a c t i o n s by means of w e t t a b i l i t y measure-ments can not be used i n the system nickel-magnesium oxide. B u l k - d i f f u s i o n of n i c k e l i n t o s i n g l e c r y s t a l s of magnesium oxide seems to take place by d i f f u s i o n along d i s l o c a t i o n s . This i s indic a t e d by the r e l a t i v e amounts of n i c k e l i n deformed and undeformed c r y s t a l s . - 69 ~ RECOMMENDATIONS FOR FUTURE WORK It i s f e l t that by using t h i s technique no fu r t h e r understanding can be gained of the d u c t i l e properties of magnesium oxide. However, c e r t a i n observations were made which may warrant f u r t h e r i n v e s t i g a t i o n . Quantitative measurement of the d i f f u s i o n of metals into magnesium oxide as a function of d i s l o c a t i o n content may indicate to what extent the presence of d i s l o c a t i o n s i s responsible f o r bulk d i f f u s i o n . Magnesium oxide could lend i t s e l f very w e l l f o r t h i s as i t might not recover from any p l a s t i c deformation introduced even i f these t e s t s were c a r r i e d out at f a i r l y high temperatures. More understanding can be gained of the s i z e , d i s t r i b u t i o n and nature of the d i s l o c a t i o n loops introduced during cleavage. The X-ray technique developed by Berg and Barrett might be applicable to t h i s problem. A study of the w e t t a b i l i t y of magnesium oxide by other metals or a l l o y s can also be of i n t e r e s t . The v a r i a t i o n s i n the wetting may be followed by motion f i l m techniques and may throw more l i g h t on the mechanisms of the hysteresis of wetting. During the X-ray fluorescent q u a l i t a t i v e analysis of the magnesium oxide, i t was noticed that s i n g l e c r y s t a l s would turn a wine-red, whereas powdered magnesium oxide would remain c o l o r l e s s . I t could be of i n t e r e s t to inve s t i g a t e t h i s phenomenon, although t h i s could carry over into the f i e l d of s o l i d state physics. - 7 0 -APPENDIX I X-ray D i f f r a c t i o n Data from Nickel-Magnesium Oxide Interface. Photographic prints of X-ray powder patterns of the i n t e r f a c i a l material and pure magnesium oxide are shown i n F i g , 3 6 . F i g . 3 6 . X-ray powder patterns Nos. 1502 and 1528. (Top - i n t e r -f a c i a l material, bottom - pure magnesium oxide). The observed X-ray data f o r the i n t e r f a c i a l m a terial and the material present as i d e n t i f i e d from the A.S.T.M. Card Index are shown i n Table IV, The reported d-values f o r the materials present are shown i n F i g s . 3 7 , 3 8 and 3 9 . Comparing the data from the A.S.T.M, cards f o r magnesium oxide magnesium n i c k e l compound and magnesium hydroxide, one can conclude that a l l these three materials are present. The l i n e s which could be responsible f o r more than one m a t e r i a l could be p a r t i a l l y due t o the presence of N i as these two l i n e s are the two strongest l i n e s reported f o r n i c k e l , as shown i n F i g , Ul' The presence of magnesium hydroxide might be explained by magnesium oxide reactingwith the moisture from the a i r or perhaps with traces of water i n the grease used i n the preparation of the specimen f o r X-ray a n a l y s i s . From the r e l a t i v e i n t e n s i t i e s of the l i n e s , one can conclude that the magnesium - 71 -TABLE IV. Data from F i l m No. 1528 Line I / I 0 d M a t e r i a l 1 15 4.7612 Mg(0H) 2 2 10 4.4357 Mg 2Ni 3 1 3.1120 Mg 2Ni 4 20 2.4351 MgO 5 5 2.3351 Mg(0H) 2 6 <1 2.2510 Mg 2Ni 7 100 2.1091 MgO 8 10 2.0359 Mg 2Ni, Ni 9 2 1.7653 Mg 2Ni, Ni, Mg(0H) 2 10 1 1.6488 Mg 2Ni 11 50 1 .4999 MgO 12 10 1.2696 MgO 13 15 1.2169 MgO 14 1 1.1315 Mg 2Ni ^ 15 5 1.0496 ; MgO 16 5 .96382 i MgO 17 10 .93956 MgO 18 10 .85786 MgO 19 5 .81007 MgO d 2. lC6 1.439 0.9119 2.4^1 '.'cu I/I. 100 52 17 10 Rad. C Dia. A 1.5405 Cut off "•ilter : . i 3oll. d A I/I, hkl d A I/I, hkl 2 .4? : 1C 1 " I/I, R*f. S . ' .AI I^CN AN :i T A T G E . d corr. abs.? X T E L . R E P O R T S , .'Bi 2.1C6 1J0 52 2CO 221-i ' :'.?70 311 Sys. Cgoic {F.C a. 4 .2 :3 b„ .) Co S.G. A OH - Fu3>. C 1.216 l.C-533 12 5 222 a Ref. B i n . Y Z 4 0 . 9 £ £ 5 .941? 2 15 t o 2V R*(. nw £ 1 . 7 3 2 ?Y 0*3.581 mp Color D I D . Sign .eio? 3 1 , , I G H P U R I T Y PHOSPHOO S A / P L E n c . i !CA NEAT AT iS00°C FOR 3 MISi - T 2fi°C To REPLACE 1-1235, 2-1207, 2-0998 £ 3 F i g . 37. A.S.T.M. Card No. 4.0829 f o r magnesium oxide. 3546 d 2 . 0 0 4 . " 2 2. 2 6 4 . " 2 1-1 . '53 I/I. 1 0 0 5 3 5 3 5 3 i /ACNL-s i i i y N I C K E L COMPOUND 1 - 1 2 6 6 Red. A 0 . 7 0 9 Filter .-"RO-, d A I/I, hkl d A 1/1, hkl Dia. 1 6 i NCHCS Cut ofl Coll. 4 . /'2 5 3 1 . 5 8 7 I/I, C A L I B R A T E D S T U I P S d corr. abs.? No 4 . . "8 1 7 1 . 5 5 3 Ref. H T 3 . 7 4 1 7 1 . A 7 7 3 . I f . 1 . 4 3 1 1 Sys. S.G. 2 . 7 8 1 3 I . O 7 a. c. A C 2 2 . 6 1 4 1 . 3 7 1 Ref. 2. 3 3 3 1 . 3 4 1 2.26 5 3 1 . 3 0 1 2 2 . 2 0 1 1 1 . 2 5 3 t o f Y Sign 2.(Xi 100 3 . 2 3 1 2V D mp Color Ref. 1 . 9 0 3 1 . 1 8 3 1 PS 3 4 5 o/o MG - I 5 o/o Ni ftLLOV 1 . 7 5 4 1 . 0 9 7 1 . 6 9 4 * Llf-'L" I S i.'iG 1 . 64 1 F i g . 38. A.S.T.M. Card No. 1.1268 f o r magnesium n i c k e l compound (MggNi). 7-2 3 9 R*d.C'jKi: t A 1 . 5 4 0 5 FikerNl Cut ofl I/I, S P E C T R O M E T E R Ret ' JBS C I R C U L A R 5 2 ? , VOLUME £ ( - . 9 5 ; 73 -Ms (CM MAG.-JESIUM H Y D R O X I D E ( B P U C I T E ) L ) S.G. D | A - P 3 M 1 c, 4 . 7 £ ? A 0 . 0 : ° / o A J C R , C U . C o l a STRI , '485 c now MGO , s; i Z ! Dx 2 . 3 7 Color C O L O R L E S S ) • A TE R H E L D * T 6 0 C ° C 1 0 , 0 0 0 P S I rofi 3 o » » S P E C I . A N A L Y S I S S M Q « S < 0 . 1 ° / C S R , T I ; < 0 . 0 0 1 a'o BA TURE T Y P E . P A T T E R N MADE AT 2 6 ° C . d A 2 . 2 £ 5 1 . 7 & 4 1 . 5 7 3 1 . 4 9 4 1 . 3 7 3 1 . 3 6 3 1 . 3 1 0 1 . 1 9 2 1 . 1 8 3 1 . 1 1 8 1 , 0 9 2 1 . 0 3 4 1 . 0 3 0 1 . 0 0 6 7 0 . 9 5 4 3 . 9 5 0 3 . 9 4 5 5 OOl 1 0 0 1 0 1 1 0 2 n o i n 1 0 3 2 0 0 2 0 1 0 0 4 2 0 2 1 1 3 1 0 4 2 0 3 2 1 0 2 1 1 0 0 5 1 1 4 2 1 2 am . = 5 2 3 . 6 6 4 3 . 5 1 5 6 . 7 8 5 5 F i g . 39. A.S.T.M. Card No. 7.239 f o r magnesium hydroxide. n i c k e l compound existed at the i n t e r f a c e or w i t h i n the magnesium oxide. I t d i d not e x i s t w i t h i n the metal drop as t h i s would require that the whole drop consisted of the compound. This could be the case i f the attack of the magnesium oxide was much greater than could be detected metaliographically. From the r e l a t i v e l i n e i n t e n s i t y i t c a n be concluded that only a small amount o f n i c k e l was removed from the n i c k e l drop i n the preparation of the powder specimen. As the l i n e i n t e n s i t i e s of the nickel-magnesium compound are at i s a s t as great as those which could be due to n i c k e l , i t can be concluded that the compound existed at the i n t e r f a c e . Also i t can be concluded that x , . i l oxide and magnesium e x i s t i n q u a n t i t i e s not large enough to be detected ,/ X-ray a n a l y s i s . Any n i c k e l oxide that formed could p o s s i b l y e x i s t i n the form of -k e l oxide-magnesium oxide s o l i d s o l u t i o n . However, as these l i n e s have ->ximately the same d-values as magnesium oxide, any evidence f o r the existence of the s o l i d s o l u t i o n i s completely obscured by the presence of the magnesium oxide. Comparison of the r e l a t i v e l i n e i n t e n s i t i e s of magnesium oxide and n i c k e l oxide-magnesium oxide (A.S.T.M. Card No. 3.0990) shows the l i n e s , at l e a s t f o r the greatest part, to be due to magnesium oxide. - 74-APPENDIX I I X-ray D i f f r a c t i o n Data from Interface Between Magnesium Oxide and N i c k e l  + 2% NiO. A photographic p r i n t of the X-ray powder pattern of the i n t e r f a c i a l m a t erial between magnesium oxide and n i c k e l + 2% n i c k e l oxide i s shown i n F i g . 40. The observed X-ray data f o r the i n t e r f a c i a l m a t e r i a l and the F i g . 40. X-ray powder pattern No. 1530 of i n t e r f a c i a l material between magnesium oxide and n i c k e l + 2% n i c k e l oxide. material present, as i d e n t i f i e d from the A.S.T.M. Card Index, are shown i n Table V. The A.S.T.M. cards f o r n i c k e l and n i c k e l oxide are shown i n F i g s . 41 and 42. It can be concluded that the material at the in t e r f a c e consists of n i c k e l oxide and n i c k e l only. The presence of the n i c k e l oxide-water compound can be explained, as f o r the presence of magnesium hydroxide (see Appendix I ) , by assuming that the n i c k e l oxide reacts with the water i n the a i r or with water i n the grease used i n the preparation of the powder specimen. The lack of evidence f o r the existence of magnesium oxide i n the i n t e r f a c i a l material supports the experimental observation that fracture occurred at the interface rather than through the magnesium oxide, as happens with drops composed of pure n i c k e l . From the r e l a t i v e l i n e i n t e n s i t i e s of n i c k e l and n i c k e l oxide can be concluded that the n i c k e l oxide, i n addi t i o n to segregating TABLE V  Data from Film No. 1530 Line I / I 0 d M a t e r i a l 1 10 2.4287 NiO 2 5 2.3374 NiO.H20 3 100 2.1044 NiO 4 100 2.0351 Ni 5 30 1.7602 Ni 6 35 1.4890 NiO 7 5 1.2621 NiO 8 10 1.2455 Ni 9 8 1.2161 NiO 10 5 1.0596 Ni 11 2 1.0505 NiO 12 1 1.0305 NiO.H 20 13 5 1.0146 Ni 14 1 .96881 NiO 15 5 .95196 NiO 16 5 .85815 NiO 17 5 .80718 Ni 18 5 .78693 Ni - 76 -d 4 - 0 8 5 4 2.034 i 1.762 1.246 2.03', M i 4 - 0 8 5 0 100 12 21 100 IJ1 C « E L Had. Cu Dia. I/I, R e t S .VAN30N A N D T A 1 G E A 1.5405 Cut off Filter Ml Coll. d corr. abs.? X F E L . R E P O R T S , Sya. C- ig ic { F . C . ) a, 3.5238 b . no./? I\B.907 mp XV Ref. S P E C T R O G R A P H i c i S.G. OH - FM3U A C Color V S I S SHO'.VS <0.0l7O E A C H At 2GV, To R E P L A C E 1-1258, 1-1260, 3-1043, 3-1051 1-1266, 1-1272, d A 2.034 100 1.762 42 1.216 21 1.0624 20 1.0172 7 0.9810 4 . 3014 14 .7880 15 4 CO 331 F i g . &L. A.S.T.M. Card No. 4.0850 f o r n i c k e l . d 4 - 0 8 3 8 2 . 0 3 8 2 . 4 1 0 1 . 4 7 6 2 . 4 1 0 NiO I/I. 4 - 0 8 3 5 1 0 0 9 1 5 7 9 1 N I C K E L O X I D E B U N S E N 1 T E Rad. Cu A 1.5405 Cut off Filter Ni CoU. I/Ii G-M COUNTER " dcorr.abs.? Ret SWANSON A N D T A T G E , X F E L . R E P O R T S , I O S _ 1950 Sya. C U B I C a, 4.1769 b. S.G. O H -A Z FM3I, Ref. I D I D . ntop D x 6 .806 mp 7?' 1 Sign S P E C . A N A L , SHOWS F A I N T T R A C E S OF f/c, S i A N D C A . AT 2 6 C C To R E P L A C E 1-1239, 2-1216, AND 3-1287 d A 2.410 2.083 1.476 1.259 1.206 1.0441 0.95B2 .9338 .8527 .9040 91 100 200 220 311 222 400 331 420 422 511 d A Fig. Zt26 A.S.T.M, Card No. 4.0835 for nickel oxide. - 77 -at the outside of the n i c k e l drop (see Experimental Observations) also segregates at the i n t e r f a c e between the n i c k e l drop and the magnesium oxide. - 78 -APPENDIX I I I Ca l c u l a t i o n of Wetting Angle from Drop Dimensions F i g . 43o Drop with necessary dimensions to ca l c u l a t e wetting angle. The r e l a t i o n between diameter D, height H, and wetting angle 0 (see F i g . 43), on the assumption that the drop i s s p h e r i c a l , i s cos 0 = D - 2 H D Taking as an example the c a l c u l a t i o n of the wetting angle of high p u r i t y Ni on the 'as cleaved' magnesium oxide 20 minutes a f t e r melting, where the fol l o w i n g measurements were made? D = 2.091 inches and H = 1.290 inches s u b s t i t u t i o n o f these values gives: cos 0 = -.229 from which 0 = 103.2 The accuracy by which the wetting angle can be determined by t h i s method i s ±2°. - 79 -APPENDIX IV Cal c u l a t i o n of Stresses at the Interface between Magnesium Oxide and  N i c k e l Drop Resulting from the Differences i n Thermal Expansion. The d i f f e r e n c e i n thermal expansion pei centimeter of in t e r f a c e equals (a^ - az)^T, where a-^  i s the c o e f f i c i e n t of l i n e a r expansion f o r magnesium oxide32 and measured to be 13.5 x 10"°^ per °C„ <x2 xs the c o e f f i c i e n t of l i n e a r expansion f o r n i c k e l ^ and equals 13.46 x 10" 6 per ° C . and i s the temperature range i n °C. through which the specimen was cooled. On cooling the n i c k e l w i l l be i n compression because of i t s lower c o e f f i c i e n t of expansion and the magnesium oxide w i l l be i n tension therefore the s t r a i n s i n these materials w i l l be of opposite s i g n . Therefore the d i f f e r -ence i n thermal expansions w i l l be equal to the sum of the e l a s t i c s t r a i n s (neglecting bending) i n each ma t e r i a l . This gives, ( a x - a e ) A T f ( l - v) + (1 - v) \ I E l E 2 J 2 i+ "] ? where E-^  i s Young's modulus of magnesium oxide and equal t o 1.72 x 10 dynes per cm.2 34 JL2 2 E 2 i s Young's modulus o f n i c k e l and equals 2.07 x UJ dynes per cm. & i s the r e s u l t i n g stress at the i n t e r f a c e , and v i s Poisson's r a t i o , here assumed to be equal to .35. Rearranging gives: ^ = E i E 2 ( a i - Q. 2)AT (E l + E 2 ) ( l - v) which gives, upon substitution' of the appropriate values, - 80 -6 = 5.78 x 1 0 ^ A T dynes per cm.2 The resolved shear s t r e s s on the s l i p planes then becomes 6 = 2.89 x 10 A T dynes per cm. 19 8 The resolved y i e l d s t r e s s 7 at room temperature i s approximately 6.90 x 10 dynes per cm. Therefore the temperature i n t e r v a l through which the specimen must be cooled f o r the stresses at the interface to exceed the y i e l d stress i s where 2.89 x 1 0 4 A T - 6.90 x 10 8 or A T - 6.90 x 10^ - 2.39 x 10^ °C. 2.89 x 10 4 As the temperature i n t e r v a l through which the specimen i s cooled i s approxima-t e l y 1.4x103°C .or only about 6% of the i n t e r v a l required f o r p l a s t i c deforma-t i o n , i t can be concluded that the di f f e r e n c e s i n thermal expansion can not cause p l a s t i c deformation. The accuracy of t h i s c a l c u l a t i o n i s of course dependent on the accuracy by which the c o e f f i c i e n t s of expansion could be determined. The c o e f f i c i e n t of l i n e a r expansion f o r n i c k e l can be assumed to be accurate. The value selected f o r magnesium oxide was considered to be the most r e l i a b l e one. Other values reported f o r the c o e f f i c i e n t of l i n e a r thermal expansion are 133 x 10~ 6 per2*-1 °C. and 13.3 x 10~ 6 per12*- °C. Calculations based on these values also show that p l a s t i c deformation could not have taken place by the differences i n contraction on cooling. - 81 -APPENDIX V Cal c u l a t i o n of Thermal Stresses at the Interface between the Nickel Drop and the Magnesium Oxide Resulting from Temperature Differences on Cooling. In p r i n c i p l e one can calculate the thermal stresses that occur on cooling by s o l v i n g the p a r t i a l d i f f e r e n t i a l equation f o r heat flow to obtain equations f o r the temperature d i s t r i b u t i o n as a function of time. From these equations the thermal stresses can then be calcu l a t e d . However, i n order to be able to solve the p a r t i a l d i f f e r e n t i a l equation f o r heat flow, one requires a d e t a i l e d knowledge of the boundary conditions. It i s experimentally impossible t o determine these. Even an approximation cannot be made, because of the of the f a s t response of the temperature of the susceptor to power input. However, an estimation of the thermal stresses can be made. Various models and temperature d i s t r i b u t i o n s can be assumed, such as a t h i n c i r c u l a r d i s c with a temperature d i s t r i b u t i o n or a body with a higher temperature than i t s surroundings, but constrained to expand. b with a temperature d i s t r i b u t i o n symmetrical about the center, as given by 35 Timoshenko and Goodier are The r a d i a l and angular stresses f o r a t h i n c i r c u l a r d i s c of radius b r o o b r j = oE - A T + 1 where a i s the c o e f f i c i e n t of l i n e a r expansion E i s Young's modulus - 82 -A T i s the temperature d i f f e r e n c e between the center of the plate and the perimeter. r i s the distance from the center to the region under consideration. Assuming a parabolic temperature dependence gives f o r the temperature d i s t r i -bution T(r) - A T r * Su b s t i t u t i o n of t h i s into the equation f o r the stresses andintegration gives (5 - aE A T / 1 - _ j £ _ \ and 4 ~4bT" ^8 \ 4 ZbZ/ which gives f o r r - 0 ir - O E A T , • 4 6 53 -3 aE A T and f o r MS T h r = b §y = 0 6 e = -1 aE AT S u b s t i t u t i n g the values, a = 13.5 x 1CT 6 per °C. and E = 1.72 x 1 0 1 2 dynes per cm.2 gives f o r r = 0 <5K= 5.84AT x 10 dynes per cm/ 6 » -17.5 A T x 10 6 dynes per cm.2 and at r = b 6 = -11.7AT x 10 dynes per cm. - 83 -the corresponding resolved shear stresses at the s l i p planes w i l l be f o r r - 0, 6 r = 2 .92^ T x 10 6 dynes per cm.2 6 3 = -8.75A T x 10 6 dyne3 per cm.2 and f o r r = b r f ^ - -5.85 T x 10 6 dynes per cm.2 The resolved y i e l d s t r e s s has been measured to be approximately 3 .1 x 10 dynes per cm.2, therefore p l a s t i c deformation w i l l occur at r = 0 when A T = 3 ,1 x IO 8 8.75 x 10 b and r = b when - 35.4 °C. A T - 3 .1 x 10 8 5.85 x 10 6 = 53°C. Another more simple c a l c u l a t i o n i s based on the s t r e s s necessary to constrain a body from expanding under increase of temperature. In t h i s case the thermal expansion equals the e l a s t i c s t r a i n due to the applied s t r e s s . This gives a A T = 6_ E where i s the r i s e i n temperature of the body. Rearranging gives 6 = aE AT = 13.5 x 10" x 1.72 x 10 A T dynes per..cm. 6 2 = 23.2 x 10 A T dynes per cm. The resolved shear stress on the s l i p planes i s then 6 2 = 11.6 x 10 A T dynes per cm. - 84 -The temperature r i s e f o r the shear stress at 1500°C to be equal to the y i e l d stress ^ A T = 3.1 x 10 - 26.7 °C. I I 5 6 x 1 ? I Assuming that these c a l c u l a t i o n s are applicable to the conditions at the i n t e r f a c e between the n i c k e l drop and the magnesium oxide, one can conclude that a temperature d i f f e r e n c e between the center of the in t e r f a c e and the perimeter of approximately 35°C can produce stresses which w i l l exceed the y i e l d s t r e s s . From the second c a l c u l a t i o n i t can be concluded that a mean temperature diffe r e n c e between the region under the drop and the region outside the drop of approximately 25 to 30°G. w i l l also cause p l a s t i c deformation. Temperature differences such as these are w e l l w i t h i n the l i m i t s of those expected. Observation of the furnace and specimen when cooling shows that the susceptor w i l l be at a temperature where i t does not emit v i s i b l e r a d i a t i o n i n le s s than t h i r t y seconds. The n i c k e l drop, however, w i l l remain red hot f o r at lea s t two or three minutes. From t h i s i t can be concluded that the areas of the magnesium oxide, not covered by the n i c k e l drop, can emit r a d i a t i o n to a v i r t u a l l y cold body, whereas the area under the drop cannot. It i s f e l t that temperature differences of the order of 100°C. are not impossible. APPENDIX VI Ca l c u l a t i o n of Density of S l i p Sources Introduced During Cleavage. This c a l c u l a t i o n i s based on the f o l l o w i n g assumptions; 1. each s l i p l i n e i s caused by the presence of one source. 2. a l l sources under the n i c k e l drop are a c t i v a t e d . 3o no s l i p i s produced by sources outside the area of the surface covered by the n i c k e l drop. 4. there i s an equal number of sources that give r i s e to s l i p l i n e s i n one [ 100J d i r e c t i o n as i n the perpendicular d i r e c t i o n . 5 . a l l the sources produce s u f f i c i e n t s l i p to be detected by o p t i c a l means. 6. a l l the s l i p l i n e s are continuous from one side of the i n t e r f a c e to the other. F i g . 8 represents an area i n the center on the i n t e r f a c e between a n i c k e l drop and magnesium oxide p l a t e l e t at a magnification of 250x. The width of the f i g u r e i s 10 cm. The diameter of the i n t e r f a c e has been measured to be .21 cm. The number of v e r t i c a l s l i p l i n e s across the width of the pi c t u r e , counting from l e f t to r i g h t , i s approximately 5 0 . These 50 s l i p l i n e s were produced by s l i p sources i n an area equal to 10 x .21 cm.2 = 8.4 x 10" 3 cm.2 250 As many s l i p sources are introduced during cleavage f o r s l i p i n one d i r e c t i o n as f o r s l i p i n i t s perpendicular d i r e c t i o n . Therefore i n t h i s area of -3 2 8.4 x 10 cm. 100 s l i p sources are introduced. This gives a s l i p source density o f 2 - 1°0 y, 12,000 s l i p sources per cm. 8.4 x 10"3 - 86 -Calculations based on the same assumptions f o r the s l i p source density of an i n t e r f a c e with a high cleavage step density, gave a density of p approximately 15,000 s l i p sources per cm, . The assumptions on which these calculations are based are v a l i d i n so f a r that i t i s reasonable to assume that each s l i p l i n e i s caused by one s l i p source only. However, i t may not be v a l i d to assume that a l l the sources under the n i c k e l drop are activa t e d as the thermal stresses w i l l be a function of p o s i t i o n and may not be high enough i n some areas to a c t i v a t e the s l i p sources. However, the experimental observation that the highest cooling rate d i d not give r i s e t o more s l i p l i n e s than a lower cooling rate would indicate that a l l the sources were a c t i v a t e d . The t h i r d , fourth and s i x t h assumptions are supported by a c t u a l experimental observation. However, the assumption that a l l s l i p sources produce enough s l i p to be detected, i s not v a l i d . The lowest s l i p l i n e that ° 29 may be detected i s of the order of 30 A high, which corresponds to a t o t a l production of about f i f t e e n d i s l o c a t i o n s . The a c t u a l p l a s t i c deformation amounts to very l i t t l e . Therefore from the point of view of amount of deformation produced, i t seems reasonable to base a c a l c u l a t i o n of s l i p source density only on those s l i p l i n e s which can be detected o p t i c a l l y . - 87 -APPENDIX VII C a l c u l a t i o n of Minimum Size of Stable D i s l o c a t i o n Loop. This c a l c u l a t i o n i s based on the assumption that the force which anchors the d i s l o c a t i o n loop i s the minimum shear stress f o r d i s l o c a t i o n g l i d e . The f o r c e which tends to contract the d i s l o c a t i o n loop a r i s e s from the tension of the d i s l o c a t i o n l i n e . By equating these forces and assuming that the loop i s c i r c u l a r gives the s i z e of the smallest loop stable i n the c r y s t a l . I Q The tension o f a d i s l o c a t i o n l i n e as given by C o t t r e l l T X/ 0.5/ t/b 2 where T i s the tension of the d i s l o c a t i o n ytxs i s the shear modulus b i s the length of the Burger's vector. The shear modulus can be calculated from S*. = E 2(1 + v) where E i s Young"s modulus and f o r magnesium oxide was measured 12 2 to be 2.51 x 10 dynes per cm. v i s Poisson's r a t i o and assumed to be 0.35. This gives 9 . 3 x 10 dynes per cm. b can be c a l c u l a t e d from the l a t t i c e parameter. This gives ft b - 2.97 x 10 cm. S u b s t i t u t i o n o f these values i n t o the equation f o r the t e n s i o n gives T = 4.1 x 10"^ dynes, l a The minimum shear s t r e s s x S f o r d i s l o c a t i o n g l i d e at room temperature was measured to be g 2 6" «= 3 . 0 x 10 dynes per cm. - 88 -Using the equation D = 2_T where D i s the diameter of the smallest 6 b stable d i s l o c a t i o n loop and T, 6 and b, as defined above, gives upon s u b s t i -t u t i o n D 1 micron The a c t u a l value of D can d i f f e r from t h i s value considerably. The calculated value i s no more accurate than the equations and values used f o r t h i s c a l c u l a -t i o n o - 89 -APPENDIX VIII Thermodynamic Ca l c u l a t i o n s. (a) D i s s o c i a t i o n of n i c k e l oxide. 2 NiO ± 5 2 Ni + 0 2 The n i c k e l oxide w i l l be just stable when the free energy change A F = 0 At 1 atm. pressure of 0 2 at 1500°C A F = +32,000 c a l . per gram mol. 0 2 At any other pressure A F - +32,000 + 4.575 x 1773 log P o 2 m 0 therefore l o g P Q =» 32.000 2 4.575 x 1773 - -3.94 therefore P n - lO"2*- atm. m .76 mm. Hg. As the t o t a l pressure i n the system i s approximately 10 to 10" mm. Hg, the n i c k e l oxide w i l l completely d i s s o c i a t e . This i s supported by the experimental observation that the n i c k e l oxide coating which formed on the n i c k e l drop, o r i g i n a l l y containing 2% n i c k e l oxide, gradually disappeared. (b) D i s s o c i a t i o n of magnesium oxide 2 MgO t^T2 Mg + 0 2 The magnesium oxide w i l l be just stable when the free energy change A F -.0 At 1 atm. 0 2 and 1500°C & F = +184,000 c a l . per gram mol. 0 2 - 90 at any other pressure A ? a +184,000 + 4.575 x 1773 log P 2 PQ or L e t t i n g Pj^ » 2 PQ 2 gives lo g 4 P 3 ^ = -22.7 l o g P 3 A = -23.3 '2 ==10 -7.8 atm. therefore -7.5 P. Mg = 10 atm. The t o t a l pressure of the system as indicated by the i o n i z a t i o n gauge i s approximately 10"•> mm. Hg or 1.3 x 10~ 8 atm. The p a r t i a l pressure of oxygen w i l l be lower than t h i s value. Also, one should consider t h a t , i n a d d i t i o n to pumping, oxygen i s continuously removed by oxidation of the molybdenum. The oxygen p a r t i a l pressure might therefore be s u b s t a n t i a l l y lower than the pressure c a l c u l a t e d . In order to s a t i s f y the e q u i l i b r i u m constant, the p a r t i a l pressure of the magnesium must be increased. The magnesium oxide w i l l therefore d i s s o c i a t e . This i s supported by the observation that magnesium oxide gradually v o l a t i l i z e s at these temperatures and pressures. (c) C a l c u l a t i o n of thickness of i n t e r f a c i a l layer of magnesium-nickel  compound. 36 This c a l c u l a t i o n i s based on a c a l c u l a t i o n by Williams and Murray f o r the system A 1 2 0 3 - Cr i n which i t was shown that the formation of a s o l i d s o l u t i o n at the i n t e r f a c e r e s u l t s i n a lowering of the surface tension. By equating the f r e e energy of formation of the compound to the energy required to lower the i n t e r f a c i a l tension to zero, the thickness of the - 91 -i n t e r f a c i a l layer can be c a l c u l a t e d . The free energy change during the formation of a s o l i d s o l u t i o n or compound A F = + RT(N]_.lnN^ + N 2 l n N 2 ) where A H i s the heat of formation and -R(N]_lnNi + N 2 l n N 2 ) i s the entropy of mixing. R i s the gas constant and N]_ and N 2 are the molar f r a c t i o n s of the compound. 37 may be assumed-^' to be approximately -2,000 c a l o r i e s per gram mole. The term -R(N3_lnNi + N 2 l n N 2 ) can be ca l c u l a t e d to be 1.28. This gives A F = -2,000 - 1.28 T at 1500°C A F = -2,000 - 2,270 = -4,270 c a l o r i e s per gram mole = -1.79 x 1 0 1 1 ergs per gram mole. As the compound i s located at the i n t e r f a c e the change i n free energy may be r e l a t e d to a decrease i n the i n t e r f a c i a l tension. The r e l a t i o n between area and thickness i s given by area x thickness = molar volume = mol. weight of compound density The molecular weight can be ca l c u l a t e d to be 107.3 grams per mol. The density can be calculated from the l a t t i c e parameters to be 3.47 grams per cc. therefore area x thickness = 107.3 = 30.9 3.47 * n *» free energy change per cc, - . = - 1.79 x 10 30.9 9 = -5.79 x 10 ergs,,per cc. The i n t e r f a c i a l t ension between n i c k e l and magnesium oxide can be determined from the w e t t a b i l i t y data. The equation r e l a t i n g the i n t e r f a c i a l - 92 tension and the surface tensions i s y •. • =:• y\, n - y„. cos © MgO-Ni " M § ° N i where • T^^Q ^ i s the i n t e r f a c i a l tension ^MgO ^ S ^ e S u r ^ a c e tension of the magnesium oxide.and calculated- 7 to be approximately 890 ergs/cm. at 1500°C. "TJJ^ i s the surface tension of l i q u i d n i c k e l and 11 2 was measured to be 1845 ergs/cm. Taking 6 = 102° gives f o r the i n t e r f a c i a l tension ^MgO-Ni = 8 9 0 + « 2 X ^ = 1270 ergs/cm, 2 For wetting to occur, Q must equal 0, i . e . , cos Q equal 1. This l i m i t i n g condition gives ^MgO-Ni = 8 9 0 " 1 8 ^ 5 = -955 ergs/cm. 2 Therefore the t o t a l decrease i n i n t e r f a c i a l tension which must take place, i f wetting i s to be achieved, i s -(1270 + 955) = -2225 ergs/cm. 2 Equating the decrease i n i n t e r f a c i a l tension to the free energy change of formation gives an i n t e r f a c i a l thickness o f 2225 cm. 5.79 x 10 9 = 38.4 A Therefore, assuming that the parameters used i n t h i s c a l c u l a t i o n are correct, 38»4 A- i s the minimum thickness of the i n t e r f a c i a l layer i f complete wetting - 93 -i s to take place. The a c t u a l thickness of the i n t e r f a c i a l layer exceeded t h i s f i g u r e s u b s t a n t i a l l y as a much th i c k e r layer f o r X-ray analysis was removed, which contained l i t t l e or no n i c k e l . - 94 -BIBLIOGRAPHY 1. White, A.E.S., Earp, E.K., Blakeley, T.H. and Walker, J., Symposium on Powder Metallurgy, 311 (1954). 2. Kingery, W.D., J. Am. Ceram. Soc, 3_6, 362 (1953). 3. Economos, G. and Kingery, W.D., J. Am. Ceram. Soc, 3_6, 403 (1953). 4. Himenik, M..Jr. and Kingery, W.D., J. Am. Ceram. Soc, 3_7_> 18 (1954). 5. Kingery, W.D., J. Am. Ceram. Soc, 3_7_, 42 (1954). 6. Johnston, P.D., J. Am. Ceram. Soc, 21> 168 (1950). 7. Allen, B.C. and Kingery, W.D., A.I.M.E. Trans., 215. 30 (1959). 8. Moore, D.G. and Thornton, H.P., J. Res. Nat. Bur. Standards, 62, 127 (1959). 9. Ellefson, B.S. and Taylor, N.W., J. Am. Ceram. Soc, 21, 193 (1938). 10. Ellefson, B.S. and Taylor, N.W., J. Am. Ceram. Soc., 21, 205 (1938). 11. Clarke, J.F., Thesis, Univ. of British Columbia, 1959. 12. King, B.W., Tripp, H.P. and Duckworth, W.H., Paper at 58th Meeting of Am. Ceram. Soc, New York City, 1956. 13. Cronin, L.J., B u l l . Am. Ceram. Soc, 3.0, 234 (1951). 14. Norton Company, Customer Bulletin, CP 843. 15. Parker, E.R., Pask, J.A., Washburn, J., Gorum, A.E. and Luhmann, W., Journal of Metals, 10, 351 (1958) 16. Gorum, A.E., Parker, E.R. and Pask, J.A., J. Am. Ceram. Soc, £1, 161 (1958), 17. Washburn, J., Gorum, A.E. and Parker, E.R., A.I.M.E. Trans., 215. 230 (1959) 18. Stokes, R.J., Johnston, T.L. and L i , C.H., A.I.M.E. Trans., 215. 437 (1959). 19. Johnston, T.L., Stokes, R.J. and L i , C.H., Honeywell Research Center, Fourth Technical Report, (Feb. 1959). 20. Handbook of Chemistry and Physics, 3_6, 2062 (1956). 21. Kingery, W.D., J. Am. Ceram. Soc, 20, 3 (1955). 22. MacFarlane, W. Jr., Metallurgical Club Journal, 8, 15 (1955). - 95 -Bibliography (cont'd.) 23. Levin, E.M., McCurdie, H.F. and Hall, F.P., Phase Diagrams for Ceramists Am. Cer. Soc. Inc. (1956). 24. Brice, and Strong, Bull. Am. Phys. Soc, 6 (1939). 25. Ditchborn, R.W., Nature, 136, 70 (1935). 26. Turnbull, R.C., Monthly Progress Report No. 214, Alfred University, Alfred, N.Y. 27. Pask, J.A., Private communication. 28. Gorum, A.E., Parker, E.R. and Pask, J.A., J. Am. Ceram. Soc., £1, 161 (1958) 29. GilLman, J.J., A.I.M.E. Trans., 212, 310 (1958). 30. Gorum, A.E., Private communication. 31. Jessop, H.T., and Harris, F.C., Photoelasticity. Cleaver-Hume Press,Ltd., London. 32. Kingery, W.D., J. Am. Ceram. Soc., £8, 3 (1955). 33. Handbook of Chemistry and Physics, 36th Ed. Chemical Rubber Publishing Co. 34. Metals Handbook, A.S.M. (1948). 35. Timoshenko, W. and Goodier, J.N., Theory of Elasticity, McGraw-Hill. 36. Williams, L.S. and Murray, P., Metallurgia 210 (1954). 37. Samis, C.S., Private communication. 38. Livey, D.T. and Murray, P., J. Am. Ceram. Soc., 21, 363 (1956). 39. GiLman, J.J., A.I.M.E. Trans. 209. 449 (1957). 40. Cottrell, A.H., Dislocations and Plastic Flow in Crystals, Clarendon Press. 41. Winkler, H.G.F., Struktur und Eigenschaften der Kristalle, Springer Verlag (1955). 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0105929/manifest

Comment

Related Items