Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Interface reactions between iron alloys and magnesium oxide single crystals Rose, Daneil Joseph 1962

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

Item Metadata

Download

Media
831-UBC_1962_A7 R65 I5.pdf [ 9.38MB ]
Metadata
JSON: 831-1.0105618.json
JSON-LD: 831-1.0105618-ld.json
RDF/XML (Pretty): 831-1.0105618-rdf.xml
RDF/JSON: 831-1.0105618-rdf.json
Turtle: 831-1.0105618-turtle.txt
N-Triples: 831-1.0105618-rdf-ntriples.txt
Original Record: 831-1.0105618-source.json
Full Text
831-1.0105618-fulltext.txt
Citation
831-1.0105618.ris

Full Text

INTERFACE REACTIONS BETWEEN IRON ALLOYS AND MAGNESIUM OXIDE SINGLE CRYSTALS  by DANIEL JOSEPH ROSE  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in the Department of METALLURGY  We accept this thesis as conforming to the standard required from candidates for the degree of MASTER OF APPLIED SCIENCE.  Members of the Department of Metallurgy. THE UNIVERSITY OF BRITISH COLUMBIA February 1962  In presenting the  this thesis in partial fulfilment  r e q u i r e m e n t s f o r an  of British-Columbia, it  freely  agree that for  available  advanced degree a t  I agree t h a t , t h e f o r r e f e r e n c e and  permission for extensive  s c h o l a r l y p u r p o s e s may  D e p a r t m e n t o r by  be  University  Library  s h a l l make  study.  I  copying of  g r a n t e d by  his representatives.  the  the  gain  s h a l l not  Department o f  Metallurgy  The U n i v e r s i t y o f B r i t i s h V a n c o u v e r 8, C a n a d a . Date  March 22,  1962  Columbia,  this  thesis my  I t i s understood  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 a l l o w e d w i t h o u t my  further  Head o f  that  be  of  written  financial  permission.  ABSTRACT The wettability of MgO by liquid iron was studied using the sessile drop technique in vacuo at 1 5 5 0 ° C .  Specimens of vacuum-cast iron were melted  on single crystals of optical-grade magnesium oxide. two minutes because of iron volatilization.  Tests were limited to  Chemical reaction at the liquid-  vapor interface caused the contact angle to decrease from 1 1 7 ° to 6 5 ° during the f i r s t minute.  Reaction between the iron and oxygen provided through the  dissociation of MgO resulted in formation of an FeO layer over the drop surface.  The FeO was drawn to the base of the drop where i t accumulated  as an annular Interfacial deposit. to magnetite  After solidification, the FeO decomposed  with separation of iron.  Within the peripheral annulus the  interface showed no sign of chemical attack.  This effect was related to  formation of a protective interfacial monolayer from electropositive impurities i n i t i a l l y present in the iron.  The major constituent of the monolayer appeared  to be silicon although positive identification was not achieved. Chemical reaction occurred in each case where alloying additions to the iron of T i , V, Cr, Nb, Ta, and Zr were made. angle with concentration was studied.  Variation of contact  Both V and Cr improved the wettability.  The effect of Zr, T i , Nb, and Ta on the wettability was not determined because oxide deposits around the basal periphery of the drops restricted expansion. Complete interfacial attack was observed with additions of Zr, T i , and V whereas additions of Nb, Ta, and Cr resulted in an annular interfacial deposit around the drop perimeter similar to the pure Fe-MgO specimens. supported the existence of an interfacial monolayer.  This phenomena  ACKNOWLEDGEMENT The author gratefully acknowledges Professor W. M.. Armstrong, Dr. A. C. D. Chaklader, and Mrs. A. M. Armstrong for their advice, guidance, and assistance during the investigation. The author is also indebted to Clayburn-Harbison Limited who • provided the necessary financial assistance for this work in the form of a Fellowship.  iii. TABLE OF CONTENTS Page 1  I. INTRODUCTION A. General Discussion of Metal-Ceramic Bonding  1  B. Metal-Oxide Binary Systems  2  C. The Sessile Drop Technique  3  D. The Effect of Solute Additions to the Liquid-Metal Drop . .  6  E. Factors influencing Selection of the Fe-MgO System and 8  Alloying Components . . . . . F. Previous Work  1 0  G. Specific Purpose of the Present Investigation  1 2  II. EXPERIMENTAL  1 3  A. Materials  1 3  1 . Magnesium Oxide  1 3  2 . Iron and Additive Metals  1 3  B. Apparatus  1 9  1 . Furnace  1 9  2 . Vacuum System  2 3  3 » Optical System  2 3  C. Specimen Preparation  2k  1 . Magnesium Oxide  2k  2 . Iron and Additive Metals D. Experimental Procedure  25 2 5  1 . Sessile Drop Experiments 2 . X-ray Diffraction  .  .  3 . Metallographic Observation . . . . . . . . . . . . . .  2 5 27 2 7  iv. T a b l e o f Contents  Continued Page  III.  RESULTS AND DISCUSSION .  .  .  .  .  .  .  .  .  .  A . Pure Fe-MgO System  29  1. R e s u l t s and C a l c u l a t i o n s  29  a) M o l t e n I r o n - - S o l i d MgO System i.  b)  29  S e s s i l e drop experiments  ii.  Thermodynamic c a l c u l a t i o n s  29 . . . . . . . .  S o l i d I r o n - - S o l i d MgO System i,  .  3k .  • Weight measurement  ii. iii.  Metallographic X-ray results  observations .  .  Iron A l l o y  3^  .  37 ^0  - - MgO System  kQ  1. R e s u l t s and C a l c u l a t i o n s a) M o l t e n Drop - - S o l i d MgO System . b)  ii.  Interfacial  VII.  14-8 56  observations  56 56  b. Microscopic  56  X-ray results  59 .  CONCLUSIONS  APPENDICES  60  65  V. RECOMMENDATIONS FOR FURTHER WORK VI.  .  a . Macroscopic  2. D i s c u s s i o n  IV.  k-8  S o l i d Drop - - S o l i d MgO System i.  3^43^-  2. D i s c u s s i o n B.  29  .  -.  BIBLIOGRAPHY  66 67  . '  95 .  V.  LIST OF FIGURES  ^  Wo.  Page  1.  Sessile Drop Parameters . . .  2.  Photographs of Apparatus  3.  Schematic Diagram of Apparatus  5  . . .  20 21  k. Susceptor Assembly and Sample Specimen  22  5.  Specimen Mounted for Metallographic Observation  .  28  6.  Contact Angle Vs. Time for Pure Iron System  .  30  7*  Sessile Drop Photographs  .  31  8.  Interface Energy Vs. Time .  33  9.  Interface Types as seen through MgO Plate  .  1 0 C o n t a c t Angle Vs. Basal Diameter  35 38  11.  Photomicrographs of Interface  39  12.  Lattice Parameter Vs. Atomic Percent Iron  ^7  13.  Contact Angle Vs. Time (Ta-alloyed Iron)  50  lk.  Contact Angle Vs. Time (Nb-alloyed Iron)  . . .  51  15.  Contact Angle Vs. Time (Cr-allpyed Iron)  . . .  52  16.  Contact Angle Vs. Time (V-alloyed Iron)  53  17.  Contact Angle Vs. Time (Ti-alloyed Iron)  5k  18.  Contact Angle Vs. Time (High Carbon Steel)  55  19.  Interfacial Deposits  57  20.  Interface Structures  21.  Plot of x Vs. /3 z Plot of/x] and/z]  22.  Wjrf = 90°  . . . . .  75 Vs. /3 .  = 90° '  Plot of xj_ Vs. / 3 b 2k.. Plot of zj_ V s . ^ b 25. Histogram of Surface Tension Variation  23.  58  76 77 78 82  26.  Vapor Pressures of Iron  86  27.  Phase Diagrams of Fe-0 and FeO-MgO Systems  90  vi. LIST OF TABLES No.  Page  I.  Free Energies of Oxide Formation  II.  Spectrochemical Analysis of MgO  III.  Properties of Materials Used  IV.  Published Surface Energies of Magnesium Oxide  V.  Spectrographs Analyses of Me'tals Used  VI.  Summary of Liquid Iron Surface Tension Values  VII.  Interface Energies at Various Times  9 . . . . . .  ik 15 l6 .  .  17 18 32  VIII. Interface Annulus Measurements  36  IX.  Sessile Drop Data  68  X.  Table of Liquid Iron Surface Tension Values  80  XI.  X-ray Data  88  INTERFACE REACTIONS BETWEEN IRON ALLOYS AND MAGNESIUM OXIDE SINGLE CRYSTALS  I. INTRODUCTION A. General Discussion of Metal-Ceramic Bonding The superior strength, corrosion, and thermal shock properties of such metal-ceramic combinations as enamelled metals, cermets, and dispersionhardened alloys have stimulated interest in the nature of metal-ceramic interactions..  Since these materials are" heterogeneous ( i . e .  composites of distinct  metal and ceramic phases) the interfacial bond strength must compare favorably with the. bulk strength of the weakest component i f the substance is to be used for industrial application.  Investigators of the bond mechanism have  employed both the crystallographic and chemical approach.  The crystallographic  approach attempts to explain interfacial properties through structural configurations: ;and interatomic bond relations, whereas the chemical approach endeavours to describe interfacial properties in terms of energetic quantities derived from thermodynamic relations.  In addition, the effects of diffusion,  mechanical stresses, and thermal stresses upon the interface have been> studied to some extent. The ultimate aim of bond investigations is the evolution of fundamental theories that w i l l aid in predicting the behaviour of systems which have not been studied." In order to avoid the complexity of multicomponent systems, most studies have treated binary systems.  These studies  have been complicated in many cases, however, by the presence of trace impurities which affect the interfacial bond.  - 2 -  B.  Metal-Oxide Binary  ment  In  comparison  of  metal-oxide  of  oxides- r e l a t i v e  in  this  of  in  interface  the  work  area of  of  such  to  the of  a  has  surface  liquid  of  minor  cermets,  importance.  warrants  results  new  low  the  However  further  developthe  abundance  investigation  one  decrease  and a u n i t  of  o  r  area  of  compounds  metal  and  (carbides,  reaction;  (or  of  formed its  of  (^SL^  to  liquid  i n the  chemical  together  and  a  and  interfacial  etc)  some  an that  impurity  i n excess  between  bulk  impurity,  through  impurities)  diffusion into  the  of  the  oxide.  liquid  area  region:  impurity  the  unit  oxide.  formed  of  define  solid  solid  reaction  gases with  The  to  form a u n i t  metal  ^  refer  required  oxide,  oxide  a n c  A l l e n and Kingery  s o l i d metal  nitrides, the  significant  metal,  containing  solutions  solid  largely  indicated.  work  surface  the  form  from d i r e c t  metal  liquid  may  the  in bringing  solid  the  energy  otherwise  area.  between  configurations  liquid  metal  unit  The  "interfacial"will  as  1  i n energy  high  resulting  the  unless  by  determined  molten metals.  report,  separation  concentration:: solid  i n this  Glasstone  the  have been  solid-liquid interface  strong bond between  oxides  liquid  the  by  by  XQL  in a  other  are  oxides  defined  as  surface A  solid  Hereafter  been  adhesion  in  k.  of  solid-liquid interface  through  3.  been  and n i t r i d e s  of  study ).  the  the 2.  and m e t a l - n i t r i d e  metal-oxide.interfaces  studies  A variety 1.  carbides  (W  energy  interface.  usually  to  has  affecting  adhesion  specifically  enlarge  metal-carbide  cermets  wettability  parameters work  to  field.  Factors through  Systems  oxide  - 3 The may b e of  interfacial  altered  eutectics,  upon  these  a  interface  low  phenomena  in a  "by n e w  p h a s e s be.-  rapid  C.  strong  cooling  The  of  Sessile  of  an  oxide  i n the  large  the  stresses  bond.  but  also  only  thermal  to  separation  changes,  etc.  interfacial  does must  liquid  due  phase  i n the  system  Not  molten  cooling  compounds,  liquid-solid  considered,  not the  stresses  region,  necessarily stresses  produced  established  during  system.  Technique  sessile  surface  i t  1.  create  i n the  subsequent  intermetallic  metal-ceramic  Drop  The  of  may  energy  result  present  s o l i d i f i c a t i o n and  formation  Since  configuration  drop  technique  is  valuable  i n determining  the  wettability  because:'  allows  the  simultaneous  determination  of  liquid  surface  3  tension the  2.  drop the  and  interface  provides  a  large  identification  energy planar  of  , interface  interfacial  which  facilitates  configurations,  diffusion,  k stresses,  etc.  ,  5 ' sessile  3»  the  k.  drop  method  theory  is  without  is  thermodynamically  adaptable apparatus  temperature  to  numerous  justified  metal-oxide  systems  (aside  possible  modification  the  shape  forces  7  experimental If  a  of  the  tending  liquid  results drop  drop  to  is  is  the  flatten  drop  reasonably  resting  the  drop  . .  are  on a  determined by  2 spheroidize  from  limitations;  ; 5.  ,  The  and  precise  solid  .  surface  as  i n Figure  an e q u i l i b r i u m between surface-tensional  1.  gravitational  forces  tending  to  •'  r e l a t i o n between  the  forces  acting  on the  peri-  8 phery stated  of by  the  solid-liquid  Dupre*  i n the  interface  was  first  equation, cos  ^SL which  is  derived  from a  <*SV horizontal  implied by  Young  and  9  ^LV resolution  of  the  energy  vectors:  later  - h ; where  JfgL o'sy^•^Y 9  = solid-liquid interface energy =• solid-vapor interface energy = liquid-vapor interface energy = contact angle'' measured through the liquid  Under the influence of the three interface energies the"drop may exist in equilibrium at any angle between 0° and l 8 0 °  .  The drop is said to "wet"  the surface i f i t possesses a contact angle less than 9 0 ° ' o Direct measurementsof 9 becomes inaccurate at angles greater than 90 • 11  In order to avoid this problem,.Bashforth and Adams (Appendix II.) for calculating'9 and O'LV ^  r o m  developed a method  dimensions x, x', z, z'  shown in Figure 1. Dorsey ^LY  r r o m  12  •  developed a simpler but less accurate method of deriving  measurement of the two dimensions x and b (also shown in Figure 1 . ) . The error in Bashforth and Adams' method has been quoted within  , 6, 13 • . .6 * 270 whereas the error in Dorsey's method may be ± yj° •  I* both cases 1  the principal error arises from the simultaneous existence of several ih 9's around.the drop periphery  different  7  .  Baes attempts to minimize this error by 1  measuring the drop dimensions from five different viewing directions. The accuracy of the sessile drop technique is slightly dependent upon the volume of the drop, which can be described by the parameter x . From z a consideration of Baes' results, which support the use of large drops, and Mack's discussion ^, which favors small drops, the optimum x value/apparently z lies somewhere between 1.15 and 1.70. 1  For 9 between 0° and 9 0 ° ,  O'LV  c a n n  ° t ^e calculated.  The contact  angle can be obtained, however, byjassuming the drop to be a spherical segment ^. 1  The contact angle is then given by 9 = 2  tan" h , x 1  where h and x are the dimensions shown in Figure 1.  - 5 -  Acute C o n t a c t A n g l e F i g u r e 1.  S e s s i l e Drop Parameters  - 6-  Ellefson and Taylor"^ state that this -assumption may be in error because 15 gravitational forces destroy the spherical relationship although Mack's results show that the maximum error introduced by gravity effects is 1°10' Advancing and receding peripheries of drops of a specific liquid , o!7 have formed equilibrium angles which differ by as much as 4 0 . Although this has been described as a wetting hysteresis effect (implying some 18 metastable configuration), Bartell and Wooley  maintain that the angle  formed in either case is a true equilibrium angle.  For the advancing case,  the final angle depends on the pretreatment of the unwetted surface.  For  the receding case, the final angle is independent of pretreatment because the liquid is receding over a previously wetted surface.  Advancing peri-  pheries are more desirable for sessile-drop experiments because they reflect the pretreated surface condition of the solid. D. The Effect of Solute Additions to the Liquid Metal Drop • Attempts have recently been made to improve the wettability of an oxide by a particular metal through solute additions to that metal. Solute atoms 1.  in general may:  form oxides through reaction with oxygen at either the solid-liquid or liquid-vapor interface,  2.  adsorb at the metal-oxide interface without entering into chemical reaction,  3.  be surface-active in the liquid, affecting V'LV-'  h.  have no effect.  Solutes of type 2. usually decrease the interface:. . energy and improve the wettability without forming new compounds having undesirable thermal, mechanical, and physical properties.  - 7 Factors  i n f l u e n c i n g the p r e f e r e n t i a l i n t e r f a c i a l  adsorption 5  of c e r t a i n s o l u t e s have been s t u d i e d by  6  1Q  several i n v e s t i g a t o r '  y  20  '  21  '  .  They have c a l c u l a t e d the e x c e s s i n t e r f a c i a l s o l u t e c o n c e n t r a t i o n u s i n g Gibb's Adsorption  Theory, w h i c h , f o r s e s s i l e drop a p p l i c a t i o n , t a k e s t h e f o r m of  the  equation  P = lkR  T  d  0  SL dlna 2  where P a ^SL T R  =  = excess i n t e r f a c i a l c o n c e n t r a t i o n = b u l k a c t i v i t y o f the s o l u t e i n t e r f a c e energy = absolute temperature = gas c o n s t a n t  (mol/cm )  20  Gamma has been i n t e r p r e t e d more c l e a r l y by Guggenheim  as t h e number of  moles o f component (per u n i t a r e a of f r e e s u r f a c e ) i n e x c e s s of what would be p r e s e n t i f t h e b u l k c o m p o s i t i o n remained u n i f o r m r i g h t up t o the face.  I f t h e s o l u t e obeys Henry's Law,  the mole f r a c t i o n can be  inter-  substituted  f o r a c t i v i t y i n the e q u a t i o n . . ,-The degree of i n t e r f a c i a l s o l u t e a t t r a c t i o n i s c o n t r o l l e d not o n l y by the p r o p e r t i e s of the s o l u t e atoms, b u t a l s o by the n a t u r e o f  the  22  oxide  surface.  s u r f a c e due The  Murray  has  shown t h a t c a t i o n s may  r e t r a c t below the  t o t h e d i r e c t i o n a l i n t e r a t o m i c bond d i s t r i b u t i o n a t t h e  oxygen i o n d o m i n a t i o n t h e n ' c r e a t e s  an e l e c t r o n e g a t i v e  surface.  l i q u i d d r o p i s p l a c e d on t h e s u r f a c e , s o l u t e atoms are a t t r a c t e d t o interface i n proportion to t h e i r e l e c t r o p o s i t i v i t y .  To a f i r s t  P r e f e r e n t i a l adsorption  of one  surface. When a the  approximation,  t h e adsorbed s o l u t e atoms f o r m an i o n i c metal-oxygen bond^ w i t h the oxygen i o n s .  oxide  s o l u t e over a n o t h e r may  surface there-  f o r e be a p p r o x i m a t e l y p r e d i c t e d t h r o u g h a c o n s i d e r a t i o n o f the f r e e energy of formation  (AF^.  ) of t h e s o l u t e o x i d e s a t t h a t t e m p e r a t u r e .  Hence s t a b l e  o x i d e - f o r m e r s such as T i , Z r , and Th would be e x p e c t e d t o adsorb p r e f e r e n t i a l l y t o C r , Mn,  and Fe which are l e s s s t a b l e  oxide-formers.  - o-  E. Factors Influencing Selection of the Fe-MgO System and Alloying Components Experimental evidence in support of the preferential interfacial adsorption theory is meager.  Results that would establish a definite cor-  relation between A F ^ ..of solute oxides and their preferential interfacial adsorption could be obtained through a study of alloying effects on some metal-oxide system.  The choice of system is limited primarily by the amount  of thermodynamic data and surface energy information available. In this work/ iron was chosen for the metal component because the 23 activities of several alloying components in liquid iron have been measured  .  Furthermore, liquid surface tension values for pure iron are known and the effects of various solute elements on this surface tension have been ascertained.  The moderate attraction between iron and oxygen makes possible the  selection from a wide range of solute elements that would adsorb on an oxygendominated surface preferentially to iron.  In liquid steels, these elements  would be referred to as deoxidizing agents. Magnesium oxide was chosen as the ceramic because of its availab i l i t y in the form of large cleavable single crystals.  There is reason  to believe that freshly-cleaved surfaces are the most contaminant-free oxide IT  surfaces obtainable  .  The solid surface energy ( X s v ) of the MgO (100)  cleavage face is more accurately known than any other oxide Xgy, although determinations of this quantity, in general, are s t i l l in a rather primitive pk  state  .  Since mechanical preparation of cleaved MgO surfaces is unnecessary,  surface energy variations are minimized.  The stability of the oxide precludes  direct chemical reduction except by the strongest oxide-former solutes. The possibility of developing Fe-MgO combinations for specific industrial applications was considered in choosing these two components. The 25  use of Ni-MgO and Co-MgO cermets : has already been reported . In view of the similarity of Fe to Co and Ni, and the numerous uses of irons and steels, development of dispersion-hardened iron alloys or cermets.: may be industrially feasible.  Solid surfaces tend to lower their surface energy by adsorption of a gas  22  .  Since the decrease in solid surface energy of MgO associated  with adsorption of various gases is not known, i t was decided that the experiments would be conducted in a moderate vacuum.  The vapor pressure of  liquid iron near its melting point prohibits the use of a high vacuum. The solute elements were chosen i n i t i a l l y from a comparison of c standard free energies of formation ( . A F - ) of the oxides at 1500°C. as 0  26  compiled by Tripp and King  .  This data is shown in Table I.  Standard free  energies do not allow accurate prediction of preferential interfacial adsorption of one solute over another.  Free energies of solution in liquid  iron, reduction of oxygen potential (associated with vacuum), interaction effects, etc. must be considered. however.  Elements chosen were T i , S i , V, Ta, Nb, Cr, Mn, and C. TABLE I. of the Oxides  -AF°  1550  Group 1, 2, 3 j BeO SrO A1 0  212  Eeduce MgO  C a 0  -  •  Table I aids an approximate selection,  2  MgO  210-  3  19k 175  Group k Th0 Ce 0o Zr0  219 211 180  TiO Ce0 Ti 0o Ti 0 Ti0 Si0 CO  166 165 "163 15U  2  2  2  Group 5  uo  2  • CH D -P W -H • H H -H a >> 0 CH rH C H CO Q •3 O PM  2  2  5  2  2  ll+5 130 129  V  00 "t «  2°3  VO Ta 0e V0 ' NbO Nb0 Nb 0c 2  2  2  2  P 0H2  V  CO  Cu 0 CuO 2  182  2  173  w  -P  Group 6, 7, 8  23 5  C0 Sn0 PbO 2  2  s  * = extrapolated values  62 58 35  2°5  P 0 As 0o Sb 03 BiO SbOp 2  5  2  2  123 117 117  llU Ilk*  MnO Cr 0o Mn 0^ Mn 0 2  3  2  3  119 101 90 73*  108* 102* 93* 76*  63* 58* 31* 25* 23*  FeO Mo0 WOo - WO2 Fe 0 CoO NiO  73  2  2  3  63 6 l 59 5^* 52 Ul  - 10 F. Previous Work , The Fe-MgO system has been studied to some extent using the sessile Kingery and Humenik^ measured the surface tension (fay)' of  drop technique.  Armco iron on MgO in helium at 1550°C. to be 1 2 ^ 0 ergs/cm^.  They observed  an interfacial discoloration of the MgO plaque implying that some chemical reaction had occurred.  Rapid volatilization of the drop in vacuum discouraged 6  further study in this direction.  Humenik and Kingery  studied the Fe-MgO  system in vacuo but their observations were again obscured by volatilization of the iron.  Furthermore, the contact angle decreased with time.  Since  Dupre's equation applied i n i t i a l l y , they extrapolated the contact angle to zero time at melt ( 1 2 3 ° ) and calculated the interfacial energy using this angle.  \^^: ---:'/'J ' '::--'i: .'-' l  :  The decreasing contact angle was attributed to  some surface reaction which was not investigated in detail. o contact angle of 1 3 0  : :r  A zero-time  was measured in helium indicating that adsorbed gas  had affected the surface energy balance.  In hydrogen .a non-metallic  surface film formed on the liquid drop which completely restricted its flow.  They surmised that the film was produced by a surface reaction or  atmospheric conditions within the furnace.  Sintered MgO plaques rather than  single crystals were used throughout their experiments.  Observations on the  Fe-BeO system by the same authors revealed that surface porosity seems to have little,.effect on the equilibrium contact angle. A variation of 2 ° with zero to ten percent porosity was noted. The surface of a sintered MgO plaque would probably display other 1  crystal faces than the ( 1 0 0 ) face as well as numerous edges and corners.  The  surface energy of the ( 1 0 0 ) MgO face has been established as 1 0 9 0 (± 20$) ergs/cm^ at 0°K (Oil)  22  .  Lennard-Jones and Taylor  have shown that MgO has an  surface energy of 3 9 ^ 0 ergs/cm^ and a crystal edge energy of 2 7 X 1 0 ^  ergs/cm.  The presence of either could thus cause a wide local variation in  solid.surface energy on a sintered plaque.  - 11 -  P r e f e r e n t i a l i n t e r f a c i a l a d s o r p t i o n has been e v i d e n t of some a u t h o r s .  Kingery  i r o n was s u f f i c i e n t  results  found i n the F e - A l ^ O ^ system t h a t 0.06$ S i i n the  t o form a s i l i c o n monolayer a t the s o l i d - l i q u i d  Humenlk and Kingery^ iron greatly  i n the  observed i n the same system t h a t  interface.  O.Ok'fo t i t a n i u m i n  d e c r e a s e d the interface:...", energy w i t h o u t a f f e c t i n g the  the  liquid  2 surface t e n s i o n . in iron.  A l l e n and K i n g e r y  d i s c o v e r e d the same e f f e c t w i t h carbon  A l t h o u g h the carbon d i d not a f f e c t the l i q u i d s u r f a c e t e n s i o n ,  carbon a d d i t i o n s of a p p r o x i m a t e l y 4$ caused bubble e v o l u t i o n fromothe drop t h a t c o m p l i c a t e d c o n t a c t a n g l e measurements.  They assumed t h e gas t o be  carbon monoxide. 7 .  Baes ; sums up the e f f e c t of v a r i o u s a l l o y i n g elements on the :  s u r f a c e t e n s i o n of l i q u i d i r o n w i t h the statement t h a t  surface-active  i m p u r i t i e s are u s u a l l y elements p o s s e s s i n g l i m i t e d s o l u b i l i t y  the  l i q u i d metal ( i . e .  the n o n - m e t a l s of group V, VI  Kingery  s u r f a c e - a c t i v e i m p u r i t i e s o f t e n have a l a r g e r atomic  19  add t h a t  -  and V I I ) .  in  K u r k j i a n and '13  s i z e and lower l i q u i d s u r f a c e t e n s i o n t h a n the s o l v e n t . found t h a t carbon and n i t r o g e n are not s u r f a c e - a c t i v e and s u l p h u r are' h i g h l y c o n c e n t r a t i o n s below  Halden and K i n g e r y  i n i r o n whereas oxygen  s u r f a c e - a c t i v e and may form s u r f a c e monolayers a t  0.1$.  k Economos and K i n g e r y  observed i n t e r f a c i a l r e a c t i o n s between  v a r i o u s m e t a l - c e r a m i c combinations i n c l u d i n g Si-MgO, T i - M g O ,  Zr-MgO and Nb-MgO.  In each of t h e s e systems they r e p o r t e d a new i n t e r f a c i a l phase which was not soluble i n either  oxide or m e t a l .  The p o s s i b i l i t y  of Ta r e d u c i n g MgO t o some  28 e x t e n t was mentioned by Shepherd getter".  because of i t s  e f f i c i e n c y as an  oxygen-  - 12 The diffusion of iron into magnesium oxide has been investigated  29 by Turnbull  using tracer techniques and more recently by Vasilos and  30 Wuensch  using X-ray and electron probe methods.  The diffusion coefficient r  °  of metallic iron in MgO determined by Turnbull between 1060  D = 9 , 5 (10~ ) 9  e  - 25,900  1  0  and 13^0  was  RT  He then substituted this value into the solution of Fick-' s equation, viz: C = G (TT^Dt)* Q  0  and studied concentration gradients.  , e  I  T5t  Vasilos and Wuensch calculated the  activation energy for iron diffusion in MgO to be 1.7 ev (compared to Turnbull's value of 1.12 ev); 31 Carter  measured the rate of diffusion of FegO^ into MgO  and claims the mechanism,to be counter-diffusion of Mg  ++  and F e  + + +  ions  through a relatively rigid oxygen lattice. 32 The FeO^MgO diffusion studies, of Rigby and Cutler  show that  vacancies diffuse into the MgO in approximate proportion to the number of Fe  + + +  already present in the MgO.  Furthermore, the area through which iron  has diffused becomes visibly discolored. G. Specific Purpose of the Present Investigation The purpose of the present investigation was to make a comprehensive wettability study of the Fe-MgO system using the sessile drop technique in vacuo at 1550°C.  The effects on wettability of various alloying elements in  the liquid iron were to be determined and compared in the hope that the pre-.; ferential interfacial adsorption trend noted by Kingery et a l . could be verified for this system.  Interfacial reaction products were to be identified  where possible by X-ray analysis and their structural configurations investigated' by metaJUographic techniques.  Mechanisms of the reactions were to be  proposed on the basis of these results.  Vacuum-cast high-purity iron and  single crystals of optical-grade MgO were to be used throughout the experiment.  - 13 II.  EXPERIMENTAL  A. Materials 1.  -  '•.'- ;."••••'•'  Magnesium Oxide O r i g i n a l l y , synthetically-produced ultra-high-purity single  c r y s t a l s of magnesium oxide (MgO), which-are now being prepared by General 33 Electric  , were to be used but these proved to be.too small f o r s e s s i l e  drop supports.  Hence high purity single c r y s t a l s of optical-grade MgO  supplied by Norton Company were used i n a l l experiments.  The  average  purity of these c r y s t a l s was determined by spectrochemical analyses on eight samples as shown i n Table I I . Adamson reagent grade MgO MgO  A purity comparison to Baker and  powder i s included.  has an ionically-bonded- rocksalt structure of two  inter-  penetrating face-centred-cubic" l a t t i c e s . . Magnesium ions occupy one set of l a t t i c e s i t e s while oxygen ions"occupy the other set. of MgO  S i g n i f i c a n t properties  and the various metals used i n t h i s investigation are l i s t e d i n Table The s o l i d surface energy  (Xgy)  o r  M  m.  S ° has been estimated by both  experimental' and.theoretical methods.' Values published by various authors are i n disagreement,  as shown i n Table IV. • The value chosen f o r t h i s work  was 1090 ergs/cm^ at 0°K, which was derived from the most recent and  thorough  22 investigation  .  2 . i I r o n and Additive Metals The high-purity iron wa.s supplied by Crucible Steel Company of America under the trade name of Ferrovac E cold-drawn 5/8; inch rod.  A  spectrographs analysis of t h i s material together with analyses of solute metals used i s shown i n Table V.  Samples of zonesrefined i r o n and  high-carbon  s t e e l shot were also analysed. The l i q u i d surface tension of pure iron has been measured by several investigators.  A summary of this'work i s presented i n Table VI.  - 14 -  TABLE II. Spectrochemical Analysis•of: MgO Samples Internal Standard - Mg 30Jk~ Si Cu •Mn " "- Fe Cr  Al •"  Element Wavelength  L  2576 , .2600 ' 2516 321+7  3082  Ca  4254  3969  Weight Per: -cent Baker and Adamson Reagent .006 Grade MgO  .0005'  .011  .006  • .001  .006  .014  Specimen  1,  .03k  .0017 .009  .020  .0013  .006  .006  2.  .029 .0014 .013  .018  .0033  .003  .007  3v  .022 ;.ooi6:> .012  .017  .0020  .005  .005  k.  .034 .0016' .008  .018  .0007  .(X>'3  .004  5.  .022 ".0041 .011  .011  .0005  .008  .008  6.  .023 : .0041 .005  .007  .0005  .006  .006,-.  .010 . .0029 .006  .007  .0011  .007  .005  .010 .0018 .007  .006  .0006  .008 ... '005  .023 .0024. • 009  .013  .0013  .006  . 7 . 8. Average Analysis  :  Data supplied "by J . H. Kelly (Steel Company of Canada Limited).  .006  - 15 TABLE III. Properties of Materials Used  Density Xy at M.P. Vapor.->Frensure M.P. (gms/cc) (ergs/cm^) (mm), 1^50°c. MgO  ElectroMetallic negativity Radius (&), Mg  2800  0—  ++  1.92 4.08 I.165  Fe  7.20  1835  '0.07  1535  Ti  4.50  i46o  0.01  1727  1.60  1.324  V  5.96  1697  < 0.001  1697  1.90  1.224  Ta  16.60  2860  <  0.001  2996  1.90  Nb  8.40  2030  <  0.001  2500  1.80  1.342  Cr  7.20  -  2.0  1900  2.20  •I.I76  Zr  6.44  -  <0.001  2127  1.60  1.597  Ref. 34  Ref. 35  Ref. 36  Ref. 36  Ref. 37  ' 1.343  Ref. 37  TABLE Published  Surface  Date  Author  IV.  Energies  of  MgO  Surface Energy (ergs /cm )  Method  2  Fricke Jura  and  Garland  Lennard-Jones Taylor  and  Kingery Livey  and  Murray  Murray  NOTE:  A l l values  for  19^3  1459  Calculation  1952  1040  Calorimetric  1953  1362  Calculation  1954  1360  1956  1090  i960  1090  (100)  crystal  ±  ±  face  20$  Estimate  ,  ,  Estimate from Lattice Energies 220 together with Heat of s o l . data.  at  0°K.  TABLE Y. Spectrographic Analyses of Metals Used  C,  Metals  o'-. 2  W  2  -V  /Fb  :,Ta  Mo  Ti  Zr,  MgO Ferra'vac Fe - 0 1 3  .0083  1  P . :. s :..Ni  Si:-  .009  .013  :Cr  • Mg  -006  Al -023  .0003 .005  ND  ND  .006  <T  .01"  bal  0.01  T  -005  T  .020  bal  ND  T  .ko  .20  .01  .01  .01  .01  .04  .01  bal  .Oh  -03  -06  .12  -05  bal  .20  -003 -013  •07  •03  .002 -02  . 0 1 bal  ND  T  -002  T  .01  - 0 0 1 ND  Zone-E Fe  -012  .005  ND  ND  Cr  ND  .006  ND  -• T  V  -05  .09  .05  Fb  .05  -07  -03  bal  Ta  -05  .02  .01  • 05  Ti  -03  .01  .05  Zr  .10  .01  .05  Hi  C Steel . 9 0  Ferrovac Fe . 0 0 8 (after.melting on MgO)  Fe" :.  j  .001  • T . • • 05 ;  :•: - 0 0 2  .-T • . 0 1 0  .002  •-T-V  -010  .002  T  -003  .h  - 0 0 3 ND ND  .09  -01  bal .15 bal . 1 8  bal -003  ND  .005  ND  ND  .007  T  ND  .005  T  .. . 0 2  TABLE VI. Summary of Liquid Iron Surface Tension Values  Ref. No.  Temp. #LV °C.  Iron Type  Method  Al  5  1550  1450  Sess. Drop  N.R.C. vac. cast  6  1550  1400  Sess. Drop  Armco  1450  11  11  11  11  Cu  Chemical Analysis FD Si C  (wt. P  io)  0  S  N  Electrolytic vac. cast .002 .01 <.01  2  1570  1700  13  1570  1720  II  1!  38  1550  1380  II  II  25  1550  1940  Estimate  3  1550  1240  Sess. Drop  39  1550  840  4o  1550  1835* 15 Sess. Drop Purified Carbonyl  This Work  1550  1426  N.RiC.  .003K.001  Ferrovac E  T  .04  .005 .002 .010  .01  .0031  .005 .00089  .0072 .005 .00051  Electrolytic Pure Armco  Pendant Drop  Sess. Drop  FerrovacE 1* 2*  <.0008  .001  .0003  .003  .01  .001 .01 .007 .002 .OO83 .005 .0003  a?  .oo4  T  <.001 <.001  1* Invoice Analysis 2* Spectrographic analysis performed by Coast Eldridge Chemists  0.01  -013 T  T <.0005  .003  - 19 B.  Apparatus 21,  The sessile drop furnace described: .'in detail by previous workers 35..> 4 l  w  a  g  1.  mo  d i f i e d slightly for this work.  Modifications were such that:  a vacuum of 10"^ mm of mercury could be attained in ten minutes,  2.  drop measurements could be made after melting at tensecond time intervals through the use of ultra.-, highspeed self-developing film.  Figure 2. shows photographs of the equipment and Figure 3. represents a schematic diagram of the apparatus. 1. Furnace The molybdenum susceptor design used by Clarke et a l .  21  was  altered to reduce radiation heating of the vycor tube and prolong the l i f e of the susceptor i t s e l f .  Dimensions were altered to:  Length (ia«)  Diameter (in.)  SUSCEPTOR  1.5 3-5  0- 75 0.75  0.020 0.005  RADIATION SHIELD  k.O  1- 75 1.50  0.005 0.005  Thickness (in.) Rose Clarke et a l .  Rose Clarke et a l . The assembly is shown in Figure k. The insertion of two sintered alumina 5.0  spacers near the ends of the radiation shield, prevented sagging and subsequent closure of the loop. Furnace power was supplied by a Lepel (model T-10-3) highfrequency induction generator,and temperatures were measured through a pyrex viewing window to ± 5°C with a Hartmann and Braun disappearingfilament optical pyrometer.  Aside from the optical-flat correction of  or  23°C. DeCleene were negligible.  showed that emissivity corrections with this apparatus  - 20 -  Figure 2. Photographs of Apparatus  OPTICAL  1. 2. 3. h.  •5. 6. 78. 910. 11. 12.  SYSTEM  FURNACE ASSEMBLY  Ground glass. 13- •Induction c o i l . lh. Vertical adjustment screw. Heating element, radiation shield and specimen Horizontal adjustment track. 15. Thermocouple gauge. 16. Ionization gauge. Focussing screw. . Adjustable bellows. 17- Gas inlet control. 18. Viewing window. Ocular lens. Objective lens,shutter and i r i s diaphram.19- Brass fittings. 20. Optical pyrometer. Vertical adjustment screw. 21. Light source interchangeable with pyrometer. Water-cooled optical flat. 22. Polaroid camera. Water-cooled brass f i t t i n g . Magnetic shutter. 23. Pivoting mirror. Vycor tube. Figure 3. Schematic Diagram of Apparatus ro h-  1  - 22 -  Figure h. Susceptor Assembly and Sample Specimen  The high vapor pressure of iron at its melting point (0.1 mm Hg) created a problem in the furnace tube.  Iron vapor condensed rapidly on  the tube interior near the cooling jackets so that the tube i t s e l f became a susceptor shortly after melting.  Experimental tests were thus limited  to two minutes in order to avoid implosion of the vycor tube. During tests, heat dissipation from the tube was accelerated by blowing a continuous stream of cold air across the tube. 2. Vacuum System The two-stage o i l diffusion pump used by Clarke et a l . was -5  inadequate for this work because i t required one hour to reach 10 and frequently became contaminated with metal vapors and oxides. therefore replaced by a Speed!vac mercury-vapor pump (model 2M2A'0-  mm Hg It was This  pump, operating in conjunction with a mechanical fore pump, produced a vacuum of 10 ^ mm Hg in ten minutes. 3. Optical System Molten iron drops magnified eight times could be viewed with ease on the ground glass plate shown in Figure 3' (# l ) «  Tne light intensity  at this magnification, however, was insufficient for photographic exposures. Clarke and co-workers found that a moderately fast.film (Ferrania Sh - ASA daylight 250) required a twenty-second time exposure at : f l l . to yield a sharp drop image.  Since their tests often ran for 45 minutes, time exposures  were satisfactory. It was discovered in preliminary investigations on the Fe-MgO system that the contact angle changed rapidly during the f i r s t two minutes. Time exposures were thus inadequate..  The optical system was then modified  to include a Polaroid Land Camera employing ultra^high-speed self-developing film (Polaroid 3000 - ASA daylight 3200) as shown in Figure 3.  - 24 At room temperature a pivoting mirror (# 23) allowed focusing of the drop on a ground glass plate (# l ) with the aid of a light source (#'21).  At  1500°C. the mirror could be quickly t i l t e d up, permitting light to reach the Polaroid camera.  Since exposures of 1 of a second at f64 were sufficient 100  to yield a sharp image, problems of vibration encountered by Clarke et a l . were eliminated.  Furthermore, the developing time of Polaroid  3 0 0 0  is ten  seconds so that photographs at fifteen-second time intervals during asingle experiment were possible. C. Specimen Preparation 1. Magnesium Oxide Materials to be used in surface tension measurements require scrupulous handling in order to avoid surface contamination.  Both magnesium  oxide and metallic specimens were prepared with care in this respect. Large single crystals of magnesium oxide were cleaved (with chisel and hammer) along -£l.00^ planes into small rectangular plates approximately 20 X 20 X 1 mm. The mechanical stresses introduced in the specimens during cleavage were relieved through annealing in vacuo at o 1100 C. for three hours.  Thermal stresses were avoided by decreasing the  temperature gradually following annealing. Adamson points out that cleaved surfaces have a wide local variation in surface energy due to surface irregularities such as edges, corners, cleavage steps, microscopic pits, etc.  Experiments were conducted to determine  a suitable etching procedure for removal of these irregularities. 42 immersion in phosphoric acid (at 100°C.) was mentioned by Evans able method for disposal of cleavage steps.  Specimen as a s u i t r  Microscopic observation revealed  that no further removal of surface irregularities occurred beyond five minutes of immersion.  A l l cleaved and annealed plates were thus etched in acid for  five minutes, rinsed quickly in water, and dried instantly with a blast of warm air in order to prevent magnesium hydroxide formation on the surface.  - 25 2. Iron and Additive Metals Preliminary experiments showed that small iron cylinders (0.250" in diameter by 0.250" in length), when melted on magnesium oxide plates, i n i t i a l l y possessed an advancing periphery.  The iron supplied in the form  of 5/8" rod was thus machined down to 0.250" on a lathe.  Sections 0.250"  in length were then cut from the rod with a jeweler's saw. The iron cylinders were rinsed in acetone, immersed in concentrated hydrochloric acid to remove oxide films, rinsed in alcohol, and dried. Specimens were placed immediately in the furnace tube and the tube evacuated to minimize further oxidation.  A l l handling was performed with metal  forceps.. Where metal additions were necessary to the iron specimen, a small hole was drilled in one end of the cylinder and a portion of the additive placed in the well as shown in Figure h.- Similar care was taken in the preparation and handling of the alloying metals. D. Experimental Procedure 1. Sessile Drop Experiments After preparation, the iron and magnesium oxide specimens were each weighed to the nearest 0.1 milligram. Alloying additions were made to the iron by weighing a portion of the additive metal to give the desired weight percent. A micrometer was used to measure the diameter of.each iron cylinder to the nearest 0.001 inches.  This measurement was recorded for purposes of  determining the photograph magnification. A magnesium oxide specimen was placed in the susceptor and upon i t an iron cylinder containing the appropriate additive was set as shown in Figure k.  The furnace tube was then assembled and the system pumped down  - 26 -  to a vacuum of one micron.  A lamp was used to project an image of the iron  specimen on the ground glass plate (# 1  in Figure 3)' Provisions for  focusing the image were made through a sliding bellows (#4).  The cylinder  image diameter on the glass plate was measured with microcalipers to allow calculation of the optical magnification.  Purified hydrogen was introduced  through a needle valve until the pressure had increased to 500 microns. Hydrogen reduction of the specimen was carried out at 500°C.  After cooling  the system was again pumped down to one micron and the mercury diffusion pump placed in operation.  Within ten minutes the pressure decreased to  -5  5 X 10  mm of Hg. The temperature was slowly increased to a point near 1000°C. where  adsorbed hydrogen remaining in the furnace tube ionized, creating a bright blue flash.  It was hoped that the ionized hydrogen would reduce a l l final  traces of oxide on the metal surface.  After a slight pause, the remaining  hydrogen was removed by the diffusion pump and the vacuum returned to 5 X 10"^mm, At this point the cold air stream was directed at the furnace tube and the temperature slowly increased to 1550°C.  Prior to melting, the incandescent  solid cylinder was viewed on the ground glass plate.  Several preliminary  experiments revealed that the time elapsed from i n i t i a l to complete melting varied from three to five seconds.  At the instant i n i t i a l melting was  observed a timer was started, the camera shutter was closed, and the mirror pivoted up to allow photography of the specimen.  Self-developing photographs  of the molten drop were taken at appropriate time intervals up to two minutes. Four seconds was subtracted from the observed time to give the true time after complete melting.  At approximately 120 seconds, the iron vapor  pressure reached 0.4 X 10  mm whereupon gas discharge occurred, the power  was decreased, and the specimen cooled to room temperature.  Dimensions for calculating contact angle and surface tension were measured on the photographs with microcalipers and dividers. 2 . X-ray Diffraction The iron drops were sheared from the'magnesium oxide plate.  The  discolored interfacial material adhering to the metal was scraped off with a sapphire rod and prepared for X-ray identification using the powder diffraction technique.  Since each alloyed specimen required a different i  target material, exposures varied from three to six hours.  Diffraction  lines on each photograph were indexed and attempts made to correlate the patterns to materials in the ASTM X-ray F i l e . 3. Metallographic Observation The iron drop (bonded to the MgO plate) was mounted in a lucite mould and half the specimen ground away as shown in Figure 5 . A crosssection of the interface was polished and examined under the microscope for visible interfacial structures.  F i g u r e 5.  Specimen Mounted f o r M e t a l l o g r a p h i c  Observation  - 29 -  III.  RESULTS AND DISCUSSION  A. Pure Fe-MgO System 1, Results and Calculations a. Molten iron-solid MgO system 1. Sessile drop experiments The sessile drop dimensions (x, z, x', z') obtained from the photographs were recorded together with time after complete melting as in Appendix I.  From this data the contact angle and liquid surface tension  values were calculated using the method outlined in Appendix II.  The  time-dependence of the contact angle was then determined as in Figure 6. Specimens melted in a gas-fired furnace with an air atmospheres possessed a constant contact angle of 2 1 ° after ten minutes. Photographs of a typical specimen (No, 1 2 ) in Figure 7. shows the decreasing contact angle with time. A wide variation in calculated liquid surface tension values was noted.  Since no trend was obvious, the possibility of a statistical variation  was investigated, (Appendix III).  A statistical analysis of the OLV  v  a  l  u  e  s  w  a  s  made  The variation followed a normal distribution with a mean 2  value of 1^26 ergs/cm  2  and a standard deviation of 1 3 9 ergs/cm .  In Dupre's equation the contact angle can decrease only i f V s L and/or  d.  e c r e a s e s  *  Assuming the liquid surface tension to be approx-  imately constant (1U26 ergs/cm ), the changing contact angle can be attributed entirely to a decrease i n ^ ^ .  ^o  a  f i r s t approximation, Dupre's equation  w i l l then relate the contact angle to  (Table VII) from which the time-  dependence may be determined as in Figure 8.  - 31 -  Figure 1.  Sessile Drop Photographs  :•  .  - 32 TABLE.VII. • " Pure Fe-MgO System  Change of Interface Energy With Time Calculated'from .tfgL = 906 - 1426 cos 9 (9's obtained from curve of 9 Vs. time) Time (sec.)  cos  9" (degrees)  9  - 0.4695 - 0.3420  0  117.0  5  109.0  10  101.0  15  93.0  -  20  -1426 cos 9  VsL ergs/cm  2  + 669.5  1575.5  • +487.7  1393.7  0.2079 *  + 296.4  1202.4  0.0698  + 99-5  1005.5  85.0  + 0.0872  - 124.3  781.7  25  80.0  + 0.1736  - 247.5  658.5  30  76.2  + 0.2385  «' 340.1  565.9  35  "73.4  + 0.2857  -.407.4  498.6  40  71.2  + 0.3223  - 459.6  446.4  45  69.5  0.3502  -4499.4  406.6  50  -67.8  + 0.3778  - 538.7  367.3  59'  ...66.8 •  + 0.3939 '  - 561.7  . 344.3  •60  66.0  + 0.4067  - 579-9  326.I  65  65.5  + 0.4147  - 591.3  314.7  70  65.1  + 0.4210  - 600>3  305.7  75  65.0  + 0.4226  - 602.6  303.4  120  65.0  + 0.4226  - 602.6  303.4  -  i6oo  Time (sec) After Complete Melting Figure 8. Interface Energy Vs. Time  33 -  The i n i t i a l c o n t a c t a n g l e ( a t z e r o t i m e a f t e r m e l t i n g ) was rained by two methods f o r comparison w i t h Humenik and K i n g e r y ' s v a l u e  deter' (123°).  A v a l u e o f 1 1 7 ° * 1 ° was o b t a i n e d from F i g u r e 6 , , w h i c h i s b a s e d e n t i r e l y on p h o t o g r a p h i c d a t a . described  T h i s v a l u e was v e r i f i e d by a second method t o be  later. ii.  Thermodynamic c a l c u l a t i o n s  The f r e e energy o f i r o n o x i d e f o r m a t i o n a t 1 5 5 0 ° C ( c o r r e c t e d oxygen p o t e n t i a l ) was d e t e r m i n e d as i n Appendix I V .  The e q u i l i b r i u m oxygen  p r e s s u r e over magnesium o x i d e , molybdenum o x i d e , and f e r r o u s o x i d e a t was c a l c u l a t e d t o be 4 . 2 3 X 10* respectively.  8  a t m . , 1.58 X 10  8  for  1550°C.  "8 a t m , , and 0.42 X 10* atm, o  The v a p o r p r e s s u r e of i r o n over l i q u i d i r o n a t 1550 C.  is  '8 63^0 X 1 0 ' a t m . (by i n t e r p o l a t i o n o f L o f t n e s s ' d a t a - see Appendix I V ) . b , S o l i d i r o n - - s o l i d • M g O system i« Weight measurements The t o t a l w e i g h t l o s s t o the Fe-MgO specimens a f t e r was 3*7 p e r c e n t (average o f n i n e ii,  Metallographic  120  seconds  specimens). observations  The i n t e r f a c e of a l l specimens, when viewed t h r o u g h t h e MgO p l a t e , p o s s e s s e d a b l a c k a n n u l u s under t h e drop o f u n i f o r m t h i c k n e s s a r o u n d t h e c i r c u m f e r e n c e as shown i n Type 1,  F i g u r e 9«  t h i n d e p o s i t o f some new i n t e r f a c i a l p h a s e . m e t a l l i c s u r f a c e was p l a i n l y v i s i b l e .  The a n n u l u s appeared t o be a I n s i d e t h e annulus a s h i n y  The i n n e r boundary o f t h e annulus  appeared t o c o r r e s p o n d w i t h t h e p o s i t i o n o f t h e m o l t e n drop p e r i p h e r y a t zero time.  A t t e m p t s were made t o v e r i f y t h i s r e l a t i o n t h r o u g h a comparison  o f m i c r o s c o p i c measurements o f the annulus d i a m e t e r w i t h b a s a l drop d i a m e t e r s measured i n t h e m o l t e n s t a t e ( T a b l e V I I I ) .  Three specimens w i t h t r u l y  c i r c u l a r p e r i p h e r i e s were examined (Specimen N o s , 13,  15 and 2 1 ) .  Type 3 Figure 9.  I n t e r f a c e Types as seen t h r o u g h Mgo  Plate  - 36  TABLE VIII. Interface Annulus Measurements Specimen 13 Magn. = 9 . 9 8  Time (sec.)  Measured Basal Diameter ( 2 x ' for 9 > 9 0 ° ) (2x for 9 < 90°) (inches)  True Basal Angle Diameter (degrees) = M.B.D. inches Magn.  13  2(1.434)  0.287  99  29  2(1.586)  0.318  77  50  2(1.691)  0.339  65  3  2(1.305)  0.295  112  12  2(1.493)  0.338  99  26  2(1.775)  0.402  76  55  2(1.896)  0.430  66  I.D. of Black Annulus (measured with microscope, inches) 0.258  Specimen 15 Magn. = 8.84  Specimen 21  0.289  Magn. = 8 . 7 5  7  2(1.265)  .0.289  106  20  2(1.415)  0.324  87  46  2(1.532)  0.351  70  78  2(1.564)  0.358  65  0.273  Contact angle was plotted versus basal diameter (Figure 10). relation was obtained.  A linear  The angle where the inside diameter of the annulus  f e l l upon the line was determined to be 117°, .116°, and 117° respectively for the three specimens mentioned above. The air-oxidized specimen possessed a completely discoloured interface similar to that of Type 3, Figure 9. When the specimens were mounted, sectioned, and observed microscopically, photographs of the interface were made.  In the center  of the drop, no visible evidence of either diffusion or new interfacial configurations was observed (Figure 11.).  Near the circumference, however,  the interfacial annulus appeared in cross-section as a new interfacial configuration (Figure 11.) which" increased in thickness from zero at the inner boundary to approximately ten microns at the drop periphery. No interfacial fractures were observed in either the vacuummelted or air-oxidized specimens.' Three of the iron drops bonded to the MgO plates were submitted to a semi-quantitative spectrographic analysis.  The average impurity  concentration is given in Table V. for comparison with the pre-melted material. No appreciable impurity reduction is evident. iii.  X-ray results  X-ray photographs of the interfacial material revealed the presence of magnetite (Appendix V).  Examination of the Fe-0 phase diagram  (Appendix VI) revealed that wustite converts to Fe^O^ with separation of metallic iron.  In support of this' observation, the photomicrograph of the  interface (Figure 11,) shows arms of what appears to be pure iron penetrating the interfacial compound.  Figure 1 0 . Contact Angle Vs. Basal Diameter  -  S k e t c h of Specimen C r o s s - S e c t i o n showing M a g n e t i t e D i s t r i b u t i o n around Drop P e r i p h e r y  Annular I n t e r f a c i a l Region  Annular I n t e r f a c i a l Region  Central I n t e r f a c i a l Region  (X2000)  F i g u r e 11.  C e n t r a l I n t e r f a c i a l Region  Photomicrographs o f I n t e r f a c e  (X2000]  39  -  - ko X-ray patterns of the air-oxidized specimen revealed that the entire drop had oxidized and solidified to Fe^O^. 2. Discussion The observed decreasing contact angle with time supports the observations of Humenik and Kingery^; ship deserves some discussion.  Evaluation of the time-angle relation-  The validity of applying Bashforth and Adams'  procedure is rather questionable since their tables are based upon equilibrium drop shapes. of error.  Angle calculations were confounded by various sources  The spreading drop prevented the acquisition of sharp photographic  images necessary for accurate dimension determination.  Iron volatilization  contributed to a lesser extent to the image obscurity.  If the MgO plate  was not placed precisely horizontal during furnace assembly, the unbalanced gravitational.force produced asymmetrical spreading subsequent to melting. Occasionally this force caused drop migration across the plate.  Thus the  combined effects of spreading, volatilization, migration, and unbalanced force distribution often created asymmetrical drops of indeterminate shape. For this reason, photographs were often discarded.  Those that appeared  symmetrical were retained for measurement but a considerable scatter of experimental points, s t i l l exists (Figure 6 . ) . Since the drop shape depends directly.on the liquid surface tension, asymmetry may account partially for the scatter i n ^ y values (Figure 25.). However other causes are probably more important in this respect.  The  decreasing angle indicates a chemical reaction at the solid-liquid and/or liquid-vapor interfaced ^3. Thermodynamic calculations show that oxygen produced through the dissociation of MgO w i l l oxidize the iron at 1550°C. (Appendix IV).  The free energy change associated with the reaction is  -8,620 cal/g-atom of oxygen.  Ferrous oxide forming on the surface of the  - kl drop w i l l be in the liquid state at this temperature.  Kozakevitch^ has  determined the liquid surface tension of FeO at 1M-20°C. to be 52'5 ergs/cm . A surface layer.of liquid FeO would therefore be expected to lower the 19  surface tension of the iron  .  The liquid surface tension determined by  statistical analysis.in this work (lk26 ergs/cm^) is much lower than other reported values (Table VI), the highest being 19^0 ergs/cm^  Since the  purity of the iron is also different (Table VI), surface-active impurities may be contributing to the decrease as well as iron oxide formation. Oxygen available for iro n' oxidation may be supplied from two sources:  the atmosphere surrounding the drop surface, and the solid-liquid  interface.  The atmosphere is continually replenished by dissociating MgO. k5  Since the associated ;magnesium vapor is insoluble in liquid iron accumulates in the furnace tube.  it  Apparently FeO forms on the drop surface.  Evidence.of FeO formation at the solid-liquid interface exists only around the periphery.  The central interfacial surface reveals no visible dis-  coloration or iron oxide deposit as does the periphery (Figure 9«)«  This  phenomenon cannot be explained satisfactorily on the basis of available experimental results but two hypotheses are presented: 1.  Oxygen liberated through the dissociation of MgO at the solid-  liquid interface does not form a ferrous oxide layer.  The formation may  be prevented by some unfavorable interfacial energy change associated with the establishment of an F ( i ) " Fe0(-]_)--Mg0£ \, structure. e  s  oxidation proceeds the FeO layer probably thickens.  As surface  The thickness is  limited by the force of gravity which attracts the slag to the basal periphery.  A thick layer of FeO would be expected to cause the ^ v °^ "^e drop  to approach the value of pure FeO (525 ergs/cm^).. The observed was reduced to approximately lk25 ergs/cm^.  however,  At the basal periphery, the  accumulating FeO affects the balance of the three interfacial energies  - 42 -  related by Dupre's equation  (^3  L?  &SV7 ^'LV^*  this effect, the contact angle must change.  I  n  o r ( l e r  '  t o  compensate for  If the advancing periphery  is not able to push the accumulated rim of liquid FeO before i t , i t may override the oxide.  The inundated FeO would continue to dissolve MgO  since they form a solid solution in both the liquid and solid state (Figure 27.). MgO raises the melting point of FeO so that partial solidification would be possible although no FeO solid solutions were detected during X-ray analysis.  The remaining liquid FeO would convert to Fe^Oi^  with separation of iron upon solidification. 2. . Subsequent to melting, electropositive impurities in the iron (such as Al,. T i , Si etc.) are attracted to the solid-liquid interface and form a stable monolayer with the oxygen-dominated surface of the MgO. The electronegative  surface is thus neutralized, which prevents further  attraction of electropositive impurities. enough to prevent dissociation of the MgO. the iron has. been proceeding since melting.  This configuration may be stable Meanwhile surface oxidation of The electropositive impurities  within the bulk continue to be oxidized until their respective deoxidation equilibria are satisfied.  The reaction products form a slag with the  surface FeO layer but their concentrations are insufficiently large to be detected by X^ray analysis.  The unlimited supply of oxygen at the slag  surface continues to oxidize the iron and the liquid FeO, attracted by gravity, builds up around the basal periphery.  Spreading then proceeds  due to the energy unbalance as described previously.  The periphery  inundates the liquid oxide and advances until a ternary equilibrium is reached at the solid-liquid-vapor interface between the liquid iron, liquid FeO, and solid MgO. 65°.  From Figure 6. this equilibrium angle appears to be  If the tests had been continued beyond two minutes, however, perhaps  this angle would have decreased s t i l l further since the air-oxidized specimen possessed an angle of 21°,  It is more likely that this angle corresponds to  the equilibrium angle between pure liquid FeO and solid MgO. The experimental results seem to support the second explanation. In view of the apparent presence of interfacial FeO around the periphery, the interfacial energy configuration of this structure seems stable. Formation of an FeO layer across the entire interface appears favorable,; therefore, unless some structure is preventing oxidation. The possibility of an interfacial monolayer forming from minor electropositive impurities in the iron is supported by the calculations in Appendix VII.  For high-purity iron containing 0,01 wt. percent silicon,  the total number of atoms present in a typical drop is 21,400 X l O ^ atoms. 1  Only 2 , 7 3 X 10^ atoms are required to form a monolayer,  Hence in order to  ensure the elimination of impurity effects upon monolayer formation, i t would be necessary to purify the iron below approximately 0.01 ppm of silicon. The interface energy (^gjO can be calculated from Dupre's equation only i f equilibrium exists between the three interfacial energies acting on the drop periphery (Figure 1.). The calculation of O'SL ^ Table VII (assuming n  constant O'LV = 1426 ergs/em^) therefore is very approximate in the region of changing contact angle.  Not only is the system far. from equilibrium, but XLV  is probably decreasing because of the influence of the surface FeO layer. After seventy seconds the contact angle remains steady at 65° so that the final value obtained ( 3 0 3 ergs/cm^) is probably reasonably accurate.  The  unknown geometrical arrangement of phases at the interface periphery prevents, a clear interpretation of this value.  It may be the interfacial energy  between Fe^N-FeO^N,  Fe^ \-Mg0^ \ , or any combination of  these three.  Fe(-]_\-Fe0( ), s  g  - kk -  The work of adhesion is given by the equation = TTe  +  + cos 9)  tf (l LV  where TT^ is the equilibrium surface pressure of a monolayer film of the liquid material on the solid surface^.  Since f|"^ is only zero when the /  liquid has a vanishing vapor pressure, the usual simplification to the equation,  cannot be made for the Fe-MgO system.  The value of,-^r' for this system e  has not been published although this quantity could be determined from a study of adsorption isotherms, which is beyond the scope of this work. Hence the work of adhesion was not calculated from the experimental results. The system under study is I n i t i a l l y governed by kinetic mechanisms. A study of.the reaction kinetics is d i f f i c u l t , however, because of the complex geometry of the system, which introduces transport problems and concentration gradients.  Eventually local equilibria are probably established in various  parts of the system but the number of components present (gases, vapors, impurities, etc.) preclude any calculations.  Hence either a kinetic or thermo-  dynamic study of the system is not feasible unless simplifications in the system geometry are made. The most important factor affecting the system appears to be the oxygen potential.  An unlimited supply of oxygen for iron oxidation is made  available through dissociation of MgO.  Magnesium oxide may sublime directly  to the molecular form and then dissociate, or dissociate at the surface and kl  vaporize as elemental magnesium and oxygen.  Kingery and Wygant  molecular vapor pressure of MgO to be., given by the equation l 0  S  MgO  P  =  - 7.320 .+ 10.25 2  claim the  - 4 5  Q  Solving the equation at 1 5 5 0 ° C , "the MgO pressureeequals (Appendix IV).  1770  X  10"  atm.  The oxygen pressure due to the dissociation of MgO was  calculated to "be 4 . 2 3 X 1 0  8 atm.  On the basis of these calculations the  48 molecular species appears to be predominant.  However Altman and Searcy  -8 state that molecular MgO vapor has a partial pressure less.than 3 X 1 0  atm.  at 2 0 3 0 ° K .  have  In support of this statement, two other investigators^,  50  shown the gaseous concentration present over solid MgO to be in the order: Mg, £>2> MgO. Hence the molecular species appears to be of minor importance. Speculation as to the relative rates of dissociation of MgO and transport of oxygen to the liquid iron surface is unfounded. The possibility of drop volatilization contributing to the decreasing angle is not excluded.  It is unlikely that a volume decrease of three percent  in two minutes would greatly affect the overall change (from 1 1 7 ° to 6 5 ° ) . The diffusion of iron into" the MgO was assessed through the application of Fick's Law (Appendix VIII).  After two minutes the" maximum  iron concentration ratio achieved, C , is 5 X 1 0 ' c  -5  42  .  Evans  found that  0  the minimum impurity concentration of iron in MgO to cause a slight yellow discoloration was approximately  0.020  wt. percent.  This corresponds to an  approximate relative iron concentration of 25 X 1 0 ~ 5 .  Hence no discoloration  from diffusion as observed by Rigby and Cutler^ would be expected.  In  support of this conclusion, neither the front view (Figure 9 ' ) o r trieccrossn  section (Figure 1 1 . ) of the interface showed visible discoloration. The black magnetite annulus In cross-section (Figure 11.) appears to be a somewhat lamellar conglomeration of iron and Fe^O^.  The layer  displaces the MgO rather than deposits on the top of the solid.  This might  be expected since the liquid FeO dissolves MgO as the periphery advances. The increasing thickness is probably due to the augmented supply of solvent (FeO) available as oxidation proceeds.  - k6 -  The structure of this interfacial layer can be explained through a consideration of phase transformations during cooling.  As the temperature  is decreased, the iron drop solidifies before the FeO so that a liquid phase exists between two solid phases (Fe and MgO).  Since iron separates from  the FeO upon solidification, solid nucleation sites would influence this separation.  The iron lamellae in Figure 1 1 .  appear' to be connected to  the iron drop so that the solid iron probably provided the nucleation sites. The nuclei then grew into the Fe^Ol). upon further cooling. Alternatively, the MgO may act as a nucleation site for Fe^O^. Since Fe^Ol). and MgO are both face-centred-cubic, the lattice parameters can be compared for purposes of matching the oxygen atom distribution on the 51  (lOO) plane of the unit cells.  Wells  "observed that an approximate linear  relationship occurs between the lattice parameter and percent iron in iron oxides (Figure 1 2 . ) for the transition ^FegO^  Fe^O^  FeO.  lattice parameter for Fe^O^ (published in ASTM X-ray File) is  If Rooksby's  substituted  for Well's value on the line as shown, the relationship is even more closely linear.  Since the Fe^O^ lattice parameter is almost exactly twice the MgO  lattice parameter, i t is conceivable that Fe^O^ nucleation would be promoted on the MgO surface. 11.)  In any case, the resulting interlocking structure (Figure  would be expected to produce a strong bond, aside from]  possible  thermal stresses. The interfacial slip lines shown in Figure 9*.(pure iron specimen) kl  reveal plastic deformation of. the MgO.  Hasselman  7  who observed similar  lines in Ni-MgO specimens, attributed the cause to thermal stresses resulting from non-uniform cooling rates across the MgO plate.  Since fractures were  not visible in any of the specimens, the ultimate strength of the MgO was not exceeded.  F i g u r e 1 2 . L a t t i c e Parameter Vs.  Atomic P e r c e n t  Iron  - 48 -  B. Iron Alloy — MgO Systems 1. Results and Calculations a. Molten drop — solid MgO system Photographs of the sessile drops were obtained at various time intervals and the dimensions recorded as in Appendix I.  In a l l cases the  alloyed drop images were more distinct than the pure iron images (compare Photograph 4 with 1, Figure •7--)«  Furthermore, the basal region of the  drops possessed an optical shadow not observed on pure iron drops.  The  height of this shadow decreased with time and occasionally the shadow disappeared within the two minute test. Experiments were conducted with alloy additions of zirconium (Zr), titanium (Ti), niobium (Nb), tantalum (Ta), vanadium (V), chromium (Cr), and drops of high carbon steel.  Subsequent to melting, gas discharge ensued  whenever additions of silicon or manganese were made.  Humenik and Kingery^  also encountered the high vapor pressure problem of silicon in vacuum at 1550°C.  Hence tests with this alloy were discontinued. Vanadium-alloyed specimens displayed an erratic behaviour after  75 seconds.  The uniformly spreading drops suddenly contracted to angles  near 65°• Photographs of the high carbon steel drops were often discarded because gas evolution followed by drop pulsation, contraction, and migration was common. Zirconium-alloyed drops seemed to explode violently after approximately twenty seconds.  Because' the scattered material dissolved the  susceptor and destroyed radiation shielding, only four tests were conducted. Consequently the results were insufficient for either time or concentration effects to be determined.  - h 9  From the drop dimensions the contact angles and liquid surface tensions were calculated (as in Appendix II).  The time-dependence of  the angles at various alloy concentrations were then plotted as in Figures 13 to 1 8 . Since no trend was observed in the liquid surface tension variation with alloy concentration;/ the values were included in the statistical analysis of Appendix III which led to the derivation of 1U26 ergs/cm  for pure iron.  In sessile drop experiments where no chemical reaction occurs, the contact angle usually remains constant with time but varies with alloy concentration in the drop.  The normal procedure is then to plot  interface energy (determined from contact angle by Dupre's equation) versus log concentration and determine interfacial alloy adsorption ( i f any) by applying Gibb.'s Theory (refer to Introduction).  Since chemical  reaction was observed in a l l cases where alloy additions were made in this work, the interpretation of interface energy remains obscure and inaccurate for reasons discussed in the previous section of this report.-  » 50 -  - 51 -  - 52 -  Time (sec) Figure 15.  After Complete Melting  Contact Angle Vs. Time (Cr-Alloyed Iron)  - 53 -  120  0  20  kO  60  00  WO  Time (sec) After Complete Melting Figure l 6 .  Contact Angle Vs. Time (V-Alloyed Iron)  120  - <jk -  - 56 tu Solid drop - - solid MgO system i.  Interfacial observations (a) Macroscopic  The interface of a l l alloyed specimens, when viewed through the MgO plate, appeared as one of the three types shown in Figure 9- Those similar to the pure iron interface ( i . e . Type l ) were the Nb-, Ta-, and Cr-alloyed specimens.  Completely discolored interfaces (i^e. Type 3)  were displayed by the T i - , V-, and Zr-alloyed samples.  A completely shiny  interface with no peripheral discoloration was observed under the steel drops«. Vertical fracture through the MgO plate around the drop periphery was observed in the high T i - and high Nb-alloyed specimens.  In a l l other  specimens, slip lines were observed at the interface (as shown in Figure 9«) (b) Microscopic In order to complete the microscopic study i t was necessary to examine the center and periphery of the interface at. both high and low alloy concentrations.  In some cases ( T i - , Z r - , Kb-, and Ta-alloyed specimens)  the interfacial deposit increased in thickness very rapidly near the periphery as shown in Specimen 32, Figure 19- In others the interfacial layer was either approximately constant in thickness across the entire interface (V-alloyed specimen 3 5 , Figure 19 )> or non-existent (steel specimen 5 6 , Figure 1 9 « ) '  No interfacial layer was observed in the center  of the Nb-, Ta-, and Cr-alloyed specimens. The interfacial deposit in the low Ti-alloyed specimen (specimen 27, Figure 20) differs from that of the high Ti-alloyed specimen (specimen 22, Figure 20.).  In specimen 27 the interfacial phase contains a precipitate  (presumably iron) which eventually grows into two distinct phases (similar  Specimen 35 Centre and Periphery (X2000)  Specimen 56 Centre and Periphery (X2000)  (in a l l photos, light region is iron, dark region is MgO)  Figure 19.  Interfacial Deposits  Specimen 27 (X2000)  Specimen 22 (X2000)  Specimen 32 (X300)  Specimen 37 (X2000) Figure 20.  Interface Structures  to specimen 22) near the periphery.  The phase in contact with the iron is  rust-brown whereas the phase adjacent to the MgO has an olive-green appearance. Only one type of interfacial "deposit was observed (specimen 37, Figure 20.) i n the Zr-alloyed specimens at a l l concentrations.  The overall  layer thickness increased with concentration. Both the Nb- and Ta-alloyed specimens displayed rather peculiar interfacial structures.  At high Nb contents the periphery appeared as in  specimen 36, Figure 20. The peripheral bead is deposited on top of the MgO rather than at the expense of the MgO as in the T l case (specimen 27, Figure 20.) The ascicular structure of the bead, greatly magnified, reveals a lamellar region as in specimen 36a, Figure 20. The Ta-alloyed specimens did not exhibit an ascicular structure in the peripheral bead (specimen 32, Figure 19«) but similar lamella/twere visible at higher magnifications (specimen; 32:, Figure 20^).  In both the Nb- and Ta-alloyed specimens, the  height of the bead (H, Figure 20.) increased with concentration. The peripheral deposits in the Cr-alloyed samples were similar to the pure iron specimens except that the precipitate was horizontal as in the V-alloyed specimens (specimen 35, Figure 19«) rather than vertical (refer to Figure 11.). i i . X-ray Eesults Attempts were made to identify the interfacial deposits at both low and high alloy concentrations by powder pattern analysis.  In some cases  the quantity of material present was insufficient for identification (Grand V-alloyed samples).  In others, distinct complex patterns were obtained  (Nb-, Ta-, and low Ti-alloyed specimens) but these could not be identified with compounds in the ASTM X-ray F i l e .  The presence of T i 0 o in both ?  - 60 -  the peripheral and. interfacial. region of the high Ti-alloyed samples was established.  This compound was present only in the peripheral deposit of  the low Ti-alloyed specimens.  The interfacial deposit yielded a complex  unidentifiable diffraction pattern.•• Since identification of reaction products was restricted, a thermodynamic assessment of the reactions occurring could not be made. Nor could a kinetic analysis of the systems be performed because of concentration gradients, geometrical factors, chemical interactions,  etc.  2 . Discussion The decreasing contact angle observed with a l l alloy additions indicates a chemical reaction in the system.  The reaction, in each case,  presumably was the oxidation of the alloy component. Since the central interface remained shiny i n the Nb-, Ta-, and Cr-alloyed samples, the. oxidation mechanism is probably similar to that postulated on page 4 2 . The oxidation species that collect over the surface of the drop in this case are greatly enriched in the alloying component. They may temporarily form a sheath which depresses iron oxide formation and/or iron volatilization. obtained.  This would account for the sharp images  As the surface oxide accumulates, gravitational forces may force  i t toward the basal periphery.  The shadowed region observed in the photor  graphs is probably due to the dissimilar emissivities of the surface oxide and iron (or iron oxide).  As surface oxidation proceeds and the alloy  concentration in the drop decreases, the rate of alloy oxidation may also decrease.  Insufficient material is formed to replenish the shadowed region  which is settling toward the base. with time.  Hence the height of the region decreases  The presumption that the oxide layer is being continually drawn  toward the basal periphery is supported by the large rim deposits observed, for example, in specimen 3 2 , in Figure 1 9 .  - 61 Where Zr, T i , and V additions were made, the oxidation mechanism must be slightly different since discoloration of the entire interface was observed after solidification.  Apparently the electropositive components  that stabilized the solid-liquid interface from attack by Nb, Ta, and Cr are not able to prevent attack by Zr, T i , and V.  Hence oxide deposits of  these metals build up across the entire interface as well as on the drop surface.  This new development does not prevent peripheral accumulation of  reaction products, however, because gravitational forces s t i l l operate on the surface layer. Because of the difficulty in making a thermodynamic assessment of these systems i t is not feasible to speculate as to which electropositive components are preventing interfacial attack in some cases and not in others.  In view of the i n i t i a l impurity concentration of the pure iron and  free energy considerations, silicon (which is present as 0.01 wt$J appears to be a major constituent of the monolayer. The erratic behaviour of the steel drops was probably due to the formation and evolution of carbon monoxide bubbles as observed.by Allen and Kingery^. Contraction of the V-alloyed drops to angles near 65° may be a result of the interfacial configuration.  The entire interface is discolored  so that presumably a vanadium oxide phase forms across the interface and the drop spreads toward some iron - vanadium oxide - MgO equilibrium angle. When the vanadium content of the drop becomes depleted, however, iron oxide, which is continually forming, dilutes the vanadium oxide layer.  The: peri-  phery reacts to this influence by returning to the iron-iron oxide-MgO equilibrium angle.  After solidification, the interfacial phase resembles  the magnetite structure observed in pure iron except that the precipitate lies parallel (specimen 35, Figure 19«) rather than perpendicular to the  - 62 -  interface.  The precipitate appears to be metallic iron but positive  identification was not possible.  The interfacial layer is probably some  complex vanadium-oxide-enriched magnetite.  The interfacial distribution  supports the hypothesis that pure FeO would extend over the entire interface in the pure iron-MgO specimens unless some monolayer was stabilizing the region as described. Figures 13.' to 1 8 . which show the contact'angle - time - concentration relation for various alloy additions illustrate three trends: 1.  Contact angles which decrease i n i t i a l l y at the same rate then  stabilize at a level dependent on concentration (Ta, Nb). 2.  Contact angles which decrease at a rate which increases with  alloy concentration and do not stabilize at a particular level (Cr, V). 3.  Contact angles which decrease at constant rate to a level  independent of.concentration after a period of time which depends on alloy concentration (Ti). The behaviour of specimens in group 1. does not manifest the wettability properties of the system.  It results from a peculiar structural  i configuration due to the accumulation of reaction products around the drop periphery as illustrated in specimen 3 2 , Figure 1 9 . The deposit on top of the MgO acts as a barrier that prevents the periphery from advancing.  Both  the rate of barrier formation and final height are dependent on the i n i t i a l alloy concentration within the drop.  At high concentrations, a high barrier  forms rapidly and restricts the drop expansion.  At low concentration a low  barrier forms slowly and allows partial '.expansion of the drop before restriction is imposed.  -.63Group 2. displays the wettability of the MgO by the alloyed iron because oxide deposits around the periphery do not form a barrier.  Oxide  accumulation may slightly retard the advancing periphery as evidenced i n the slope change of the curves after one minute.  This effect is more  pronounced in the Cr-alloyed specimens because more oxide is available for accumulation around the periphery.  In the V-alloyed specimens, the oxide  formed is distributed uniformly across the interface whereas in the Cr-alloyed specimens, the oxide accumulation is confined to the periphery-. The accelerated rate of angle decrease with allpy concentration suggests that the equilibrium angle for the.iron - (V or Cr) oxide - MgO system is probably much smaller than the iron - iron oxide - MgO system. The reason for the behaviour of group 3' is less certain. Apparently the equilibrium angle for the iron - Ti oxide - MgO system at low concentrations is 1 0 5 ° . The precipitated microstructure in specimen 27, Figure 20. (which exists across the entire interface) resembles the pure iron counterpart. Near the periphery the accumulating oxide appears to dissolve the MgO rather than deposit on top of i t as in the' Ta- and Nb-alloyed s'amples.  At high  concentrations, however, the oxide probably deposits at the periphery too rapidly for complete dissolution of the MgO so that a barrier is formed which restricts the periphery.  At intermediate concentrations (e.g. curve  for 0.46$, Figure 17») a barrier forms i n i t i a l l y because of the high rate of T i oxide formation. This barrier begins to dissolve MgO and effectively decreases in height.  Since the depleted drop cannot replenish the oxide  supply and strengthen the barrier, the periphery begins to advance and eventually the barrier disappears at the expense of the MgO. o then advances to its equilibrium value of 105 .  The periphery  - 6h The curve for the steel sample (Figure 18.) resembles the curve for the pure Fe-MgO system except that the final angle is much higher ( 8 5 ° ) . The carbon is probably retarding iron oxide formation and hence preventing the periphery from advancing to the iron - iron oxide - MgO equilibrium angle of 6 5 ° . Since thermodynamic and kinetic calculations were confined by lack of information and complexity, and the reaction products were unidentifiable in almost a l l cases, the precise oxidation mechanism for a particular alloy addition was not determined.  Furthermore, because phase diagrams  of the binary systems of iron oxide and oxides of the alloying elements have not been published, not even prediction of reaction products was possible.  IV. CONCLUSIONS Where chemical reaction occurs, the sessile drop technique does not necessarily manifest the wettability properties of the system.  The  drop behaviour may be governed by (a) interfacial energy relationships or (b) the nature of the reaction product deposit.  Furthermore, the cal-  culation of wettability properties and interfacial energies is inaccurate because of the kinetic character of the parameters involved.  Kinetic  analysis of the reaction mechanism is complicated by geometry factors, concentration gradients, transport problems, etc.  The quantity of reaction  products formed in sessile-drop experiments is insufficient for positive identification and subsequent prediction of possible mechanisms.  A macro-  scopic study of the interface would probably yield more definite information about the chemical reactions involved. Chemical reactions occurringthroughout this work prevented any quantitative study of the preferential interfacial adsorption of electropositive' alloy components originally intended.  The iron reacted with  oxygen released through the dissociation of MgO and formed a.surface oxide layer which reduced the liquid surface tension and decreased the interfacial energy.  The oxide accumulated around the basal periphery of the drop and  solidified as wustite, which decomposed to magnetite and iron.  Interfacial  attraction and monolayer formation of electropositive impurities in the iron may have occurred since no evidence of reaction was observed within the annular oxide deposit.  Results of alloying experiments revealed complete interfacial  attack with alloy additions of T i , Zr, and V, and annular deposits similar to pure iron where additions of Ta, Nb, and Cr were made.  A consideration of free  energy data together with i n i t i a l impurity concentration of the iron suggests that silicon is probably one of the major constituents stabilizing the interface.  - 66 Chemical reaction was evident, when the iron was alloyed with Ta, Nb, Cr, V, T i , and Zr. Presumably the reactions were oxidation of the alloying component.  The wettability of the MgO by the iron improved with  increasing concentration of Cr and V.  In the case of Nb, Ta, T i , and  Zr additions, however, no wettability conclusions could be drawn because the drop behaviour was complicated by peripheral oxide deposits that restricted drop expansion. 'r  V. RECOMMENDATIONS FOR FURTHER WORK Further work on the Fe-MgO system could probably be carried on to greater advantage with iron of higher purity.  A macroscopic system  would provide sufficient quantities of reaction products for identification and kinetic mechanisms could be investigated without complication from geometrical factors, transport.problems, and concentration gradients. In the case of alloy additions to the iron, preliminary studies that would establish phase diagrams for the iron oxi'de--alloy oxide systems would be useful.  The effect of various atmospheres (particularly oxygen)  on the reaction kinetics could then be studied.  Once this information is  gained, a sessile drop study could then be made to determine the. reaction behaviour of single crystals of MgO.  The nature of concentration gradients  and diffusion around the interfacial area.could be studied using a microprobe analyzer. established.  Perhaps then the existence of a monolayer could be  VI. APPENDICES APPENDIX I . S e s s i l e Drop Data  - 68 TABLE IX. Sessile Drop Data Pure Fe-MgO Specimen Time After Mag.' of No. Complete Melting.' Photo. ( ) s e c  10  11 12  13 14 15  16  17 18 19 20 21  10 25 50 110 17 35 80 2 9 4o 105 13 29 50 24 40 80 2 12 26 55 5 24 50 120 16 37 33 44 9 l 7 20 46 78  x  z  7-78 1.452 I.626 1.810 1.885 1.364 I.526 1.650 7.90 1.332 7.90 1.387 1.609  8.84 8.84 7.70  9.05 7.88 8.75  ' ,  - inches -  m  9.98  x  1.763 1.4.34 1.586 I.691 1.525 1.607 1.642 I.365 I.513 1.775 I.896 1.420 1.644 1.788 1.830 1.428 1.635 1.690 1.754 1.334 1.330 1.300 1.415 1.532 1.564  ^  1.181 1.429 1.228 I.185 1.208 1.330 1.204 1.050 1.120 1.281 1.215 1.360 1.155 1.118 1.267 1.255 1.075 1.300 1.170 1.025 1.197 1.305 1.256 1.493 1.369 1.241 I.190 1.372 1.307 1.175 1.167 1.400 1.204 1.295 1.242 1.165 -1.311 1.134 1.267 1.133 1.265 1.350 1.076 .991  10 Determinations  ,  z  k ' Q °LV . AV erg/cm degrees  I..336  .1285  1.444 1.417  1308 1520 1428  1-535 I.370  1545 1235  1.495  1535  1.383 1.464 1.362  1355 1465 1352  Average  103 74 66 65 88 77 65 113 102 71 65 99 77 65 81 72 64 112 99 76 66 110 77 67 65 89' 73 75 70 103 11^ 106 87 70 65  1403 ± 10$  - 69 -  Fe + T i - MgO Mag. of Specimen f> T i Time No. Photo. in Fe After Complete Melting (sec.) 22  2k 25  26  27 28 29 30 31  3.85  15 4o 1 3.85 6 90. 1.00 2 7 21 38 44 57 0.46 l • •16 30 36 90 0 . 2 2 18 22 2 0.22 7 0.22 30 70 0.11 90 2 0.11 5 20 50  8.00 8.00 8.71 8.71 8.71 8.52 8.52 8.52 •8.52 8.52 8.52 7.52 7.52 7.52 7.52 7.52 8.78 8.78 7-55 7-55 8.97 8.97 7.73 7.73 7.73 7-73 7-73  z  X  X'  z'  erg/cm inches 1.257 1.084 1.264 1.086 1.282 1.130 1.304 1.147 1.310 1.138 1.215 1.072 " 1.215 1.072 1.215 1.072 1.246 1.108 1.297 1.131 1.327 1.144 1.276 1.078 1.276 1.078 1.329 1.1341.382 I.160 1.382 1.160 1.233 I.096 1.245 1.092 1.270 1.090 1.270 1.090 1.391 1.221 1.393 1.220 1.348 ' 1.138 1.246 -1.060 i . 2 9 8 1.132 1.373 1.130 1.372 1.130  1.092 I.069 1.079 1.107 i.087 1.070 1.070 1.070 1.137 1.233 1.296 1.108 1.108 1.268 I.352 1.352 1.139 1.195 1.084 1.084 1.368 1.367 1.314 1.173. 1.234 1.346 1.346  1.590 1.588 1.718' 1.718 1.710 1.578 1.578 1.578 1.524 1.436 1.378 1.588 1.588 1.464 1.380 I.380 1.510 1.413 I.626 1.626 1.462 1.460 1.387 1.377 1.434 1.300 1.301  AV degrees 9  1385 1350 1470 1485 1345 . 1415 .1415 1415 1580 1455 1355. 1375 1375 1610 1545 1545 i44o 1325 1525 1525 1585 1555 1425 1310 1760 1250 1255  133 133 133 133 134 132 132 132 120 112 105 132 132 . 114 105 105 119 112 134 134 104 105 105 115 111 104 104  Fe + Ta - MgO $ Ta 32  5.00  39  0.08  40  0.01  4l  0.01  60  0.03  61  0.21  20 90 l 10 20 45 60 5 15 35 50 85 20 33 75 40 80  8.97 8*97 7.98 .7.98 7.98 7.98 7.98 8.36 8.36  8.01 8.01 8.01 8.27 8.27  1;472 1.472 1.355 1.350 1.415 1.317 1.315 1.468 1.479 1.489 1.486 1.368 • ' 1.427 1.452 1.457 1.419 1.422  1.267 1.424 1.589 1.267 1.424 1.589 1.148 1.312 1.458 1.142 "1.313 1.450 1.154 1.387 1.340 1.092 1.297 1.292 1.097 1.295 1.290 1.253 1.425 1.570 1.252 1.434 1.486 1.291 1.290 1.185 1.192 1.410 1.369 1.226 1.449 1.322 1.229 1.451 1.313 1.191 1.379 1,482 1.190 1.369 1.483  1475^ 1475 l4io 1395 1185 114.5 1180 1605 1525  1410 1570 1565 1375 1320  109.5 109.5 112 111 106.5 104 104 UO.5 IO6.5 82 82 82 102 95-5 95 109 110  - 7 0 -  Fe + Nb - MgO Specimen $Nb. Time No. i n Fe A f t e r Complete Melting (sec.)  36  42  43  •58.  8.80  5  15 50. 90 0.01 7 .25 85 0.005  5 '  12 33 60 100 0.0k 1 0 15 4 0.30 1 0 5  59  40  70 .  Mag. o f Photo.  X  x'  . z'  1.490 1.490 1.490 1.490 1.397  1.533 1.533 1-533 1.533.  1390 1390 1390 1.390 , 1535  1.333 1.412  1.486  1.4o8  1680 1110  1.350 1.405 1.405 1.390 1.390 1.390  1.442 .1.372 1.372 1.496 I.496 1.496  1505 1215 1215 1495 1^95 1495  z inches  8.55 8.55 8.55  8.55  8.93 8.87 8.87  1.527 1.527 1.527 1.527 1.434 1.474 1.476 1.384 I.436 1.540  1.546  8.90 8.90 8.90 8.71 8.71 8.71  1.545 I.380 1.420 1.421 1.428 1.428 1.428  1.274 1.274 1.274 1.274  1.244  1.359 1.359 1.221 1.191 1.252 1.228 1.228 1.205 1.198 1.199 1.230 1.230 1.230  :  1.424  AV degrees  ^LV ergs/cm  9  '  107 107 107 107 106.7 85 85 107.5 103.5 78 77 77 105 101 101 107 107 107  Fe + V - Mg50  35'  6.00  44  1.10  45  6.13  46  47  0*55  0.04  5 20 50 6 11 25 5 13 1 6 21 41 10 22  8.95  40  48  49 50  50 6 8.32 15 43 0.65 0 5 20 4 0.42 : 23 43 75  8.89  .1.510 1.750 1.799 1.610 1.680 1.802 1.517 1.744 1.295 1.387 1.588 I.658 1.491 1.538 1.591 1.660  82 59 45. 76 71 60 81  1.320 P. 9 9 6 0.740  1.268 1.194 1.041 1.290 1.096 1.157 1.228 1.254 1.005 0.872 1.315 1.152 1.012 0.995 1.548 1 . 2 1 0 1.642 0.956 •1.739 0.740 1.238 1.114 1.171 1.398 1.345 1.628 1.034 1.416 1.312 1 . 5 8 3 I.O65 1.685 0.938 1 . 7 5 3 O.766  64  l.4ll  1620  109  84  65 56 83 74  65  62 76 60  46  1.424  1655  111.5 •8 8 65 86 68 58, 47  - 71 -  Fe + Gr - MgO Specimen f Cr Time No. in Fe After Complete Melting' (sec.)  33  34  51  52  57  4.00  5 " 15 30 90 0.19 2 5 • 15 45 8o 9-72 0 6 25 4o 65 2.00 2 11  4o 115 0.03 4 22 50 100  Mag. of Photo.  X  z  x'  z'  inches  8.90  8.96 8.96  8.57  "8,93  8,88  1.507 1.802 I.916 I.968 1.384 1.439 I.819 I.925 1.937 1.333 1.554 1.702 1.668 1.666 1.269 1-578 I.628 1.940  1,302 1.821 1.882 1.897  1.251 1.087 0.852 0.735 1.225 1.244 1.442 I.038. O.988 1.149 1.308 D.756 0.640 0.620 1.141 1.211 0.754 0.798 1,128 1.322 I.067 0.995  AV degrees 9  erg/cm  1.501 1.338  1225  95 62 48  1.314 I.588 1.401 1.502  1685 1490  1*248 I.529  1355  4i 112 106 77 57 54 117 80 48  42 41  1.180 1.433  1725  1.260 1.398  1250  O.936 0.928 0.927 0.892 0.943 0.993  1540 1620 1825 1200 1490 . 1530.  112 75 50 45 :.io8 72 59 55  Steel - MgO  12 20 4o " 1 7 20 115 240  8.91 8,91 8.91 8.37 8.37 8.37  0.900 0,900 0.889 0.872 0,906 0,930 1.004 O.965 0.984 0.925  0.967 0.963 0.941 0.965 0.988 1.015  I.123 0.100 0.988 1.208 1.204 1.158.  Average ^ y 1551 Cast Fe (.6 det'ns)  107 106 99.5 II7.5 .112 105.5  . ,88  86.. 5  L  Fe + Zr. - MgO  io Zr  37 38  3-73'  5 12 .41 50  • 8.91 1.320 1.132 I.063 I.780. 8.9I 1.348 I.I70 1.056 1.850 8.90 1.3l3 1.148I.O96 I.698  1190 1360 1370  147.5 147.5 133.5  1565  109 106  Zone Refined Iron - Mg .64  10 14  40 60 90  9.03 9.03  I.518 1.308I.465 1.629 1.508 1.582 1.541 1.319 1.573 1.455 1.732 1.297 I.831 1.232  I5IO  84  74 68  APPENDIX II. Bashforth and Adams Method  - 73 Bashforth and Adams Method For purposes of error determination, specimen number 36 (Fe + 8.80$ Nb) was examined.  Measurements on the photograph yielded: x = z = . x' = z'=  x  1.527 ± .001 inches 1.274 ±-.003 1.490 * .001 " 1.533 ± .001 "  = I..1I99 ± .003  z  From Figure 21. 1.1+40 ± .030 where fe = gdb  2  From Figure 222. x b  = 0.852 * .002  z b  g = gravity., constant d = density of liquid Fe b = radius of curvature . of drop apex  = 0.711 ± .004  It becomes apparent that the accuracy of z and z, is much less -than b that of x and x . Hence most accurate value of b b = I.527 '* .001 = 1.792 ± .006 inches 0.852 * .002 . b = 3.211 ± .022 (inches)  ^  2  = ; db .2  g  4o Most accurate value of density (d) for iron is 7«2 g/cc constant^ g = 980.7 cm/sec . 2  8.55 * .03.  Hence Y °  .  One inch - 2.54 centimeters.  Magnification =  = 980.7(7.2) (3-211 * .022)(2.54)  2  T T r  L  Y  (8:55 *.03) (i.44 r  = 1390 ± 49 = 1390 * 3.-52^ To find contact angle ,0: x'  = O.83I ± .004  z' b  = O.855 * .004  .03)  Gravity. •  - lk Bashforth and Adams Method Continued From Figure 23 • Q = 107 ± 3 ° From Figure 2k.  s  o  ,o  9 = 106.5 * 1  Hence the more accurate value is obtained from the z' measurement.  7 error factor at x = 1.199 is 6.8. Since the uncertainty z = 0.25$, the uncertainty in y^y using Baes' method is 0.25(6.8) = 1.7$' The Baes  in x z This method thus gives an approximate estimation of the error from direct analysis of x . z Since ^ g y l s known within ± 20$, and ^ y within * 25$, the calculated from Young and Dupre's equation (applied.to non-equilibrium conditions) is probably within * 50$  At angles close to 90°, the point of curvature becomes, difficult to determine and hence the accuracy in determining z rapidly decreases. Total accuracy therefore decreases.  APPENDIX I I I .  S t a t i s t i c a l A n a l y s i s of L i q u i d I r o n S u r f a c e Tension Values  TABLE X. Liquid Iron Surface Tension Values Specimen N o <  10 12 13 15 16 19 20 21 22 24 25  26  27 28 29 30 31  \/ y , „ /y p «LV 0LV-1U26 CflLV-lU26) N  1285 1308 1520 1428 1545 1235 1535 1355 1465 1352 1385 1350 1470 1485 1345 1415 1415 1415 1580 1455 1355 1375 1375 1610 1545 1545 l44o 1325 1525 1525 1585 1555 1425 1310 1760 1250 1255  -141 -118 + 94 + 2 +119 - 91 +109 - 71 + 39 - 74 - 4l - 76 + 44 + 59 - 81 - ll - ll - ll +154 + 29 - 71 - 51 - 51 +184 +119 +119 + 14 -101 + 99 + 99 +159 +129 - l -116 +334. -176 -171  19,881 13,924 8,836 4 14,161 8,281 11,881 5,04l 1,521 5,476 1,681 5,776 1,936 3,481 6,561 121 121 121 23,716 841 5,041 2,601 2,601 33,856 14,161 14,161 196 10,201 9,801 9,801 25,281 16,641 l 13,456 111,556 30,976 29,241  Specimen . . N o  32 39  40 60 61 36  42 43 58 59 46 49 33 34 51 52 57 37 38  v  y ^ \2 OLV OLV-1426 («LV-l426)  1475 1475 1410 1395 1185 1145 1180 1605 • 1525 1410 1570 1565 1375 1320 1390 1390 1390 1390 ' 1^35 1680 1110 1505 1215 1215 1495 1495 1495 1620 1655 1225 1685 1490 1355 1725 1250 . 1190 1360 1370  + 49 + 49 - 16 - 31 -24l -281 -246 +179 + 99 - 16 +144 +139 - 51 -106 - 36 -.36 - 36 - 36 +104 +254 -316 + 79 -211 -211 + 69 + 69 + 69 +194 +229 -201 +259 + 64 - 71 +299 -176 -236. - 66 - 56  2,401 2,401 256 96l 58,081 78,961 60,516 32,o4l 9,801 256 20,736 19,321 2,601 11,236 1,296 • 1,296 1,296 1,296 10,816 64,516 99,856 6,24l 44,521 44,521 . 4,761 4,761 4,761 37,636 52,441 4o,4oi 67,081 4,096 5,04l 89,401 30,976 55,696 4,356 3,136  - 81 Results of Statistical Analysis of Liquid Iron Surface Tension Values Mean  tf  LV  = £ XLV N  Standard Deviation,  (5  106988 = 1426 erg/cm .75 ~  / %S  2  ^LV-1426)' N  l , Ma, 201 75  138.7 erg/cm  Experimental Curve of Author  2  Theoretical Normal Distribution Curve  Portion of Surface Tension Values Within:  * <T  68.0$  68.26$  1426 * 2()"  94.7$  95.46$  1426 ± 3 ^  100.0$  99.7 $  1426  Number of Measurements in Region 0 . 5 0 " Designated ro  1019.9  ON  00  O  H  ro  H ..ON  CD  ro o  UJ _  *l ure  1158.6  tr  K Hcn d0 CD O d" CD 4  1297.3  P O CD H j  •!p.  H' CO O d" P S» cn d HO cn Hj d Hf O H'  H)  P? f3  1-3  •  ro  1  CD  ro  vn  1.  1426,0  H  CD  U  P  CD  1  H-  o p  CD  4  1564.7  o  CD H O H3 P CD P O cn H>  ro ro  X CD  1703.4  +  ro ro  +  IVJl  1842.1  + OJ  ro  - S8 "  cn CO d -  H  H-  4 O P  s  CD P c+ P  APPENDIX IV. Thermodynamic Calculations  - 84 Thermodynamic  Calculations  Oxygen p r e s s u r e over magnesium o x i d e : Magnesium o x i d e u s u a l l y d i s s o c i a t e s i n t o i t ' s gaseous elements i.e. For t h i s  MgO  C s )  -*Mg  i 0  ( g 0 +  2  reaction, AF°  = 176,500 - U7.5 T  ^  Assuming s t o i c h i o m e t r i c d i s s o c i a t i o n , pMg = AF° =  2p0 -RT l o g ( p M g ) ( p O 2') 2  ?  -RT l o g 2 ( p 0 3/2 )  =  2  a t T = 1823°K p0 2.  A  2(s)  — *  M  F° = 1 3 7 , 0 0 0  =  o  8  atm.  °2(g)  ( s )  +  -  39.4T  -RT l o g p 0  2  p0  2  = 1.57  2  Oxygen p r e s s u r e over f e r r o u s Fe0 AF°  (  l  — >  )  = =  3  1823°K  at T =  3.  4.23 X 1 0 "  =  2  Oxygen p r e s s u r e over molybdenum o x i d e : M o 0  (  • Fe  56,830  (  l  )  - II.94  -RT l o g ( p 0 )  8  oxide: ^  +  X 1 0 " atm.  (  T  g  )  2  3  2  2  a t T = I.82 3°K P  0  2  =  0.42 X 10  a  t  m  .  2  g  )  Thermodynamic C a l c u l a t i o n s Continued h.  Iron vapor pressure over l i q u i d i r o n : 36  The vapor pressure of metals was studied "by Loftness • , who reports the temperatures at which the i r o n vapor pressure equals 0.001, 0.01, 0.1, 1.0, and 10 mm- of mercury. These were p l o t t e d i n Figure 2 6 . o , and the pressure at 1550 C. i n t e r p o l a t e d from the graph to "be O . O M - 7 mm. of mercury. 5.  Vapor pressure of molecular magnesium oxide: MgO log  -27,320 T  P, MgO  +  10.25  47  at T =•1823°K -8 = 1 7 7 0 X 1 0 atm.  pMgO 6.  Reduction of MgO by i r o n : MgO  + Fe  (s)  A  F  = =  (1) A(RT  FeO  (1)  log p 0 ) 2  U.575(l823)(-0.99) - 8 » 2 6 0 cal/gmole  + Mg  (g)  APPENDIX V. X-ray Data  -88 -  TABLE XI. X-ray Data  Co  Kc< (Fe f i l t e r )  Experimental Values Line d i/io 1  20  4.82  2  6o  2.95  3  100  30 KV, 10 ma, 5 hours A.S.T.M. Published Values MgO  Fe  •  2.52  3°4 4.86  30  2.97  60  2.53  4  20  2.427  2.431  5  90 .  2.097  2.106  6.  30  7  2.425  10  2.097  50  1.713  1.714  4o  6o  1.614  1.615  60  8  90  1.486  1.484  70  9  10  1.325  1.326  10  10  20  1.278  1.279  30  ll  10  1.266  1.266  10  12  20  1.120  1.120  20  13  40  1.089  1.091  50  14  20  1.046  1.048  20  5.  10  .987  .988  10  16  30  .967  .966  .968  • 4o  17  40  • 937  .942  .938  30  1  1.489  10  100  100  52  4  1.270  1.053  5  2  Cu Tarf•set L a t t i c e parameter of P e g ^ determined from experimental r e s u l t s .  4, = 8.378 %  Co Tart set  APPENDIX  VI.  Phase Diagrams  - 90 -  Atom % oxygen 3000  2600  2200  1800  1400  1000  \J 0  26  0? 0.4 22  Wt. % oxygen  F«0-M 0 8  -1  1  r-  Liquid  zooo  / 1600  1200  0  FeO  Magnesio-wustite  zo  _i  10  i  i 60  i  I 80  i_  100  MgO  Figure 21. Phase Diagrams of Fe-0 and FeO-MgO Systems  - 91 -  APPENDIX VII. Calculation of Monolayer Formation from Impurities in the Iron  - 92 C a l c u l a t i o n of Monolayer F o r m a t i o n from Impurities  To f i n d i n t e r f a c e a r e a f o r  i n the  a typical  Iron  drop!:  weight = 1 g x dimension = 1.365  inches  Magnification = 8.84 Hence d i a m e t e r  =  Area  1f  =  2(l.365)(2."54) 8.84 (O.786)  0.485 cm  =  2  Take s i l i c o n as a sample i m p u r i t y . analysis).  If  2  I n i t i a l concentration  0.01 w t . $  (by  structure  t h e n each oxygen atom on the s o l i d s u r f a c e r e q u i r e s \  atom.  spectrographic  = 0.786; cm  is  a monolayer forms i n the SiO^  To f i n d the number of s u r f a c e oxygen atoms on the  silicon  (100) f a c e  of  the MgO: The MgO l a t t i c e S i n c e MgO i s  parameter = 4.213  &  (ASTM X - r a y  File)  face-centered-cubic  2 oxygen atoms occupy (4.213 X 1 0 ) - 8  2  = 17.7 X 1 0 ^ cm - 1  2  -l6 1 oxygen atom t h e r e f o r e  o c c u p i e s 8 . 8 5 X 10  cm  e q u a l s 0.485  8T85T10- )  The number of oxygen atoms a t i n t e r f a c e  c  16  + l U  5 . 4 7 X 10  2  atoms 14  Si0 = 2.73 X 10  Number of s i l i c o n atoms r e q u i r e d f o r Number of atoms i n drop i n i t i a l l y  2  = .0001 ( 6 . 0 2 X 10 ^) 2  14 21,400 X 10 /atoms 2  Thus o n l y t o form a monolayer.  0.01$ of i m p u r i t y  8  ,  atoms=  1  atoms p r e s e n t i n i t i a l l y  is  sufficient  APPENDIX VIII. Diffusion Calculations  Diffusion Calculations Calculation of diffusion distance of Fe into MgO (after Turnbull)  •i C= C  Fick'-s  Law  only valid i f  1.  uniformly • infinite layer of Fe  2.  infinitely thick diffusion medium.  Q  (Dt)  - x , j k Dt 2  2 e  is  (Condition 1. would be upset by an impurity monolayer). Turnbull concludes elements with atomic radii. ^> 1.32 w i l l not diffuse.  37 The atomic radius of Si = 1.32. Turnbull measured diffusion of solid iron into MgO in temperature  range of 134-0° -*• 1060°C.  For purposes of illustration, this calculation  w i l l involve extrapolation of his results to melting point even though experiments were carried out at 15° above the melting point. D = 9.5 (10 )  ^  Y  = 9.5 d o " )  -25,900 RT  p  cm /sec.  =9.5 do" ) - '  -fZ^2£  9  e = 9.5 (io~ )  9  1.98(1808)  7  e  (.00072)  9  -12 2 = 6.83 X 10 cm /sec. Let t  = 120 sec.  Fick's law now becomes C = 5.08(10"^) o u  Hence at x = 0 2 = 5.08 (10~5) o C  ;  3.28 X 1 0 ~  9  2k  - 95 -  VII..BIBLIOGRAPHY 1.  Glasstone, S., "Textbook of Physical Chemistry", Van Nostrand, N.Y. (1946).  2.  Allen, B. C , and Kingery, W. D., Trans. A.I.M.E. 215, 30 (1959).  3.  Kingery,. W.D., and Humenik, M. J r . , J . Phys. Chem. 57, 359 (1953).  4. Economos, G . , and Kingery, W. D., J . Am. Cer. Soc. 36 [ 1 2 ] 4 0 3 (1953). 5.  Kingery, W. D., J . Am. Cer. Soc. ^7, 42 (1954).  6. Humenik, J . , and Kingery, W. D., J . Am. Cer. Soc. 37, l8 (1954).. 7. Baes, C. F . , and Kellogg, H. H . , J . Metals, 5, 6 4 3 (1953). 8. Young, T . , Phil Trans. 5_4, l805; works (ed. Peacock), Vol I . , p. 432. 9. Dupre, A . , "Theorie Mechanique De La Chaleur," p. 369, 1869. 10.  Heine, J . H. D., J . Am. Cer. Soc. 2 1 , [6] 213 (1938).  11.  Bashforth, F . , and Adam, J . C , "An Attempt to Test Theories of Capillarity" Cambridge University Press, London, 1883.  12.  Dorsey, N. E . , J . Wash. Acad. S c i . , 18 [19] 505 (1928).  13.  Halden, F . A . , and Kingery, W. D., J . Phys. Chem. 59_, 557 (1955)-  14.  Mack, G. L . and Lee, D. A . , J . Phys. Chem. 40 [2] 169 (1936).  15.  Mack, G. L . , J . Phys. Chem. 4 0 [2] 159 (1936).  16. 17-  Ellefson, B. S. and Taylor, N. W., J . Am. Cer. Soc. 2 1 [6] 205 (1938). Adam, N. K . , "The Physics and Chemistry of Surfaces", Oxford'University Press., 3rd E d . , London, 194l.  18.  Bartell, F. E . , and Wooley, A. D., J . Am. Cer. S o c ,  19.  Kurkjian, C. R., and Kingery, W. D., J . Phys. Chem., 6_0, 961 (1956).  20.  Guggenheim, E. A . , J . Chem. Phys. 4, 689 (1936).  21.  Clarke, J . F . , M.A.Sc. Thesis, The University of British Columbia, i960.  22.  Murray, P., Powder Met.,, No. 5, 64 (i960).  23-  Philbrook, W. 0., and Bever, M. B . , "Basic Open Hearth Steelmaking", A.I.M.M.E. Publication, N.Y., 1951.  24.  Adamson, A. W., "Physical Chemistry of Surfaces", Interscience Pub., N. Y., i960.  25.  TInklepaugh, J . R., and Crandall, W. B . , "Cermets", Reinhold Pub., N.Y. (I960).  55, 3518 (1933).  -  Bibliography Continued 26.  Tripp, H. P., and King, B. W., J . Am. Cer. S o c ,  27..  Lennard-Jones, J . E . , and Taylor, P. A . , , Proc. Royal Soc. London, 109A 476.  3 8 . [.12]  9  6 -  H-32 ( 1 9 5 5 ) .  (1925).  28.  Shepherd, W. G . , "Study of Electrical and Physical Characteristics of Secondary Emitting Surfaces",_WADC report TR 5 7 - 7 6 0 , 1 9 5 8 .  29.  Turnbull, R. C , "Study of Basic Mechanisms of Diffusion into Ceramic Materials", Monthly Progress Report No. 21k, Alfred University, Alfred,. N.Y.,  30.  Vasilos, T.., and Wuensch, B. I . , Am. Cer. Soc. B u l l . ,  31.  Carter, R. E . , "The Mechanism of Solid State' Reaction Between Magnesium Oxide and Aluminum Oxide and Between Magnesium Oxide and Ferric Oxide", G. E. Research Lab. Report 6O-RL-2I+85M, i 9 6 0 .  32.  Rigby, E . B . , and Cutler, I . B . , "Iron Oxide Diffusion into Magnesium Oxide Single Crystals", Paper l U - B - 6 l , XVIIIth Int. Congr. of Pure and Applied Chem., Montreal, August 6 , 1 9 6 1 .  33.  Carter, R. E , , Private.Communication.  3k'.  Handbook of Chemistry and Physics ( 3 7 t h Ed.), Chemical Rubber Pub. Co. Cleveland, 1 9 5 5 .  35.  DeCleene, M. L . A . , M.A.Sc. Thesis, University of-British Columbia, i 9 6 0 .  36.  Loftness, R. L . , U. S. Atomic Energy Commission Report NAA-SR-132, 1952..  37«  Darken, L . S., and Gurry, R. W., "Physical Chemistry of Metals", McGraw H i l l , N. Y . , 1 9 5 3 . /  38.  Becker, G . , Hardus, F . , and Kornfeld, H . , Arch. Eisenhutten,  39.  Davis, J . K. and Bartell, F. E . , Anal. Chem. 2 0 1 1 8 2 ( 1 9 4 8 ) .  1+0.  Kozakevitch, P., and Urbain, G . , J . Iron and Steel Inst. 1 8 6 , 1 6 7 ( 1 9 5 7 ) .  kl.  Hasselman, D. P. H . , M.A.Sc. Thesis, University of British Columbia, 1 9 5 9 -  k2.  Evans, D. G . , M.A.Sc. Thesis, University of British Columbia, 1 9 6 1 .  1+3.  Cline, R. W., Fulrath, R. M . , Pask, J . A . , J . Am. Cer. Soc. kk [ 9 ] 1+23 ..  k_0 [k] 1 9 1  (1961).  2 0 , 363 ( 1 9 U 9 ) .  (I961).  kk.  Kozakevitch, P., Rev. Metaliurgic, k6, 505 (19I+9).  U5.  Metals Handbook, A.S.M. (19I+8).  1+6.  Bondi, A . , . Chem. Rev. 5_2, I+17 ( 1 9 5 3 ) .  1+7.  Kingery, W. D., and Wygant, J . F . , Cer. B u l l . , 3 1 , 2 5 1 ( 1 9 5 2 ) .  Bibliography  48.  4-9.  A l t m a n , R. L . , and S e a r c y , A . W., " V a p o r i z a t i o n of Magnesium Oxide and i t s R e a c t i o n w i t h A l u m i n a , " Symp. on Chem. a n d Thermo. P r o p , at High Temp., X V I I I t h I n t . Cong, of Pure and A p p l i e d Chem., M o n t r e a l August 6 , 1 9 6 1 . Ihghram,  M. C ,  and Drowart.,  Temperature Temp. 50.  Porter,  R.  1347 51.  Continued  Wells,  A.  1950.  J . , "Mass  Chemistry",  Advanced  Technology,  Stanford  F.,  W. A . ,  Chupka,  Spectrometry Papers  Research  and  of  Applied to  Int.  Institute,  Inghram,  M. G.,  J .  Symp. pp.  on  High High  338-378,  1959.  Chem.Phys. 2 3 ,  (1955). F.,  "Structural  Inorganic  Chemistry",  Oxford  Press,  Clarendon,  

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.831.1-0105618/manifest

Comment

Related Items