UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Thermodynamics of hydrogen in electroslag remelting Chattopadhyay, Subrata 1986

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

Item Metadata

Download

Media
831-UBC_1986_A1 C52_7.pdf [ 6.11MB ]
Metadata
JSON: 831-1.0078430.json
JSON-LD: 831-1.0078430-ld.json
RDF/XML (Pretty): 831-1.0078430-rdf.xml
RDF/JSON: 831-1.0078430-rdf.json
Turtle: 831-1.0078430-turtle.txt
N-Triples: 831-1.0078430-rdf-ntriples.txt
Original Record: 831-1.0078430-source.json
Full Text
831-1.0078430-fulltext.txt
Citation
831-1.0078430.ris

Full Text

THERMODYNAMICS OF HYDROGEN IN ELECTROSLAG REMELTING by Subrata Chat topadhyay B.E. , R.E. C o l l e g e , Durgapur, INDIA, 1978 M .Tech . , Indian Institute of T e c h n o l o g y , Kanpur, 1980 A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L M E N T O F T H E R E Q U I R E M E N T S FOR T H E D E G R E E OF D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E S T U D I E S Department of Metal lurgical Engineer ing W e accept this thesis as c o n f o r m i n g to the required standard T H E U N I V E R S I T Y OF BRITISH C O L U M B I A M a r c h , 1986 © S u b r a t a C h a t t o p a d h y a y , 1986 T8 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head o f my department o r by h i s o r her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department o f The U n i v e r s i t y of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 1 Date DE-6 (3/81) ABSTRACT Oxide ion act iv i t ies in binary C a F 2 - C a O and ternary, C a F 2 -C a O - A l 2 0 3 and C a F 2 - C a O - S i 0 2 , s l ags were de te rmined by C 0 2 - s l a g equi l ibr ium exper iments at 1 4 0 0 ° C . The carbonate c a p a c i t y of these s lags i w a s c o m p u t e d and c o m p a r e d with the s u l f i d e 1 capac i ty data avai lable in the l i terature. The s imi lar i ty in their t rends s u g g e s t s the p o s s i b i l i t y of character iz ing carbonate capac i ty as an al ternat ive b a s i c i t y index for f l u o r i d e - b a s e d s l a g s . The opt ica l bas ic i ty of these s l a g s is d i f f icu l t to d e f i n e , and with the a s s u m p t i o n of C a F 2 as an inert d i luent , this parameter d id not s h o w any dist inct re lat ionship with carbonate c a p a c i t y . S l a g - D 2 0 equi l ibr ium exper iments were p e r f o r m e d at 1 4 0 0 ° C wi th binary ( C a F 2 -C a O ) and ternary ( C a F 2 - C a O - A l 2 0 3 and C a F 2 - C a O - S i 0 2 ) s l ags to de termine water so lub i l i t y at two di f ferent partial p r e s s u r e s of D 2 0 . A new technique o f s l a g s a m p l i n g was e m p l o y e d using a quartz tube. A new and re l iable m e t h o d of water ana lys is in ESR s lags w a s d e v e l o p e d . The new water so lub i l i t y data were higher than the prev ious data by an order of magn i tude , h o w e v e r , the so lub i l i ty data s h o w e d a linear re la t ionship with the square root of water vapour partial p ressure . A l s o , the so lub i l i t y was at a m a x i m u m in binary s lags and a m i n i m u m in ternary s l a g s conta in ing S i 0 2 with N n / N . _ <2. C a O S i 0 2 T h e s e exper imental f ind ings were used a l o n g with literature data to generate a new equi l ibr ium ratio for hydrogen distr ibuted be tween the s l a g and the metal during an ESR p r o c e s s . Th is in fo rmat ion w a s further ii iii ex tended to c o m p u t e the m a x i m u m p e r m i s s i b l e water content in an initial ESR s lag for the p roduct ion of an ESR ingot wi th an acceptab le level of h y d r o g e n . Table of Contents A B S T R A C T ii L ist of T a b l e s vi i List o f F igures ix List o f S y m b o l s xiv A C K N O W L E D G E M E N T S xvi 1.0 I N T R O D U C T I O N 1 2.0 L I T E R A T U R E REVIEW 7 2.1. H Y D R O G E N IN ESR S L A G S 7 2.1.1 F O R M OF H Y D R O G E N P R E S E N T IN E S R S L A G S 7 2.1.2 S O L U B I L I T Y OF H Y D R O G E N IN ESR S L A G S 9 2.1.3 F A C T O R S A F F E C T I N G S O L U B I L I T Y OF H Y D R O G E N IN F L U O R I D E S L A G S 17 2.2. H Y D R O G E N IN N O N - E S R S L A G S 18 2.2.1 W A T E R SOLUBIL ITY IN G L A S S 20 2.2.1.1 W A T E R SOLUBIL ITY IN B I N A R Y G L A S S S Y S T E M S 20 2.2.1.2 W A T E R SOLUBIL ITY IN T E R N A R Y G L A S S S Y S T E M S 25 2.2.1.3 M E C H A N I S M S O F W A T E R D I S S O L U T I O N 32 2.2.2 W A T E R SOLUBIL ITY IN M E T A L L U R G I C A L S L A G S 35 2.2.2.1 B I N A R Y S L A G S „...".'.„.'..'...:. 35 2.2.2.2 T E R N A R Y S L A G S 39 2.3. H Y D R O G E N IN ESR I N G O T S .49 2.3.1 R E A C T I O N S OF H Y D R O G E N T R A N S F E R 52 2.3.2 F A C T O R S C O N T R O L L I N G H Y D R O G E N IN T H E INGOT 57 2.4. T E C H N I Q U E S OF M E A S U R E M E N T O F H Y D R O G E N IN S L A G 58 iv V 2.4.1 S A M P L I N G OF S L A G 58 2.4.2 M E T H O D S OF A N A L Y S I S 59 2.5. T H E R M O D Y N A M I C P R O P E R T I E S OF ESR S L A G S 65 2.5.1 P H A S E D I A G R A M S 66 2.5.2 B A S I C I T Y A N D A C T I V I T Y O F ESR S L A G S 76 2.5.2.1 C O N C E P T OF B A S I C I T Y 76 2.5.2.2 A C T I V I T Y IN ESR S L A G S 79 2.5.3 SOLUBIL IT IES 85 2.5.3.1 S O L U B I L I T Y OF G A S E S IN M O L T E N S L A G S 85 2.5.3.2 S O L U B I L I T Y OF G A S E S IN F U S E D S A L T S ..86 3.0 O B J E C T I V E S 87 4.0 E X P E R I M E N T A L 88 4.1. C O j - S L A G EQUIL IBRIUM 88 4.2. D 2 0 - S L A G EQUIL IBRIUM 92 4.3. W A T E R A N A L Y S I S 96 4.3.1 F A B R I C A T I O N OF HIGH V A C U U M A P P A R A T U S 97 4.3.2 P R E P A R A T I O N FOR C A L I B R A T I O N 99 4.3.3 C A L I B R A T I O N A N D S L A G A N A L Y S I S FOR W A T E R 100 5.0 R E S U L T S A N D D I S C U S S I O N S 104 5.1. O X I D E ION A C T I V I T Y IN ESR S L A G S 104 5.1.1 C A R B O N A T E EQUIL IBRIUM IN -FLUORIDE S L A G S 104 5.1.2 V E R I F I C A T I O N OF C A R B O N A T E EQUIL IBRIUM 112 5.1.3 A C T I V I T Y C A L C U L A T I O N F R O M T H E S U L F I D E C A P A C I T Y D A T A 113 5.1.4 I M P O R T A N C E OF T H E C A R B O N A T E EQUIL IBRIUM 116 5.1.4.1 C A R B O N A T E C A P A C I T Y A N D B A S I C I T Y O F S L A G 117 5.1.4.2 C A R B O N A T E C A P A C I T Y A N D O P T I C A L B A S I C I T Y 124 vi 5.2. W A T E R S O L U B I L I T Y IN ESR S L A G S 124 5.2.1 A D V A N T A G E S A N D D I S A D V A N T A G E S OF T H E D E U T E R I U M T R A C E R D E T E C T I O N T E C H N I Q U E _ 126 5.2.2 W A T E R S O L U B I L I T Y IN B INARY A N D T E R N A R Y F L U O R I D E - B A S E D S L A G S 127 5.2.3 W A T E R S O L U B I L I T Y A N D W A T E R V A P O U R P R E S S U R E IN ESR 134 5.2.4 T H E R M O D Y N A M I C A N A L Y S I S OF T H E W A T E R S O L U B I L I T Y D A T A 136 5.2.5 P R E D I C T I O N OF H Y D R O G E N L E V E L IN ESR I N G O T S 142 6.0 S U M M A R Y 146 6.1. C O N C L U S I O N S 146 6.2. S U G G E S T I O N S FOR F U T U R E W O R K 147 R E F E R E N C E S 149 A P P E N D I X I 156 I. I. C H E M I C A L A N A L Y S I S OF S L A G 156 1.1.1 F L U O R I D E A N A L Y S I S 156 1.1.2 A N A L Y S I S FOR C A L C I U M , S IL ICON A N D A L U M I N U M 157 1.1.3 A T O M I C A B S O R P T I O N M E T H O D S 157 A P P E N D I X II 159 II. 1. DEFINITION OF O P T I C A L B A S I C I T Y 159 11.2. C A L C U L A T I O N OF O P T I C A L B A S I C I T Y 160 List of Tables Tab le Page Tab le 2.1 Repor ted data on water so lub i l i ty in ESR s lags 10 Tab le 2.2 I o n - o x y g e n attract ion in var ious ox ides 34 Tab le 2.3 Water so lub i l i t y in d i f ferent s i l icate s l a g s 4 1 50 Tab le 4.1 V e r i f i c a t i o n of water ana lys is by i s o t o p e tracer technique 103 Tab le 5.1 A c t i v i t y of C a O in C a F , - C a O s y s t e m at 1 4 0 0 ° C 106 Tab le 5.2 A c t i v i t y of C a O in C a F 2 - C a O - A l 2 0 3 s y s t e m at 1 4 0 0 ° C 106 Tab le 5.3 A c t i v i t y of C a O in C a F 2 - C a O - S i 0 2 s y s t e m at 1 4 0 0 ° C 107 Tab le 5.4 Var ia t ion of 7 r < _ 2 in binary a luminate and s i l ica te s lags at 1 5 0 0 ° C r°JL 114 Tab le 5.5 C o m p a r i s o n of exper imental and ca lcu la ted ac t iv i ty of C a O ; ca lcu la t ions b a s e d on sulphide capac i ty m e a s u r e m e n t s " in C a F 2 - C a O - A l 2 0 3 s l ags at 1 4 0 0 ° C ; 7 =3.4 115 Tab le 5.6 Var ia t ion of oxide ion act iv i ty in N a 2 0 - S i 0 2 s l a g s at 1 3 7 3 ° K obta ined f r o m carbonate equi l ibr ium and E M F m e a s u r e m e n t s 9 2 . 120 Tab le 5.7 Water so lub i l i ty in binary f luor ide s l a g s at 1 4 0 0 ° C 128 T a b l e 5.8 C h e m i c a l ana lys is o f b inary s lags 128 Tab le 5.9 Water so lub i l i ty in ternary f luor ide ( F l - C - A ) s l a g s at 1 4 0 0 ° C ..129 vii vii i Tab le 5.10 C h e m i c a l ana lys is of ternary ( F l - C - A ) s lags 129 Tab le 5.11 Water so lub i l i ty in ternary f luor ide ( F l - C - S ) s lags at 1 4 0 0 ° C . 131 Tab le 5.12 C h e m i c a l a n a l y s i s of ternary ( F l - C - S ) s lags 132 Tab le 5.13 C o m p a r i s o n of the new water ana lys is with V E W ( A u s t r i a ) - a n a l y s i s 133 Tab le 5.14 A c t i v i t y c o e f f i c i e n t s of hydroxy l ion , 7 , at 1 4 0 0 ° C in f luor ide s lags obta ined f r o m exper imental data of ox ide ion act iv i ty and water so lub i l i t y 138 Tab le 5.15 C a t i o n - A n i o n attract ion in var ious sal ts 139 List of Figures Figure Page F i g . 1.1 S c h e m a t i c d iagram of E l e c t r o s l a g Remel t ing P r o c e s s 2 F i g . 1.2 H y d r o g e n f lak ing in an ESR ingot of AISI 4340 s t e e l , 7 0 0 m m dia. , T r a n s v e r s e s e c t i o n , as c a s t , nital etch 3 F i g . 1.3 S c h e m a t i c d iagram of hydrogen m o v e m e n t in ESR 6 F i g . 2.1 H y d r o g e n contents in ESR s lag and metal under d i f ferent c o n d i t i o n s 4 11 F i g . 2.2 D is t r ibut ion coe f f i c ien t of h y d r o g e n be tween s lag and metal during ESR as a funct ion of s lag b a s i c i t y 1 13 F i g . 2.3 Var ia t ion of water so lubi l i ty in s lags with water vapour partial pressure at 1 6 5 0 ° C 1 4 14 F i g . 2.4 E f f e c t o f s u p e r i m p o s e d DC on the h y d r o g e n level in ESR p r o c e s s 4 . 19 F i g . 2.5 E f f e c t o f temperature on water so lub i l i t y in alkali and alkal ine earth d is i l icate g l a s s e s 2 8 22 F i g . 2.6 E f f e c t o f temperature on water so lub i l i t y in s i l icate s lags obta ined at p = 146torr 3 1 23 HjO F i g . 2.7 Water so lub i l i t y as a funct ion o f N a 2 0 content in N a 2 0 - S i 0 2 s y s t e m 3 0 .24 F i g . 2.8 Water so lub i l i t y as a funct ion of L i 2 0 content in L i 2 0 - S i 0 2 s y s t e m 5 0 .26 ix X F i g . 2.9 Water so lubi l i ty as a funct ion of m o l e percent base in var ious s i l icate s lags at p = 146 t o r r 3 1 27 H 2 0 F i g . 2.10 H y d r o g e n solubi l i ty in var ious s i l ica te s lags conta in ing ac id oxide at 1 5 0 0 ° C , C a O / S i O 2 = 0 . 5 9 , p u _ = 289 t o r r " 28 HjO F i g . 2.11 H y d r o g e n solubi l i ty in var ious s i l ica te s lags conta in ing amphoter ic oxide at 1 5 0 0 ° C , C a O / S i O 2 = 0 . 5 9 , p = 2.89 t o r r " 29 H 2 0 F i g . 2.12 Water so lubi l i ty in the C a O - S i 0 2 - L i 2 0 s y s t e m 3 4 30 F i g . 2.13 Water so lubi l i ty in the C a O - S i 0 2 - S r O s y s t e m 3 4 31 F i g . 2.14 C o m p a r i s o n of di f ferent water so lub i l i t y data in C a O - S i 0 2 s y s t e m 36 F i g . 2.15 E f f e c t of temperature on hydrogen so lub i l i ty in 63 S i 0 2 - 3 7 C a O s l a g 3 3 37 F i g . 2.16 Water so lub i l i ty in T e O ' - S i 0 2 s y s t e m 3 ' 38 F i g . 2.17 Water so lub i l i ty (ppm H 2 0 ) in C a O - S i 0 2 - A l 2 0 3 s y s t e m at 1 5 5 0 ° C and p u =289 to r r 2 2 .40 H 2 0 F i g . 2.18 Relat ionship between o x y g e n dens i ty and water so lub i l i ty in C a O - S i 0 2 - A l 2 0 3 s y s t e m at 1 5 0 0 ° C and p =190 t o r r 3 6 42 H 2 0 F i g . 2.19 Water so lub i l i ty in C a O - F e O - S i 0 2 s y s t e m at 1 6 0 0 ° C , C a O / S i 0 2 = 1 and p = 760 t o r r 3 8 .44 H 2 0 F i g . 2.20 Water so lubi l i ty (ppm H 2 0 ) in C a O - S i 0 2 - F e O s y s t e m obta ined at 1 5 5 0 ° C , p H Q = 2 8 9 to r r 2 2 45 F i g . 2.21 Water so lubi l i ty as a funct ion of s l a g bas ic i ty and m a g n e s i a content at 1 5 5 0 ° C and p =289 to r r 4 3 .47 n 20 F i g . 2.22 Water so lub i l i ty as a funct ion o f s i l i c a act iv i ty in the C a O - M g O - S i O j s lags at p =760 to r r 4 4 .48 H 2 0 xi F i g . 2.23 H y d r o g e n equi l ibr ium be tween s lag and metal in an ESR o p e r a t i o n , s lag is 40 C a F 2 - 3 0 A l 2 0 3 - 3 0 C a O , T = 1 7 8 0 ° C , p =7 tor r 1 54 H ,0 F i g . 2.24 H y d r o g e n solubi l i ty in ESR ingots as a funct ion of water vapour pressure in the a t m o s p h e r e 4 8 '. 56 F i g . 2.25 S c h e m a t i c d iagram of the apparatus for determin ing h y d r o g e n in ESR s l a g s 5 1 61 F i g . 2.26 Appara tus for determining h y d r o g e n in s l a g s 1 7 63 F i g . 2.27 S c h e m a t i c d iagram of h y d r o g e n a n a l y z e r 5 2 64 F i g . 2.28 Phase d iagram for C a O - C a F 2 s y s t e m 68 F i g . 2.29 D e p r e s s i o n of f reez ing point of C a F 2 by C a O , M g O and S r O " . .69 F i g . 2.30 L iquidus line of C a F 2 - A l 2 0 3 m e l t s 6 4 70 F i g . 2.31 Phase d iagram for the s y s t e m * C a F 2 ' - A l 2 0 3 ; * C a F 2 ' = C a F j + 2wt% C a O 72 F i g . 2.32 Phase d iagram for C a F 2 - C a O - A l 2 0 3 s y s t e m 5 6 73 F i g . 2.33 Phase d iagram for C a F 2 - . C a O - S i 0 2 s y s t e m ; Cr = C r i s t o b a l i t e , Tr = T r i d y m i t e 5 6 75 F i g . 2.34 L o n g - l i f e oxygen s e n s o r for measur ing oxygen potent ia l o f s l a g 7 5 . 78 F i g . 2.35 A c t i v i t y of C a O in C a O - C a F 2 s y s t e m at 1 5 0 0 ° C .81 F i g . 2.36 A c t i v i t y of C a O in C a F 2 - C a O - A l 2 0 3 s y s t e m : , 1 5 0 0 ° C 6 7 , 1 4 2 7 ° C " ; ^ 83 F i g . 2.37 Isoact iv i ty of C a O in the C a F 2 - C a O - S i 0 2 s y s t e m at 1 4 5 0 ° C . 84 Fig . 4.1 S c h e m a t i c diagram of the apparatus for carbonate equi l ibr ium s tud ies 89 F i g . 4.2 T y p i c a l cal ibrat ion plot for carbonate ana lys is 91 F ig . 4.3 S c h e m a t i c diagram of the apparatus for D 2 0 - s l a g equi l ibr ium water vapour pressure 93 F i g . 4.4 C o m p a r i s o n of theoret ical and exper imental equi l ibr ium water vapour pressure 95 F i g . 4.5 S c h e m a t i c diagram of the apparatus for water a n a l y s i s of s l a g . ...98 F i g . 4.6 T y p i c a l cal ibrat ion plot for water ana lys is 101 F i g . 5.1 C a O act iv i ty in the C a O - C a F 2 - b i n a r y s lag s y s t e m 108 F i g . 5.2 C a O act iv i ty in the C a O - A l 2 0 3 - C a F 2 ternary s l a g s y s t e m 109 F ig . 5.3 C a O act iv i ty in the C a O - S i 0 2 - C a F 2 ternary s lag s y s t e m 110 F i g . 5.4 C o m p a r i s o n of carbonate capac i ty and sulphide c a p a c i t y 9 0 of s i l i c a t e , aluminate and f luor ide s lags 118 F i g . 5.5 Re la t ionship between carbonate capac i ty and oxide ion ac t iv i ty in f luor ide s lag at 1 4 0 0 ° C 122 F i g . 5.6 Relat ionship between carbonate capac i ty and s o d i u m oxide ac t iv i ty in N a 2 0 - S i 0 2 s lag at 1 2 0 0 ° C . ' 3 123 F i g . 5.7 Re la t ionship between opt ica l b a s i c i t y and carbonate capac i ty in binary s i l ica te , aluminate and ternary f luor ide s l a g s 125 F i g . 5.8 So lub i l i t y of water in F l u o r i d e - b a s e d s lags as a f u n c t i o n of water vapour pressure 135 F i g . 5.9 Re la t ionship between hydroxy l capac i ty and ox ide ion act iv i ty in var ious f luor ide s l a g s 141 xiii F i g . 5.10 The m o d i f i e d h y d r o g e n equi l ibr ium between s lag and m e t a l , b a s e d on the present work 143 List of Symbols carb A l 3 0 3 Inter- ionic d is tance Bas ic i ty on the b a s i s of carbonate capac i ty C a O Carbonate capac i ty of a s lag co>-c s e Fl A G ° hyd I K L Si M N No O B P R S Carbonate capac i ty of a re fe rence s lag Su l f ide capac i ty Interaction parameter C a F 2 Standard free energy change Hydroxy l ion I o n - o x y g e n interact ion Equi l ibr ium react ion constant A v o g a d r o ' s number Distr ibut ion c o e f f i c i e n t of h y d r o g e n b e t w e e n s lag and metal A v e r a g e molecu la r weight M o l e f ract ion M o l e f ract ion o f o x y g e n a t o m s in a melt Opt ica l B a s i c i t y Partial pressure G a s constant , 1.987 c a l o r i e s / d e g r e e . m o l e S i O , xiv XV T : Temperature X : M o l e f ract ion x : Pauling e lec t ronegat iv i ty Z , : va lency of anion Z 2 : va lency of cat ion a : A c t i v i t y 7 : A c t i v i t y c o e f f i c i e n t p : Dens i ty <j> : Diameter ( ) : s lag phase [ ] : metal phase { } : gas phase ACKNOWLEDGEMENTS I w o u l d like to express m y s incere gratitude to P r o f e s s o r A . Mi tche l l for his gu idance through out the c o u r s e of this work . Helpful d i s c u s s i o n s with other facu l ty m e m b e r s and f e l l o w graduate students a l s o contr ibuted to the s u c c e s s o f this s tudy . I am indebted to P r o f e s s o r R. Butters and Mr . Rudy C a r d e n o for all the technical a s s i s t a n c e at d i f ferent s tages of the exper imenta l work . T~ I w i s h to grateful ly a c k n o w l e d g e the f inancia l a s s i s t a n c e of N S E R C , C a n a d a and the C o n s a r c C o r p . xv i CHAPTER 1 INTRODUCTION The E l e c t r o s l a g Remel t ing (ESR) p r o c e s s has been w i d e l y used as a s e c o n d a r y re f in ing step for structural const ruc t iona l s t e e l s , s ta in less s tee ls and var ious other a l loys such as Inconels and H a s t e l l o y s . A s c h e m a t i c of the p r o c e s s is shown in f igure 1.1, where a c o n s u m a b l e e lec t rode is remel ted in a w a t e r - c o o l e d crucib le through a mol ten p o o l of bas ic s l a g . The thermal energy of this p r o c e s s is supp l ied by the res is tance heating of the s lag p o o l for which the e lectr ica l c o n n e c t i o n s f r o m a t rans former are made b e t w e e n the e lec t rode and the w a t e r - c o o l e d b a s e - p l a t e . The f inal remel ted p r o d u c t s , s o o b t a i n e d , are not on ly c h e m i c a l l y re f ined by the s l a g - m e t a l reac t ions , but they a lso p o s s e s s reproduc ib le i so t rop ic mechan ica l proper t ies and g o o d sur face f i n i s h . M o r e o v e r , the p r o c e s s o f f e r s the advantage of p rec ise a l loy addi t ions and contro l o v e r the f inal ingot structure. H o w e v e r , c o m p a r i n g the features o f other s e c o n d a r y ref in ing p r o c e s s e s , such as V a c u u m A r c Remel t ing (VAR) and E l e c t r o n B e a m Me l t ing (EBM), the E S R p r o c e s s s tands wel l over others in its potent ia l to p roduce ingots o f bigger s i z e s . Th is advantage w o u l d lose its meri t , if the p r o b l e m of h y d r o g e n , as o b s e r v e d in the f o r m o f hairl ine c racks around the ingot basal m i d - r a d i u s r e g i o n , is not a v o i d e d (figure 1.2). The p r o b l e m with h y d r o g e n is m o r e no t iceab le in ESR ingots with d iameters exceed ing about 5 0 0 m m . The sma l le r ingots have a favourab le thermal h is tory to permit adequate d i f f u s i o n b e f o r e crack f o r m a t i o n p r o b l e m s d e v e l o p . 1 2 POWER SUPPLY TRANSFORMER OR RECTIFIER ELECTRODE CRUCIBLE COOLING WATER SLAG POOL METAL POOL LIQUID + SOLID SOLIDIFIED INGOT SLAG SKIN «— 4j =-£^_-=^--_T-^ — - --. I - BASEPLATE Fig. 1.1 Schematic diagram of Electroslag Remelting Process 3 Fig 12 Hydrogen flaking in an ESR ingot of AISI 4340 s tee l , 700mm dia., Transverse sect ion, as cast, nital etch. The circled area shows a typical hydrogen f lake. 4 The to lerable level of hydrogen in s tee ls is on ly around 2 p p m . A n increase b e y o n d this level not on ly g i v e s r ise to hairl ine c r a c k s , but a l s o reduces the fat igue strength and fracture t o u g h n e s s of s t e e l s . T h e s e hairl ine c r a c k s , a lso known as f l a k e s , are genera l ly very v is ib le o n a fracture or meta l lographic sur face and are in a l m o s t all c a s e s readi ly detectab le in a s - c a s t and heavy s e c t i o n f o r g e d s t e e l s . M o r e o v e r , these d e f e c t s are a l s o o f ten detected by u l t rasonic ^test ing of fo rg ings f r o m ESR ingots as r e f l e c t i o n s are produced f r o m the d i s c o n t i n u t i e s . B e f o r e the cracks d e v e l o p the o n l y w a y to sa lvage a c o s t l y ESR ingot wi th a high level of hydrogen is to t ransfer the ingot qu ick ly to an anneal ing furnace and soak the ingot fo r a su f f i c ien t length of t ime (depending on the c r o s s - s e c t i o n of the ingot) . The extra fac i l i t i es and the t ime invo lved in this operat ion w o u l d make the ESR p r o c e s s f inanc ia l l y less v iab le . The so lu t ion to this p r o b l e m l ies in the opera t ion of the p r o c e s s i tsel f where the hydrogen level in the ingot shou ld be m o n i t o r e d as the remel t ing p r o g r e s s e s . Unfor tunate ly , there is no s a f e m e t h o d o f s a m p l i n g the l iquid s tee l , which is b e l o w a p o o l of l iquid s l a g , wi thout endanger ing the qual i ty of the ingot . Whi le s a m p l i n g with a quartz tube , any broken p iece of the tube, if it remains in the l iquid s t e e l , wi l l ruin the w h o l e ingot . O n the other hand, s a m p l i n g the s lag and moni tor ing its c o m p o s i t i o n are re lat ively e a s y and are sa fer o p e r a t i o n s . In a d d i t i o n , p rev ious s t u d i e s 1 ' 2 ' 3 indicate that there is an equi l ibr ium part i t ion o f h y d r o g e n b e t w e e n the s lag and the meta l . T h u s , the mon i to r ing of s lag c o m p o s i t i o n w o u l d give an indirect m e a s u r e m e n t o f h y d r o g e n in the ESR ingot . 5 There are three major s o u r c e s of h y d r o g e n during an ESR p r o c e s s : the c o n s u m a b l e e lec t rode , the a tmosphere on top of the s lag and the s lag i tse l f . The hydrogen level in the e lec t rode is norma l ly low (=*2ppm) and in the ideal c a s e , the h y d r o g e n content in an ESR ingot s h o u l d not be m o r e than that of the e l e c t r o d e . M o s t ESR s l a g s conta in a c o n s i d e r a b l e amount of C a O which is very h y g r o s c o p i c in nature. C o n s e q u e n t l y , any s l a g , not dr ied and s tored proper ly b e f o r e it is u s e d , w o u l d inherently conta in a substant ia l amount of h y d r o g e n . In add i t ion , the mois ture in the a t m o s p h e r e , wh ich is in direct contact wi th the s lag during remel t ing , in f luences the hydrogen level in the s l a g . The hydrogen t ransfer m e c h a n i s m during an ESR p r o c e s s invo lv ing highly bas ic s lag can be dep ic ted as in f igure 1.3. The free oxide ions in the mo l ten s lag react with water vapour present in the a tmosphere and stabl i l ize hydroxy l ion in the s l a g : {H20} + ( O 2 ) = 2 ( O H ) 1.1 The f inal t ransfer of hydrogen f r o m the s lag to the ingot takes place a c c o r d i n g to the f o l l o w i n g reac t ion : 2 ( O H ) = 2[H] + ( O 2 ) + [O] 1.2 C o n s i d e r i n g the a b o v e react ions and the need to moni tor s lag c o m p o s i t i o n , the impor tance o f the reaction(1.1) is c lear ly evident in cont ro l l ing the hydrogen in the s lag and subsequent ly in the ingot. T h e r e f o r e , the present work is d i rected pr inc ipa l ly t o w a r d the invest igat ion o f ox ide ion ac t iv i ty and the hydroxy l ion capac i ty of E S R s l a g s . 1.3 S c h e m a t i c d iagram of hydrogen m o v e m e n t in E S R CHAPTER 2 LITERATURE REVIEW 2.1. H Y D R O G E N IN ESR S L A G S In the E lec t ros lag Remel t ing p r o c e s s , the s lag p l a y s an unique ro le , be ing at the s a m e t ime , the source of rthermal requ i rements and a l s o of chemica l reac t ions to produce a c h e m i c a l l y re f ined and p h y s i c a l l y s o u n d ingot . A s far as the m o v e m e n t of hydrogen is c o n c e r n e d , the cri t ical role of ESR s lags has been studied by many authors. 1 " 7 The data reported here are obta ined f r o m laboratory exper iments except in the c a s e s of Jaeger et a l . 4 and Chuiko et al . ' who invest igated the h y d r o g e n prob lem in industrial ESR units. 2.1.1 F O R M OF H Y D R O G E N P R E S E N T IN ESR S L A G S H y d r o g e n can be present in ESR s lags as d i s s o l v e d gas , in the f o r m of p ro ton H + and hydroxy l ion , O H \ T o date there are no literature reports of the ex is tence of hydrogen d i s s o l v e d in its molecu lar f o r m . M o r e o v e r , the S iever t ' s law re lat ionship as o b s e r v e d in m o s t s t u d i e s , s h o w i n g the squareroot d e p e n d e n c y of the partial pressure of water vapour on the so lub i l i t y of hydrogen in s l a g s , d o e s not support any molecu lar d i s s o l u t i o n of h y d r o g e n . Never the less , it i s , in pr inc ip le , p o s s i b l e to have a minute quantity of hydrogen d i s s o l v e d in s l a g s , but p r o b a b l y because of the d i f f i cu l t i es in the determinat ion techn ique , su f f ic ien t p r e c i s i o n is not ava i lab le to detect it. 7 8 The presence of hydrogen as p ro ton (H +) in s tee l mak ing s lags is s u g g e s t e d by W a l s h et a l . ' because of the pro ton 's sma l l ionic s i z e . In blast furnace s lag ( 4 0 S i O 2 - 4 0 C a O - 2 0 A l 2 0 } ) the direct e v i d e n c e of D* (in a i s o t o p e tracer s tudy) is repor ted by K o b a y a s h i et a l . . 1 0 S imi la r c o n s i d e r a t i o n is made by Forno et a l . 1 1 in ESR s lags because of the e l e c t r o c h e m i c a l nature of react ions occur r ing during this p r o c e s s . [H] = H* + e 2.1 The a b o v e anod ic react ion is p o s s i b l e in an ac id ic s l a g , but in the ESR s y s t e m , m o s t s lags are highly bas ic in nature and thus doubt is ra ised as to the feas ib i l i t y of that reac t ion . Holzgruber et a l . 6 s u g g e s t e d that for s lags with high O 2 " anion c o n c e n t r a t i o n , hydrogen is t ranspor ted in the s lag by the jumping of cat ion H* f r o m one anion to another e v e n though s o m e h y d r o g e n m a y be present as OH" ion . The reac t ions occur as f o l l o w s . {H20} + ( 0 » - ) = 2 ( O H ) 2.2 ( O H ) + ( O 2 ) = ( O 2 ) + (H +) + ( O 2 ) 2.3 A more feas ib le p roposa l is made by M e d o v a r et a l . 1 2 regarding the f o r m of ex is tence of hydrogen in s l a g s . T h e y p r o p o s e d that the presence of both f o r m s of i o n s , such as OH" and H + , and their p ropor t ion w o u l d depend on the c o m p o s i t i o n of the s l a g . The increase of hydrox ide f o r m e r s such as ox ides and sal ts of C a , M g , K and Na in the s lag w o u l d result in the p r e d o m i n a n c e of hydroxy l ion (OH - ) . The p resence of hydroxy l an ions in bas ic ESR s lags is a lso just i f iab le o n the bas is of the ex is tence of s u l f i d e 1 3 , f luor ide and c y a n i d e 1 4 ions which a l s o have s imi lar negat ive charges and ionic s i z e s . Other a u t h o r s 7 ' 1 5 have ma in ly c o n s i d e r e d the p r e s e n c e o f hydroxy l ions in their r e s p e c t i v e s tud ies c o n c e r n i n g h y d r o g e n in 9 ESR s l a g s . 2.1.2 S O L U B I L I T Y OF H Y D R O G E N IN ESR S L A G S Whatever the f o r m of h y d r o g e n may be , there are repor ts ava i lab le on its so lubi l i ty in f luor ide s l a g s , as tabulated in table 2.1. Other data are not c o m p l e t e in nature: a lso d isc repanc ies can arise through the d i f f e r e n c e s in sampl ing and determinat ion techniques e m p l o y e d . B e c a u s e of the doubt a s s o c i a t e d with the f o r m of hydrogen present in ESR s l a g , s o m e a u t h o r s 6 ' 1 6 preferred report ing the equivalent water so lub i l i t y whereas others 1 ' 3 ' 4 ' 1 2 " 1 7 reported it as the hydrogen content d i s s o l v e d in the s l a g . The latter m e t h o d s e e m s more popular s ince it a l lows a direct c o m p a r i s o n with h y d r o g e n level in an ingot whereas the water so lub i l i t y data g ives cor re la t ion with the moisture level in the a t m o s p h e r e , in contact wi th the s l a g . The hydrogen solubi l i ty data can be c o n v e r t e d to equivalent water so lub i l i t y by mul t ip ly ing the fo rmer by a fac tor of 9, without any no t iceab le e r ror . " where the hydrogen concentra t ion w a s found to be 30 to 60 p p m (figure 2.1). Due to the h y g r o s c o p i c nature of these s l a g s , proper s torage and handl ing s y s t e m s are required to keep the hydrogen level under c o n t r o l . The increase of l ime in the s lag increases the d i f fus ib i l i t y of h y d r o g e n . The h y d r o g e n a lso increases with the decreas ing grain s ize of the s lag as a result o f increasing the s p e c i f i c sur face area. meta l b e c o m e s a lmost constant after remel t ing for more than an hour. For a s l a g wi th 30 -40% C a O and the humidi ty o f the a tmosphere o f 14 to 16 Jaeger et a l . 4 s tudied main ly C a O - A I 2 0 3 - C a F 2 t y p e s of s lag The hydrogen distr ibut ion c o e f f i c i e n t be tween s l a g and 10 Tab le 2,1 ReDorted data on water so lub i l i t y in ESR s l a q s ( re fererences in b racke ts ) S l a g C o m p o s it ion(wt.%) Prl,0 T e m p . (HP) N o . C a F 2 C a O A . 2 0 3 S i 0 2 M g O torr ° C p p m 1 77.1 21.6 0.1 0.4(2) 1.0 8.0 17.0 166.5 432 558 2 67.2 3.3 28.1 0.8(2) 1.0 8.0 17.0 51.3 144 198 3 40 30 30(1) 6.97 1780 216 4 60 20 20(1) 0.98 1730 315 5 50 25 25(3) 6 252 6 75 25(3) 6 49.4 7 43.5 30.2 1.5 21.2 0.9(3) 6 41.7 8 20 40 40(5) 16 418 9 60 15 25(5) 15.6 18 330 110 10 51 24.5 19.5(17) 150 1600 225 11 80 H in slag,ppm60 H in metal, ppm. slag remeltinf, slag teeming into the mould 6 10 14 Remelting Time , hr. Ingot dia. 800-1000 mm. F i g . 2.1 H y d r o g e n conten ts in ESR s l a g and meta l under d i f f e ren t c o n d i t i o n s 4 12 g m / N m 3 , the distr ibut ion constant attained a value be tween 3.5 and 4.5 w h i c h is in c l o s e agreement with the va lues of Shurmann et a l . 1 (figure 2.2). The hydrogen level is further reduced whi le remel t ing under a p ro tec t ive h o o d conta in ing a dry a t m o s p h e r e , a f ind ing which is a lso repor ted by other authors. 8 * 1 2 A detai led invest igat ion invo lv ing f luor ide s lags of c o m p o s i t i o n s C a F 2 - 2 5 % A l 2 0 3 , C a F 2 - 2 5 % C a O , C a F 2 - 2 5 % C a O - 2 5 % A l 2 0 3 , C a F 2 - 3 0 % C a O - 2 0 % S i O 2 and C a F 2 - 3 0 % C a O - 1 0 % S i O 2 - 1 0 % A I 2 O 3 is carr ied out by Nakamura et al . . 3 The distr ibut ion ratio of hydrogen (L^) is indicated to vary f r o m 1.9 to 10.5, the m a x i m u m being in a binary s lag conta in ing C a O and the m i n i m u m in the quarternary s lag having both S i 0 2 and A l 2 0 3 . Th is rat io is not dependent on the hydrogen level of the e lec t rode and the water vapour pressure on top of the s l a g . H o w e v e r , the hydrogen content of the s lag can be e x p r e s s e d as a linear funct ion of the hydrogen content o f e lec t rode and the square root of water vapour pressure (figure 2.3). Many S o v i e t i n v e s t i g a t i o n s 2 ' 8 ' 1 9 - 2 1 invo lve the s lag des igna ted as A N F - 6 (67 .2%CaF 2 -3 .3%CaO-28 .1% A l 2 0 3 - 0 . 8 % S i O 2 ) . Nik i lski i et a l . 1 9 o b s e r v e d that the hydrogen distr ibut ion c o e f f i c i e n t reached an equi l ibr ium value of 6 whereas accord ing to Latash et a l . 2 , it w a s 3.8 and independent o f water vapour p ressure . No s p e c i f i c temperature is g iven in either of these repor ts . O n the other hand, Rebrov et a l . 2 1 indicated that a m a x i m u m hydrogen d is t r ibut ion ratio of 3 is attained after approx imate ly 30 minutes of remel t ing opera t ion , and subsequent ly the d istr ibut ion ratio s h o w e d a gradual d e c r e a s i n g trend with t ime . It is a lso s u s p e c t e d that C a F 2 has an e f fec t in d e c r e a s i n g the so lub i l i t y o f water in s l a g , but no proper reason is c i ted to support this F i g . 22 D is t r ibut ion c o e f f i c i e n t o f h y d r o g e n b e t w e e n s l a g and meta l dur ing E S R as a funct ion o f s lag b a s i c i t y 14 F i g . 2.3 Var ia t ion of water so lubi l i ty in s lags with water vapour part ial p ressure at 1 6 5 0 ° C 2 4 15 in f luence of C a F 2 . There is s o m e uncerta inty regarding the potent ial in f luence o f i ron , manganese and c h r o m i u m oxide on the so lub i l i t y of water in ESR s l a g . The a s s u m p t i o n of C a F 2 as an inert diluent and the o b s e r v a t i o n on the e f fec t of addi t ion of F e O and T i 0 2 on water so lub i l i ty in n o n - f l u o r i d e s l a g 2 2 support this t rend. The addi t ion o f C a O - c l i n k e r (containing s o m e F e 2 0 3 and a trace amount of T i 0 2 ) in C a F 2 - b a s e d s lag reduces the hydrogen content by a fac tor of ten . 2 3 E v i d e n t l y , further fundamenta l s tud ies are needed to es tab l ish this aspect of f luor ide s l a g s . The e f fec t of adding Y and rare earth (spec ia l ly Ce and La) ox ides and f luor ides is quite favourab le as far as the reduct ion of hydrogen in s lag is c o n c e r n e d . 2 0 A d d i n g o n l y rare earth o x i d e s result in a 30% reduct ion of hydrogen in remel ted metal whereas the rep lacement by rare earth f luor ides is not found to be as e f f e c t i v e . The hydrogen content of A N F - 6 s lag is 4 0 - 1 1 5 c m V l O O g m be fore any preheat ing . 1 9 It can be reduced to 2 0 - 8 0 c m 3 / 1 0 0 g m depending on the hea t ing /s to r ing pract ice . Chuiko et a l . ' measured a much lower level of h y d r o g e n (17-19 c m 3 / 1 0 0 g m ) after treat ing the A N F - 6 f lux at 4 0 0 ° C for 4 hours . The p r o c e s s of mel t ing the flux with a graphite e lec t rode further reduces the hydrogrn to 10-15 c m 3 / 1 0 0 g m . In a remel t ing p r o c e s s , a m a x i m u m level o f h y d r o g e n , around 3 0 c m 3 / 1 0 0 g m , is repor ted by other authors 2 * 2 1 . V e r y little in format ion is ava i lab le on h y d r o g e n so lub i l i ty in binary f luor ide s l a g s . Ma in ly C a F 2 - C a 0 3 ' 2 4 and C a F 2 - A I 2 0 y ' 3 ' 5 s y s t e m s have been invest iga ted reveal ing that s lags conta in ing C a O have a greater water capac i ty than those conta in ing A l 2 0 3 . In the C a F 2 - C a O s y s t e m at 1 6 0 0 ° C water so lub i l i t y increases f r o m 71 p p m to 414 p p m as the l ime 16 c o n c e n t r a t i o n changes f r o m 1.6% to 21.6% 2 4 , whereas in 30% A l j 0 3 binary f luor ide s lag it is on ly 79.2 p p m at 1 7 8 0 ° C . The higher so lub i l i t y of water in l ime s lag is expla ined by the p r e s e n c e of free oxide ions and the water v a p o u r - s l a g react ions as g iven in equat ion 2.2. In the case of A l 2 0 3 b inary s l a g s , the c o r r e s p o n d i n g water so lub i l i t y react ion is not as c lear . Rather the binary s l a g m a y t r a n s f o r m into a ternary melt ( C a F 2 - A l 2 0 3 - C a O )f as a result o f the f o l l o w i n g reac t ion at high temperature: 3 C a F 2 + A l 2 0 3 = 2 A I F 3 + 3 C a O 2.4 In such a c a s e , the f inal water so lub i l i ty w o u l d be d ic ta ted by the ternary c o m p o s i t i o n . There is no in format ion ava i lab le on the water so lub i l i t y in C a F 2 . - S i 0 2 b inary . A s far as the opera t ion o f ESR is c o n c e r n e d the proper t ies of this b inary are not at all f a v o u r a b l e . M o r e o v e r , s tudy ing this s y s t e m as a ^strict b inary ' is i m p o s s i b l e because of the o c c u r r e n c e of the f o l l o w i n g react ion at high temperature: 2 C a F 2 + S i 0 2 = 2 C a O +S iF 4 (g ) 2.5 The avai lable water so lub i l i t y data on ternary s y s t e m s are p resented in table 2.1 where it is ev ident that under s imi la r s i tua t ions , C a F 2 - C a O - A l 2 0 3 w o u l d have a higher concent ra t ion of h y d r o g e n than the C a F 2 - C a O - S i 0 2 s y s t e m . In m o s t c a s e s , the water content d o e s not increase b e y o n d 0.1%. H o w e v e r , s o m e reports indicate a much higher va lue , such as in C a O - A l 2 0 3 - C a F 2 s l ags which conta in water o f the order o f 0.2% to 0.5%6 and a s lag o f c o m p o s i t i o n C a F 2 - 2 5 % C a O - 2 5 % A I 2 0 3 with 1600 17 p p m of h y d r o g e n at 1 5 5 0 ° C and 188 torr water vapour p r e s s u r e 3 . 2.1.3 F A C T O R S A F F E C T I N G S O L U B I L I T Y OF H Y D R O G E N IN FLUORIDE S L A G S H y d r o g e n is mainly t ransferred f r o m the mois ture in the a tmosphere to the s lag even though there are s o m e other s o u r c e s such as hydrogen in the e lect rode and the initial h y d r o g e n level of the s lag i tself (whatever pretreatment is done to ESR s l a g , rthere is a lways s o m e residual hydrogen in the start ing slag) . A l l the a u t h o r s 1 - 5 ' 1 2 ' 2 4 ' 2 5 repor ted the square root of water vapour pressure d e p e n d e n c y on the so lubi l i ty of hydrogen in s lag (Sievert 's law). C o n s e q u e n t l y , operat ing the p r o c e s s in a dry a tmosphere 2 * 4 or b lowing argon through the s l a g 1 2 should reduce the level of hydrogen s i g n i f i c a n t l y . D i f ferent pretreatments of s lag such as ca lc in ing it under a neutral a t m o s p h e r e 1 2 having a m i n i m u m content water vapour and other h y d r o g e n o u s c o m p o u n d s ( C H 4 , C 2 H 2 etc.), or premel t ing the s lag 4 * 8 wi th a graphite e lec t rode are r e c o m m e n d e d to obta in a low hydrogen s l a g . The importance of the g a s - s l a g water react ion , as a l ready been m e n t i o n e d , and the locat ion of the reac t ion , i.e. the g a s / s l a g in ter face , b e c o m e further evident when we cons ider the product ion of bigger d iameter ESR ingots . In this s i tuat ion , the larger g a s / s l a g interface a r e a 2 5 w o u l d enable m o r e hydrogen transfer react ion in a shorter t ime. T h e r e f o r e , the remel t ing ra te 5 wi l l l imit the hydrogen concent ra t ion in the s l a g . Due to the e lec t rochemica l nature of ESR r e a c t i o n s , it is s u g g e s t e d that the hydrogen is r e m o v e d f r o m the s lag by e l e c t r o l y s i s . 4 U s i n g the n o n - c o n s u m a b l e e lec t rode and s u p e r i m p o s e d D C s u p p l y , hydrogen 18 can be reduced f r o m the s lag (figure 2.4), but the p r o b l e m of c o r r o s i o n of the e lec t rode by the f luoride s lag is repor ted to be the real hindrance against this idea . ' 5 Fina l ly , the s lag c o m p o s i t i o n 2 0 " 2 5 wh ich d ictates the act iv i ty of ox ide ion is a very important factor in cont ro l l ing the h y d r o g e n level in the s l a g . A n y reduct ion in free o x i d e - i o n concent ra t ion w o u l d result in a lower hydrogen level in the s l a g , but in industrial p rac t ice , the change in the s lag c o m p o s i t i o n has to be ba lanced with decreas ing desul fur iz ing power of the s l a g . P r o b a b l y , the S i 0 2 / S i and A l 2 0 3 / A l react ions are more important , because the most cr i t ical s t e e l s are a lso very r ig id ly c o n t r o l l e d for [Si] and [Al ] . A s lag made ac id ic by high S i 0 2 and A l 2 0 3 leads to e x c e s s i v e p i c k - u p of [Si] and [Al ] . M o r e o v e r , c o m p o s i t i o n changes alter the p h y s i c a l proper t ies of s l a g , such as permeab i l i t y and v i s c o s i t y , w h i c h can a l s o inf luence the hydrogen t ransfer to the s l a g . 2 4 The m o s t direct in f luence , h o w e v e r , is of c o m p o s i t i o n on the furnace s lag operat ing temperature . 2.2. H Y D R O G E N IN N O N - E S R S L A G S Extens ive studies have been carr ied out o n the so lub i l i t y o f water in p e n o l o g i c a l mater ia ls such as m a g m a s and granite invo lv ing a very high pressure of water vapour ( > > 1 a t m ) . 2 6 In recent y e a r s , interest has a lso been s h o w n in water so lub i l i t y in g l a s s e s and meta l lurg ica l s lags where the partial pressure of mois ture is b e l o w a t m o s p h e r i c p ressure . In the present context the latter case is of part icular interest to the present 19 E C L Q_ C cn o i_ TJ >-SZ Q> c D SZ O 2--2-Legend A negotive polority x positive • no superimposed DC -4- T 2 T " 3 -1 current density, A/cm" Fig. 2.4 Effect of superimposed DC on the hydrogen level in ESR process* 20 w o r k . 2.2.1 W A T E R S O L U B I L I T Y IN G L A S S There are r e p o r t s 2 7 ' 1 ava i lab le on water so lub i l i t i es main ly in b inary s i l icate g l a s s e s conta in ing ox ides such as , L i 2 0 , Z n O , N a 2 0 etc . The parameters wh ich are found to be of pr ime interest in cont ro l l ing the so lub i l i t y are c o m p o s i t i o n , temperature and p a h i a l pressure of water vapour . The exact nature of the inf luences is d i f f icu l t to c o n f i r m because of the exper imenta l p r o b l e m s ar is ing out of high mel t ing po in t , high v i s c o s i t y and l o s s of vo la t i le c o n s t i t u e n t s . 3 1 2.2.1.1 W A T E R S O L U B I L I T Y IN B I N A R Y G L A S S S Y S T E M S M o s t repor ts , except the work of T o m l i n s o n 2 9 , s h o w a linear re la t ionship of the water so lub i l i ty wi th the square root of water vapour p r e s s u r e . N a 2 0 . 2 S i 0 2 s y s t e m is s tud ied by this author and it is c o n c l u d e d that p o s s i b l y there are t w o r e a s o n s fo r the apparent dev ia t ion f r o m l ineari ty at higher partial pressures of water vapour . F i rst , the accuracy of the a n a l y s i s is c o m p a r a t i v e l y poor ( ± 1 5 % ) ; and s e c o n d l y , the gas f l o w rate e m p l o y e d during equi l ibrat ion exper iments is substant ia l ly high (1 lpm), w h i c h m a y result in an unsaturated water vapour in the gas s t r e a m . In N a 2 0 - P 2 0 5 and N a 2 0 - 2 B 2 0 3 m e l t s 2 7 , . the integral ext inct ion funct ion of OH bands (a measure of water in quenched s lag ) a lso s h o w s l inearity with the water vapour partial p ressure . In N a 2 0 - S i 0 2 m e l t s , water so lub i l i t y increases with decreas ing tempera tu re . 2 8 " 3 0 Even though the s a m e trend is noted in all three repor ts , the amount of d i s s o l v e d water is found to be c o n s i d e r a b l y lower in the report o f T o m l i n s o n 2 ' (300 ppm as o p p o s e d to 800 p p m around 1 2 0 0 ° C 21 a c c o r d i n g to the other two reports) . The s imi lar a f fec t of temperature is r e c o r d e d in K 2 0 - S i O 2 > C s 2 0 - S i 0 2 mel ts , 2 * * 3 0 whereas the o p p o s i t e trend is o b s e r v e d in L i 2 0 - S i 0 2 , S r O - S i 0 2 , B a O - S i 0 2 s y s t e m s 2 8 (figure 2.5). U y s and K i n g 3 1 d id not see any not iceab le e f f e c t of temperature in Z n O - S i 0 2 s y s t e m s (figure 2.6) . A probable reason for these c o n f l i c t i n g reports may be that the analyt ica l technique d i f fe red in the T v a r i o u s s tud ies . Russel and Kurk j ian 2 8 * 3 0 purged v dry ' oxygen (assumed dry with respect to water s o l u b i l i t y ) through the mel ts to e v o l v e the d i s s o l v e d water . Th is m e t h o d , as R u s s e l 2 8 po ints out, d o e s not g ive the total d i s s o l v e d water in the mel t , but the amount of water r e m o v e d by v d r y ' o x y g e n . T o m l i n s o n 2 9 ana lyzed water in the g lass by e v o l v i n g it in a vacuum at 1 2 0 0 ° C and measur ing the partial pressure m a n o m e t r i c a l l y . S u r p r i s i n g l y , the evo lu t ion takes a c o n s i d e r a b l e t ime (10hrs) at this temperature . In the vacuum f u s i o n techn ique , as reported by U y s and K i n g , 3 1 the e v o l v e d water is reduced to h y d r o g e n by an a luminium fo i l and subsequent ly this gas is ana lyzed in a thermal conduct iv i ty ce l l . The major p r o b l e m of this vacuum technique is that the alkali ox ides are a lso v a p o r i z e d during the evo lu t ion of water and c o n d e n s e d around the c o l d zone o f the apparatus, where the c o n d e n s a t e w o u l d absorb part of the e v o l v e d water . The e f fec t of c o m p o s i t i o n on water so lub i l i t y is a lso not very c o n s i s t e n t accord ing to di f ferent a u t h o r s . 3 0 3 2 For the N a 2 0 - S i 0 2 s y s t e m Kurkjian and R u s s e l 3 0 o b s e r v e d a m i n i m u m water so lub i l i t y around 25 m o l e percent N a 2 0 , (figure 2.7), while S c h o l z e et a l . 3 2 repor ted a decreas ing water so lub i l i t y with the decrease of base in the g l a s s . S imi la r con f l i c t ing trends are a l s o evident in a L i 2 0 - S i 0 2 mel t , wh ich is a l s o s tud ied by U y s and 22 0.30 0 .25-Legend A L i , 0 - 2 S i 0 , x K 0 - 2 S i 0 . 6^ 0.20H Ii "£ o . ^ O O i_ CD "5 £ 0.10-0.05-o.oo-• N o , 0 - 2 S i 0 , B S r 0 - 2 S i 0 , E BaO-_2SjOA x C s i 0 : i 2 S i 0 2 T 6 8 (1/T)x104 °K I 9 Fig. 2.5 Effect of temperature on water solubility in alkali and alkaline earth disilicate glasses" 23 10000 E CL CL O O i_ 1000 Legend A 7 2 . 3 m o l e % L i , 0 + CaO x 1 9 - 2 1 m o l e % L i , 0 & 4 4 - 4 6 m o l e % L i , 0 • 5 7 - 5 9 m o l e % ZnO E 6 5 - 6 6 m o l e % ZnO -x •-I ts-100 5.2 5.4 i i i 5.6 5.8 6 ( l / T ) x 1 0 4 ' 0 K 6.2 6.4 6.6 F i g . 2.6 E f f e c t of temperature on water so lub i l i t y in s i l ica te s lags obta ined at p = 146torr 3 ] HjO 24 0.22 0.06 -f 1 1 1 1 1 10 20 30 40 50 60 Na 2 0, mole% Fig. 2.7 Water solubility as a function of Na,0 content in Na 20-Si0 5 system 3 0 25 K i n g 3 1 (figure 2.8 and 2.9). A decreasing" trend of so lub i l i ty is o b s e r v e d , but the trend is changed around 50 mole percent b a s e . Within the smal l range of c o m p o s i t i o n in Z n O - S i 0 2 and C a O - S i 0 2 , the e f fec t of c o m p o s i t i o n is r e v e r s e d . 2.2.1.2 W A T E R S O L U B I L I T Y IN T E R N A R Y G L A S S S Y S T E M S In S i O j - C a O - ac id oxide mel t 's 3 3 , at a constant molar ratio of C a O / S i 0 2 , the addi t ion of acid o x i d e s , such as P 2 0 5 , B 2 0 3 and G e 0 2 , increases the water so lub i l i t y , the e f fec t be ing a m a x i m u m with P 2 0 5 and a m i n i m u m with G e 0 2 (figure 2.10). The addi t ion of amphoter ic ox ides such as A l 2 0 3 and T i 0 2 s h o w s a m i n i m u m so lub i l i ty behavior at around 20% amphoter ic o x i d e , and this c o m p o s i t i o n p o s s i b l y demarca tes the t ransi t ion f r o m bas ic to ac id behavior (figure 2.11). Iguchi et a l . 3 4 s tudied the e f f e c t of addi t ions of alkali metal ox ides and a lka l ine -ear th metal ox ides on the so lub i l i ty of water in l iquid s i l icate me l ts (f igure 2.12 and 2.13). In all c a s e s , a m i n i m u m so lub i l i t y is o b s e r v e d around the metas i l icate c o m p o s i t i o n . H o w e v e r , the addi t ion of an a l k a l i - m e t a l ox ide such as L i 2 0 , N a 2 0 and K 2 0 increases the so lub i l i ty o f water , the e f f e c t of K 2 0 being" the s t rongest and that of L i 2 0 being the w e a k e s t . On the other hand, the addi t ion of an a l k a l i n e - e a r t h - m e t a l ox ide such as BaO and S r O d e c r e a s e s the water so lubi l i ty except for M g O addi t ions where the e f fec t is r e v e r s e d . The e f fec t of B a O is larger than that of S r O . A l l these ternary mel ts s h o w little e f fec t of temperature on 0.20 Li20, mole% Fig. 2.8 Water solubility as a function of Li 20 content in Li 30-Si0 2 system 3 0 27 10000-C (D "c 1000-o O X • BI Legend L i , 0 - S i 0 , const. CoO-Li,0-SiO CoO-const. Li.O-SiO CoO-SiO, ZnO-_SiO; CoO-SiO * FeO-SiO - -H 100- -T-70 10 -T-20 30 "~r~ 40 ~1~ 50 60 mole% base 80 F i g . 2.9 Water so lub i l i ty as a func t ion o f m o l e percent b a s e in var ious s i l i ca te s l a g s at p = 146 torr H g 3 1 H jO 28 acid oxide, wt% Fig. 2.10 Hydrogen solubility in various silicate slags containing acid oxide at 1500°C, CaO/SiO,=0.59, p = 289 torr 3 3 H jO 48.5 F i g . 2.11 H y d r o g e n so lub i l i t y in var ious s i l icate s l a g s c o n t a i n i n g amphote r ic ox ide at 1 5 0 0 ° C , C a O / S i O , = 0 . 5 9 , p = 2.89 t o r r 3 3 H,0 3500 3000-2500 E CL CL >?• 2000-Q O V) L-6 1500-1000-500-Legend A Coo-SiO,—41.5mole% Li ,0 x CoO-SiO -35mole% Li.O • CoO-SiO, -27.5mole% L i ,0 B CoO-SiO„-17.5mole% L i ,0 0.2 0.4 0.6 0.8 N CaO/ N Si0 2 1.2 Fig. 2.12 Water solubility in the CaO -S iO j -L i , 0 system 3 4 440T 420- Legend A 2.8 mole% SrO 400 - X 5.9 mole% SrO E - • 9.1 mole% SrO B 12.4 m o l e % SrO 280 H 1 1 1 1 J 0.2 0.4 0.6 0.8 1 1.2 N Ca0 / N S10 2 F i g . 2.13 W a t e r so lub i l i t y in the C a O - S i O j - S r O s y s t e m 5 4 32 Water solubi l i ty . ' 1 ' 3 3 ' 3 ' 1 2.2.1.3 M E C H A N I S M S OF W A T E R D I S S O L U T I O N Infrared a b s o r p t i o n spect ra s tudies and the o b s e r v a t i o n of m i n i m a of water so lub i l i ty in these mel ts at least suggest that more than one m e c h a n i s m of water d i s s o l u t i o n operates to dictate the so lub i l i t y l eve l . F incham and R i c h a r d s o n 3 5 s u g g e s t e d that in l iquid p o l y m e r i c ox ides such as s i l ica tes o x y g e n exists in three f o r m s : doub ly b o n d e d 0 ° , s ing ly bonded O", and free oxygen ions 0 2 \ T h e y equi l ibrate with each other , with concent ra t ions depending upon mel t ing temperature , character is t ics of the cat ion and the oxide c o m p o s i t i o n . There are four d i f ferent react ions of water p o s s i b l e with these three o x y g e n s p e c i e s depend ing on the bas ic i ty of oxide melt . The reac t ions are as f o l l o w s 3 3 (the equivalent express ions for s i l icate mel ts are a lso shown) : In the ac id s l a g , I I I - S i - O - S i - + H 2 0(g ) = 2—Si - O H 2.6 I I I 0 ° + H 2 0(g) = 2 0 H ° In the bas ic s l a g , I I I 2 ( -S i - 0-) + H 2 0(g ) = - S i - O - S i - + 2 0 H - 2.7 I I I 2 0 - + H 2 0(g ) = 0 ° + 2 0 H -I I 2 ( - S i - 0 ) + H 2 0 = 2—Si - OH + 0 2 - 2.8 I I 2 0 - + H 2 0 (g ) = 2 0 H ° + 0 2 ' 33 In the highly bas ic s lag , 0 2 + H 2 0 ( g ) = 2 0 H -One interesting feature of all the react ions is that the water so lub i l i t y is proport ional to the square root of p in all the c a s e s . T h u s , the e v i d e n c e of this re lat ionship d o e s not ref lect the type of m e c h a n i s m prevalent in di f ferent c o n d i t i o n s . f R u s s e l 2 8 c o n s i d e r e d the f i e ld strength of the cat ion to explain the temperature coe f f ic ien t of so lub i l i t y in d is i l icate g l a s s e s . The so lub i l i t y d e c r e a s e s with increasing temperature f o r ca t ions with low f ie ld strength whereas the o p p o s i t e is true with ca t ions of high f ie ld st rength. I o n - o x y g e n attract ion (I) is another parameter c o n s i d e r e d by U y s and K i n g 3 1 to explain the e f fec t of c o m p o s i t i o n and the general so lub i l i t y behav ior . This parameter I, can be de f ined as : I _ i 2 . 2 - 9 a 2 Where Z is the cat ion va lency and a is the internuclear d is tance be tween the ca t ion and oxygen in the appropr iate c o - o r d i n a t i o n . In bas ic m e l t s , ca t ions wi th smal ler i o n - o x y g e n at t ract ions d i s s o l v e more water and in ac id s y s t e m , larger i o n - o x y g e n attract ion s h o w s higher water so lub i l i t y . The i o n - o x y g e n attraction for amphoter ic ox ides lies be tween these two ex t remes (table 2.2). On this b a s i s , water i tsel f can be c o n s i d e r e d as an amphote r i c ox ide . Thus it w o u l d act as an ac id in a b a s i c melt and as 34 T a b l e 2.2 Ion-Oxygen A t t r a c t i o n i n v a r i o u s Oxides Oxides Ion-Oxygen A t t r a c t i o n ( I ) P 2 ° 5 3.31 s i o 2 2.44 B 2 ° 3 2.34 Ge0 2 2.14 T i 0 2 1.85 A 1 2 ° 3 1. 66 F e 2 ° 3 1.50 H.O 1.05 MgO 0.95 CoO 0.89 ZnO 0.88 FeO 0.87 MnO 0.83 CaO 0.70 L i 2 0 0.50 Na 20 0.36 K 20 0.27 35 base in an ac id melt . 2.2.2 W A T E R S O L U B I L I T Y IN M E T A L L U R G I C A L S L A G S The type of s lags inc luded in this c a t e g o r y are main ly l i m e - s i l i c a t e conta in ing var ious amounts of T e O ' , A l j 0 3 , M g O and M n O which are o b s e r v e d in the blast furnace and at d i f ferent s tages of s t e e l m a k i n g . The water solubi l i ty invest iga t ions include both industrial s lags and synthet ic s l a g s , which are made f r o m pure oxide ingredients . A l l s tudies s h o w a proport ional re lat ionship of water so lub i l i ty wi th the square root of water vapour pressure . 2.2.2.1 B I N A R Y S L A G S M a n y r e p o r t s 9 ' 3 3 ' 3 6 ' 3 7 are avai lable on the water vapour d i s s o l u t i o n in C a O - S i 0 3 s y s t e m s around 1 5 0 0 - 1 6 0 0 ° C . Wi th the increase of b a s i c i t y , the so lub i l i ty increases s l ight ly (650 to 750ppm) (figure 2.14) except in the c a s e studied by Fukushima et a l . 3 3 , where a m i n i m u m is o b s e r v e d around the metas i l icate c o m p o s i t i o n . Fukushima et a l . 3 3 a l s o repor ted the increase of so lub i l i ty with increas ing temperature , but the e f fec t w a s not great (figure 2.15). In manganese s i l icate m e l t s , as repor ted by W a l s h et a l . 9 , the hydrogen so lub i l i t y is s l ight ly lower c o m p a r e d to that of l ime s i l i ca tes on a molar b a s i s . A t 64 mo le percent bas ic o x i d e , the h y d r o g e n so lub i l i ty in manganese s i l ica te is 65 p p m as against 75 p p m for the l ime s i l icate when the water vapour pressure is 1 a tm. A further reduct ion in hydrogen so lub i l i ty is o b s e r v e d in the T e O ' - S i 0 2 b inary 3 1 * 3 8 (f igure 2.16). U y s and K i n g 3 1 repor ted an a lmost constant h y d r o g e n so lub i l i ty up to the or thos i l ica te c o m p o s i t i o n in the "FeO' rich s ide and then the so lub i l i ty is 0 .90 -r 0 . 8 5 0 . 8 0 E >: 0 .75 w 0 .70 ^ 0 . 6 5 -0 .60 Legend A Wolsh el ol,(9) 1500°C x Sochdev et al.(36) 1550°C • Iguchi et ol,(34) 1600 0C B Coutures and Peraudeau(49) 0 . 5 5 -0 .3 0.4 0.5 0.6 N S i 0 2 0.7 0.8 F i g . 2.14 C o m p a r i s o n of d i f ferent water so lub i l i ty data in C a O - S i 0 2 s y s t e m 37 4.31-t— 1 1 r 5.2 5.3 5.4 5.5 5.6 5.7 (l/T)X10* °K F i g . 2.15 E f f e c t of temperature on h y d r o g e n so lub i l i ty in 63 S i 0 5 - 3 7 C a O s l a g " . 38 80 Fig. 2.16 Water solubility in "Feo'-SiOj system 3 1 39 inc reased very sharp ly ; but in the same zone Wahls ter and R e i c h e l 3 8 o b s e r v e d a cont inuous and gradual increase of so lub i l i t y . In s i l icate s y s t e m s the trend of decreas ing solubi l i ty in C a O - S i 0 2 , M n O - S i 0 2 and T e O ' - S i 0 2 is very s imi lar to that of the i o n - o x y g e n attract ion parameter , (I) (table 2.2), and I is thus a helpful indicator o f water so lub i l i ty . In binary a luminates , C a O - A l 3 0 3 3 9 s h o w s a c o n s i d e r a b l y higher so lub i l i ty than that of S i 0 2 - A l 2 0 3 . 4 0 A t 35 m o l e percent A l 2 0 3 the water so lub i l i ty of ca lc ium aluminate is 1200 ppm as c o m p a r e d to around 500 p p m for the s i l i ca -a lumina te s y s t e m . A g a i n so lub i l i t y increases with l ime in C a O - A l 2 0 3 , but in S i 0 2 - A l 2 0 3 it s h o w s an initial increasing trend with a lumina and later reaches a m i n i m u m around 90% A l 2 0 3 . Th is unusual trend is explained in te rms of A l / S i o x y b r i d g e s and a p h e n o m e n o n re lated to sur face act iv i ty of S i 0 2 in the S i 0 2 - A l 2 0 3 m e l t s . In the F e O - C a O 3 8 s y s t e m , the water so lub i l i ty is s l ight ly higher than F e O - S i 0 2 m e l t s , but the so lub i l i ty s h o w s a s imi lar re lat ionship wi th c o m p o s i t i o n . 2.2.2.2 T E R N A R Y S L A G S Extensive studies have been done on water so lub i l i ty in the C a O - S i 0 2 - A l 2 0 3 s y s t e m . 9 , 2 r 3 3 ' 3 6 ' 4 1 E v e n though there is a dispute over the abso lu te quantity of water d i s s o l v e d (in 40 C a O - 4 0 S i O 2 - 2 0 A l 2 0 3 s lag 720ppm H 2 0 ' against 5 4 0 p p m 4 1 at 1 4 0 0 ° C ) , the increase of bas ic i ty ( C a O / S i 0 2 ) s h o w s an increase of so lub i l i ty as repor ted by all the authors (f igure 2.17). S a c h d e v et a l . 3 6 and Fukushima et a l . 3 3 a lso o b s e r v e d a m i n i m u m solub i l i ty around 20 wt% A l 2 0 3 at a constant b a s i c i t y . The e f fec t o f temperature is found to be insigni f icant in this s y s t e m 1 " and the 40 Si02 Al 2 0 3(wt%) Fig. 2.17 water solubility (ppm H20) in CaO -Si0 2- Al 20 3 system at 1550°C and p =289 torr" HjO 41 operat ing react ions with water vapour , a c c o r d i n g to Dav ies and S p a s s o v 4 1 are: I I I 4 ( - S i - 0 ) + H 2 0 = 2 ( - S i - 0" H - O - S i ) - + 0 J - 2.10 I I I I I I 2 ( - S i - 0 ) + H 2 0 = ( - S i - O . . . . H - 0 - S i - ) + OH" 2.11 I I I S a c h d e v et a l . 3 6 cons idered a new structural parameter , o x y g e n d e n s i t y , in v i e w of the fact that these s lags conta in more than 90 v o l . pet of o x y g e n ions . The o x y g e n densi ty is de f ined as : p L No O x y g e n densi ty = — — — 2.12 M where p = dens i ty of the melt in g m / c . c . L = A v o g a d r o ' s number No = m o l e f ract ion of oxygen a t o m s in the melt M = average molecu lar weight of the melt ca lcula ted f r o m the molar concentra t ion of each melt c o m p o n e n t . The water so lub i l i ty l inearly decreases with the increase o f o x y g e n dens i ty (figure 2.18) C o n s i d e r i n g the structural aspec ts o f a l u m i n o s i i i c a t e s , the s izes of the a luminum ion ( r=0 .5A) and the s i l i c o n ion ( r=0.41A) and their charge leve ls are c l o s e l y re lated. Consequent ly like a SiO 4 , - te t rahedron, A l 3 + ion has a lso been s u g g e s t e d 3 6 to f o r m AIOJ- as wel l as A I O I - structures in l iquid s lags (even though s imp le r ionic c o m p l e x e s such as A l O j and AIO 3 ^ are p lausib le at high temperatures) . When a s m a l l amount o f A l 2 0 3 is present in C a O - S i 0 2 - A l 2 0 3 s l a g , a luminium ions exist on ly in tetrahedral c o o r d i n a t i o n . Thus s imi lar to s i l i con i o n s , A l 3 + acts like a network 42 60 55-E CL D_ 50 "c o u c: cu cn 45 O JZ 40-• 35 r 390 Legend A 5wi% A l 2 0 3 x 10wt% Al 2 0, • 20wt% Al 2 0, v • \ A > X \ \ X " V A • V 395 400 405 410 oxygen density, x10 415 -20 420 425 Fig. 2.18 Relationship between oxygen density and water solubility in CaO - S i O , - AljOj system at 1500°C and p =T90 torr" H 2 0 43 connec to r and its further addit ion reduces the water so lub i l i t y . A b o v e 20 wt .pct . A l , 0 3 , 3 6 this tetrahedral bonding b e c o m e s unstable caus ing a change to octahedra l b o n d i n g , AIO 9 , - . A s a resul t , a propor t iona l amount of c a l c i u m ions is re leased f r o m the structure mak ing more free o x y g e n ions ava i lab le for water so lub i l i ty reac t ion . , Thus the so lub i l i t y of water is increased at this s tage . A s imi lar c o n c l u s i o n can be drawn by c o n s i d e r i n g i o n - o x y g e n attract ion (I) when the coord ina t ion number for the a luminum ion changes f r o m 4 to 6 result ing in a weaker at tract ion of A l 3 * i on . The reports on water so lub i l i t y in C a O - S i 0 2 - T e O ' s y s t e m are not at all c o n s i s t e n t , p o s s i b l y because of the d i f f i cu l ty in de termin ing the f o r m of T e O ' present in the s l a g . Iguchi and F u w a 4 3 repor ted a tendency of decreas ing water so lub i l i t y with the addi t ion of T e O ' in C a O - S i 0 2 s l a g , whereas Wahls ter and R e i c h e l 3 8 o b s e r v e d an initial increase f o l l o w e d by a decrease upto 20 m o l e percent F e O , where the so lub i l i ty starts r is ing again upto 60 m o l e percent F e O , and b e y o n d that c o m p o s i t i o n the so lub i l i ty decreases (figure 2.19). Iguchi, et a l . 2 2 repor ted s o m e w h a t s imi lar behavior of T e O ' with regard to the water so lub i l i t y (figure 2.20). It w a s s u s p e c t e d that the increase of T e O ' act iv i ty resul ts in partial ox idat ion of T e O ' to F e 2 0 3 wh ich d e c r e a s e s the f inal water so lub i l i t y . A n increase of temperature is f o u n d to increase the so lub i l i ty of water in these s l a g s . 4 2 C a O - S i O j - M g O , another important ternary s y s t e m , s h o w s s o m e apparent con f l i c t on the e f fec t of M g O on the water so lub i l i t y . Iguchi and F u w a 4 3 reported an increase of the water so lub i l i t y with increasing M g O a d d i t i o n , but Zuliani et a l . 1 8 w h o s tud ied the s a m e s y s t e m by a 80-1 OH 1 1 — — 1 1 r — 1 0 20 40 60 80 100 120 'FeO', mole% Fig. 2.19 Water solubility in CaO -FeO-Si0 2 system at 1600°C, CaO /SiO, = 1 and p. _ = 760 torr" H,0 45 Si0 2 FeO<wt%) Fig. 2.20 Water solubility (ppm HjO) in CaO -Si0 2-FeO system obtained at 1550°C, p u =289 torr" HjO 46 thermograv imet r ic m e t h o d did not f ind any e f fec t of M g O , when it is rep laced by C a O on a molar b a s i s , al though both the reports agreed upon a m i n i m u m solub i l i ty around unit bas ic i ty which is d e f i n e d as ^QQQ + N ^ Q ) / N g j Q (figure 2.21). S o s i n s k y et a l . 4 ' f ound the inadequacy of this empi r i ca l l y de f ined bas ic i ty index, and instead they c o n s i d e r e d the act iv i ty of s i l i ca to r e s o l v e the a b o v e ambigui ty (figure 2.22). T h e y a lso obta ined an empir ica l re la t ionship to est imate water;, so lub i l i t y in te rms of the act iv i ty of s i l i c a , and it w a s g iven b y : (H 2 0)ppm = (1095-2180O. + 3 1 4 6 a 2 - 1 3 6 5 a 3 )p° u 5 2.13 o i U 2 oiUj oiUj n 2U Zuliani et a l . 1 8 ' reported the independency of temperature f r o m the water vapour so lub i l i ty over the range of 1475 to 1 5 7 5 ° C , whi le Iguchi and F u w a 4 3 ob ta ined , d i f fe rent , though not very s ign i f icant e f f e c t s of temperature in the ac id and bas ic ranges of s l a g s . In the ac id range, the so lub i l i ty of water d e c r e a s e s with increasing temperature , whi le the o p p o s i t e is true in the bas ic range. S o m e resul ts on s lags used for d i f ferent industrial p r o c e s s e s are avai lab le . 9 " 1 1 ' 4 5 ' 4 6 Water d i s s o l v e d in blast furnace s lags is c l o s e to an equi l ibr ium l e v e l 4 6 and it ranges f r o m 11 to 48 c .c . / IOOgm depending on the locat ion in the furnace f r o m which the s lag is s a m p l e d . The water so lub i l i ty in these s lags is a lso found to vary l inearly with b a s i c i t y , which is d e f i n e d as the ratio of ( C a O + M g O ) to ( S i 0 2 + A l 2 0 3 ) and is studied in the range of 0.9 to 1.0. In ac id o p e n hearth s l a g s 9 , the water content is c o n s i d e r a b l y lower than that in the s lags of other steel making p r a c t i c e s , and the water content d o e s not reach an equi l ibr ium level wi th the furnace a t m o s p h e r e . 700 650 600 E CL 2 550 o >- 500 X! O w 450 Legend A 25wt% MgO x 20wt% MgO • 15wi% MgO 400 350 0.4 0.6 0.8 1.2 1.4 1.6 basicity, (NMgO + N C a O ) / S i 0 ; 1.8 F i g . 2.21 Water s o l u b i l i t y as a funct ion o f s l a g b a s i c i t y and m a g n e s i a content at 1 5 5 0 ° C and p , =289 to r r 4 3 H 2 0 1100 1000 900-800-700- \ A E E Legend 5wt% MgO 10wt% MgO 15wt% MgO 20wt% MgO 25wt% MgO tr—. -A I I I I / / / / / 600--0 T 0.2 0.4 0.6 1 _ a Si02 0.8 Fig. 2.22 Water solubility as a function of silica activity in the CaO -MgO-SiOj slags at p =760 torr 4 4 H jO 49 Th is is , p r e s u m a b l y , due to the high v i s c o s i t y of s i l ica te s l a g s . The water level in these s lags is repor ted to be o n l y around 115 p p m , about one third lower than the basic o p e n hearth s l a g s 9 " 4 5 wh ich s h o w a l m o s t the equi l ibr ium value towards the end of a heat. E lect r ic furnace s l a g s 9 " 1 3 conta in about 200ppm water during the ox id iz ing per iod and are a l s o c l o s e to equi l ibr ium with the furnace a t m o s p h e r e . In s o m e c a s e s , 9 the water level i s repor ted to decrease during the heat. T h e s e m u l t i c o m p o n e n t c o m p l e x industrial s l ags are d i f f icu l t to s tudy in order to extract the inf luence of d i f ferent var iab les such as c o m p o s i t i o n . Thus s o m e i n v e s t i g a t i o n s 4 1 ' 4 6 i n v o l v e d m u l t i c o m p o n e n t synthet ic s lags such as C a O - S i 0 2 - A l 2 0 3 - M n O 4 1 and C a O - S i 0 2 - A l 2 0 3 - M g O 4 6 . Dav ies and S p a s s o v 4 1 obta ined a smal l but s ign i f icant e f f e c t of rep lac ing C a O by M n O in the above quarternary s y s t e m (table 2.3). A t a constant M g O , the other quarternary s lag has s h o w n an a l m o s t s imi la r e f fec t wi th the increase of a lumina and the subsequent decrease o f l ime. 2.3. H Y D R O G E N IN ESR I N G O T S The acceptab le level of hydrogen for s tee l ingots dest ined for fo rg ings with a re lat ive ly smal l fo rg ing reduct ion is around 2 p p m . B e y o n d this c o n c e n t r a t i o n , wh ich is not that u n c o m m o n in ESR ingots , e s p e c i a l l y with a large c r o s s - s e c t i o n , it is p o s s i b l e to produce smal l hairl ine c r a c k s , a lso k n o w n as hydrogen f l a k e s . The exact m e c h a n i c s of this crack f o r m a t i o n and propagat ion are debatab le , al though the cr i t ical role of h y d r o g e n is very wel l r e c o g n i z e d . 50 T a b l e 2.3 Water S o l u b i l i t y i n D i f f e r e n t S i l i c a t e S l a q s S l a g Composition, wt% Temperature Water Content CaO s i ° 2 A 1 2 ° 3 M n 0 ( D e9"' c ) S l a g 40 40 20 0 / 1400 0.536 L i q u i d u s Temp.: 1350 1450 0.536 1500 0.539 1550 0.546 1600 0.417 30 50 20 0 1450 0.495 L i g u i d u s Temp.: 1350 1500 0.498 1550 0.502 1600 0.500 20 60 20 0 1450 0.476 L i q u i d u s Temp.: 1390 1500 0.473 1550 0.465 1600 0.472 35 40 20 5 1450 0.532 30 40 20 10 1450 0.521 20 40 20 20 1450 0.504 0 38 0 62 1350 0. 379 51 H y d r o g e n , having a smal l eovalent radius (0.32A), can d i s s o l v e as an interstit ial so l id so lu t ion in s tee ls and other m e t a l s . Its so lub i l i ty increases with temperature and attains a m a x i m u m in the austeni te phase because of its F C C structure. A l s o , the so lub i l i t y var ies l inear ly with the square root of the pressure of H 2 , s u g g e s t i n g that h y d r o g e n d i s s o l v e s a t o m i c a l l y . Hydrogen tr ies to e f fuse when -fit exceeds its so lub i l i t y limit during rapid c o o l i n g of an ingot. In the case of large s e c t i o n s , the d i f f u s i o n distance being c o n s i d e r a b l e , h y d r o g e n can be t rapped in defect s i tes such as v a c a n c i e s , d i s l o c a t i o n s , grain boundar ies and i n c l u s i o n - m a t r i x i n t e r f a c e s 4 1 . The fo rmat ion 6f hydrogen m o l e c u l e s at those s i tes w o u l d give r ise to a very high pressure w h i c h , a long with the thermal and t rans fo rmat ion s t r e s s e s , may result in rupture which appears as hydrogen f l a k e s . The molecular d i f f u s i o n of hydrogen be ing very s l o w (d i f fus iv i ty in iron at 1 0 0 ° C is 5x10" 5 c m V s e c ) and the d i s s o c i a t i o n reac t ion , H 2 — > 2 H , being highly endothermic(103.3 K c a l / m o l ) , it is ex t remely d i f f icu l t to r e m o v e this h y d r o g e n f r o m the t rapped s i t e s . The s l o w c o o l i n g of ingots a lso d o e s not help the escape of hydrogen s ince the hydrogen s h o w s a tendency to segregate at the center . The isothermal heating at higher temperature ( > 2 0 0 ° C ) is found to be a better m e t h o d for hydrogen r e m o v a l , even though it takes a great length of t ime . Thus most of the prev ious invest iga t ions I ' ' T , n , " ' 4 ! ' 4 9 in this area have f o c u s s e d on understanding the react ion m e c h a n i s m and the fac tors 52 cont ro l l ing the f inal level of hydrogen in ESR ingots . 2.3.1 R E A C T I O N S OF H Y D R O G E N T R A N S F E R In v i e w of the fact that the re f ined l iquid metal in the ESR p r o c e s s is in cont inuous contact with the l iquid s l a g , many a u t h o r s 1 2 , 2 5 , 4 9 suggest that the f inal t ransfer of h y d r o g e n to the metal takes p lace by the d i s s o c i a t i o n react ion of hydroxyl ion f r o m the s l a g , and this react ion can be writ ten a s : 2(OH-) + (Fe 2*) = [Fe] + 2[0] + 2[H] 2.14 M o r e o v e r , Medovar et a l . 1 2 c o n s i d e r e d that the e l e c t r o c h e m i c a l d ischarge of hydroxy l ions at the anode c a u s e d the s imu l taneous increase of hydrogen and oxygen in the ingot, (OH-) = [H] + [O] + e- 2.15 Holzgruber et al .* p r o p o s e d a di f ferent h y d r o g e n t ransfer s c h e m e f r o m the s lag to the m e t a l , which can be wri t ten a s : ( O H ) + e- = ( O 2 ) + [H] 2.16 In case of the p resence of protons (H +) in the s l a g , the f o l l o w i n g ca thod ic react ion is a l s o c i t e d 1 2 : (H +) + e- = [H] 2.17 H o w e v e r , c o n s i d e r i n g the e lec t rochemica l reac t ions in E S R , the a b o v e react ion is highly unl ikely s ince the over r id ing e l e c t r o c h e m i c a l react ion in this s y s t e m w o u l d be : 53 Fe = F e 2 * + 2e~ 2.18 The react ion (2.18) w o u l d not on ly cont ro l the o x y g e n potent ia l in E S R , but a lso limit the other react ions having higher potent ia ls . A l s o the p r e d o m i n a n c e of ( O H ) s p e c i e s in these s l a g s , as o b s e r v e d in infrared s t u d i e s , w o u l d make the last react ion (2.17) least important for further c o n s i d e r a t i o n . T" F r o m all the above reac t ions , it is very evident that in the case of a t h e r m o d y n a m i c equi l ibr ium that a constant d is t r ibut ion ratio of hydrogen in the s lag and metal w o u l d be reached . In a pi lot plant study by Schurmann et a l . 1 , the d ist r ibut ion c o e f f i c i e n t , (H)/[H], between the C a F j - C a O - A I 2 0 3 s lag and the metal is invest igated (figure 2.23). A f t e r about 30minutes , this c o e f f i c i e n t , wh ich is a funct ion of the basic i ty , o f the s l a g , reaches a constant va lue . S imi la r o b s e r v a t i o n s are a l s o reported by other a u t h o r s 2 , 1 9 , 2 1 whi le s tudy ing other s l a g - m e t a l s y s t e m s . The initial hydrogen level in the m e t a l , as o b s e r v e d at the b o t t o m of an ingot , is normal ly higher than the equi l ibr ium level which is reached after a certain t ime of o p e r a t i o n 2 . If we c o n s i d e r a s imple r f o r m of h y d r o g e n t ransfer react ion such as : 2 ( O H ) = ( O 2 ) + [O] + 2[H] 2.19 and the p resence of lower o x y g e n potential at the initial s tage of reme l t ing , the a b o v e react ion w o u l d shift towards the right g iv ing r ise to a higher h y d r o g e n level at the b o t t o m of the ingot . B a g s h a w 4 8 s i m p l i f i e d the hydrogen transfer p r o c e s s , c o n s i d e r i n g the overa l l g a s - s l a g - m e t a l equi l ibr ium and the o b v i o u s f o l l o w i n g reac t ion ; 54 70 Legend A Hs \ x Hm_ \ D Hs/Hm \ / X / x . • • - — • " X — - X — 60 50 E CL QL V) I -40 2-i 10 —1 1 1— 20 30 40 time, min. —T" 50 -30 --20 60 F i g . 2.23 H y d r o g e n equi l ibr ium be tween s l a g and meta l in an ESR o p e r a t i o n , s l a g is 40 C a F , - 3 0 A l , 0 3 - 3 0 C a O , T = 1 7 8 0 ° C , p =7 to r r 1 HjO 55 H,0(g) = 2[H] + [0] 2.20 The o x y g e n potential in the s y s t e m which d i rec t ly c o n t r o l s the o x y g e n level in the ingot a lso in f luences the hydrogen level in the m e t a l . For the a b o v e equi l ib r ium, at a constant temperature , increas ing the o x y g e n potent ia l w o u l d reduce the hydrogen content in the meta l . H o w e v e r , this w o u l d happen at the expense of lower desul fur iz ing p o w e r and higher ox ida t ion l o s s e s of the a l loy ing e l e m e n t s . The consequent dependence of hydrogen in the metal on the water vapour pressure (figure 2.24) has been o b s e r v e d in d i f ferent s tud ies , 2 * 3 ' 4 8 but by mere ly looking at this equ i l ib r ium, the vital capabi l i ty o f s lag in cont ro l l ing hydrogen transfer can not be e s t i m a t e d . The es t imat ion of equi l ibr ium hydrogen so lub i I ity(%) can be m a d e by the f o l l o w i n g e x p r e s s i o n " : where = the equi l ibr ium constant of so lub i l i ty in i ron, e' = the interact ion parameter b e t w e e n hydrogen and the e lement i H [i] = the concent ra t ion of the e lement i, %. S u r p r i s i n g l y , in the a b o v e fo rmu la , the partial pressure of h y d r o g e n , p , is H 2 c o n s i d e r e d instead of a more relevant term i.e. water vapour p ressure , p H 20' The temperature dependence o f the equi l ibr ium constan t , K^, teract ion parameter , e' , are g i v e n b y 1 9 , H = -1730/T - 1.688 2.22 t "fThe e x p r e s s i o n shou ld b e , log = -1730/T - 1 688 56 hydrogen content, ppm Fig. 2.24 Hydrogen solubility in ESR ingots as a function of water vapour pressure in the atmosphere4' 57 V a t o l i n et a l . 1 5 ana lyzed the hydrogen t ransfer reac t ion using the avai lable t h e r m o d y n a m i c data to obta in an e x p r e s s i o n of the h y d r o g e n d is t r ibut ion ratio between the s lag and the metal as a func t ion of temperature and the concent ra t ion of F e j 4 in the s l a g . Un for tuna te ly , the e x p r e s s i o n pred ic ted much lower va lues than those o b s e r v e d in p rac t ice , sugges t ing s o m e f l a w s in the theoret ical a n a l y s i s . 2.3.2 F A C T O R S C O N T R O L L I N G H Y D R O G E N IN T H E INGOT S i n c e the main sources o f hydrogen in ESR are the mois ture in the s lag and in the a tmosphere on top of the s l a g , reducing these t w o inputs w o u l d reduce the hydrogen in the remel ted meta l . T h u s , all the f a c t o r s l imi t ing the hydrogen so lub i l i ty in s l a g , as d e s c r i b e d p r e v i o u s l y , w o u l d in turn contro l the hydrogen level in the meta l . M o r e o v e r , Masui et a l . " f o u n d that the g a s / s l a g inter face area , e lec t rode h y d r o g e n content , remel t ing rate and s lag quanti ty are a l s o cont ro l l ing f a c t o r s with regard to the hydrogen level in the ingot . The s imi lar in f luence of remel t ing rate and m e t a l / s l a g interface area is not ev ident in the work of Holzgruber et a l . . 6 T h e y o b s e r v e d that the h y d r o g e n at the b o t t o m of the ingot is cont ro l led by the mois ture of the flux and at the top ma in ly by the hydrogen in the e lect rode and water vapour pressure in the a t m o s p h e r e . Th is is p r o b a b l y because in their s y s t e m the initial H 2 0 in the flux is very high. The argon b low ing through the s lag and liquid meta l he lped M e d o v a r el a l . " to reduce the final h y d r o g e n content by 2-2 .5 t i m e s . In a regular p r a c t i c e , such a step w o u l d be t o o dangerous to imp lement . 58 P o c k l i n g t o n 7 did not get any benef i t us ing direct current as the mel t ing s o u r c e , whereas Holzgruber et a l . 6 f ound the use of s u p e r i m p o s e d D C c o m p o n e n t to be the m o s t e f f e c t i v e m e a n s of reducing the h y d r o g e n level in a laboratory sca le ESR unit. In A C remel t ing p rac t i ce , P o c k l i n g t o n 7 s t r e s s e d cont ro l l ing the iron oxide leve ls in the s lag for o p t i m u m concent ra t ion of hydrogen in the ingot . Both the s t e p s , m e n t i o n e d a b o v e , alter the oxygen potent ial of the s y s t e m , t 'which in turn, as d i s c u s s e d b e f o r e , inf luence the f inal hydrogen leve l . 2.4. T E C H N I Q U E S OF M E A S U R E M E N T OF H Y D R O G E N IN S L A G 2.4.1 S A M P L I N G OF S L A G The p r o b l e m of hydrogen determinat ion in s lag begins at the s a m p l i n g s tage . Prev ious workers used m o s t l y a c o p p e r 3 or tungsten 2 * 2 4 bar to co l lec t representat ive s lag s a m p l e s f r o m an exper iment . The ESR s lags be ing highly h y g r o s c o p i c are prone to contaminat ion whi le e x p o s e d in the air during handl ing. A l s o , due to the conduc t ing nature of these s lags and the harsh envi ronment around the s a m p l i n g area in an ESR unit, the e lectr ica l power must be d i s c o n n e c t e d during s a m p l i n g . Th is pract ice in t roduces ripple marks on the sur face of a s o l i d i f i e d ingot which is o therwise remarkably s m o o t h . A n o t h e r e f fec t is the change in thermal reg ime of the mol ten s lag poo l which changes the important phys ica l and c h e m i c a l character is t ics of the s l a g . C o n s i d e r i n g all these a s p e c t s , s lag s a m p l i n g us ing a quartz tube is e n v i s a g e d , 3 but th is has not p r e v i o u s l y been used due to the 59 s u s p i c i o n that the s i l i ca might d i s s o l v e into the s l a g . 2.4.2 M E T H O D S OF A N A L Y S I S The accurate determinat ion of h y d r o g e n in ESR s lag is a major o b s t a c l e to any study of the behavior of h y d r o g e n in this s y s t e m . The s i m p l e weight l o s s method of e v o l v i n g water f r o m the s lag by heating around 5 0 0 - 6 0 0 ° C is not suf f ic ient to r e m o v e all the hydrogen f r o m the s l a g . S o m e chlor ide sal ts are found to retain water even at 1 0 0 0 ° C 5 0 . A l s o f l u o r i d e - s l a g s hydro lyze at high temperature and generate HF g a s , thus mak ing the weight loss ana lys is more c o m p l i c a t e d and uncerta in . In the ESR s l a g , the h y d r o g e n , present as hydroxy l ion , is c h e m i c a l l y b o n d e d to the structure and s o is present as a stable s o l u t i o n at the high temperature . On the other hand, hydrogen f o r m s a p h y s i c a l s o l i d so lu t ion with s t e e l ; the e f f e c t i v e s p e c i e s , be ing the hydrogen a t o m , 1 1 can d i f f u s e out of steel eas i l y when heated in a vacuum s y s t e m . T h u s , for the h y d r o g e n a n a l y s i s of s l a g , it is n e c e s s a r y to melt the s lag in a v a c u u m to extract the d i s s o l v e d hydrogen either in the f o r m of molecu lar hydrogen or water vapour which shou ld be ana lyzed immedia te ly in a sens i t i ve apparatus. Var ious techniques have been d e v e l o p e d to improve the c o n f i d e n c e and accuracy of measurement . In the l i terature, most of the m e t h o d s can be c l a s s i f i e d into two c a t e g o r i e s ; vacuum extract ion and extract ion into a stream of carrier g a s . 5 1 A g a i n , a m o n g the v a c u u m ' extract ion m e t h o d s , there are two g r o u p s : vacuum h e a t i n g 5 2 and vacuum f u s i o n of s l a g . 3 ' 5 3 A m a s s spec t romet r ic s t u d y 5 3 revea ls that the h y d r o g e n e v o l v e d f r o m s lag exists both as H 2 0 and molecu la r h y d r o g e n . T h e r e f o r e , 6 0 for quanti tat ive measurement , all of it has to be either c o n v e r t e d to H 2 with the help of reducing agents such as a l u m i n u m , f e r r o m a n g a n e s e , ca lc ium carbide or o x i d i z e d to water vapour by s o m e oxidant , such as C u O . In determin ing hydrogen in f luor ide s lags by the carrier gas m e t h o d , i n v e s t i g a t o r s 1 3 ' 5 1 de tected the p resence of S i F , and HF in the gas s t ream e v o l v i n g out of the s lag at 1 2 0 0 ° C because o f the f o l l o w i n g r e a c t i o n s : C a F 2 + H 2 0 = C a O + 2HF 2.24 4HF + S i 0 2 = S i F , + 2 H 2 0 2.25 The in ter ferences of HF and S i F 4 in the f inal a d s o r p t i o n tower are taken care of by pass ing the gas through a c a l c i u m oxide tube, kept at 5 0 0 ° C , and these gases result in the f o l l o w i n g c o n v e r s i o n s ; S i F , + 2 C a O = S i 0 2 + 2 C a F 2 2.26 2HF + C a O = H 2 0 + C a F 2 2.27 T h u s , the total hydrogen is de te rmined grav imet r ica l ly in the f o r m of H 2 0 by adsorb ing in a m a g n e s i u m perchlorate tower (figure 2.25). The standard state d e c o m p o s i t i o n temperature of ca lc ium hydrox ide is ' 5 8 0 ° C . 5 4 T h e r e f o r e , keeping the ca lc ium ox ide tube at 5 0 0 ° C w o u l d g ive r ise to s o m e amount of mois ture being t rapped in the ca lc ium o x i d e ; and a l s o , the c o l d parts of tub ing , be fore the a d s o r p t i o n t o w e r , w o u l d a d s o r b s o m e mois ture . T h e s e m a y be the r e a s o n s fo r gett ing a low value of h y d r o g e n in ESR s lags as ana lyzed by S c h m i d t . 5 1 61 Fiq 2 25 Schemat ic d iagram of the apparatus for determin ing hydrogen in ESR s l a g s - . LCarr ie r gas (0 2 ) 2. f l o w m e t e r 3. H 2 S 0 4 4. furnace 5. s o d a l i m e + 1 ^ 9 ( 0 1 0 ^ 6. Py thagoras tube 7. C a O tube 8-9. absorp t ion tubes 10. f l o w m e t e r 62 In a m o d i f i e d m e t h o d , 1 7 (figure 2.26) the s lag is mel ted in an M g O cruc ib le induct ive ly under the f l o w of o x y g e n as a carrier gas . The e v o l v e d gas is then p a s s e d through l ime , kept at 5 0 0 ° C , and c o p p e r o x i d e , kept at 6 5 0 ° C , to obtain hydrogen in the f o r m of water vapour . The quanti tat ive detect ion of water vapour is done in a X e i d e l ' e lec t ro ly t i c ce l l . The cel l is equipped with a constant vol tage source and plat inum e lec t rodes c o a t e d with P 3 O s which reacted with water to generate H 3 P 0 4 , wh ich is a lso the e lec t ro ly te of the c e l l . A t a constant potent ia l , the' change in current f l o w in the cel l is proport ional to the water reacted and hence to the h y d r o g e n e v o l v e d f r o m the s l a g . The detect ion of hydrogen as water is less des i rab le because of its low vapour pressure and the high adsorp t iv i ty of water vapor on the w a l l s o f the tubes of an analyzer . In this regard , the a n a l y s i s of hydrogen p o s e s fewer d i f f icu l t ies whi le us ing a pressure g a u g e , 5 2 a m a s s s p e c t r o m e t e r 3 ' 5 3 or a c o m m e r c i a l l y avai lable gas analyzer such as L E C O R H - 1 E and R H - 2 2 and O M K F - 7 1 1 . In d e v e l o p i n g a vacuum heating technique wh ich requires , re la t ive ly , lower temperatures which do not melt the s l a g , L l o y d and S h a n a h a n 5 2 f ound that a cons iderab le amount of mois ture f r o m the a t m o s p h e r e a d s o r b e d on the bas ic s lag sur face and the s y s t e m required initial heat ing of the sample at 1 5 0 ° C in a v a c u u m . Reproduc ib le results are obta ined in the vacuum apparatus, by e v o l v i n g hydrogen f r o m the s lag s a m p l e at 9 0 0 ° C and measur ing the resul t ing pressure changes by a p i s t o n - o p e r a t e d M c L e o d gauge (figure 2.27). A n y water vapour , if present in the e v o l v e d g a s e s , is conver ted to hydrogen by p a s s i n g it through f e r r o m a n g a n e s e . The p r o b l e m a s s o c i a t e d with this type of method is that it 6 3 r F i g . 2.26 Appara tus for de termin ing hydrogen in s l a g s 1 7 . I .oxygen 2. P j 0 5 3. rotameter 4. 4 - w a y s t o p c o c k 5. sample 6. induct ion furnace 7. C a O - r e a c t i o n furnace 8. C u O - r e a c t i o n furnace 9. cal ibrator 10. E lec t ron gauge 11. 'Ke ide l ' cel l 64 hot extraction iirnacc ,specimen tubs J _ 29/32 therrno, couple 14/23 C 40/36 mprcyry vapour trap containing dry ice m acetone 19/26 2 I J 26 D Off! mercury _ J || dillusion 1, pump J—I 1 o o l , pypn I 1 furnace J to atmosphere P i rani gauge gas injection bulb 14/23 McLeod gauge reservoir Fig. 2.27 Schematic diagram of hydrogen analyzer" 65 is d i f f icu l t to c o n f i r m the c o m p l e t e "evolut ion of h y d r o g e n in a s o l i d s a m p l e , and s ince it normal ly takes a longer t ime to a n a l y z e , there is the p o s s i b i l i t y of los ing s o m e hydrogen due to its very high d i f f u s i o n c o e f f i c i e n t . The vacuum f u s i o n m e t h o d , even though the m o s t c o m p l e x , is used by many au thors 3 ' 9 , 2 4 to determine hydrogen both in s tee lmak ing and ESR s l a g s . Nakamura and H a r a s h i m a 3 ana lyzed T the h y d r o g e n in ESR s lags by induct ive ly mel t ing the s lag in a m o l y b d e n u m crucib le a long with a luminum f o l l o w e d by a c o l l e c t i o n of the generated gas in a reservo i r wh ich is later c o n n e c t e d to a m a s s analyzer to detect the amount of h y d r o g e n present . The use of a m o l y b d e n u m crucib le is not des i rab le as far as the a n a l y s i s of hydrogen is c o n c e r n e d , s ince the metal can f o r m vo la t i le o x i d e 9 wh ich can adsorb hydrogen and result in a lower determinat ion of hydrogen in the s l a g . 2.5. T H E R M O D Y N A M I C P R O P E R T I E S OF ESR S L A G S The quantitat ive in format ion of d i f ferent t h e r m o d y n a m i c p roper t i es , such as phase d i a g r a m s , chemica l ac t iv i t ies of d i f ferent c o m p o n e n t s in binary and ternary s lags and so lub i l i t i es of g a s e s is very helpful in se lec t ing a proper c o m p o s i t i o n of s lag for an ESR opera t ion . B e c a u s e of the highly c o r r o s i v e nature of ESR s lags at higher temperature , it is very d i f f icul t to carry out a cont ro l led exper iment for accurate determinat ion of the above proper t ies . Di f ferent exper imenta l techniques are e m p l o y e d by a number of workers for a better understanding of phase d iagrams and other t h e r m o d y n a m i c behavior of ESR s l a g s . Unfor tunate ly in s o m e c a s e s , there are w ide d isc repanc ies a m o n g the repor ted resu l ts , p r e s u m a b l y due to l imi tat ions and the di f ferent leve ls of accuracy in the 66 var ious exper imental techniques . Errors m a y a l s o creep in because of the high instabi l i ty of f luor ide s lags in an o p e n s y s t e m with respect to c h e m i c a l c o m p o s i t i o n . Th is aspect is due to the presence of vo la t i le s p e c i e s , such as C a F 2 , A I F 3 and S i F 4 , and the feas ib le chemica l reac t ions are as f o l l o w s (equation 2.28 is s a m e as equat ion 2.24): C a F 2 + H 2 0 = C a O + 2HF 2.28 3 C a F 2 + A l 2 0 3 = 2A IF 3 + 3 C a O 2.29 There fo re exper iments not des igned to analyze the f inal c o m p o s i t i o n of s lag w o u l d probab ly give er roneous resu l ts . 2.5.1 P H A S E D I A G R A M S For a s u c c e s s f u l opera t ion during e lec t ros lag remel t ing , s lag shou ld have a mel t ing point at least 1 0 0 ° K lower than that of meta l . The phase d iagram is an important guide to judge this cr i ter ion and a l s o the p r e s e n c e of any misc ib i l i t y gap which w o u l d alter the p h y s i c a l and chemica l character is t ics of the melt dras t ica l ly . A l r e a d y phase d iagrams of C a F 2 - b a s e d s l a g s , both for binary and ternary s y s t e m s , are repor ted by many a u t h o r s " - 6 1 H o w e v e r , cons iderab le d i s c r e p a n c i e s exist with regard to the ex is tence of di f ferent phases and the l iquidus l ines of var ious s y s t e m s s i n c e they are obta ined by exper iments in s o m e c a s e s , such as D T A and quenching t e c h n i q u e s , 6 1 and s o m e t i m e s deduced f r o m other r e s u l t s . " In the present contex t , only the aluminate and s i l icate s lags conta in ing C a F 2 wi l l be r e v i e w e d . C a O - C a F , S o far it has been c lear ly es tab l ished that this s y s t e m f o r m s a eutect ic at 1 3 6 0 ° C and around 16% C a O . A l s o , other authors (except 67 B a a k " ) agree wel l with the Iiquidus line on the C a F 2 rich s e c t i o n be fo re the eutec t ic . B a a k " reported the ex is tence of a m i s c i b i l i t y gap in the range of 1-8% C a O on the bas is of the m e a s u r e m e n t s of e lectr ica l conduc t i v i t y (figure 2.28). The reason for this d i s c r e p a n c y m a y be due to the f o r m a t i o n of C a C 2 p h a s e , which is most l ikely fo r an exper iment carr ied out in a graphite c ruc ib le . There fore the results obta ined by c r y o s c o p i c s tudies as carr ied out by Koj ima and M a s s o n 6 3 can be taken as a more rel iable re fe rence in this region (figure 2.29). A l s o there is w ide scatter of data b e y o n d the eutect ic po in t , 5 6 but the reports by Kor and R i c h a r d s o n 5 5 and Ries and S c h w e r d t f e g e r 6 4 agree wel l to se rve as a guidel ine in this zone . C a F , - A U O , This s y s t e m can be c o n s i d e r e d binary on ly when exper iments are carr ied out in a sea led c e l l . O therw ise the f o l l o w i n g react ion tends to make it a C a F 2 - C a O - A I 2 0 3 ternary( s a m e react ion as in 2.29): 3 C a F 2 + A l 2 0 3 = 2 A I F 3 + 3 C a O The above react ion is a typ ica l d i s p l a c e m e n t equi l ibr ium react ion in a s imple rec iproca l salt s y s t e m where the var iat ion in c o m p o s i t i o n with respect to all four i o n s , such as , C a , A l / O , F changes the p r o p o r t i o n s of sal ts in the mixture. Th is var iat ion can be proper ly represented by four c o m p o s i t i o n axes with po ly thermal p r o f i l e s . H o w e v e r , the representat ion b e c o m e s much more c o m p l i c a t e d when there are more than t w o sa l ts and when s o m e stable c o m p o u n d s are f o r m e d . Ries and S c h w e r d t f e g e r 6 4 ve r i f i ed the ex is tence of the a b o v e react ion by per fo rming exper iments both in an o p e n cel l and a sea led ce l l (figure 2.30). The s a m e reason can be attr ibuted to the d i s c r e p a n c i e s in 2100 / / / / / Legend Baak, T.(62) / / / I j Ries and Schwerdtfeger(64) Budnikov and Tresvyotsii(66) 20 40 60 80 100 mass% CaO 2.28 Phase diagram for CaO - CaF2 system U 6 0 - I 1320 H 0 .00 0.05 0.10 0.15 0.20 mole fraction, oxide 0.25 g. 2.29 D e p r e s s i o n of f reez ing point of C a F 2 by C a O , M g O and S r O 70 C J o CD D " 5 2000 1900-1800 1700-1600 •"A Primary A l 2 0 3 CD CL E CD 1500 f Primary C A 6 U00- 1 1300 1200 20 40 60 wt% Al 0 2 3 80 100 Fig. 2.30 Liquidus line of CaF 2 - A l j 0 3 melts according to Ries and Schwerdtf eger. 6 4 The melts are probably substantially contaminated with CaO. 71 phase d iagrams reported by Kuo et a l . 6 5 and Mi tchel l et a l . 6 0 (f igure 2.31). The extent of d i f fe rence w o u l d depend on the amount of C a O f o r m e d due to the a b o v e reac t ion . 70wt% C a F 2 - 30wt% A l 2 0 3 , a w i d e l y u s e d ESR s l a g , is invest igated by Ries and S c h w e r d t f e g e r . 6 4 The evapora t ion o f A I F 3 at temperatures 1 6 0 0 - 1 7 5 0 ° C changes the s y s t e m to a ternary and the l iquid separa tes into a t w o - p h a s e mixture ar is ing out of the m i s c i b i l i t y gap which wi l l be dealt wi th in detail in the next s e c t i o n . CaF„ - C a O - A I , Q , Th is ternary s lag s y s t e m is very w i d e l y used in e l e c t r o s l a g remel t ing and s o has been invest iga ted by severa l authors to s tudy its equi l ibr ium phases and the melt behavior at d i f ferent tempera tures . Both the o p e n 5 9 " 6 4 ' 6 7 and c l o s e d " s y s t e m s are c o n s i d e r e d but Ries and S c h w e r d t f e g e r 6 4 repor ted no s igni f icant d i f fe rence to be o b s e r v e d due to these d i f ferent c o n d i t i o n s . H o w e v e r , under open cel l c o n d i t i o n , the s y s t e m should be c o n s i d e r e d as quarternary t ype , C a F 2 - C a O - A I F 3 - A I 2 0 3 . 5 6 Dif ferent c o m p o u n d s , s o far ident i f ied in this s y s t e m 5 6 are: C A 6 , C A 2 , C A , C 3 A , C 1 2 A 7 , C n A 7 F I and C 3 A 3 F I (figure 2.32). N a f z i g e r 5 8 s tud ied this ternary s y s t e m under a partial hel ium a tmosphere , but did not o b s e r v e any m i s c i b i l i t y gap which is c o n f i r m e d to exist by other r e p o r t s 5 6 ' 6 4 both under open and c l o s e d cel l c o n d i t i o n s . Further indirect e v i d e n c e of m i s c i b i l i t y gap can be deduced f r o m the ul t rasonic m e a s u r e m e n t s , i s o d e n s i t y , i s o v i s c o s i t y and i soac t i v i t y contours in that c o m p o s i t i o n r a n g e . 5 6 The electr ical c o n d u c t i v i t y is measured by Mi tche l l and C a m e r o n , 6 8 and it is revealed that at higher C a F 2 c o n t e n t s , networks are b a s e d on A I 0 2 F | - and A I O F 2 , and that l ime addi t ions to these mel ts re lease the f luor ide ions and produce AIO?f and A I 0 2 s t ructures. 1700 -i Fig. 2.31 Phase diagram for the system * CaF 2 ' - A l 2 0 3 ; * CaF 2 ' = + 2wt% CaO 73 CaF, 80 60 146 40 20 mass % C a O Fig. 2.32 Phase diagram for CaF2 - CaO - AljO, system" 74 A t low CaFj c o n t e n t s , 6 ' the m o l e c u l e s of A l 3 0 , - are bound by c a l c i u m ions and any addi t ions of C a F 2 w o u l d tend to break the ne tworks and subsequent ly the f luor ide ions w o u l d f i l l the vacant s i t e s . The structural in format ion further can be der ived f r o m the u l t rasonic studies on the mel ts . A t low levels of C a F 2 (0-30%) the melt c o n s i s t s of main ly A I - 0 structures which p o s s e s s a cons iderab le amount of cova lent b o n d i n g . A t high levels of C a F 2 (70-100%) the structure is p redominant ly ionic and at an intermediate concent ra t ion (30-70% C a F 2 ) both cova lent and ionic types of structure c o - e x i s t . CaFn - C a O - S i O , Severa l s tudies are r e p o r t e d 5 9 " 6 1 ' 7 0 - 7 3 fo r this ternary s y s t e m . Under o p e n - c e l l c o n d i t i o n s , it b e l o n g s to a s t r ic t ly rec iproca l s y s t e m , C a F 2 - C a O - S i F 4 - S i 0 2 , but for all pract ical p u r p o s e s , the es tab l ished ternary phase d iagram can be c o n s i d e r e d (figure 2.33). The c o m p o u n d s f o r m e d in this s y s t e m are: C S , C 3 S 2 , C 2 S , C 3 S , C S 3 F I , C 4 S 2 F I and C 9 S 3 F I . The d i f ferent exper imental c o n d i t i o n s and techniques e m p l o y e d to s tudy the phase d iagram have resul ted in s o m e d i s c r e p a n c i e s . S a l t 5 9 reported two separate t w o - l i q u i d r e g i o n s , but other a u t h o r s 5 6 ' 6 1 ' 7 0 o b s e r v e d o n l y one misc ib i l i t y gap over a large c o m p o s i t i o n range c o v e r i n g both the reg ions as p r o p o s e d by S a l t . 5 9 The t h e r m o c h e m i c a l r e s u l t s 7 2 a lso support the wider and s ingle zone m i s c i b i l i t y gap . A d d i t i o n of C a F 2 in ac id ic s i l icate melt is b e l i e v e d to act as a network m o d i f i e r , whereas in a bas ic melt it apparent ly behaves as a diluent as o b s e r v e d in weight l o s s and v i s c o s i t y m e a s u r e m e n t s . 7 1 Th is p h e n o m e n o n is expla ined by Sh inmei et a l . 7 1 by the nature of c o m p e t i t i o n be tween F- and 0 2 ~ on the s i l icate network. The f luor ide ion is far weaker moss V« CoO 2.33 Phase diagram for CaF2 - CaO -SiO, system; Cr =Cristobalite, Tridymite" 76 than oxide ion in this respect and also S i - F is highly volatile at high temperatures. The cryoscopic studies by Baak and Olander 7 0 and Suito and Gaskel l 7 3 suggest the formation of trimeric ring metasilicate ion SijO6," and free F" ions when N _ /^N >1.5. The latter report also suggests the CaO o i O 2 formation of polyf luorosilicate anions such as S i 0 3 F 3 - , SiOjF2.- or SiOF 3 when N ^ /N „ <1.5 in the melts, but there is no experimental evidence CaO S t 0 2 to prove the existence of these complex anions. 2.5.2 BASICITY AND ACTIVITY OF ESR S L A G S 2.5.2.1 CONCEPT OF BASICITY There is no single method or index to define the basicity of metallurgical slags which are mainly a mixture of different oxides. Different practical methods are only applicable to a certain range of compositions and components. The use of the V-ratio is limited to a range within Blast Furnace related slags which normally contain varying proportions of CaO , SiOj , A l 2 0 3 and MgO. Even at the extreme ranges of the above components (acid or basic), these concepts lose their significance as a proper guide to basicity. An improved version is the concept of 'excess base', which is defined as fol lows: Excess Base = 2.30 all all basic oxide acid oxide where N refers to mole fraction and I is the Ion-oxygen attraction for a 77 particular c a t i o n . The prob lem here is to d is t inguish be tween ac id and bas ic c o m p o n e n t s spec ia l l y for a amphoter ic oxide such as A l 2 0 3 . A n o t h e r drawback is wi th the I o n - o x y g e n at tract ion parameter which cannot be appl ied for a n o n - o x i d e c o m p o n e n t , such as C a F 2 . In this regard, the recent idea of opt ica l b a s i c i t y 1 4 is more e laborate and re f lec ts much more of the fundamental nature of a s l a g . It is d i rect ly related to the chemica l nature of oxide ions present in the s lag and theoret ica l ly can be es t imated f r o m the known Paul ing e lec t ronegat iv i ty of c a t i o n s . H o w e v e r , for the t ransi t ion metal o x i d e s , the direct ca lcu la t ion of opt ica l bas ic i ty is not p o s s i b l e . A l s o in a s lag conta in ing ha l ide , such as C a F j - C a O , the concept of opt ica l bas ic i ty is not e s t a b l i s h e d . For oxide s l a g s , the opt ica l bas ic i ty is found to be independent o f temperature . The measurement of the o x y g e n potent ial o f a s lag (related to the oxide ion act iv i ty ) is a lso c o n s i d e r e d in the s a m e w a y as measur ing the pH of an aqueous so lu t ion . Th is c o n c e p t is the m o s t sc ien t i f i c and is w i d e l y a c c l a i m e d . H o w e v e r , the major p r o b l e m is related to the d e s i g n of a proper e lec t rode which can be used at high temperatures and in all k inds of s lags without any chemica l attack. A g a i n , like water , there is no re ference so lvent avai lable as a s tandard . H o w e v e r , the latest d e v e l o p m e n t of a l o n g - l i f e oxygen s e n s o r " (figure 2.34) using a Z r 0 2 - M g O s o l i d e lec t ro ly te l o o k s p r o m i s i n g . The measurement o f o x y g e n potent ia l is d i rect ly related to the oxide ion act iv i ty due to the f o l l o w i n g re la t ion: 1/2 0 2 + 2e = O 2 - 2.31 N o w for a known concent ra t ion of oxide ion , N , the ac t iv i ty • Q 2 - ' data w o u l d a l low us to calculate 7 , an indicator for the dev ia t ion f r o m 78 Fig. 2.34 Long-l i fe oxygen sensor for measuring oxygen potential of s lag 1 5 7 9 ideality that reflects upon the ionic structure of the slag. Wagner 7 6 proposed a new definition for basicity on the basis of carbonate capacity of a slag, defined as; B = C / C * 2.32 carb CO 7 " COj-where the carbonate capacity of a given slag C can be expressed as: wt% C0\-2.33 C 0 2 CO 7 " p. The C is the carbonate capacity in a reference slag. The choice of the CO?f reference slag i.e. 0.4 CaO +0.4SiO 3 + 0.2 A l 2 0 3 ,as proposed by Wagner, 7 6 seems unjustified, 7 7 since from the optical basicity relationship this slag is likely to have a very negligible carbonate capacity. This concept of basicity should be very useful if 7 r n 2 does not vary much with composition, as expected due to the big ionic size of a carbonate ion. Unfortunately, there are very few experimental data available in the literature for verification. Like the carbonate capacity, the sulfide, phosphate and hydroxyl capacities depend on the oxide-ion activity. Therefore, the variation of oxide ion activity should show a similar trend in these capacities and as a consequence they also should be interrelated. 2.5.2.2 ACTIVITY IN ESR SLAGS The information on activity of a component in the slag not only corroborates the previously mentioned phase boundaries but also reflects the stability and the possibility of a chemical reaction under different conditions occurring during electroslag remelting. The experimental techniques employed and the assumptions made for the determination are very critical to obtain reliable activity data. In a way similar to the 80 exper iments related to the phase d iagram de te rmina t ion , the a n a l y s i s o f f inal s lag c o m p o s i t i o n is essent ia l for f l u o r i d e - b a s e d s l a g s . Lack o f these precaut ions has resulted in a wide var ia t ion in act iv i ty data repor ted in the literature. H o w e v e r , s o m e act iv i t ies and the trend of var ia t ion are rel iable enough to d i s c u s s as a guide line for future work . C a F , - C a O Dif ferent authors have ca lcu la ted the act iv i ty of l ime in this binary by di f ferent m e t h o d s . Instead of exper iments , M u r a t o v , 7 8 Mehrot ra et a l . 7 9 , and Kor and R i c h a r d s o n " e m p l o y e d part icle interact ion energy , s tat is t ica l t h e r m o d y n a m i c s , and bas ic t h e r m o c h e m i c a l da ta , r e s p e c t i v e l y , to calculate the l ime act iv i ty in this s y s t e m . The values g iven by Muratov are not acceptab le s ince the author used e r roneous melt ing po ints in the melt data. The other two va lues , even though s imi la r , do not corre la te p roper ly wi th the phase d iagram. Kor and R i c h a r d s o n " der ived the act iv i ty data at any temperature using the f o l l o w i n g re la t ion: log a ( l iquid) = - 6 - 6 ( 2 8 7 3 - T ) 2.34 a C a O v ; 4.575 T Hawkins et a l . 8 0 es t imated l ime act iv i ty by a s s u m i n g regular so lu t ion behav ior and measur ing the C a O saturat ion at 1 5 0 0 ° C . T h e y obta ined o. Q = 1 at N Q =0.26 which matched wel l wi th the phase d iagram. E d m u n d s and T a y l o r 6 7 measured the p for the equi l ibr ium reac t ion : C a O + 3C = C a C 2 + CO(g) 2.35 U s i n g a known equi l ibr ium constant and a s s u m i n g negl ig ib le so lub i l i ty o f C a C j in the mel t , these authors es t ima ted a l ime act iv i ty which was very s imi lar to that es t imated by Hawkins et a l . 8 0 (figure 2.35). H o w e v e r , at lower concent ra t ions of C a O , increasing so lub i l i ty of C a C 2 can af fect this 81 mole fraction, CaO Fig. 2.35 Activity of CaO in CaO - CaF2 system at 1500°C 82 e s t i m a t i o n . " C r y o s c o p i c measurments by Ko j ima and M a s s o n " are a rel iable m e t h o d for es t imat ing act iv i ty in this s y s t e m , at least in the dilute s o l u t i o n r e g i o n . C a F , - C a O - A U O , The equi l ibr ium react ion (2.35) is a lso e m p l o y e d to determine l ime act iv i ty in this ternary by Edmunds and T a y l o r . 6 7 S imi la r act iv i ty data are a l s o obta ined by Z h d a n o v s k i 8 1 w h o c o n s i d e r e d the three binary phase d iagrams in order to compute the equivalent act iv i t ies in the ternary. Un for tuna te ly , both these works are not in a c c o r d with the ternary phase d iagram e s p e c i a l l y the misc ib i l i ty gap reg ion . A l l iber t and C h a t i l l o n 8 2 c o n s i d e r e d the f o l l o w i n g reac t ion: C a O + AIF(g) + C a F 2 = 2CaF(g) + AIOF(g) 2.36 and used a m a s s spec t romete r , c o n n e c t e d to an e f f u s i o n cel l to determine the act iv i ty of C a O at 1600 and 1700K. There is a large d i s c r e p a n c y be tween these results and those o f E d m u n d s and T a y l o r , 6 7 except when C a O act iv i t ies are high (figure 2.36). A l s o , s o m e of the ac t iv i t ies do not c o r r e s p o n d to a h o m o g e n e o u s l iquid, but rather to two phase reg ions as ev ident in this f igure. C a F , - C a O - S i O , S o m m e r v i l l e and K a y 7 3 de te rmined the s i l ica ac t iv i ty in this s y s t e m by mon i to r ing p of the f o l l o w i n g reac t ion : W W S i O j + 3C = S i C + 2CO(g) 2.37 S u b s e q u e n t l y , the act iv i t ies of C a O are ca lcu la ted by app ly ing the ternary G i b b s - D u h e m re lat ionship (figure 2.37). Unfor tunate ly , there are no direct independent l ime act iv i ty data to use f o r c o m p a r i s o n . H o w e v e r , the f reez ing ig. 2.36 Activity of CaO in CaF, - CaO - ALO. system-...,1427°C,J • 1 * * 7 liq. 84 Fig. 2.37 Isoactivity of CaO in the CaF} - CaO -Si0 3 system at 1450 C 85 point d e p r e s s i o n measurements are carr ied out by Baak and O l a n d e r , 7 0 K o i i m a and M a s s o n 6 3 and Sui to and G a s k e l l 7 3 to determine a „ _ . The f irst J C a F 2 work s h o w s a smal ler act iv i ty value c o m p a r e d to the other t w o and the d i f f e r e n c e b e c o m e s more s igni f icant at lower C a F 2 contents in the s l a g . 2.5.3 S O L U B I L I T I E S 2.5.3.1 S O L U B I L I T Y OF G A S E S IN M O L T E N S L A G S In this area, pr inc ipa l ly the a b s o r p t i o n of h y d r o g e n , sulphur, carbon and n i t rogen have been s tud ied . The s tudies regarding the absorp t ion of h y d r o g e n have been d i s c u s s e d in detail in a prev ious s e c t i o n . Kor and R i c h a r d s o n , " Hawkins et a l . 8 0 and Raschev et a l . 8 3 invest iga ted the sulphur absorp t ion in C a F 2 - C a O and C a F 2 - C a O - A l 2 0 3 s l a g s . For C a F 2 - C a O s l a g , Hawkins et a l . 8 0 obta ined a much higher sulphur capac i ty value c o m p a r e d to the other two g r o u p s , but the d e p e n d e n c e on C a O content is cons is ten t in all the repor ts . A max imum sulphur capac i ty (0.15%)" is o b s e r v e d around 0.25 mo le f rac t ion of C a O . The rep lacement of Ca by M g and Sr lowers the su l f ide capac i ty . The increase of temperature a lso lowers the su l f ide capac i ty , but the e f fec t is not that s i g n i f i c a n t . 8 0 In C a F 2 - C a O - A l 2 0 3 , Raschev et a l . 8 3 o b s e r v e d a 0.15% sulphur level at 1 8 5 0 ° C which hardly var ied with temperature . S c h w e r d t f e g e r and S c h u b e r t 1 4 de termined the ni trogen and carbon so lub i l i t y in f luor ide s lags at 1 6 0 0 ° C . Carbon is d i s s o l v e d in the s lag as C N - and C^-, whereas ni t rogen is d i s s o l v e d as N 3 - and C N \ In a 86 C a F 2 - C a O - A l 2 0 3 s l a g , the total so lub le carbon is 0.25% and the c o r r e s p o n d i n g ni t rogen level is 0.13%. Ni t rogen w a s a l s o f o u n d to o b e y the S iever t ' s law re la t ionship . The addi t ion of C a F 2 in C a O - A l 2 0 3 s l a g s resul ts in higher cyanide and carbon so lub i l i t i es . 2.5.3.2 S O L U B I L I T Y OF G A S E S IN F U S E D S A L T S The noble gases (He, Ne , A and X e ) are spar ing ly so lub le in mol ten f luor ide sal ts because of their very weak in teract ions with the s o l v e n t s . The so lubi l i ty increases with the increase of pressure (Henry's law), temperature and with the decrease in the s ize of the gas a t o m . For e x a m p l e , at 1 atm gas pressure , the so lub i l i ty of He in 0 .53mole N a F + 0 . 4 7 m o l e ZrF« melt is only 2 1 . 6 x 1 0 - ! m o l e / m l of melt at 6 0 0 ° C and increases upto 42.0x10"' m o l e / m l of melt at 8 0 0 ° C . M Simi la r s imple so lu t ions are f o r m e d when carbon d iox ide d i s s o l v e s in mol ten alkali hal ides and the so lub i l i t y inc reases with the increasing free v o l u m e of the salt . For ins tance , at 1 atm p r e s s u r e , the so lub i l i ty of C 0 2 at 9 5 0 ° C is 7.0x10" 6 m o l e / m l of KCI c o m p a r e d with 4 . 0 X 1 0 " 6 m o l e / m l of N a C l . 5 0 The molar vo lume of KCI is 51.4ml and that of NaCl is 38.9ml. The gases which react with mol ten sal ts g ive r ise to much higher so lub i l i t i es . 0.75 mo le of f luor ine can be d i s s o l v e d in the melt of 0 .25mole CuCI 2 + 0.75mole KCI because of the f o l l o w i n g r e a c t i o n 5 0 : 2CuCI 2 + 6KCI + 6 F 2 = 2 K 3 C u F 6 + 5CI 2 2.38 S i m i l a r l y , HF is a lso highly so luble in N a F - Z r F 4 mel ts and its so lub i l i t y d e c r e a s e s with the increase in temperature. 87 CHAPTER 3 OBJECTIVES The f o r e g o i n g rev iew revea ls that the t h e r m o d y n a m i c s of reac t ions related to the hydrogen transfer in ESR s y s t e m are sti l l not very wel l d e f i n e d . T h e r e f o r e , the ob jec t ives of this work are d i rected toward the f o l l o w i n g : 1. Independent invest igat ion of the oxide act iv i ty of binary and ternary f l u o r i d e - b a s e d s lags and a c o m p a r i s o n of the results with the avai lable literature data. 2. D e v e l o p m e n t of a new and rel iable s a m p l i n g technique of ESR slag for subsequent water a n a l y s i s . 3. D e s i g n , fabr ica t ion and operat ion of an i m p r o v e d apparatus to carry out water ana lys is in s lags s o that the o n - l i n e mon i to r ing of h y d r o g e n in an ESR ingot can be made in future. 4. A n a l y s i s of the p rev ious and new water so lub i l i ty data to def ine the cont ro l l ing parameter during E l e c t r o s l a g remel t ing with respect to h y d r o g e n in the ingot . 5. A p p l i c a t i o n of t h e r m o d y n a m i c s to c lar i fy the cri t ical react ions invo lved during the p r o c e s s and a s s e s s m e n t of d i f ferent s lags in that p e r s p e c t i v e . 87 CHAPTER 4 EXPERIMENTAL 4.1. C C - S L A G EQUILIBRIUM For all exper iments , synthet ic s lags of d i f ferent c o m p o s i t i o n s were prepared by weighing appropriate amounts of reagent grade C a F 2 , A I 2 O j , C a C O j and S i 0 2 . A l l the equil ibria runs were carr ied out inside a vert ical super 3 Kanthal furnace conta in ing a recrys ta l l i zed a lumina tube (36"x1^-"0 O.D). Both ends of the tube were c l o s e d and water c o o l e d by brass p lates and c o p p e r tubes r e s p e c t i v e l y . The bo t tom end w a s c o n n e c t e d both to a hel ium and to a ca rbon d iox ide cy l inder through a t w o - w a y va lve . The o n - l i n e f l o w and pressure were c h e c k e d by a ca l ibrated f l o w m e t e r and a mercury m a n o m e t e r . The hel ium line w a s connec ted through an absorp t ion tube conta in ing ' A S C A R I T E ' to e l iminate C 0 2 f r o m the gas s t ream (figure 4.1). A l l the exper iments related to f luor ide based s lags were done at 1 4 0 0 ° C ± 5 ° C mainta ined by a temperature contro l ler (Honeywel l 'Cont inuous B a l a n c e ' type) . B e f o r e each exper iment , the s e t - u p was f lushed with C 0 2 at least for an hour and the temperature of the furnace w a s s tab i l i zed at the set temperature . A 5c.c . plat inum crucib le conta in ing 200mg s l a g int roduced f r o m the b o t t o m end of the furnace w a s ra ised to the hot zone with the help of a p lat inum wire and a magnet a s s e m b l y . During the exper iment , the 88 89 temperature recorder temperature controller magnet quartz tube cold trap Gas I controller CrrccrKitograph recorder CO2 Helium cylinder cylinder Fig. 4.1 Schematic diagram of the apparatus for carbonate equilibrium studies 90 C 0 2 flowrate was maintained constant at 500c.c/min. Each run lasted for an hour after which the liquid slag in the crucible was quenched by quickly lowering it onto the water cooled brass plate. The C 0 2 solubility in the slag was determined in the same furnace which was purged with helium overnight. The next morning the gas composition was checked for any trace of C 0 2 by a Pye-Unicam gas chromatograph (series 104). Actual analysis was postponed until all the C 0 2 in the assembly was flushed out. To analyze C 0 2 in the slag, the platinum crucible containing the equilibrated slag was again raised to the hot zone of the furnace, but with helium flowing at the rate of 200ml/min. The gases exiting from the furnace were delivered directly to the Gas Chromatograph sample ioop(2ml). The evolved C 0 2 from the slag was cold trapped in that loop which was dipped in the liquid nitrogen. The C 0 2 evolution was found to be completed within 30 minutes. This was confirmed by extending the time of cold trapping and also by adding some B 2 0 3 to the slag. Later, the frozen C 0 2 was evolved by removing the cold trap and its quantity analyzed in the Gas Chromatograph. " The analyzer column of the Gas Chromatograph was modified by a 1/4 "<f> copper tube, 2" of which was filled with silica gel (mesh 80/100) in order to improve the detection limit and the response time. Before each set of experiments the instrument was calibrated repeating the same procedure, but taking different amounts of C a C 0 3 instead of slag in the platinum crucible (figure 4.2). The analysis error is found to be within 10% of the content when the attenuation was at 20x, carrier gas (helium), flowrate 50ml/min, detector oven temperature 100°C , column temperature 100 Fig. 42 Typical calibration plot for carbonate analysis. 92 7 0 ° C and the detector f i lament current 100 m A . For chemica l ana lys is of s l a g s , identical C 0 2 equi l ibr ium exper iments were repeated. The s lags were ana lyzed by the standard f u s i o n m e t h o d ( N a 2 C 0 3 + Z n O ) where C a , S i and A l were ana lyzed by a t o m i c a b s o r p t i o n spec t rophotomete r (Perkin E lmer 306) and the f luor ine a n a l y s i s w a s done by f l u o r i d e - i o n se lec t ive e lec t rode (Appendix I). 4.2. D , 0 - S L A G EQUILIBRIUM The vert ical super kanthal furnace which was used for the p r e v i o u s set of exper iments was m o d i f i e d to carry out the water v a p o u r - s l a g equi l ibr ium (figure 4.3). A bigger plat inum crucib le (50c.c.) w a s p l a c e d in the hot zone on a smal le r diameter a lumina tube. Wi th a bucket i ike arrangement it was p o s s i b l e to qu ick ly retr ieve the crucible f r o m the top with a m o l y b d e n u m wire h o o k . The top port , p r e v i o u s l y used for the magnet ic lower ing and raising d e v i c e , w a s conver ted to a s a m p l i n g por t . A constant temperature magnet ica l ly st irred water bath w a s made to generate di f ferent levels of D 2 0 vapour partial p ressures in the fu rnace . He l ium, used as a carrier gas for D 2 0 vapour , w a s f irst dr ied by m a g n e s i u m perchlorate be fore entering the water bath, w h o s e temperature w a s mainta ined by a T h e r m i s t e m p Tempera ture Contro l ler (model 71). The temperature of hel ium was a lso ra ised by p a s s i n g it init ial ly through a co i l of c o p p e r tubing (1/4"<6) which was i m m e r s e d in the water bath. Next, the carrier gas w a s p a s s e d through a ser ies of c l o s e d test tubes conta in ing D 2 0 at greater than 99.9% purity and bubbl ing through a gas d ispers ion tube in each c a s e . T o prevent c o n d e n s a t i o n o f the D 2 0 saturated gas s t r e a m , the 93 rherrrocpuple sampling port MoSi2 furnace Platinum crucible Water bath temperature controller flowmeter Furnace controller to manometer F i g . 4.3 S c h e m a t i c d iagram of the apparatus for D 2 0 - s l a g eui l ibr ium water vapour pressure 94 remainder of the copper tubing c o n n e c t e d to the fu rnace , w a s heated a b o v e 1 0 0 ° C by means of a heating tape. Initial tr ials were c o n d u c t e d to ve r i f y the generat ion of saturated water vapour pressure in the gas s t ream by c o l l e c t i n g the water vapour at the end o f the copper tube in an absorp t ion c o l u m n conta in ing m a g n e s i u m perchlorate . The results (figure 4.4) at d i f ferent f lowra tes of he l ium were found to be quite s a t i s f a c t o r y wi thin the range of this inves t iga t ion . Bulk s lags were prepared by m e c h a n i c a l l y mix ing di f ferent pure reagent grade ingredients such as C a F 2 , C a C 0 3 , A l 2 0 3 and S i 0 2 in a mechan ica l shaker for 30 minutes. A l l the binary and ternary s lags were equi l ibrated fo r one hour at two di f ferent partial p ressures of D 2 0 (0.042 and 0.122 atm) at 1 4 0 0 ° C and constant f l o w rate of hel ium (300ml /min) . A t the end of each run, a s lag sample was c o l l e c t e d through the s a m p l i n g port in a quartz tube (5-7 mmc/>) by app ly ing suc t ion through a rubber bu lb . T h i s s imp le m e t h o d w a s simi lar to the metal s a m p l i n g technique, but has not been tried be fo re because of the p o s s i b i l i t y of s i l i ca contamina t ion of the s l a g . For this r e a s o n , during later s l a g a n a l y s i s , S i a n a l y s i s w a s c o n d u c t e d for n o n - s i l i c a t e s l a g s . There w a s no S i de tec ted in those s lags c o n f i r m i n g the va l id i ty of this s i m p l e s a m p l i n g technique . A l l the s lag s a m p l e s , s o o b t a i n e d , were stored in a d e s i c c a t o r for further c o m p o s i t i o n and water so lub i l i t y determinat ions. The chemica l ana lys is of s lag w a s carr ied out by the s a m e 95 0.25 0.20-^ 0.15-"o O CN X °- 0.10 0.05 0.00-r 20 i 30 Legend A theoretical x experimental helium flow: 3 0 0 - 5 0 0 m l / m i n 40 50 60 70 80 bath temperature, C Fig. 4.4 Comparison of theoretical and experimental equilibrium water vapour pressure 96 f u s i o n m e t h o d as d e s c r i b e d in the s e c t i o n 4.1. 4.3. W A T E R A N A L Y S I S S ince there was no unique m e t h o d for the a n a l y s i s of water in f luor ide s l a g s , and in the literature it was o b s e r v e d that d i f ferent techniques y ie ld d i f ferent resul ts , it w a s c o n s i d e r e d n e c e s s a r y to des ign and fabr icate a new experimental technique with due regard to the drawbacks of the prev ious m e t h o d s . The trace quantity of water c o u l d be detected either as the total water e v o l v e d f r o m the s lag or as the equivalent hydrogen obta ined by reducing the water . The de tec t ion of water as water had certain p r o b l e m s : water w o u l d c o n d e n s e on its w a y to the detect ion chamber if all the area is not adequate ly heated; m o r e o v e r , m o s t mater ia ls a b s o r b e d a certain quantity of water and it w a s d i f f icu l t to ensure that a s y s t e m was \ v a t e r - f r e e ' . The o b v i o u s c o n c l u s i o n f r o m the a b o v e f a c t s w a s that it w o u l d be p o s s i b l e to des ign an apparatus in which the water c o u l d be ana lyzed as h y d r o g e n . For detect ion of any e lement as a trace quant i ty , the m a s s spec t romete r w a s an ideal t o o l , wh ich unfortunate ly requires a high vacuum env i ronment . A g a i n , in a vacuum s y s t e m , hydrogen w a s f o u n d to be one of the predominant s p e c i e s (small w o n d e r s ince it m a k e s up m o r e than 90% of all a toms of the universe!) b e s i d e s ni t rogen and water . T h e r e f o r e , the high background level of hydrogen again made this m e t h o d ques t ionab le . For tunate ly , hydrogen has two i s o t o p e s , deuter ium and t r i t ium, and i s o t o p e de tec t ion techniques were not on ly c lean and rel iable but a lso r e m o v e d the doubts surrounding the or ig in of the test mater ia l . The h y d r o g e n , bes ides be ing introduced during equi l ibr ium exper iments , m a y a lso 97 contamina te the sample as H 2 0 during its handl ing. On the other hand, the natural abundances of these i s o t o p e s (e.g 0.015% for deuter ium) are too low to have any e f fec t on the ana lys is . A m o n g these t w o i s o t o p e s , the c h o i c e of deuter ium w a s quite s t ra ight forward s i n c e it is n o n - r a d i o a c t i v e and re la t ive ly much cheaper and eas i ly ava i lab le . 4.3.1 F A B R I C A T I O N OF HIGH V A C U U M A P P A R A T U S Even though the high vacuum s y s t e m was to be c o n n e c t e d to a meta l l ic analyzer head of a m a s s s p e c t r o m e t e r , a g lass apparatus w a s built (figure 4.5) s ince it w a s a much cleaner s y s t e m and a lso it cou ld be eas i ly c o n n e c t e d to a s i l i ca react ion tube where s lag s a m p l e s wou ld be induct ive ly f u s e d . A water c o o l e d mercury d i f f u s i o n p u m p , which was backed by a mechan ica l rotary p u m p , was c o n n e c t e d to the apparatus to maintain a vacuum of the order of 10-' torr. A l iquid n i t rogen trap w a s e m p l o y e d to i m p r o v e the e f f i c i e n c y of the s y s t e m . A heating tape w a s wrapped around the g lass tube c l o s e to the analyzer head to c lean up the residual gases qu ick ly and to keep the absorpt ion of h y d r o g e n - r e l a t e d s p e c i e s to a m i n i m u m . A g l a s s - K o v a r joint enabled the vacuum tight c o n n e c t i o n of the 7 0 m m sta in less steel f lange of the analyzer head which de tec ted di f ferent g a s e s on the bas is of their m a s s number with the help of a contro l uni t (VG M I C R O M A S S 1/2). The other end of the g lass tube w a s c o n n e c t e d to the s i l i ca react ion tube through a met rosea l leak, a high vacuum va lve and a quickf i t joint (size 60/71) . The react ion tube area w a s a l s o equ ipped with a s a m p l i n g ho ld ing area and as a resul t , this area c o u l d be c o m p l e t e l y isolated f r o m 98 T C Mercury diffusion pump. Ionization gauge Thermocouple gauge & Ionization control \ Analyser head 1 overpressure 1/ relay Recorder \ M. S. controller Heater - Reaction tube (silica) .Moly- funnel • induction coil graphite crucible •alumina tube F i g . 4.5 S c h e m a t i c d iagram of the apparatus for water a n a l y s i s of s lag 99 the high vacuum m a s s spect rometer area and back f i l l ed by argon to enable s a m p l e loading and f lushing of the s y s t e m . Th is area w a s a lso c o n n e c t e d to the mechanica l pump to attain a vacuum around 5 x 1 0 J t o r r . A graphite crucible was u s e d as a s u s c e p t o r fo r induct ion heating of s a m p l e s . Th is crucible was wrapped by a graphite felt p laced inside an a lumina tube to reduce the overheat ing of the s i l i ca tube. A water c o o l e d copper c o i l , connec ted to a high f requency induct ion s u p p l y , w a s p laced around the quartz tube. In the s a m p l e ho ld ing area, one magnet ic pusher w a s p l a c e d , s o that the sample cou ld be d r o p p e d into the graphite crucible without disturbing the v a c u u m . 4.3.2 P R E P A R A T I O N FOR C A L I B R A T I O N For proper cal ibrat ion of this c o m p l e x apparatus it was best to f o l l o w the s a m e s teps as was done for a s l a g . For this reason Ca(OD) 2 w a s prepared by the f o l l o w i n g m e t h o d . 50 c .c . o f pure D 2 0 was taken in a beaker and p l a c e d inside a n i t rogen f lushed des icca tor (no des iccant inside) . A b o u t 35gm of C a C 0 3 was taken in a crucible and heated to 1 2 0 0 ° C for 3 hours ins ide a muf f le furnace to generate pure C a O . The pure l ime, s o ob ta ined , w a s t ransferred into the beaker conta in ing D 2 0 , this beaker being left inside a ni trogen f lushed g love box overnight for the c o m p l e t e c o n v e r s i o n to Ca (OD) 2 . The equi l ibr ium constant of this react ion involv ing H 2 0 (2 .7x10 n at 300K) 5 4 st rong ly favours the product ion of Ca (OD) 2 . Later this c o m p o u n d w a s dr ied o f f on a hot plate and s tored carefu l ly inside the g love box . For ca l ib ra t ion , a smal l k n o w n quantity of C a ( O D ) 2 was p laced in a capsule a long with a suf f ic ient amount of reducing agent , such as 100 a l u m i n u m . A f te r initial t r ia ls , about 0.5" l o n g , one end c l o s e d tubes were made by dri l l ing a 0.089" bore in a 1/8" mi ld steel r o d . A m e a s u r e d quanti ty of Ca (OD) 2 and a luminum w a s p a c k e d inside the tube and the o p e n end w a s sea led o f f by a 3 /32" ball bear ing . Th is procedure ensured the c o m p l e t e reduct ion of D 2 0 to D 2 s ince no peak was o b s e r v e d at m a s s 20 whi le using hel ium as the back f i l l ing g a s . 4.3.3 C A L I B R A T I O N A N D S L A G A N A L Y S I S FOR W A T E R For both the cal ibrat ion and the s lag ana lys is the f o l l o w i n g s teps were taken. The high vacuum area w a s w a r m e d up and c leaned s o that the pressure w a s s teady at a I0~ 6torr range and it was iso la ted f r o m the s i l i ca tube area which was then induct ive ly heated for 5 minutes . The crucib le was then c o o l e d for 15 m i n u t e s . C o n s e q u e n t l y , the blank gas a n a l y s i s of the s i l i ca tube area was m a d e by open ing the high v a c u u m va lve which c o n n e c t e d the m a s s s p e c t r o m e t e r area. A s imi lar procedure w a s f o l l o w e d after d ropp ing a sample into the graphite cruc ib le . Be fore each set of s lag ana lys is a ca l ibrat ion plot of p ve rsus the amount of C a ( O D ) 2 w a s obta ined (figure 4.6) for es t imat ion of water in the s l a g . A l s o at the end of a set of s lag s a m p l e s , a ca l ibra t ion capsule was introduced and the s y s t e m was c h e c k e d fo r any i n c o n s i s t e n c y . For s lag a n a l y s i s , about 100 -300mg of ground s lag (depending on the es t imated water level ) was w r a p p e d in a c leaned (acetone and carbon tetrachlor ide) a luminum f o i l . A f t e r ana lys ing 10-15 s lag s a m p l e s it w a s n e c e s s a r y to c lean the s i l i ca tube wh ich was by then c o a t e d with a substant ia l quantity of vapor i zed a l u m i n u m . It was p o s s i b l e for this a luminum to absorb s o m e H 2 and D 2 and thus y ie ld ing lower va lues of partial pressures of these two s p e c i e s . F i g . 4.6 T y p i c a l ca l ibra t ion plot for water a n a l y s i s 102 This new analytical tool was subsequently verified by analysing water solubility in known binary non-fluoride systems (table 4.1). In both cases, the agreement was quite good with respect to the previous data. A l s o , the repetition of the same slag samples showed that the reproducibility of the analysis was within 10% of the content. A few slags (both binary and ternary) were also equilibrated with H 20 under the same conditions which were maintained for D 2 0 - s l a g equilibrium. These slags were also sampled in quartz tubes and later sealed in evacuated glass tubes to prevent any moisture pick-up from the atmosphere. These encapsulated slag samples were sent out to Vereinigte Edelstahlwerke (VEW), Austria, for an independent analysis of water solubility in ESR slags by a different method. The analyses, so obtained, are discussed in the next chapter. 103 Table 4.1 Verification of Water Analysis by Isotepe Tracer Technique Composition Literature This method Slag No X CaO X A1203 X Si02 (HjO) (B^O) ppm torr ppm 1 50.4 49.6 150 900^ 9 ^  1178 2 37 63 289 A29C2) 526 CHAPTER 5 RESULTS AND DISCUSSIONS 5.1. O X I D E ION A C T I V I T Y IN ESR S L A G S F r o m the ionic theory of s l a g , we formula te aQaQ-a = a x a 5.1 C a O C a 2 * O 2 " = CL , if a . , = 1 O 2 - ' C a 2 + In the present s y s t e m s this a s s u m p t i o n is quite val id s i n c e the ESR s lags s tud ied are highly bas ic and the on ly cat ion is the c a l c i u m i o n . A l s o , in the ternary s y s t e m s contain ing A l 2 0 3 , it is highly probab le that the cat ions remain on ly as c a l c i u m , s ince a luminum is p robab ly present on ly as c o m p l e x anions such as A I O ; , AIO?,-. The s a m e reason ing is app l icab le to the s i l ica te s y s t e m s used in ESR s l a g s . 5.1.1 C A R B O N A T E EQUILIBRIUM IN FLUORIDE S L A G S The determinat ion of ox ide ion act iv i ty can be done by s tudy ing the f o l l o w i n g equi l ibr ium; C 0 2 ( g ) + ( O 2 ) = ( C O 2 - ) 5.2 f o r w h i c h , the equi l ibr ium constant i s , a N y K = C ° I " C O 2 - - 7 C 0 j - 5 3 .art, p ^ .a ' 2 W ^ W 2 In the presence o f c a l c i u m , the change in the f ree energy can be writ ten 104 105 a s 8 5 : A G 0 = -40248.6 + 34.4T c a l / m o l e where T is temperature in K. T o calculate the act iv i ty of C a O , the act iv i ty c o e f f i c i e n t of carbonate ion is es t imated by c o m b i n i n g the c r y o s c o p i c data of Koj ima and M a s s o n 6 3 and the data f r o m the binary s l a g - C 0 2 equi l ibr ium exper iment of the present w o r k , and the s lag is a s s u m e d to be regular in v iew of the fact that these coplanar carbonate ions are too large to have any s ign i f icant interact ions with other anions such as A I 0 2 . The f inal determinat ion of the oxide ion act iv i ty in binary f luor ide s lags s h o w s a p o s i t i v e dev ia t ion , whereas the dev ia t ion is negat ive in ternary, C a F 2 - C a O - A I 2 0 3 and C a F 2 - C a O - S i 0 2 , s l ags (table 5.1 to 5.3). It is a lso o b s e r v e d that the act iv i ty of l ime is a s t rong funct ion of b a s i c i t y (more s p e c i f i c a l l y , the rat ios of l ime and s i l i ca or a lumina depending on the type of s lag) . The act iv i ty o f C a O can a lso be d e d u c e d f r o m the in format ion avai lable in the l iterature. T h e s e data are p resented in f igure 5.1 to 5.3, but s ince their c o m p u t a t i o n i n v o l v e s certain a s s u m p t i o n s , they are not very rel iable in certain c a s e s . In the C a F 2 - C a O binary s y s t e m , the act iv i ty of C a O at 1 4 0 0 ° C is c o m p u t e d by the G i b b s - D u h e m Integration m e t h o d (figure 5.1). The c r y o s c o p i c exper imental data of Ko j ima and M a s s o n 6 3 are used and the s y s t e m is a s s u m e d to f o l l o w regular so lu t ion behav ior . The exper imental 106 Tab le 5.1 A c t i v i t y of C a O in C a F , - C a O s y s t e m at 1 4 0 0 ° C CaO CaO ' CaO 0.136 0.24 1.77 0.138 0.25 1.82 0.17 0.29 1.76 0.19 0.38 1.99 Table 5.2 A c t i v i t y of C a O in C a F , - C a O - A U O T s y s t e m at 1 4 0 0 ° C N CaO CaO CaO CaO A l 2 0 3 0.403 0.05 0.13 3.05 0.47 0.09 0.183 3.14 0.52 0.09 0.19 3.08 0.56 0.1 0.18 3.1 0.22 0.08 0.386 3.9 0.38 0.17 0.448 4.99 0.41 0.18 0.439 6.27 0.32 0.22 0.687 7.02 0.26 0.24 0.92 7.67 107 Table 5.3 Activity of CaO in CaF, - CaO -SiO, system at 1400°C CaO ECaO YCa0 CaO/N, 0.62 0.11 0.18 3.3 0.52 0.17 0.33 3.02 0.37 0.18 0.48 3.1 0.22 0.13 0.60 3.4 0.51 0,013 0.025 2 0.45 0.019 0.041 2 0.38 0.098 0.26 2.3 0.26 0.03 0.102 2.9 108 1.0 0.9 0.8 0.7 o 0.6 o o 0.5 •t— > OA < 0.3 02 0.1 0 I 1— 2, 1 : calculated from lit. (63) 2 : Edmunds and Taylor (67) 0 0.1 0.2 0.3 OA 05 0.6 0.7 0.8 0.9 1.0 Mole fraction. CaO Fig. 5.1 CaO activity in the CaO - CaF2 -binary slag system 109 0.8 c5 0 6 ocr c o 0.4 0.2 Legend A calculated for C/A=3(94) x from literature (67) • calculated for C/A=1(94) B from literature (67) s^ X^ A / / / I 1 i r 1— 0 0.2 0.4 0.6 0.8 mole fraction, CaO Fig. 5J2 CaO activity in the CaO - A l s 0 3 - CaF2 ternary slag 110 0.12 0.10-0.08-C (3 0.06-D 0.04-0.02-0.00 - r 0 x / / x7 / / / / Legend A c a l c u l a t e d for C/S=3(94) x f r o m l i te ra ture (72) • c o l c u l o t e d for C/S=1(94) B f r o m l i tera ture (72) g 3 i 0.2 0.4 0.6 0.8 mole traction, CaO Fig. 5.3 CaO activity in the CaO -Si0 2- CaF2 ternary slag 111 data of E d m u n d s and T a y l o r 6 7 are a l s o c o n s i d e r e d here. The equi l ibr ium they s tudied i s ; C a O + 3C = C a C j + C O 5.4 It is a lso a s s u m e d that C a C 2 is c o m p l e t e l y inso luble in the s l a g , but this is not true in f luor ide slag s y s t e m s . 8 6 The C a F 2 - C a C 2 s y s t e m is reported to f o r m a eutect ic at 1 2 4 0 ° C and N = 0.14. T h u s , a s s u m i n g unit act iv i ty C a C 2 for C a C 2 in the above equi l ibr ium react ion w o u l d result in higher values for l ime act iv i ty data. A s imi lar argument is app l icab le in the determinat ion of s i l i ca act iv i ty in the C a F 2 - C a O - S i 0 2 s y s t e m by the f o l l o w i n g e q u i l i b r i u m 7 7 : S i 0 2 + 3C = S i C + 2 C O 5.5 E v e n though the so lub i l i ty of S i C in C a F 2 - s l a g is highly unl ike ly , the ava i lab le C a O wou ld react to f o r m C a C 2 and make the s lag more than a 3 c o m p o n e n t s y s t e m . Hence the act iv i ty m e a s u r e m e n t s obta ined by measur ing the P w o u l d be in error . H o w e v e r , the ac t iv i ty data s o obta ined can sti l l be c o n s i d e r e d for c o m p a r i s o n . In the C a F 2 - C a O - A I 2 0 3 ternary s y s t e m , C a O act iv i ty is ca lcula ted f r o m the data of the C a O - A I 2 0 3 b inary by a s s u m i n g that C a F 2 is an inert diluent (figure 5.2). These data m a y not be very rel iable because of the c o m p l e x i t y of the aluminate s y s t e m conta in ing f luor ide ions , as o b s e r v e d by Hara and O g i n o . 8 7 H o w e v e r , the data can serve as a re ference for c o m p a r i s o n with the experimental data and the data f r o m E d m u n d s and T a y l o r . 6 7 112 The experimental data s h o w a substant ia l negat ive dev ia t ion e v e n at a molar ratio of C a O / A I 2 0 3 = 3 and approach ideal i ty on ly at higher molar rat ios (table 5.2). The ca lcu la ted data and the va lues f r o m the l i terature 6 7 (figure 5.2) s h o w a s imi lar trend but the ca lcula ted data approaches ideal i ty at a ratio lower than 3. W h e n the molar ratio is 3, both of these data s h o w a pos i t i ve dev ia t ion . S i n c e the ionic radii of o x y g e n (1.4A) and f luorine (1.36A) an ions are s imi la r , it can be p r e s u m e d that the f luor ine w o u l d gradual ly replace the o x y g e n f r o m an aluminate b o n d , wi th the p r o g r e s s i v e addit ion of C a F 2 . H o w e v e r , the structure is much m o r e c o m p l e x in C a F 2 saturated mel ts where two separate l iquid phases exist (figure 2.32, ternary diagram of C - A - F I ) . In the C a F 2 - C a O - S i O j ternary s y s t e m , the act iv i ty of C a O is ca lcu la ted f r o m the data of C a O - S i 0 2 b inary , a s s u m i n g C a F 2 is an inert diluent (figure 5.3). The s imi lar behav ior of C a F 2 in this ternary is a l s o o b s e r v e d by Sh inmei et a l . 8 8 and Hara et a l . . 8 9 S o m m e r v i l l e and K a y 7 2 studied the reaction(5.5) to determine the s i l i ca ac t iv i ty . The m o d i f i e d G i b b s - D u h e m equat ion is used to calculate the l ime act iv i ty in this ternary s y s t e m . A l l these act iv i ty data s h o w a substant ia l negat ive dev ia t ion at all molar rat ios of C a O / S i 0 2 . A l s o the inert diluent character is t ics of C a F 2 in the s i l icate melt are evident in the exper imental data. 5.1.2 V E R I F I C A T I O N OF C A R B O N A T E EQUIL IBRIUM In order to val idate the a s s u m p t i o n that the carbonate ions behave regular ly in s lag so lu t ion , s imi lar carbonate equi l ibr ia exper iments are carr ied out at 1 5 0 0 ° C , in the binary s y s t e m s with k n o w n oxide ion ac t iv i ty such a s , C a O - A I 2 0 3 and C a O - S i 0 2 . The results obta ined in these exper iments are s h o w n in table 5.4. The act iv i ty c o e f f i c i e n t o f the 113 carbonate ion s h o w s apparent regular s o l u t i o n behavior in the binary a luminate and s i l icate s l a g s , but the ac t iv i ty c o e f f i c i e n t is a d i f ferent order o f magni tude in di f ferent t y p e s of s l a g s , be ing smal ler in the a luminate s l a g s . The act iv i ty c o e f f i c i e n t in f luor ide s lag is be tween the values obta ined in the aluminate and s i l icate s l a g s . 5.1.3 A C T I V I T Y C A L C U L A T I O N F R O M T H E S U L F I D E C A P A C I T Y D A T A The ca lculat ion of the act iv i ty of C a O f r o m the su l f ide capac i ty d a t a " in the C a F 2 - C a 0 - A I 2 0 3 s y s t e m can be done a s s u m i n g the regular so lu t ion behavior of C a S . The resul ts in table 5.5 are c o m p a r e d with the exper imenta l ly de termined v a l u e s . It is c lear ly ev ident that the d i s c r e p a n c y which exists at lower N / N rat ios gradual ly d i m i n i s h e s at higher ra t ios . In this c a s e , the reason m a y be that the a s s u m p t i o n of the regular so lu t ion behavior of C a S in this melt is not c o m p l e t e l y va l id . Th is argument can further be just i f ied after compar ing the ionic radii of O 2 ( 1 . 4 0 A ) and S 2 ( 1 . 8 4 A ) and their e lec t ron ic conf igura t ion and the subsequent var iat ion of 7 Q 3 Q . o b s e r v e d in the C a F 2 - C a 0 - A I 2 0 3 s y s t e m (table 5.2). The d i f f e rences in act iv i ty data a l s o occur because s o m e of the p rev ious studies are based on the initial s lag c o m p o s i t i o n , and do not a l low for changes during the equi l ibrat ion. In the case of f luor ide s l a g s , substant ia l c o m p o s i t i o n changes occur during the course of an exper iment because of the presence of volat i le s p e c i e s such as C a F 2 , S i F , , A I F 3 e t c ; there fore , all high temperature equi l ibr ium exper iments concern ing f l u o r i d e - b a s e d s lags require f inal ana lys is of the mel ts to determine more accurate t h e r m o d y n a m i c data. The present work is cons is ten t in this w a y . The o n l y p r o b l e m in this technique is to extract the oxide ion act iv i ty at a 114 Table 5.4 V a r i a t i o n of Y r n 2 - i n B i n a r y Alumina te C 0 3 and S i l i c a t e Slags a t 1500 C Sl a g No. X C B Q a C a Q N c yQ0 CA-1 0.58 0.2 4.57 x 10 0.12 CA-2 0.65 0.4 5.798 x 10~ 3 0.19 CA-3 0.68 0.7 9.2 x 10~ 3 0.21 CS-1 0.57 0.02 6.28 x IO" 3 8.8 x 10 CS-2 0.5 0.005 2.74 x 10~ 3 5.1 x 10 CS-3 0.4 0.003 3.49 x 10~ 3 2.4 x io' 115 Comparison of Experimental Activity of CaO with calculated one Obtained from Sulphide Capacity Measurements^ 5 ^  in CaJ^-CaO-A^O o Slags at 1400 C; y_ - 3.4 Experimental Calculated Slag No. NCaQ NCa0/NAU0, aCa0 aCaO 1 0.4 3.05 0.05 0.16 2 0.47 3.14 0.09 0.19 3 0.52 3.08 0.098 0.199 4 0.56 3.1 0.1 0.198 5 0.22 3.9 0.09 0.106 6 0.38 4.99 0.17 0.275 7 0.41 6.27 0.18 0.368 8 0.32 7.02 0.22 0.309 9 0.26 7.67 0.24 0.23 116 s p e c i f i c c o m p o s i t i o n of a s l a g , s ince it is d i f f icu l t to es t imate the f inal c o m p o s i t i o n f r o m the start ing material for a part icular equi l ibr ium exper iment . H o w e v e r , these act iv i ty data w o u l d be usefu l part icular ly for a region where act iv i t ies are not very c o m p o s i t i o n s e n s i t i v e . 5.1.4 I M P O R T A N C E O F THE C A R B O N A T E EQUIL IBRIUM The study of the carbonate equi l ib r ium, in spi te of the fact that it d o e s not have the direct impl ica t ion of s tudies re lated to P and S , can be corre la ted to s o m e important t h e r m o d y n a m i c p roper t i es . A l r e a d y , it has been s h o w n that the data on carbonate equi l ibr ium can be used to obta in the oxide act iv i ty . o f the s l a g , w h i c h , in a w a y , ind icates the behav ior of the s lag with respect to d i f ferent c h e m i c a l r e a c t i o n s . A l s o , c o m p a r e d to other equi l ibr ia , such as those of su l f ide and p h o s p h a t e , the carbonate equi l ibr ium is much more s i m p l i f i e d in nature s ince at unit a tmospher ic pressure of C 0 2 , we have to c o n s i d e r on ly the carbonate ion act iv i ty and the react ion equi l ibr ium constant to determine the oxide ac t iv i ty . M o r e o v e r , it is sa fe to a s s u m e regular s o l u t i o n behav ior of the carbonate ion because of its large ionic size and low so lub i l i ty in many s lag s y s t e m s . A s a c o n s e q u e n c e , the major o b s t a c l e in s tudy ing this equi l ibr ium is the accurate detect ion of carbonate in the s l a g . The exper imental des ign should be done in such a w a y that the m i n i m u m C 0 2 e v o l v e d can be ana lyzed accurate ly in an apparatus operat ing within to lerable exper imental error. The partial molar heat of d i s s o l u t i o n of ox ide can a lso be c o m p u t e d if the c o r r e s p o n d i n g act iv i ty data are avai lable over a temperature range. H o w e v e r , in the present c a s e , due to the const ra in ts of the operat ing temperature range of the apparatus and the l imi ted range of 117 l iquid phase reg ion of the s lag s y s t e m s s t u d i e d , this t h e r m o d y n a m i c parameter can not be c o m p u t e d . 5.1.4.1 C A R B O N A T E C A P A C I T Y A N D B A S I C I T Y OF S L A G The carbonate equi l ibr ium a lso can be e m p l o y e d to calculate carbonate capac i ty of a s l a g , which is de f ined a s : N coi-C ^ = 5.6 carb p „ ^ C 0 2 W a g n e r 7 6 p r o p o s e d that the bas ic i ty of s l a g s , invo lv ing 0 J - ions as reactants , can be d e f i n e d as( same as in equat ion 2.32): C B carb carb = — 5.7 C carb where C is the carbonate capac i ty in a re fe rence s l a g . He a lso carb s u g g e s t e d that the s l a g , 0 . 4 C a O + 0 . 4 S i O 2 + 0 .2A I 2 O 3 , be d e e m e d as a re ference s l a g . In a recent invest iga t ion , S o s i n s k y et a l . 7 7 c o n c l u d e d on the bas is of opt ica l bas ic i ty that this s lag w o u l d have too low a carbonate capac i ty to be c o n s i d e r e d as a reference s l a g . Unfor tunate ly , there is no direct exper imental e v i d e n c e to ver i fy this and to se lec t s o m e other s lag as a re fe rence to d e f i n e bas ic i ty accord ing to the equat ion (5.7). fn such a c a s e , carbonate c a p a c i t i e s o f di f ferent s lags can be c o m p a r e d and their var ia t ion with the change of c o m p o s i t i o n p lo t ted (figure 5.4 ). S i n c e the su l f ide c a p a c i t y 9 0 is a l s o a s imi lar funct ion of oxide ion ac t iv i ty , the change in carbonate c a p a c i t y with respect to the c o m p o s i t i o n s h o w s a s imi lar t rend. The carbonate capac i ty is the highest in the binary C a O - C a F 2 s l a g , and the addi t ion of C a F 2 in the s i l icate and aluminate s lags increases the carbonate c a p a c i t y . S imi la r e f f e c t s were rea l ised in the su l f ide capac i ty inves t iga t ions . 118 0-Legend A CoO-CoT. -1- x CoO-Cof.-AI.O. O CoO-AI.O. / • Co0-S»0. -2- / U log ( / -s- a 9 I -4- / s-( 3 0.2 0.4 0.6 0.6 1 mole frocfion bose -2-o o o O -3--4--5-Legend CoO - C o f . CoO-AI.O. CoO-SiO. CoO-A I .O . - C o r , CoO-S iO j -CoF , —I— 0.2 - 1 — 0.4 - I — 0.6 -I— o.e mole fraction base Fig. 5.4 Comparison of carbonate capacity and sulphide capacity 9 0 of silicate, aluminate and fluoride slags 119 F r o m these observa t ions on carbonate and su l f ide c a p a c i t i e s , it can be sa id that the addi t ion of C a F 2 o p e n s up the structure of s l a g , resul t ing in the greater capac i t ies of S 2 _ and COf - ions in the s l a g . A s ment ioned b e f o r e , carbonate equi l ibr ium data can be ut i l ized to deduce the oxide ion act iv i ty of the s l a g . S o m e a u t h o r s 9 1 have a l s o tr ied to determine oxide ion act iv i ty f r o m the E M F measurements of a concent ra t ion ce l l . S ince m o s t s lags do not s h o w ideal behav ior , the ca t ions cannot be a s s u m e d to have constant ac t iv i ty . C o n s e q u e n t l y , the re lat ion as f o l l o w s is not v a l i d : a " O 5 " a " ^ — = a a O 2 - N a 2 0 where a " and a ' are the oxide ion act iv i t ies of s lags p laced in a O 2 - O 2 " concent ra t ion c e l l , a " and a' are the act iv i t ies of N a , 0 of the N a 2 0 l s l a 2 0 respec t ive s lag in the s a m e ce l l . The a b o v e equat ion can be true o n l y when <* N a + = <^ a +- G o t o et a l . 9 2 p e r f o r m e d both the carbonate equi l ibr ium and E M F measurements on the N a 2 0 - S i 0 2 s l a g . T h e s e data can be e m p l o y e d to calculate a ^ 2 and ^ (table 5.6) and the d i s c r e p a n c y b e c o m e s evident as the concent ra t ion c h a n g e s . The contr ibut ion f r o m any irregular so lu t ion behavior of carbonate i o n , even if there is any , w o u l d be much smal le r than the d i f f e r e n c e s o b s e r v e d in the values of a and a , ^ as a result O 2 " ^s la 2 0 o f the var iat ion in the ca t ion act iv i ty . Rewrit ing the express ion for ox ide ion act iv i ty f r o m the equation(5.3): 120 T a b l e 5.6 V a r i a t i o n o f Oxide Ion A c t i v i t y i n Na^O-SlO^ S l a g s a t 1373K o b t a i n e d by Carbonate E q u i l i b r i u m and 92 EMF Measurements Carbonate EMF E q u i l i b r i u m , Experiment, XNa O l o g X C 0 2 XC0*" A c t i v i t y , 0*' A c t i v i t y , Na.,0 0.5 -1.93 0.012 1.00X10" 6 l . O O x l O " 6 0.6 -0.9095 0.123 1.03X10" 6 3.20X10" 5 0.7 -0.592 0.256 2.10X10" 5 i . o o x i o " 3 0.8 -0.456 0. 350 2.90X10" 5 0.01 121 or . N 7 C 0 J " C O 2 -l o 9 a Q 2 - = | Q 9 3 + l o 9 j ^ - 5 - 1 0 C 0 2 Here the term N / P r n ' s equivalent to the carbide capac i ty (C ) and 7 /K is a constant for a s p e c i f i c s lag s y s t e m , thus \* \J 3 log a = log C ^ + Q i 5.11 0 2 " carb where Q = I 0 9 ( 7 C 0 | . / K ) T h e r e f o r e , the plot of l o g a versus logC . w o u l d be a straight line with O 2 " carb unit s l o p e and the intercept Q is equal to - l o g C . at unit ox ide ion carb ac t iv i ty . The carbonate capac i ty and the oxide act iv i ty data in C a O - C a F 2 , C a O - A I 2 0 3 - C a F 2 and C a O - S i 0 2 - C a F 2 at 1 4 0 0 ° C are p lo t ted in f igure 5,5. The s l o p e of the line is very c l o s e to unity and f r o m the intercept 7 r n 2 . is ca lcu la ted to be 0.043. In a s imi lar e x e r c i s e , M a e d a et a l . 9 3 p lo t ted the act iv i ty of N a 2 0 at 1 2 0 0 ° C , and the graph s h o w s a s imi lar s l o p e (figure 5.6). The act iv i ty c o e f f i c i e n t of carbonate ion , 7 r n 2 , can a lso be ca lcu la ted here a s s u m i n g the only cat ion to be N a 4 , in which c a s e K is equal to 2.4x10 4 a t m - 1 . T h u s , 7 r r w . is found to be 15.1, represent ing a pos i t i ve dev ia t ion and a s igni f icant interact ion with the mel t , wh ich is not at all expected cons ider ing the big s ize of carbonate i o n s . In C a O - S i O j s l a g , 7 r n 2 is ca lcu la ted f r o m the C 0 2 so lub i l i ty w \J 3 0 - 0 . 5 -o cn o -1 Legend - C a O - C a F 2 x C a O - C a F 2 - A I 2 0 3 • CaO-CaF - S i O . / v x £ - 1 . 5 -•2-T-- 5 "TJ - 2 . 5 - 2 l o 9 c carb -1 .5 -1 F i g . 5.5 Relat ionship be tween carbonate c a p a c i t y and ox ide ion ac t iv i ty f luor ide s lag at 1 4 0 0 ° C 123 F i g . 5.6 Relat ionship between carbonate capac i ty and s o d i u m oxide act iv i ty in N a 2 0 - S i 0 3 s lag at 1 2 0 0 ° C 9 3 124 exper imenta l m e a s u r e m e n t s at 1 5 0 0 ° C . It is in the range of 2 x 1 0 ' to 9 x 1 C r \ 5.1.4.2 C A R B O N A T E C A P A C I T Y A N D O P T I C A L B A S I C I T Y In oxide s l a g s , opt ica l b a s i c i t y ( O B ) has been re lated to su l f ide capac i ty data ( - l ogC^) 7 4 and thus c o n s i d e r e d to be a parameter fo r clear character izat ion of a s l a g . A s s h o w n earlier,' the su l f ide capac i ty and the carbonate capac i ty exhibit s imi lar t rends in both f l u o r i d e - f r e e and f luor ide b a s e d s l a g s . It is expected that opt ica l bas ic i ty a lso can be cor re la ted with the carbonate capac i ty of d i f ferent s l a g s . A t this s tage , it is not p o s s i b l e to calculate opt ica l b a s i c i t y of C a F j - C a O s l a g . T h e r e f o r e , on ly ternary f luor ide s l a g s , such as C a F 2 - C a O - S i 0 2 and C a F 2 - C a O - A I 2 0 3 , are c o n s i d e r e d where C a F 2 is a s s u m e d to play an inert role with respect to opt ica l b a s i c i t y . The binary s i l ica te and aluminate s lags for which the carbonate equi l ibr ium data are ava i lab le are a lso included in this p lot . The f inal f igure re f lec ts an increas ing trend of opt ica l bas ic i ty with increase of carbonate capac i ty (figure 5.7). H o w e v e r , no s imp le linear re lat ionship can be der ived as o b s e r v e d in oxide s lags with respect to su l f ide capac i ty . It s e e m s that opt ica l bas ic i ty is not a very sens i t ive parameter in highly bas ic and f luor ide s l a g s , whereas oxide ion act iv i ty or carbonate capac i ty can be c o n s i d e r e d to be more sui table parameter to character ize a s l a g . 5.2. W A T E R S O L U B I L I T Y IN ESR S L A G S The newly d e v e l o p e d deuter ium tracer de tec t ion technique , as d e s c r i b e d in the p rev ious chapter , was e m p l o y e d to ana lyze water so lub i l i t y in all ESR s l a g s . A l l so lub i l i ty data vary f r o m about 9000 to 1000 p p m - 1 .6 -1 .8 - 2 - £ - 2 . 2 -O O O CD O - 2 . 4 -- 2 . 6 -- 2 . 8 --3 - , r 0 .60 0 .65 X A A X A X X A • A A A Legend A C a F 2 - C a O-AI 2 0 5 x C a F 2 - C a O - S ! 0 2 • C a O - A I 2 0 , H CaO-S iO . n r T 0.70 0 .75 0 . 8 0 0 .85 0 .90 0 .95 optical basicity F i g . 5.7 Re la t ionship between opt ica l b a s i c i t y and carbonate capac i ty binary s i l i c a t e , aluminate and ternary f luor ide s lags 126 and are much higher than those repor ted in the l iterature. S i n c e the use of this new technique of so lub i l i ty measurement is the major reason for the d i s c r e p a n c y , it is n e c e s s a r y to detail both the advantages and d r a w b a c k s of this n o v e l method of a n a l y s i s . 5.2.1 A D V A N T A G E S A N D D I S A D V A N T A G E S OF THE D E U T E R I U M T R A C E R D E T E C T I O N T E C H N I Q U E t. The major c o n c e r n for a regular water ana lys is of s lag is the c o n t a m i n a t i o n of the s a m p l e . In a normal s i tuat ion , if the s lag after s a m p l i n g is e x p o s e d to air, it a d s o r b s mo is tu re . The same p r o b l e m ar ises during the storage of the s a m p l e , thus requiring a m o i s t u r e ~ f r e e , inert a t m o s p h e r e in the storage area. H o w e v e r , in the case of deuter ium having natural abundance on ly 0.015%, one can be sure about the source of this i s o t o p e being ana lyzed . O n c e the s lag has been equi l ibrated with D 3 0 , there is no p o s s i b i l i t y of contaminat ing the s a m p l e with any measurable quantity of deuter ium. M o r e o v e r , the rel iabi l i ty of a n a l y s i s of a s p e c i e s in a m a s s s p e c t r o m e t e r depends on the reso lu t ion of the instrument. If there is no hel ium s o u r c e present in the s y s t e m , the separat ion of m a s s e s of H2 and D 2 (the percentage m a s s d i f fe rence is c o n s i d e r a b l e ) can eas i l y be done in a s i m p l i f i e d v e r s i o n of a m a s s s p e c t r o m e t e r . A l s o the ana lys is of deuter ium is much more re l iable , s ince there is no not iceab le blank cor rec t ion required for this measurement . Deuter ium is not a rad ioact ive i so tope like t r i t ium, there fore this ana lys is d o e s not require any spec ia l handl ing, loca t ion or l i c e n s e d p e r s o n n e l . 127 On the other hand, this technique ca l ls for spec ia l preparat ion of the ca l ibra t ion s a m p l e , to ensure that C a ( O D ) 2 is c o m p l e t e l y reduced to D 2 instead of D 2 0 . M o r e o v e r , the apparatus is des igned in such a w a y that the graphite c ruc ib le , wh ich heats up the s lag and c a p s u l e s , gradual ly b e c o m e s loaded with used s lag and s t e e l . A l s o , a luminum vapour d e p o s i t s with inc reased p ropor t ions around the s i l i ca tube. The c o m b i n e d e f fec t is the dev ia t ion of the apparatus f r o m ca l ibra t ion l eve l , but this b e c o m e s evident o n l y after analyz ing around 15 s a m p l e s . Thus the technique requires very f requent , t i m e - c o n s u m i n g and e laborate s e t t i n g - u p p r o c e d u r e s . 5.2.2 W A T E R SOLUBIL ITY IN B INARY A N D T E R N A R Y F L U O R I D E - B A S E D S L A G S Binary s lags conta in ing d i f ferent p ropor t ions of C a F 2 and C a O , which have been equi l ibrated at 1 4 0 0 ° C and water vapour p ressures of 31.8 torr and 92.5torr, are ana lyzed for water so lub i l i ty (table 5.7-5.8). The so lub i l i ty of water increases with the increas ing C a O content of the s lag and the water vapour partial p ressure . The so lub i l i ty level ranges f r o m about 2400 to 8700ppm. There are no data avai lable in the literature to c o m p a r e the so lub i l i ty levels for this t ype o f s l a g . H o w e v e r , by c o m p a r i n g the su l f ide capac i ty data and s imi lar ox ide ion behav ior , these s lags can be p r e s u m e d to have a max imum level of water so lub i l i ty . Ternary s lags of the type C a F 2 - C a O - A l 2 0 3 are a lso ana lyzed fo r water so lub i l i ty after p e r f o r m i n g the equi l ibr ium exper iment at 1 4 0 0 ° C and water vapour partial p r e s s u r e s of 31.8torr and 92.5 torr (table 5.9-5.10). In the ranges of c o m p o s i t i o n of these s l a g s , ana lyzed water so lub i l i t y var ies f r o m 1800 to 4200 p p m . The m a x i m u m solub i l i ty is obta ined at a higher partial pressure o f mois ture and at 0.21 m o l e f rac t ion 128 Table 5.7 Water Solubi l i t y In Binary Slags at 1400 C Slag No. (H 20), ppm i \ o " * l s torr V 9 2 ' 5 torr SL-1 .2400 + 120 2500 + 130 SL-2 2500 A600 SL-3 3100 3000 SL-4 5400 8700 Table 3.8 Chemical Analysis of Binary, { F l - C V . S I A G « Slag No. X CaF X CaO SL-1 88.1 n .9 SL-2 83.5 16.5 SL-3 81.5 18.5 SL-4 78.6 21.4 129 Table 5.9 Water Solubility In Ternary (Fl-CWA) Slags at 1400 C Slag No. (H20), ppm i torrf" \o " 92 *5 t o r r SL-6 3800 ± 200 4200 ±200 SL-7 1800 2300 SL-8 2900 3400 Table 5.10 Chemical Analysis of Ternary(Fl-C-A) Slags Slag No. Z CaF2 Z CaO Z A1203 NCaO NCa0/NAl203 SL-6 78.6 15.8 5.6 0.21 5.1 SL-7 62.8 24.7 12.5 0.32 3.6 SL-8 47.4 32.4 20.2 0.42 2.9 130 of C a O . The reason for such a high level of so lub i l i ty at a low C a O content is the ratio of C a O to A l 2 0 3 , wh ich is m a x i m u m at that c o m p o s i t i o n . Eight ternary s lags of the type C a F 2 - C a O - S i 0 2 are ana lyzed for water so lub i l i ty after they have been equi l ibrated at 1 4 0 0 ° C and 31.8 and 92.5 torr of water vapour p r e s s u r e s (table 5.11-5.12). T h e s e s lags exhibit m i n i m u m water so lubi l i ty c o m p a r e d to data for all the s lags a n a l y z e d . A l s o the solubi l i ty increases with the increase of N _ / N _ _ y y C a O S i 0 2 ratio and the partial pressure of water vapour . The m i n i m u m so lub i l i t y is obta ined at 0.31 mole f ract ion of C a O and the molar ratio of C a O / S i 0 2 at 2.2. P r e s u m a b l y , the oxide ion act iv i ty and the free avai lable 0 2~ ions are m i n i m u m at this condi t ion and they g ive the lowest level of water so lub i l i t y . In table 5.13, the water a n a l y s e s done by V E W ( A u s t r i a ) are c o m p a r e d with the present ly d e v e l o p e d i so tope tracer technique at U B C . The V E W m e t h o d ana lyzed water as H 2 0 as o p p o s e d to equivalent D 2 in the present m e t h o d . In binary s lags the agreement is better than that of C a F 2 b a s e d s i l ica te s l a g s . A l s o , the higher va lues of water so lub i l i ty in the V E W technique can be accounted for in binary s lags by c o n s i d e r i n g the greater p o s s i b i l i t y of adsorb ing moisture f r o m the a tmosphere in the s o l i d s lag s a m p l e s be fo re it has been encapsula ted in an evacuated tube. In the isotope tracer techn ique , "water" ana lyzed in the s l a g is D 2 0 . By c o n s i d e r i n g the t h e r m o d y n a m i c data on H 2 0 and D 2 0 and a s s u m i n g the s a m e degree of d i f fe rence for the t h e r m o d y n a m i c data on Ca (OH) 2 and C a ( O D ) 2 , it is p o s s i b l e to est imate the free energy of f o r m a t i o n of C a ( O D ) 2 . A t . 1673K this d i f fe rence due to i s o t o p i c var iat ion is on ly 223 c a l o r i e s . Thus for all pract ical p u r p o s e s , the measured D 2 0 so lub i l i ty can be regarded e Table 5.11 Water Solubility In Ternary (Fl-C-S)Slags at 1400 C (H2Q), ppm Slag No. PH 0 - 31.8 torr % 0 " 9 2 , 5 t o r r 2 2 SL-9 430Q ± 200 6200 SL-10 3000 4200 SL-11 4700 6800 SL-12 2000 2500 SL-13 1900 3800 SL-14 1000. 1800 SL-15 1500 2000 SL-16 1900 2800 132 Table 5.12 Chemical Analysis Ternary (Fl-C-S) Slags Slag No. X CaF 2 X CaO X S i 0 2 HQaQ ^ C a 0 ^ { SL-9 79.0 17 A 0.22 A.6 SL-10 64.A 26.8 8.8 0.33 3.3 SL-11 A8.1 38.8 13.1 0.A5 3.2 SL-12 A1.8 39.A 18.8 0.45 2.2 SL-13 81.3 12.9 5.8 0.17 2.A SL-1A 62.1 25.5 12.A 0.31 2.2 SL-15 45.1 36.2 18.7 0.A2 2.1 SL-16 35 A2.7 22.3 0.A8 2.1 133 T a b l e 5.13 Comparison of the Pr e s e n t A n a l y s i s w i t h the  VEW-Analysis S l a g P r e s e n t Work VEW, A u s t r i a ( H 2 0 ) , ppm (H 2 0 ) , ppm SL - 4a 5400 5500 SL - 4b 8700 9100 SL - 16a 1900 1000 SL - 16b 2800 1200 Note: a: For p„ = 31.8 t o r r H 20 b: F o r p„ _ = 92.5 t o r r H 2 ° 134 as the H 2 0 solubi l i ty in the s l a g . In light of the a b o v e c o m p a r i s o n , the main reason for the d i f f e r e n c e s of water solubi l i ty data in the literature and the va lues obta ined in this work can be attributed to the s a m p l i n g m e t h o d e m p l o y e d . The normal method of using a metal rod for f reez ing out s o m e l iquid s lag m a y result in substant ia l loss of mois ture during its c o o l i n g per iod for w a t e r - s a t u r a t e d s l a g s . In the present s a m p l i n g technique , this p o s s i b i l i t y d o e s not exist and consequent ly it is c o n s i d e r e d to be a better s a m p l i n g technique for s l a g s . 5.2.3 W A T E R SOLUBIL ITY A N D W A T E R V A P O U R P R E S S U R E IN ESR Even though the so lub i l i t y data are higher by an order of magni tude , the straight line re la t ionship with the square root of water vapour pressure is obta ined (figure 5.8). Th is S iever t 's law re la t ionship a l s o s h o w s the fact that the binary s lags have the highest water so lub i l i t y and the C a F 2 - C a O - S i 0 2 s lags have the least . W a l s h et a l . ' s u g g e s t e d that this type of re lat ionship ex is ts b e c a u s e of either of the f o l l o w i n g r e a c t i o n s : {H20} + ( 0 J ) = 2 ( O H ) 5.12 {H20} = 2(H*) + ( O 2 ) 5.13 H o w e v e r , the s e c o n d react ion can be ruled out in this c a s e because of the f inding that the so lub i l i t y increases with the increase of ox ide ion ac t iv i ty . Both these re la t ionsh ips can on ly be s a t i s f i e d with the f o r m e r equi l ibr ium react ion which w o u l d cont ro l the water so lub i l i t y level in the s l a g . 135 F i g . 5.8 So lub i l i t y of water in F l u o r i d e - b a s e d s lags as a func t ion o f water vapour pressure 136 The water vapour pressure c o n s i d e r e d here is the equi l ibr ium water vapour pressure generated by saturat ing the gas s t ream at d i f ferent w a t e r - b a t h temperatures . H o w e v e r , this gas is further heated up be fo re react ing with the s lag at 1 4 0 0 ° C . The error due to c o r r e s p o n d i n g d i s s o c i a t i o n of water vapour at high temperature can be c h e c k e d by c o n s i d e r i n g the equi l ibr ium d i s s o c i a t i o n r e a c t i o n : H 2 0 = H 2 + 1/2 0 2 5.14 A t 1 4 0 0 ° C , s ince the equi l ibr ium constant is on ly 1.6x10' 5 a t m 1 ' 2 , the m a x i m u m pressure of p . p ° ' 5 is 7x10" 7 when p is equal to 0.04 a tm. H 2 0 2 H 2 0 C o n s e q u e n t l y , the error invo lved in a s s u m i n g that the equi l ibr ium water vapour pressure of the constant temperature bath remains unchanged on top o f the s l a g , is neg l ig ib le . H o w e v e r , the va lues w o u l d have been d i f ferent for a s i tuat ion where a m o l y b d e n u m or graphite crucib le is u s e d to hold the s l a g . In this c a s e , the oxygen potent ia l is g o v e r n e d by the ox ida t ion potent ia l o f the c ruc ib le , result ing in a substant ia l d i s s o c i a t i o n of water vapour . 5.2.4 T H E R M O D Y N A M I C A N A L Y S I S OF T H E W A T E R S O L U B I L I T Y D A T A In highly bas ic s l a g s , o x y g e n is present as f ree O 2 - i o n s . The water vapour s lag react ion in this type of s lag can thus be represented by equation(5.12). The c o r r e s p o n d i n g equi l ibr ium c o n s t a n t , K^y^ i s : y d = 5.15 p a %0 O 2 -•y2 N 2 ' O H - ' O H -H H , 0 O 2 -137 The free energy change for the above react ion invo lv ing only C a 2 * ca t ion is g iven b y 5 4 : AG 0 = -27868 + 75.6T - 6.14TlnT c a l / m o l e 5.16 hyd T h u s , at a particular temperature the equi l ibr ium cons tan t , K^y^, c a n D e ca lcu la ted by the re la t ionship: - A G h y d / R T A t a particular water vapour p ressure , p^ ^ , and k n o w n , as de te rmined by carbonate equi l ibr ium, the water so lub i l i t y data can be used to es t imate 7 , the act iv i ty c o e f f i c i e n t of O H - i o n . In binary s lags it is OH" ca lcu la ted to be in the range of 0.2-0.3, whereas in ternary s lags it is f o u n d to vary f r o m 0.07-0.09 (table 5.14). Unfor tuna te ly , there are no data ava i lab le in the literature to compare with these f i n d i n g s . In the l iterature, the i o n - o x y g e n at t ract ion, I, is c o n s i d e r e d to be an important parameter in cont ro l l ing water so lub i l i t y in d i f ferent oxide s l a g s . A more general ized term w o u l d be the c a t i o n - a n i o n at t ract ion, A , w h i c h c o u l d be de f ined in the s a m e f a s h i o n i.e A = Z , Z 2 / a 2 , where Zx is the an ion v a l e n c y , Z 2 is the cat ion va lency and 'a ' is the inter ionic d is tance . In table 5.15 this parameter A has been calculated for d i f ferent sal ts which are of impor tance in ESR slag s y s t e m s . The addi t ion of A l 2 0 3 in C a O -C a F 2 , which resul ts in a lower water so lub i l i t y , can be expla ined by this parameter , A , wh ich inf luences the so lub i l i ty in an inverse re la t ionship . On this b a s i s , sal ts conta in ing s i l i c o n s h o w the max imum va lues o f A and thus this parameter c o n f o r m s with the o b s e r v a t i o n that the s i l ica te s lags exhibit m i n i m u m water so lub i l i t y . 138 T a b l e 5.14 A c t i v i t y C o e f f i c i e n t o f H y d r o x y l I o n . a t 1673K i n F l u o r i d e S l a g s o b t a i n e d  from e x p e r i m e n t a l d a t a o f O x i d e i o n  A c t i v i t y and Water S o l u b i l i t y S l a g No. N C a 0 N C a o ' / N S i 0 1 < « H A 1 4 0 > N 0H- OH -S L - 1 0.16 B i n a r y 4.85x10 0 . 2 8 0 0 . 2 0 0 S L - 3 0.24 ' B i n a r y 6 . 2 8 X 1 0 " 3 1.000 0.280 S L - 6 0.21 5.1 ( A 1 2 0 3 ) 7 . 8 2 X 1 0 " 3 0.100 0.073 S L - 8 0.42 2.9 ( A 1 2 0 3 ) 5.80X10 - 3 0. 070 0 . 0 8 0 S L - 9 0.22 4 . 6 ( S i 0 2 ) 8.52X10 - 3 0 . 2 0 0 0. 093 S L - 11 0.45 3.2 ( S i 0 2 ) 8 . 3 7 x l 0 ~ 3 0.180 0. 090 SL - 12 0.45 2.2 ( S i 0 2 ) 3.49X10" 3 0.019 0. 070 T a b l e 5.15 C a t i o n - Anion A t t r a c t i o n i n V a r i o u s S a l t s S a l t C a t i o n - Anion A t t r a c t i o n A CaF 2 0.36 A1F 3 0.87 S i F 4 1.28 Ca(OH) 2 0.32 A1(0H) 3 0.73 CaO 0.70 A 1 2 0 3 1. 66 s i o 2 2.44 140 C o n s i d e r i n g the water so lub i l i t y react ion(5 . l2 ) aga in , the equi l ibr ium constant , c a n be wr i t ten as in equat ion 5.15. Tak ing logar i thms on both s ides and rearranging: log a = 2log(N /p °'5 ) + log ( 7 2 /K ) O 2 - v OH- H 2 O y w O H h y d 7 5.18 S imi la r to carbonate c a p a c i t y , the hydroxy l ion c a p a c i t y . C ^ ^ , can be d e f i n e d as N /p°J A l s o for a particular s y s t e m , 7 /K w o u l d be OH HjO O n -c o n s t a n t , s a y , B = log ( 7 0 H _ / K ) 5.19 T h u s , the oxide ion can be expressed a s : log <i = 2logC. + B 5.20 O 2 " hyd T h e r e f o r e , the logar i thmic plot of ox ide ion act iv i ty and hydroxy l ion c a p a c i t y w o u l d be a straight l ine. Unfor tuna te ly , the oxide ion act iv i ty as de te rmined by the carbonate equi l ibr ium d o e s not c o r r e s p o n d for all the s lags for which the hydroxy l ion capac i t i es have been d e t e r m i n e d . H o w e v e r , s o m e literature data and s o m e ext rapola t ion of the present data ass is t in m a k i n g the plot for di f ferent logC , in f luor ide based s lags (figure 5.9). hyd The uncerta ini t ies in oxide act iv i ty data make the plot m o r e tentat ive (or qual i tat ive!) . H o w e v e r , s o m e important features can be p o i n t e d out very c lea r ly . A l l the highly bas ic s lags such as C a O - C a F 2 , C a O - C a F 2 - S i 0 2 with N / N _ . _ >3 and C a O - C a F 2 - A l 2 0 3 s h o w d i f ferent straight l ines C a O S i 0 2 but of s imi lar s l o p e , suggest ing the opera t ion of the s a m e equi l ibr ium react ion with water vapour and s l a g . In a less bas ic s l a g such as C a O -C a F 2 - S i 0 2 where N _ / N - 2, the operat ing react ion might deviate as C a O S i 0 2 the s l o p e of the line changes . A l s o , 7 for a s p e c i f i c s l a g can be OH" c o n s i d e r e d to be a constan t , but it is a d i f ferent value for d i f ferent s lag 0 - 0 . 5 --1 o b cn o - 1 . 5 -- 2 -• 2 . 5 - r - 2 . 2 ^ E ^ B B^^B^B Legend A n-c x F I - C - A • F I -C -S ,Nc /Ns=3 B F I -C -S ,Nc /Ns=2 - 2 - 1 . 8 -1.6 -1.4 log C n y d -1 .2 F i g . 5.9 Relat ionship be tween h y d r o x y l c a p a c i t y and ox ide ion ac t iv i ty va r ious f luor ide s lags 142 systems as shown earlier. The above discussion reveals two important facts. The oxide ion activity and the activity coefficient of hydroxyl ion, 7 , are two Un-critical parameters for characterizing slags with respect to water solubility in ESR slags. This investigation resulted in some useful data for both oxide ion activity and 7 in some fluoride-based slags used in ESR. A more complete and comprehensive study of these two parameters would help us to estimate water solubility accurately in all ESR slags. 5.2.5 PREDICTION OF HYDROGEN LEVEL IN ESR INGOTS So far, the discussion has been mainly directed toward the determination and estimation of the hydrogen level in the slag. However, the final aim of the ESR process is to produce an ingot with a predictable and tolerable hydrogen content. As mentioned before, the hydrogen content in the ingot reaches a steady state value after about 30 minutes from starting the process. The ratio of hydrogen between the slag and the metal (L^) also attains a constant value around the same length of time. Thus the water solubility data can help to predict the final hydrogen content in the ingot, depending on the temperature of operation, gaseous atmosphere on top of the slag and the slag composition. The previous studies 1 ' 3 indicate that Lj_j changes from 2 to 10 depending on the slag basicity. However, in the present investigation, owing to the higher solubilities of water in fluoride slags, this ratio is higher by an order of magnitude. In that respect, the solubility data in CaF 2 - CaO - A l 2 0 3 slag can be used to modify figure 2.23, assuming that the hydrogen in the ingot is the same. The resulting figure (figure 5.10) shows that the L^ varies from about 20 to 45, as expected earlier. Thus, the exact values of and the available 143 50 500 40H E & 30-E x E ^ 20-to X 4 / Legend A Hs x Hm  • Hs/Hm 400 V 300 E CL D_ </f X 1-200 10 h 100 "X--x x x — x 0-+-0 10 20 30 40 50 time, min. 60 F i g . 5.10 The m o d i f i e d hydrogen equi l ibr ium between s lag and m e t a l , b a s e d o n the present work 144 data on water solubility in ESR slag would enable us to predict the final level of hydrogen in an ESR ingot. At this stage, the constraints of the present water analysis technique and the unreliability of the conventional sampling and analytical methods do not permit us to establish an on-line hydrogen monitoring system in an ESR operation. However, the experimental data obtained during this study can be utilized to estimate the limiting partial pressure of moisture in the gas phase. For instance, the production of a 25ton ESR ingot containing a maximum of 2ppm hydrogen would tolerate a maximum partial pressure of moisture of 8.6x10 3 atm, which is equivalent to a dew point of approximately 5 ° C . This assumes that 1 ton of 50 CaF 2 -25 CaO -25 A l 2 0 3 slag is employed during the process and that an equilibrium between the metal and the atmosphere is maintained at 1600°C, for which is taken as 45. The corresponding values for 7 and a ^ 2 are considered to be 0.08 and 0.05 respectively. Again, considering the above case, operating with some protective atmosphere, the starting slag should not contain more than 8l0ppm water. Unfortunately, the problem of hydrogen is most severe at the initial period of operation, where the value is also substantially low. Taking, for example, the figure 5.10, during the starting period is around 20 and thus the maximum allowable moisture level in the starting slag is only 360ppm. The commercial slag, such as S2022 (product of Wacker-Chemie), reported to contain 500ppm of H 2 0, is unsuitable for the production of low hydrogen ESR ingot. This water level also corresponds to the analysis done at 650°C and is thus definitely not the absolute level of water in that slag. 145 Thus the present study not on ly l o o k s into the fundamenta ls of h y d r o g e n transfer during ESR operat ion but a l s o s e r v e s the funct ion of es t imat ing the l imit ing parameters to produce an acceptab le ESR ingot . CHAPTER 6 SUMMARY 6.1. C O N C L U S I O N S The c o n c l u s i o n s that can be drawn f r o m the present work are s u m m a r i z e d b e l o w . The oxide ion ac t iv i ty in the f luor ide based binary and ternary s lags has been determined and c o m p a r e d with the avai lable literature data. The carbonate equi l ibr ium is carr ied out by equi l ibrat ing the s lag with C 0 2 . It is not on ly an independent and rel iable technique but is a l s o helpful in ca lcula t ing the carbonate capac i ty of d i f ferent s l a g s . In bas ic ESR s l a g s , carbonate capac i ty exhibits a s imi lar re la t ionship wi th the s lag c o m p o s i t i o n , as p r e v i o u s l y repor ted , wi th respect to su l f ide c a p a c i t y . H o w e v e r , there is no c lear re la t ionship estab l ished b e t w e e n the carbonate capac i ty and the opt ica l b a s i c i t y . The latter parameter s e e m s to be less sens i t ive in f l u o r i d e - b a s e d s l a g s , most p r o b a b l y due to the inherent l imi tat ions o f the opt ica l bas ic i ty in character iz ing n o n - o x i d e s l a g s . The water v a p o u r - s l a g equi l ibr ium is s tudied in s imi lar s lags at 1 4 0 0 ° C at t w o di f ferent partial pressures of mo is ture . The saturated D 2 0 vapour in the s t ream of pure hel ium is used for the equi l ibra t ion . A l l s lags are s a m p l e d by a new techn ique , sampl ing in a quartz tube , wh ich retains the equi l ibr ium water level in the s l a g . 146 147 A new technique , e m p l o y i n g a s i m p l e m a s s s p e c t r o m e t e r , has been d e v e l o p e d to analyze water in the s l a g . Th is m e t h o d is better and more rel iable than those repor ted in the l i terature, but is more c o m p l e x . It has been ver i f i ed by a n a l y s i n g standard n o n - f l u o r i d e s lags and c o m p a r i n g the resul ts of s imu l taneous a n a l y s e s of f luor ide s lags p e r f o r m e d both in our laboratory and at V E W , A u s t r i a , where a d i f fe rent , independent m e t h o d is u s e d . The water so lub i l i t y , de termined by this new iso tope tracer techn ique , s h o w s values an order of magnitude higher than those repor ted in the l iterature. The major r e a s o n s for these d i f f e r e n c e s are the better s a m p l i n g technique and the rel iable i s o t o p e tracer a n a l y s i s e m p l o y e d . H o w e v e r , the so lub i l i ty data s h o w the square root re la t ionship with water vapour p ressure , re f lec t ing the type of m e c h a n i s m operat ing in a bas ic s l a g . T h e s e data can be used together wi th literature in format ion to generate a m o d i f i e d equi l ibr ium ratio of h y d r o g e n be tween ESR s lag and an ingot . F ina l ly , these resul ts can be s u c c e s s f u l l y e m p l o y e d to est imate the l imi t ing cont ro l l ing parameters for the p roduc t ion of an ESR ingot with an acceptab le level of h y d r o g e n . 6.2. S U G G E S T I O N S FOR F U T U R E W O R K In light of the present f i n d i n g s , further s tudies in the f o l l o w i n g d i rect ions are ind icated . 1. The partial heat of so lu t ion of the carbonate ion in d i f ferent ESR s lags shou ld be de te rmined . There are no data on the temperature dependence o f the equi l ibr ium in f luor ide s l a g s . Thus exper iments on C 0 2 - s l a g equi l ibr ium at d i f ferent temperatures wi l l lead to 148 better understanding of carbonate solution thermodynamics. 2. The C 0 2 - s l a g equilibrium experiments should also be extended to other slags containing MgO and rare-earth oxides. The direct determination of carbonate capacity would help to better characterize different ESR slags and these data can further be translated to compute their desulfurizing potentials. 3. The data on solution thermodynamics of water in fluoride slags are sparse. Thus the water vapour-slag equilibrium experiments at different temperatures would be highly desirable not only for a better definition of hydrogen thermodynamics of ESR slag but also to determine an optimum temperature of operation of the ESR process. 4. The present water analysis technique is not yet suitable for on-l ine monitoring of hydrogen during an ESR process. Hence, the method has to be modified to meet the final g o a l — the on-line control of hydrogen during the ESR process. The high background level of hydrogen in the vacuum apparatus makes the direct analysis of water as hydrogen impractical. The analysis of water as H 2 0 or the isotope dilution method using deuterium are the two viable alternatives to be explored for direct monitoring of water in the slag. REFERENCES 1. Sch'urmann, E. et al .: A r c h . E isenh'ut tenwes, V o l . 52, A u g u s t 1981, pp 303 -310 . 2. L a t a s h , Yu.V. et al .: P r o c . 7th International C o n f e r e n c e on V a c u u m Meta l lu rgy , N o v e m b e r 1982, T o k y o , J a p a n , p p 1 2 3 7 - l 2 4 3 . 3. Nakamura , Y and Harash ima , K.: T e t s u - t o - H a g a n e , V o l . 63, A u g u s t 1977, pp 1235-1243. 4. Jaegar , H. et al.: " Invest igat ions regarding the cont ro l of H y d r o g e n and A l u m i n i u m content in E S R - i n g o t s ' , Boehler B r o s . & C o . Ltd (Austr ia) . 5. M a s u i , A . : P roc . 5th International S y m p . on E S R , Part 1, 1974, P i t tsburgh, pp 284-305 . 6. Holzgruber , W . et al .: S p e c i a l E lec t rom eta l lu rgy , 1976, pp 161-174. 7. P o c k l i n g t o n , D.N.: J . of Iron and S tee l Institute, V o l . 211, June 1973, pp 4 1 9 - 4 2 5 . 8. C h u i k o , N .M. et al .: Meta l lu rgy , Ju ly 1977, pp 2 1 - 2 2 . 9. W a l s h , J .H . , C h i p m a n , J . , K ing , T .B . and Grant N.J. : T r a n s . A I M E , V o l . 206, 1956, pp 1568-1576. 10. K o b a y a s h i , M . et al.: Spr inger S e r i e s in Chemica l P h y s i c s , V o l . 19 (S IMS3) , 1982, pp 361-364. 11. F o r n o , A . E . J . et al .: J . o f Iron and S tee l Institute, V o l . 209, D e c . 1971, pp 966 -968 . 12. M e d o v a r , B.I. et al .: S p e c i a l E l e c t r o m e t a l l u r g y , 1976, pp 101-117 . 13. Bek tu rsunov , S h . S h . et a l . : T rans la t ion f r o m Izvest. V U Z - C h e r n . Met. , S e p t . 1961, pp 4 4 - 5 3 . 14. S c h w e r d t f e g e r , K and Schuber t , G . : Me t . T rans . B, V o l . 8B , D e c . 1977, 149 150 pp 689-691 . 15. V a t o l i n , A . N . et al .: F i z . - K h i m . Iss led . Meta l lurg . P r o t s e s s o v , S v e r d l o v s k , V o l . 10, 1982, pp 144-152. 16. J a r o s l a v , S . et a l . : Zvaran ie , V o l . 26, M a y 1977, pp 149-152. 17. T. El G a m m a l and W a l a d a n , M. : S e c o n d International S y m p . on Metal lurgical S lags and F luxes , N o v . 1984, Lake T a h o e , N e v a d a , pp 453 -465 . 18. Zu l ian i , D.J. et al . : ISS T r a n s a c t i o n s , V o l . 1, 1982, pp 6 1 - 6 7 . 19. N iko l ' sk i i , V . S . et al .: Stee l in the U S S R , D e c . 1979, pp 619 -622 . 20. S a e n k o , V . Y a . et al .: P r o b l . S p e t s . E lec t rometa l l . , V o l . 17, 1982, pp 3 2 - 3 3 . 21. Rebrov , L.B. et al . : P rob l . S p e t s . E lec t rometa l l . , V o l . 11, 1979, pp 8 - 1 3 . 22. Iguchi, Y . et al .: T rans . Iron and Stee l Institute of J a p a n , V o l . 9, 1969, pp 189-195. 23. Hi tachi , Ltd. , J p n . T o k k y o Koho J P 82 32, 698 (CI. C22B9 /18 ) , 13 J u l y 1982. 24. La tash , Y u . V . et a l . : Izv. A k a d . Nauk S S S R , N o . 3, 1979, pp . 4 0 - 4 4 . 25. M a s u i , A . et al .: T e t s u - t o - H a g a n e , V o l . 63, 1977, pp 2181 -2190 . 26. N i c h o l l s , J . : Short Course in app l ica t ion of T h e r m o d y n a m i c s o f Pe t ro logy and Ore d e p o s i t s , Apr i l 1977, V a n c o u v e r , pp . 160-184. 27. K lohne , C . and N o w a c k , N.: A r c h . E isenh i i t tenwes , V o l . 52, J u l y 1981, pp. 271 -273 . 28. R u s s e l , L.E.: J . S o c . G l a s s T e c h . , V o l . 41 , 1957, pp. 3 0 4 - 3 1 7 T . 29. T o m l i n s o n , J .W.: J . S o c . G l a s s T e c h . , V o l . 40 , 1956, pp 2 5 - 3 1 T . 151 30. Kurkjian, C.R. and R u s s e l , L.E.: J . S o c . G l a s s T e c h . , V o l . 42, 1958, pp 130-144. 31. U y s , J . M . and K ing , T .B. : T rans . A I M E , V o l . 227, A p r i l 1963, pp 492 -500 . 32. S c h o l z e , H. et al .: A d v a n c e s in G l a s s T e c h n o l o g y , 6th International C o n f . G l a s s T e c h n o l o g y , W a s h i n g t o n , 1962, pp 230 -245 . 33. Fukush ima , T . et al.: T rans . I.S.I.J., V o l . 6, 1966, pp 19-26 . 34. Iguchi, y . et al .: S y m p . on C h e m i c a l Meta l lu rgy of Iron and S t e e l , ISI, B S C Corpora te Lab , U. of S h e f f i e l d , 1971, pp 2 8 - 3 0 . 35. F i n c h a m , C . J . B . and R ichardson , F.D.: P r o c . R o y . S o c . , A 2 2 3 , 1954, pp 4 0 - 6 2 . 36. S a c h d e v , P.L. et al .: Met . Trans. , V o l . 3, 1972, pp 1537-1543. 37. C o u t u r e s , J .P . et al . : C.R. S e a n c e s A c a d . S c i . , Ser . 2, V o l . 293, 1981, pp 1049-1052. 38. Wah ls te r , V . M . and Re iche l , H.H.: A r c h . Eisenh'uttenwes, V o l . 40, J a n . 1969, pp . 19-25 . 39. S c h w e r d t f e g e r , K. and Schuber t , H.G.: Me t . T rans . B, V o l . 9B, March 1978, pp 143-144. 40. C o u t u r e s , J .P . et al .: Rev. Int. Heutes T e m p e r . Refract . Fr., V o l . 17, 1980, pp . 351 -361 . 41. D a v i e s , M.W. and S p a s s o v , A . : J . of Iron and Stee l Institute, O c t . 1967, pp 1031-1033. 42. Imai, M . et al . : S tud ies in Meta l lu rgy , 1969, pp 6 6 - 7 5 . 43. Iguchi, Y . and Fuwa, T . : T rans I.S.I.J., V o l . 10, 1970, pp 2 9 - 3 5 . 44. S o s i n s k y , D .J . et al . : Me t . T r a n s . B, V o l . 16B, 1985, pp 6 1 - 6 6 . 152 45. Wah ls te r , V . M . and Hi lper, A . : Stahl und E i s e n , V o l . 89, 1969, pp 710 -716 . 46. N o v o k h a t s k i y , I.A. et al .: A k a d e m i i a Nauk. S S S R , Izvestia. , M e t a l l y , N o . 5, 1968, pp 22 -27 . 47. G i b a l a , R. and D e M i g l i o , D.S.: P r o c . of Third International C o n f . on E f fec t of Hydrogen on Behaviour of Mater ia ls , M o r a n , W y o m i n g , A u g u s t , 1980, pp 113-122. 48. B a g s h a w , T.: Proc on Third International S y m p . on E S R , P e n n s y l v a n i a , M e l l o n Institute, 1971, pp 183-211 . 49. Cou tures , J .P . and Peraudeau, G . : Rev . Int. Hautes T e m p . Refract . , V o l . 18, 1981, pp 321-346 . 50. B l o o m , H.: The C h e m i s t r y of M o l t e n S a l t s , W A . Benjamin Inc., 1967, p 23. 51. S c h m i d t , J . : A r c h . Eisenh'uttenwes, V o l . 46, 1975, pp . 711 -713 . 52. L l o y d , M.H. and Shanahan, C . E . A . : J . o f Iron and Stee l Institute, S e p t . 1973, pp 615 -621 . 53. Kal inyuk, N.N.: Industrial Lab. , V o l . 44, June 1977, pp 740 -742 . 54. Handbook of Chemis t ry and P h y s i c s : C R C Press . Inc., F lo r ida , 63rd E d . , 1982-1983. 55. Kor , G .J .W. and R i c h a r d s o n , F.D.: T r a n s . Met . S o c . A I M E , V o l . 245, F e b . 1969, pp 319-327 . 56. M i l l s , K.C. and Keene , B.J . : International Met . Rev. , N o . 1, 1981, pp 2 1 - 6 9 . 57. M i t c h e l l , A . : Canadian Met . Quart . , V o l . 20, 1981, pp 101-112. 58. Nafz iger , R.H.: High T e m p . S c i e n c e , N o . 5, 1973, pp 4 1 4 - 4 2 2 . 59. Sa l t , D.J. : P roc . 1st International S y m p . on E l e c t r o s l a g c o n s u m a b l e E lec t rode Remel t ing and C a s t i n g T e c h . , V o l . 1, Carneg ie M e l l o n U., 153 P i t tsburg , 1967, pp 1-29. 60. M i t c h e l l , A . et al .: J . Iron and Stee l Institute, A p r i l , 1970, p 407. 61. Mukher jee, J . : J . A m e r . C e r a m i c S o c . , V o l . 48, 1965, pp 2 1 0 - 2 1 3 . 62. Baak, T.: A c t a C h e m . S c a n d . , V o l . 8, 1954, p 1727. 63. K o j i m a , H. and M a s s o n , C.R.: C a n . Chem. , V o l . 47, 1969, pp 4221-4228 . f 64. R i e s , R. and S c h w e r d t f e g e r , K.: A r c h . Eisenh'uttenwes, V o l . 51, Apr i l 1980, pp . 123-129. 65. K u o , C.K. et al .: A c t a C h i m . S i n i c a , V o l . 30, 1964, pp 381 -385 . 66. Budn ikov , P.P. and T r e s v y a t s i i : D o k l . A k a d . Nauk. S S S R , V o l . 89, 1953, p 479. 67. E d m u n d s , D .M. and T a y l o r , J . : J . of Iron and Stee l Institute, V o l . 210, 1972, pp 280 -283 . 68. M i t c h e l l , A . and C a m e r o n , J . : M e t . Trans. , V o l . 2, D e c . 1971, pp 3361-3366 . 69. E v s e e v , P.P.: A u t o m . W e l d . (USSR) , V o l . 20, 1967, p 42. 70. Baak, T . and Olander , A . : A c t a C h e m . S c a n d . , V o l . 9, 1955, pp 1350-1354. 71. S h i n m e i , M . et al . : C a n a d . Met . Quart . , V o l . 22, 1983, pp 5 3 - 5 9 . 72. S o m m e r v i l l e , I.D. and Kay , D A . R . : Me t . Trans . V o l . 2, June 1971, pp 1727-1732. 73. S u i t o , H. and G a s k e l l , D.R.: Me t . T r a n s . B, V o l . 7B, D e c . 1976, pp 567 -575 . 74. S o m m e r v i l l e , I.D. and S o s i n s k y , D .J . : 2nd International C o n f . on Meta l lurgical S l a g s and F luxes , Lake T a h o e , N e v a d a , N o v . 1984, 154 T M S - A I M E , pp 1015-1025. 75. Nagata , K. and G o t o K.S.: T r a n s . Iron and Stee l Inst, o f J a p a n , V o l . 25, 1985, pp 2 0 4 - 2 1 1 . 76. W a g n e r , C . : Met . T rans . B, V o l . 6B, Sept . 1975, pp 4 0 5 - 4 0 9 . 77. S o s i n s k y et al .: Presentat ion at the 24th C IM C o n f . , V a n c o u v e r , A u g u s t 1985. 78. M u r a t o v , A . M . : U D C 669, 891787*546.16.669, 0.46, 584, 1973, pp 55 -56 . 79. Mehro t ra , G . M . et al .: A r c h . Eisenh'ut tenwes, V o l . 47, D e c . 1976, pp . 7 1 9 - 7 2 3 . 80. Hawk ins , R.J . et al .: J . of Iron and Stee l Institute, V o l . 209, A u g . 1971, pp 6 4 6 - 6 5 7 . 81. Z h d a n o v s k i , A A . : P rob . S p e t s . E lec t romet . , N o . 9, 1977, pp 114-119. 82. A l l i b e r t , M . and Chat i l l ion , C : C a n a d . Me t . Quart . , V o l . 18, 1979, pp 349 -354 . 83. R a s c h e v , Z . et al . : A r c h . E isenh'ut tenwes, V o l . 53, J a n . 1982, pp. 1-4. 84. B lander , M. : M o l t e n Salt C h e m i s t r y , In terscience, N e w Y o r k , 1964, pp 127-237. 85. K u b a s c h e w s k i , O . and A l c o c k , C .B. : Meta l lurg ica l T h e r m o c h e m i s t r y , 5th ed. , P e r g a m o n P r e s s , 1979, p379. 86. M i t c h e l l , A . : T r a n s . Met . S o c . of A I M E , V o l . 242, D e c . 1968, pp 2507-2511 . 87. Hara, S . and O g i n o , K.: C a n a d . Me t . Quart . , V o l . 20, J a n . 1981, pp 113-116. 88. M o r i n a g a , K. et al . : T M S paper s e l e c t i o n , A 7 9 - 1 8 , The Met . S o c . of A I M E , pp 1-13. 155 89. Hara , S . et al .: P r o c . of f irst International S y m p . on M o l t e n Salt C h e m i s t r y and T e c h n o l o g y , K y o t o , J a p a n , Apr i l 1983, pp 257 -260 . 90. R i c h a r d s o n , F.D.: P h y s i c a l Chemis t ry of Me l ts in Me ta l lu rgy , A c a d e m i c P r e s s , New York , 1974, V o l . 2, pp 291 -304 . 91. F r o h b e r g , M . G . et al .: A r c h . E isenhut tenwes , V o l . 44, A u g . 1973, pp . 585 -588 . 92. G o t o , K.S. et al . : S e c o n d International S y m p . of Met . S l a g s and F luxes , Lake T a h o e , N e v a d a , N o v . 1984, pp 4 6 7 - 4 8 1 . 93. M a e d a , M . et al . : Met . T rans . B, V o l . 16B, Sept . 1985, pp 561 -566 . 94. T u r k d o g a n , E.T.: P h y s i c a l Chemis t ry at High Temperature T e c h n o l o g y , A c a d e m i c P r e s s , N.Y., 1980, pp 131-133. 95. Fraser , M.E. and M i t c h e l l , A . : M e t a l - S l a g - G a s React ion P r o c e s s e s , E d . b y : F o r o u l i s , Z . A . and Smel tzer , W.W. , E l e c t r o c h e m . S o c . Inc., P r i n c e t o n , N.J . , 1975, pp 199-209. A P P E N D I X I 1.1. C H E M I C A L A N A L Y S I S OF S L A G 1.1.1 FLUORIDE A N A L Y S I S 100 mg of a dried and p o w d e r e d s a m p l e is w e i g h e d in a p la t inum cruc ib le . A l s o 500mg N a 2 C 0 3 and lOOmg Z n O are w e i g h e d in it and mixed thoroughly with the s a m p l e . The mixture is then f u s e d at 9 0 0 ° C for 30 minutes . The t ime and temperature are to be kept constant as they are cr i t ical to obtain proper f u s i o n and to keep f luorine l o s s to a m i n i m u m . The f u s e d product is d i s s o l v e d in a p las t ic beaker conta in ing 50ml d is t i l l ed water and 4ml c o n e . HCI. The so lu t ion is f i l tered and made up to 100ml in a vo lumet r ic f lask . Th is stock so lu t ion is di luted 10 t imes in water . 10ml o f this s o l u t i o n a long with the equal v o l u m e of buf fer is taken in a p las t ic beaker . The so lu t ion is st irred s l o w l y by a magnet ic stirrer and the potent ia l through the f luor ide ion e lec t rode is r e c o r d e d to the nearest O . l m V . A l l potent ia l measurements are done after 3 minutes of i m m e r s i o n of the e lec t rode to obtain c o n s i s t e n t l y s tab l ized E M F data. The s a m e m e t h o d is app l ied to measure potent ia ls of the s tandards . Both the temperature and the st i rr ing rate are mainta ined constant fo r all m e a s u r e m e n t s . The standard potent ia ls are p lo t ted on a s e m i l o g graph paper , keep ing the concent ra t ion in the l o g - s c a l e . The straight line s o obta ined 156 157 can be u s e d to determine f luor ide c o n c e n t r a t i o n in an unknown s a m p l e . The s l o p e of the line should be c l o s e to 56mV for each d e c a d e change of c o n c e n t r a t i o n s . B u f f e r : (1.0M s o d i u m citrate - 0.2M p o t a s i u m nitrate) 294 g m o f s o d i u m citrate and 20 gm of potas ium nitrate are d i s s o l v e d and di luted to make 1 liter s o l u t i o n in water . S t o c k F luor ide : 2.22 gm of dry , reagent grade NaF d i s s o l v e d and di luted to 1 liter wi th water . W o r k i n g S t a n d a r d s : The s tock f luor ide is di luted to make standards conta in ing 4, 16, 20 and 40 m g / L F luor ine . Appropr ia te amounts of N a 2 C 0 3 , Z n O , A l s tock so lut ion (if needed) and Si s tock s o l u t i o n (if needed) are added to maintain the same leve ls as in the s a m p l e s . 1.1.2 A N A L Y S I S FOR C A L C I U M , S IL ICON A N D A L U M I N U M Fresh graphite c ruc ib les ( s p e c - g r a d e graphite) are p r e - i g n i t e d at 1 0 0 0 ° C for 30 minutes. 50 mg p o w d e r e d sample and 300mg reagent grade l ithium metaborate are weighed and m i x e d thoroughly in a graphite c ruc ib le . The mixture is fused at 1 0 0 0 ° C for 15 minutes . The melt is sw i r l ed to c o a l e s c e and immedia te ly poured into a beaker conta in ing 25ml 8% H N 0 3 . The so lu t ion is st irred to d i s s o l v e c o m p l e t e l y and di luted to 50 ml with water in a vo lumet r ic f lask . S imi la r s t e p s are f o l l o w e d to make a blank conta in ing 300mg L i B 0 2 . 1.1.3 A T O M I C A B S O R P T I O N M E T H O D S S i l i c o n : S i l i c o n is ana lyzed f irst s i n c e it is m o s t s e n s i t i v e to l o s s e s . S tandards conta in ing 5, 10, 20, and 50 m g / L Si are made up. T h e y a lso conta in 100 m g / L C a , 4 % H N 0 3 and 0 .6%LiBO 2 . The s a m p l e s o l u t i o n s are run 158 in an A t o m i c A b s o r p t i o n S p e c t r o p h o t o m e t e r 'as i s ' in a ni trous o x i d e - a c e t y l e n e f l ame with the f i lament current at 40 m A and burner angle 0 ° . The s tandards are run at the beg inn ing and at the end to check the s tab i l i ty of ca l ibra t ion . The blank and a r e f e r e n c e so lut ion conta in ing known amounts of all the element are a lso a n a l y z e d . C a l c i u m : C a l c i u m standards of 5, 10, 15, and. 20 m g / L Ca are made by di lut ing one s i l i c o n s tandards . The s a m p l e s are di luted 1/50 and run in a f l ame of ni t rous o x i d e - a c e t y l e n e with the f i lament current at 1 0 m A and burner angle of 0 ° . The blank and a re fe rence so lut ion are a l s o tested during the run. A l u m i n u m : A l u m i n u m standards of 20, 60, 80 and 100 m g / L A l are made and these a l s o contain 100 m g / L C a , 4% H N 0 3 , 0.6% L i B 0 2 and 1% L a . S a m p l e s are di luted 1/2 to 1/5 to conta in 1% La and run in nitrous o x i d e - a c e t y l e n e f l ame with the f i lament current atv 18mA and burner angle at 0 ° . A P P E N D I X II 11.1. DEFINITION OF O P T I C A L B A S I C I T Y The concept of L e w i s bas ic i ty is related to the e lec t ron donor p o w e r of the oxygen present in a s y s t e m . H o w e v e r , there is no numer ica l s c a l e to def ine this parameter quant i tat ive ly . A n opt ica l sca le has been p r o p o s e d to measure the Lewis bas ic i ty of var ious oxide g lass s y s t e m s and s o m e mol ten s a l t s 1 . Th is opt ica l sca le of bas ic i ty is te rmed as opt ica l b a s i c i t y . When a trace quantity of p robe i o n s , such as T l " , P b 2 + or B i 3 \ are in t roduced in a s y s t e m , f requency s h i f t s , depend ing on the bas ic i ty are r e c o r d e d in their ultraviolet s - p s p e c t r a . The greater the degree of red sh i f t , greater is the bas ic i ty of the s y s t e m . The red shift can a lso be v i e w e d as a result of the expans ion of the outer s and p orbi ta ls of the probe ion . The opt ica l bas ic i ty of var ious l o w temperature oxide g l a s s e s (for example , N a 2 0 - B 2 0 3 and L i 2 0 - B 2 0 3 ) and sulf a t e - c h l o r i d e g l a s s e s is m e a s u r e d 1 ' 2 . H o w e v e r , in s o m e g lass s y s t e m s and meta l lurg ica l s l a g s , exper imenta l measurements are di f f icul t and s o m e t i m e s not f e a s i b l e due to o p a q u e n e s s in the ul traviolet reg ion . In these c a s e s , the theoret ical ca lcu la t ions of opt ical bas ic i ty p r o v e d to be a g o o d a l ternat ive 3 ' 4 . The next s e c t i o n s h o w s how to do these theoret ical ca lcu la t ions on the bas is of e lec t ronegat iv i ty of d i f ferent c a t i o n s , norma l ly present in meta l lurg ica l s l a g s . 159 160 11.2. C A L C U L A T I O N OF O P T I C A L B A S I C I T Y The opt ica l b a s i c i t y , O B , o f a s lag can be ca lcu la ted by the f o l l o w i n g e x p r e s s i o n : (OB) >s,ag = V ° B A + V ° B k where (OB) ^ and (OB) are the opt ica l bas ic i t i es of d i f ferent oxide c o m p o n e n t s in the s l a g , X is the equivalent cat ion f rac t ion which can be ca lcu la ted f r o m the mo le f ract ion of c o m p o n e n t s , such as in A - B s l a g : _ mo le f rac t ion of c o m p . A x no . of oxygen a t o m s in A ^ ^ A . Z m o l e f ract ion of c o m p . i x no . of o x y g e n in i i =a,b In CaFj b a s e d s l a g s , the cont r ibut ion f r o m C a F 2 is ignored , and thus on ly the oxide c o m p o n e n t s are c o n s i d e r e d . For e x a m p l e , in a C a F 2 - C a O - S i 0 2 mel t : S i O , = J N + 2N C a O S i O , The opt ical bas ic i ty of a pure oxide c o m p o n e n t is related to the Paul ing e lec t ronegat iv i ty , x, o f the ca t ion by the f o l l o w i n g e x p r e s s i o n : 0 74 O B = — - 11.4 (x-0.26) For e x a m p l e , for C a O , the x of Ca is 1.0 there fore : < O B t a O = y ° S i m i l a r l y , cons ider ing the x of S i and A l as 1.8 and 1.5 r e s p e c t i v e l y , the opt ica l bas ic i t i es of S i 0 2 and A l 2 0 3 are ca lcu la ted to be 0.48 and 0.61 in 161 that order . R E F E R E N C E S 1. D u f f y , J . A . and Ingram, M.D.: J . A m e r . C h e m . S o c . , Vo l .93 , Dec .1971 , pp .6448-6454 . 2. D u f f y , J . A . and Ingram, M.D.: P h y s . C h e m . G l a s s e s , Vo l .16 , Dec.1975, pp .119 -123 . 3. D u f f y , J . A . and Ingram, M.D.: J . Inorg. N u c l . C h e m . , Vo l .37 , 1975, pp .1203-1206 . 4. D u f f y , J . A . et a l : J . C h e m . S o c . Faraday T r a n s . I, Vo l .74 , 1978, pp .1410-1419 . 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0078430/manifest

Comment

Related Items