McMaster UniversityDigitalCommons@McMaster

Open Access Dissertations and Theses Open Dissertations and Theses

10-1-1976

Attack of Magnestic Refractories by SteelmakingSlagsStan-Man Kim

Follow this and additional works at: http://digitalcommons.mcmaster.ca/opendissertationsPart of the Metallurgy Commons

This Thesis is brought to you for free and open access by the Open Dissertations and Theses at DigitalCommons@McMaster. It has been accepted forinclusion in Open Access Dissertations and Theses by an authorized administrator of DigitalCommons@McMaster. For more information, pleasecontact scom@mcmaster.ca.

Recommended CitationKim, Stan-Man, "Attack of Magnestic Refractories by Steelmaking Slags" (1976). Open Access Dissertations and Theses. Paper 3121.

,

(

.I

ATTACK OF MAGNESITE REF~~CTORIES BY

STEELMAKING SLAGS

•

•/

(

,.

...

•

ATTACK OF MAGNESITE REFRACTORIES BY

STEELMAKING SLAGS

by

SUNG-MAN KIM, B.Sc.

A Thesis

\

\

Submitted to the School of Graduate Studies

in Partial Fulfillment of the Requirements

for the Degree

Doctor of Philosophy

McMaster University

Oct~r 1976

\l()(' 1'(:;: OF PHI {,O;:';O\,1l\ (l0'1(,)Ml' tot ! 1U l' g ~ .

•

~Id.iMnTU UNl VERSITYII anh 1 t (l n, On let rI 0

. <t

TITLE.

i\IlTHOH

SUPEHVISOHS'

NmiHEH OF PAm·;;:.;

Atta\.'k or ~bgnf'Sltl) lkfractori0sby Stpc}m,lklng Slags

Sl;llg-\Ltn Kun, H.Se. (SPOll} t\ationall:n l\l'n,j ty, KOl'PCl)

PI'i.lfp~<-;()rs W-K. Lu and P.S. Nld101s011

(X\l), H;2

.~

\

tAB~THACT

\

,

IExpprim(;ntal inv('stigatlons of the attack of carbon-fl'ee

• 10/

and carbon-bearing magnesiLe refractories by synthetic stl'el-

making slags at 1600"C and relevan·t thcoretirul" analysis ar0

documented in this thesis.

The attack of carbon-f.rc(~ rnagnt~slt(' brick by steelmaking

slags was studied by immersing cylindrical specimE.::'ns in molten

slag contained in a noble metal crucible. The major variable

of study was the composition of the slag, i.e., the alumina

and magnesia contents and the lil11e-silica ratio.

The reacted samples are examined macroscopically in terms

of elongation, slag climb and penetr.ation. Microscopic examina­

tion by petrographfc techniques and microprobe analysis further

l} lust ra te the macroscop i c observations aSRoe i ated wi th the chem-

ical reactions taking place between the refractories and the

penetrating slag.

The role of carbon in extending the lining life of BOF

carbon-bearIng refractorIes is Lho main objective of this re~

search. Test crucibles were reacted with slags. The reacted

snmp10s ~0r0 mic~oscn~ira]ly st\zdiprt. The formation nnd dcs-

tructioh of a dense M~O layer in reacted specimens was stUdied.

Theoretical analysis nf the f0rmat{on of the dense MgO layer,

\ras undert ukcn based on the assurnp l ion, that the- format ion of

,~lg0 is the resul t of a gaseous reac.tion between magnesium

vapor and carbon dibxide.

ThiS work has been carried out under condi tions relevant to

refractory pl'oblems invoh"ed ill a. few major steelmaking vE'ssels.

The ~:esul ts obtained here have ~hed light on the' kinetics a.nd m<:ch-

3nism of the at~ack of rnaglwsi te refract0riE:~by steelmaking 5J ags ... . .

)

•ACKNOWLEDGMENTS

r

The author is greatly indebted to his supervisors,. .---Dr. W-K. Lu, Dr. P.S. Nicholson and Dr. A.E. Hamielec for

, I

their advic~!and guidance throughout the coqrse of this work.

studieS'. Similarly, the author? J{

also extends his appreciat~on to Dr. S.G~ Whiteway of Atlantic

The a~thor extends his gratitude to Mr. D. Hayes ofII

Canadian R~fractories Division~ G~e~~~lle, Quebec, for his

invaluablelhelP on petrographic,

Regional Laboratory of NRC, Halifax, Nova Scotia, for his

effort of measuring the viscosity of the slag used in this

work, Which, unfortunately, was not successful.

/<""'< The financial assistance of the American Institute of

Iron and Steel (AISI)'and the Canadian National Research

Council (NRC) through Research Contrac~ No. 43-236 and Research

Grant ,A28yi, respecti VelY', are gratefully acknowledged. The'. .

author wishes to thank Mr. G. Anthony of Youngstown Sheet and

Tube Company, Youngstown, Ohio~ and Mr. C.E.' Osterholtz of

\Wheeling-Pittsburgh Steel Corporation~'steubenville,Ohio, for

their kind cooperation as repre~entatlves of the Refractory

Panel of the AISI General Research Committee. ~'

The author also thanks the fa~ulty, staff and f~llow grad-

uate students of the Department of Metallur.gy and Materials

Science, McMaster University, for many helpful discussion§ andr;.

aid, anti Mrs. A. Neumayer for her excellent typing of this

thesis.

Finally, the author commends his wife, Sookja, for all her

sacrifice and moral support throughout the course of this work.

iv

"

TABLE OF CONTENTS

Page

1-

4

4

5 :'57

121212

1314141516

17

222224242626272929

29

30

333334343535

38

39

41

434343

Carbon-Bearing'­BrickCarbon-Free Magnesite'

~ITERATURE REVIEW

, >!;f.

I NTRODUC'TI ON

2.5.22.5.32.5.4

c'HAPTER I

CHAPTER I I}

2.1 INTRODUCTION

2.2 BASIC OXYGEN FURNACE STEEL~~KING AND SLAGMAKING2.2.1 Basic Oxygen Furnace Steelmaking2.2.2 Basic Oxygen Furnace Slagmaking

2.3 REFRACTORIES FOR BASIC OXYGEN FURNACE2.3.1 Classification of BOF Brick

2.3.1.1 Pitch-Bonded and TemperedMagnesite Brick .

2.3.1.2 Burned Magnesite Brick2.3.1.3 Pitch-Impregnated Magnesite Brick

2.3.2 Raw Materials of Magnesite Brick2.3.2.1 Magnesite and Periclase2.3.2.2 Pitch

'2.4 IMPURITIES IN MAGNESITE

2.5 CARBON IN BOF MAGNESITE BR'!'CK2.5.1 Role of Carbon in Brick

2.5.1 1 Nonwetting of Carbon by Molten Slag2.5.1.2 ~as Evolution2.5.1.3 Dense Layer FormationCarbon CoverageCarbon StructureDecarburization2.5.4.1 Decarburization by Furnace

. Atmosphere2.5.4.2 Decarburization by the Slag

Cqmponents2.5.4.3 Decarburization by MgO and Si02

2,6 THE FAILURE OF MAGNESITE BRICK2.6.1 Spalling During Burn-in2.6.2 Abrasion from Scrap Impact2.6.3 Sheet Spalling2.6.4 Slag Attack

2.6.4.1 Attack of..Magnesi te

2.6.4.2 Attack ofBrick

2.6.5 The Zonal Lining io a BOF

2.7 . THE EFFECTS OF STEELMAKING PRACTICE ONLINING PERFORMANCE2.7.'1 The Operational Contrcl of Steelmaking

2.7.1.1 The Iron Oxide Content of Slag2.7.1.2 End-Point Control

,.

",

v

Page

2.7.2 Hot Metal Composition 412.7.2.1 Hot Metal Silicon Content 442.7.2.2 ~he Minimization of Carried-Over 45

Slag2.7.3 Fluorspar Subst i tutes 45

2.7.3.1 A1203-Based Materials 462.7.3.2 MgO-Based Materials 46

.,CHAPTER III EXPERIMENTAL

3~ INTRODUCTION

3.2 FURNACE ASSEtffiLY3.2.1 Molybdenum Wire-Wound Furnace3.::2 Reaction Tube Assembly3.2.3 Crucible Support Assembly3.2.4 Refractory Specimen-Holder

.3.2.5 Temperature Control3.2.6 Furnace Atmosphere3.2.7 Globar Furnace

3.3 MATERIALS3.3.1 Refractory, Materials3.3.2 Slag Preparation

3.3.2.1 Slag Composition3.3.2.2, Wustite3.3.2.3 Prefusion of Slag

3.3.3 80% Platinum - 20% Rhodium Crucible,3.4 EXPERIMENTAL PROCEDURE

3.4.1 Immersion Test3.4.2 R~fractory Crucible Test in a Graphite

Container3.4.3 Refractory CruciBle Test in Air Atmosphere3.4.4 Examination of Reacted Specimens

3.4.4.1 Mac~oscopic Observations3.4.4.2 Microscopic Observations3.4.4.3 Elec~ton Probe Analysis

CHAPTER IV EXPERIMENTAL RESULTS

49

49

505050525353'5355

55555757575759

595962

6263636364

66 ..

4.1 INTRODUCTION 66

4.2 IMMERSION TESTS OF CARBON-FREE BRICK 664.2.1 Effect of Temperature 714.2.2 Slag Penetration and Climb 714.2.3 An Index of Hot Streng~h and the 72

Failure of Specimens4.3 MICROSCOP1C EXAMINATION OF CARBON-FREE SPECIMENS 73

4.4 ELECTRON MICROPROBE ANALYSIS OF CARBON-FREE 78SPECIMENS

vi

4.5 SUMMARY OF IMMERSION TESTS WITH CARBON-FREEBRICK SPECIMENS

4.6 CRUCIBLE TESTS OF CARBON-BEARING REFRACTORIES4.6.1 Tests with Oxidizing Slags in FlowIng

Air Atmosphere4.6.2 Tests with Iron Oxide-Free Slags in

Air Atmosphere4.6.3 Tests in Reducing Atmosphere

4.7 MICROSCOPIC OBSERVATIONS OF CARBON-BEARINGSPECIMENS4.7.1 Empty CrucIble Reacted with Flowing Air

Atmosphere4.7.2 Crucibles Reacted with Iron Oxide­

Con~aining Slags4.7.3 Crucibles Reacted with Iron OXIde-Free

Slags4.7.4 Crucibles Reacted with Iron-Oxide

Cont~ining Slag in Reducing Atmosphere.4.8 ELECTRON MICROPROBE ANALYSIS OF CARBON-BEARI~G

SPECIMENS4.8.1 Empty Crucible in Flowing Air Atmosphere4.8.2 Crucibles Reacted with Iron Oxide-

Containing Slags4.8.3 Crucibles Reacted with Iron Oxide-Free

Slags4.8.4 Crucibles Reacted with Iron Oxide­

Containing Slag in Reducing Atmosphereg

4.9 SUMMARY OF CRUCIBLE TESTS WITH CARBON-BEARIN~o" ~BRICK "

·CHAPTER V DISCUSSION

79

7980

82

82

84

84

85

87,

87

88

8889

90

90

\

5.1 INTRODUCTION5.1.1 A1203-MgO-Containing Slag in Steelmaking5.1.2 Carbon-Free Refractory and Slag as a

Single System5.1.3 Attack of Carbon-Free Refractories by Slag5.1.4 Capillary Penetration of Slag into

Carbon-Free Brick

929298

100102,

5.2 EFFECTS OF CO~WOSITIONAL CHANGE IN AGGRESSIVENESS 103OF SLAG5.2.15.2.25.2.3

5.2.45.2.55.2.6

Tbe Detrimental Effects of A1203 in SlagBeneficial Effects of MgO in SlagsThe Comb~ned Effects of A1203 and MgOAddi tion'S ,The Effect of Oxidation State of SlagEffects of Basicity of SlagThe Effects of Iron Oxide in the Slag

vii

103107108

109110112

4.5 SUMMARY OF IMMERSION TESTS WITH CARBON-FREEBRICK SPECIMENS

4.6 CRUCIBLE TESTS OF CARBON-BEARING REFRACTORIES4.6.1 Tests with Oxidizing Slags in Flowing

Air Atmosphere4.6.2 Tests with Iron Oxide-Free Slags 1n

Air Atmosphere4.6.3 Tests in Reducing Atmosphere

4.7 MICROSCOPIC OBSERVATIONS OF CARBON-BEARINGSPECIMENS4.7.1 Empty CrucIble Reacted with Flowing Air

Atmosphere4.7.2 Crucibles Reacted with Iron Oxide­

Con~aining Slags4.7.3 Crucibles Reacted with Iron OXide-Free

Slags4.7.4 Crucibles Reacted with Iron-Oxide

Cont~ining Slag in Reducing Atmosphere

4.8 ELECTRON MICROPROBE ANALYSIS OF CARBON-BEARI~G

SPECIMENS4.8.1 Empty Crucible in Flowing Air Atmosphere4.8.2 Crucibles Reacted with Iron Oxide-

Containing Slags4.8.3 Crucibles Reacted with Iron Oxide-Free

Slags4.8.4 Crucibles Reacted with Iron Oxide­

Containing Slag in Reducing Atmosphere'" Q

4.9 SUMMARY OF CRUCIBLE TESTS WITH CARBON-BEARIN~o ~BRICK 0

•

'CHAPTER V DISCUSSION

79

7980

82

82

84

84

85

87.

87

88

8889

90

90

929298

100102,

, 5.1 INTRODUCTION5.1.1 A1203-MgO-Containing Slag in Steelmaking5.1.2 Carbon-Free Refractory and Slag as a

Single System5.1.3 Attack of Carbon-Free Refractories by Slag5.1.4 capillary Penetration of Slag into

Carbon-Free Brick

5.2 EFFECTS OF CO~WOSITIONAL CHANGE IN AGGRESSIVENESS 103OF SLAG5.2.15.2.25.2.3

5.2.45.2.55.2.6

The Detrimental Effects of A1203 in SlagBeneficial Effects of MgO in SlagsThe Comb~ned Effects of Al 20 3 and MgOAddi t ion7:> .The Effect of Oxidation State of SlagEffects of Basicity of SlagThe Effects of Iron Oxide in the Slag

vii

103107108

109110112

/

TABLE

I

2

3

5

6

7

8

LIST 01" TABLES

TITI:.E

MagnesIum and calclum silicates in bInaryequillbrium wIth MgO

Influence of CaO/S102 ratio on the phasedIstrlbutlon of E'lements·

Influence of composItion and microstructureon bondIng

ChemIcal analYSIS and physIcal data ofrefractory brIcks

Composltlon of slags

Specimen elongatIon, slag clImb andpenetratIon

Calculated partial pressures Qf Mg(g) andCO in contact with carbon

-Fractions 'of decarburization by MgO and CO2

PAGE

17

21

21

....56

58

68

117

146

--

LIST OF FIGURES•

FIGURE

1

TITLE

Schemat1c representation of progress ofrefining In a BOF. ('1)

PAGE

6

2 Typical changes in iron-oxide content and 8temperature wIth blowing tfme.(3)

3 Typical changes In slag co~osltion ~ith 8blowing time.(3)

4 The specifIc aggressiveness as function of 10'tlme.(8)

5 Phase dIagram for the system CaO-MgO-Si02 .(12) 18

6 Photomicrographs illustrating the structure of 23three periclase grains. In each case, thelarge gray crystals are MgO and the sil~cate

locat"ion 1S indicated. A 95 percent MgO grainis shown contairling appreciable silicate aroundthe MgO crystals. A 98 percent MgO grain con­ta1ns less silicate, but the silicate is stilllocated around the crystals. The 97 percent MgOgrain wIth a higjer CaOjSi02 ratio containsSIlicate as concentrated areas and not as a film

. around MgO crystals. Reflected light, X52.(3)

7 ComparIson of chemical compositions of peri- 23clase brick used wit~ and without pitch.(3)

8 Schematic of wear of indicated brick typ~s.(3) 25

9 Illustration of wetting of pores of brick 25with and ~ithout carbon.(3)

. ,10 Relative slag erosion versus density of peri- 28

clase brIck made wIth indicated pitches.(3)

11 Effect of heat treatment of pitch on crystal- 28linity of carbon residue.(3)

12

13

14

Typical wear pattern of BOF lining.(38)

Relation of residual carbon content and rela­tive slag erosion for indicated tempered­periclase and magnesite brick.(3)

Typical zonal lining of BOF.(38)

28

36

36

"

~r

) 0-

,FIGURE -

TITLE PAGE"

15 Lining number versus lining life.(3) 42

16 Frequency curves of turndown 'temperature on 42. BOF campaigns;(23)~.

17 Effect of S.i. and Ti contents of toni on 42lining 1if(>.(3)

\

18 Schematic diagram of furnace. 51.'~ 4

19 Temperature profile' of furnace at 1600°C 54

20 Specimen temperature and heating and cooling 61. <

schedule.

Figures 21 to 60 inclusive are located at the end of the thesis.

21"

22

Carbon-ftec magne'si te brick sPecimens I from left: I

after immersion in ma:;ter slag (46% CaO, 30% Si02,17/0 FeD) aDd '7% MnO) fnr 20) 40, 60, 80 and 100minutes in t~at order. '

Carbon-fr('~~ magnesIte bli ~k specimens, from leftfresh specimen, and th0se 01 5 minute immersionIn slags, A, 0, C, L, c;.n, H, J, K, L, N, P, Q andR in that order (Table·5, page 58, for slag compo-sition). '

23

24

Partially sectioned carbon-free magnesite brickspecimens, from left~ fresh specimen, and those ofSO second lmmersion in slags A, B, C, D, E, F, G,H, J, K, L, N, P, Q and R in that orqer (Table 5,pag~ 58, for slag composition)...

•Partially sectioned carbon~free magnesite brickspecimens, from left, fresh specimen, and those~f

5 minute immersion in slags'A, B; C, D, E, F, a, H,J, K, L, N. P, Q and R in that order (Table 5, page58, for slag composition).

"

25

'. I

26

27

Carbon-free magnes~te brick speci~ens reacted at1550°C, from left, after 5 minute immersion in slagsA, C and F; partially sectioned specimens after 30seconds in slags A, C and F; and partially sectioned

.sp€cimens after S'rninutes in slags A, C and F: slag A(master slag),~~~ag C (10% A120S), ahd slag F (15%'A1 20S-8% MgO): -

Speci~~n e~ongation an4 slag climb vs. s!ag composition.~

Slag p~netration vs. slag composition.

FIGURE~

28

29

30

•31

TItLE

Magnesite brick fragments which remained in slag D(15% A1203) for ano~her'30 minutes at 1600°C after5 minute immersion test; dark areas are slag phasesand'MgO grains have suffered slag penetration alongthe subgrain boundarie~ (13X).

Photomicrographs (33X) of as-received burned magnesitebrick; (a) in plain light and (b) under crossed nicols:P, periclase; 8, silicate; V, pore.

Thin sectlon photograph (4.2X) of carbon-free specimenafter 5 minute immersion in slag A (master slag);brown regions are slag-penetrated matrix and lightareas are void spaoes .

Thin section photograph (4.2X) of carbon-free specimenafter 5 minute, immersion in slag K (8% MgO); brownregions are slag-penetrated matrix and light areasare void spaces. w

32 Thin section photograph (4.12X) of carbon-free specimenafter 30 seconds immersion in slag C {lO% A1203);brown regions are slag-penetrated matrix and lightareas are void spaces.

33 Ca-Al-si1icate compound (dendrites) developed in thematrix of carbon-free brick specimen after 5 minuteimmersion in slag C (84X).

34 ' Thin section photograph (4.2X) of carbon-free specimenafter 5 minute immersion in slag F (15% A1203-8% ·MgO);brown regions are slag-penetrated matrix and lightareas are void spaces. The thin section has been madefrom the portion remaining in the specimen holder.

35- Thin section photograph (4.2X) of carbon-free specimenafter 5 minute, immersidn in slag G (10% A1203-13% MgO);brown regions are slag-penetrated matr~x and lightareas are void spaces:

36 Ca-Al-silicate compound (dendrites) developed in thematrix of carbon-free brick specimen after 5 minuteimmersion in slag G (84X).

37 Thin section photograph (4.2X) of carbon-free specimen­after 5 minute immersion in Fe203-containing slag P(46% CaO, 30% 8i02, 17% Fe203, 7% MnO): brown'regionsare slag-penetrated 'matrix and light areas are voidspaces.

FIGURE TITLE

38 Thin section photograph (4.2X) of carbon-free specimen~a>-j;.ter 5 rllinute immersion in Fe203 sla.g Q (10% A1202):brown regions are slag-penetrated matrix and lightareas Are void spac0s. The thin section has been madefrom the portion remaining in the holder.

39 Thin section microphotograph (84X) showing dendriticcrystallization in the matrix of cylindrical carbon­free brick specimen after 5 minute immersion in Fe203slag Q (10% A1203).

40 Thin section photograph (4.2X) of cylindrical carbon­free specimen after 30 second immersion in Fe203 slagR (15% AI203-8% MgO): brown regions are slag-penetratedmatrix and light areas are void spaces.

41 Thin section photograph (4.2X) of carbon-free specimenafter 30 second immersion in slag N (CaO/Si02 ~ 1);brown regions are slag-penetrated matrix and lightareas are void spaces.

42 Monticellite (CMS) d~veloped in t~e matrix of carbon­free brick specimen after 5 minute immersion in slag N(CaOjSi02 =1); 84X ..

43 Thin section photographs (4.2X) of cylindrical carbon­free brick specimen which has been withdrawn by standardprocedure after 30 second immersion in slag H; (a) inplain light and (b) under crossed nicols .. Brown regionsare Fe- and Mn-rich slag-penetrated outer layer. Brightareas are the matrix penetrated mainly by silicate ofslag.

44 Thin section photographs (4.2X) of cylindrical carbon­free brick specimens which have been withdra\m immed­iately after (a) 30 second and (b) 5 minute immersionsin slag G (10% A1203-l3% MgO).

45 Composite strip charts~of electron microprobe analysisover cross-sections of cylindrical carbon-free brickspecimens; ea) as-received and (b) after 30 secondimmersion in slag H (distance from the surface of thespecimen) .

46 Cross-sections of carbon-bearing brick cruciblestested in air atmosphere for 2 hours, from left, withoutslag, with slags C, F and N at 1600°C and with slag Vat 1650°C in that order.

47 Cross-sections of carbon-bearing brick cruciblestested at 1600°C in air atmosphe~e for 2 hours, fromleft, without slag and with siags T and U in that order.

xiii

--

..

II< •

. ,.

FIGURE

48

TITLE

Cross-sections of carbon-bJaring brick cruciblestested at 1600°C for 30 minutes in graphite container,from left. without slag and with slags A, D and H inthat order.

50

-

49 Thin section photograph (4.2X) of pitch~impregnatcd

magnesite brick coked in a graphite container at 1600°Cfor 30 minutes.

Thin section photograph (4.2X) of crucible wall madeof carbon-bearing magnesite brick maintained in airatmosphere at l600°C for 2 hours; carbon-bearingceqtral region is ~urrounded by light decarburizedzones an~ thin white layers are noticed at the'boundaries between these two regions.

51 Carbon-bearing magnesite brick crucibles reacted inair atmosphere for 2 hours with slag N (a) andslag C (b) at 1600°C and.slag V (c) at l650°C.

52 Thin section photomicrographs (33X) of carbon-bearingbrick crucibles reacted in air atmosphere for 2 hourswith (a) slag N (CaO/S102 = 1) (.b) slag C (CaO/Si02 = 1.5)at 1600°C and (c) slag V (CcaO/SiOZ = 2) at 1650°C.

53 Thin section photographs (4.2X) of sectioned cruciblewall made of carbon-bearing brick reacted with (a)irap oxide-free slag T (10% AlZ03) and (b) U.(15% AlZ03­8% MgO) in air atmosphere at l600°C for 2 hours.

54 Thin section photographs (4.2X) of sectioned cruciblewall made of carbon-bearing brick reacted with slag D(15% Al z0 3 ) in graphite container at 1600°C for 30minutes; (a) in plain light and (b) under crossed nicols.

55 Composite strip charts of microprobe analysis over apitch-impregnated magnesite brick sample coked at 1600~C

for 30 minutes in a graphite container (distance fromthe surface of the specimen).

56 Composite strip charts of microprobe analysis over acarbon-bearing magnesite brick sample tested in airatmosphere at 1600°C for Z hours (distance is fromthe surface o~ the specimen).

57 Composite strip charts of microprobe analysis alongthe line GHI of Figure 51(b) over a well-defined denseMgO layer formed in a carbon-bearing magnesite brick /crucible reacted with slag C in air atmosphere at1600°C for 2 hours; slag, right and refractory, left.

xiv