Note: Descriptions are shown in the official language in which they were submitted.
10~5~53~ .
SPI~C~FICATION
. . ._.___
ACKGROUND OF TE~E INVENTION
This invention relates to improvements ;n high fired basic,
direct bonded, magnesia-chrome ore, also referred to as magnesite-chrom ,
refractory shapes and the method of manufacture of such improved shapes.
More specifically, refractories made in accordance with this invention are
provided which yield significantly improved service life in severe wear zone
of industrial furnaces as compared with the performance of a variety of re-
fractory cornpositions and products commercially available today.
Basic, direct bonded, magnesite-chrome refractory brick
represent an important, if not lhe most important, class of refractory
employed as a furnace lining.
; The state of the art of cGnventional direct bonded ref~ ctories
presently so relied upon by industry, is well developed in the United '.tates.
The introduction of direct bonded refractories early in the 1960's was made
possible by the a~ailability of relatively high purity raw materials, especiall
washed or concentrated chrome orës wherein the SiO2 content was reduced
from 4 to 6% down to below 2%, and as low as 1% for ore of ~frican origin
known as Transvaal concentrates. Similarly, beneficiated chrome ore with
- 20 SiO2 content of 1. 5 to 3% became availaule from the Philippines.
These ores in combination with synthetic periclase or dead
burned high purity magnesite containing less than 1. 5%, preferably less than
1%, SiO2, could be processed into brick shapes in the conventional manner
and fired at temperatures higher than 1650C (3000F) without excessive
slumping or sticking. As is well described in the literature, the direct bond so
-2-
'1~ . '.
. . _
' 1 1~5953~; ~
¦ developed is the result of high tempcrature interactions between the chromite
¦ and magnesia, involving solid statc reactions, solution-precipitation reaction¦ and redistribution of silicates which were present in the raw materials as
¦ accessory mineral phases. Most desirably, the periclase crystals are
¦ sintered directly to chrome ore, periclase is bonded to periclase, and secon-
dary spinels bond periclase crystals. Additiona]ly, some silicate bonding
may co-exist.
In the manufacture of direct bonded brick, size graded
magnesia and chrome ore are mixed with temporary binders and pressed at
pressures exceeding 5000 psi, for example as high as 16, 000 psi, dried. and
fired at temperatures above 1650~C. Materials, izing, and processing are
adequately described in U.S. Patent No. 3,180, 744.
Refractories important in the 1950's, such as silica brick
and chemically bonded basic brick. have l~rgely been replaced by high lired
50, 60 and 70% MgO direct bonded brick in open hearth roofs, walls and up-
takes, electric arc furnace walls, and copper converter linings. The newer
special processing units such as vacuum degassing and Argon-Oxygen De-
'carburization ~AOI)) vessels are currently lined extensively with direct bonde
brick. The 60% MgO class is dominant because it represents a desirable
economic balance between costs, chemical resistance and physi'cal propertie~ .
Service life or productivity has' been generally improved in
many of the furnaces employing direct bonded brick thereby leading to new
efforts to balance furnace wear by zoning with improved products. Many
materials have been tried, such as fused cast basic, rebonded fused grain
~; ' brick, and direct bondecl brick of higher MgO class, but in most cases each
"'i' 'I -3-
.', , . ' _. .
.~ ~ . '
., ___ . ., ___...... . .~.. . . . . , . . .. .. . . _.. '___ . _ .. _
1059536
¦ was found to have inherent disadvantagcs.
I The more expensive fusion cast basic brick, while similar in
¦ chemical composition to conventional direct bonded basic, is extremely dens
¦ and essentially free of micropores. Althougl~ this product is high]y resistan
S ¦ to slag erosion, it often fails due to spalling or cracking and bu~k loss initiat d
by thermal shock stresses. These fe~tures detract from its use in modern
high production open hearth roof center sections and backwalls as well as AO
tuyere lines and the like,
Rebonded fused grain brick products are even more expensive
than the premium priced fused cast shapes for most conventional applicationc ,
While somewhat less slag resistant than fusion cast brick these refractories
where not p-one to spalling are often uneconomical at nearly two time s the
price of conventional direct bonded brick.
Another approach known in the refractories art to proAuce
basic brick with improved service life is to prereact or sinter together the
periclase or MgO source, which can be rnagnesium hydroxide, magnesium
carbonate or caustic magnesiaS with ground, sized chrome ore at high
temperature, e.g., above 170ûC (3100F), toform direct solid-solid bond-
ing in grains prior to sizing the aggregate for brick forming. In most cases
the shape is fired at temperatures above 1600~C (2910F). The method of
manufacture and properties of basic bri~k formed from prereacted grains
are outlined in Austrian Patent No. 189, 113 and corresponding U. S. Patent
No. 3, 429, 723,
Unfortunately, it is commercially difficult to produce brick
from prereacted grains due to the fact that it is often undesirable to con-
~ .
--4--
.. , _ ' ' .
~, , ., _ _ _ . .. .. . , , .. _ , ._
.
~ S953~;
',i' I
¦ taminate a periclase ~- ain-producing plant by intr~>ducing chrome ore. In
¦ a conventional brick plant, the use of a prereacted grain often means an addi
¦ tional, costly processing line to avoid chrom~ contamination of the periclase
¦ grindin~ and batching system. In any eventJ while such prereacted brick
I possess desirable strength and slag resistance, they are less resistant to
¦ thermal shock than conventional direct bonded brick and ther~fore have not
proven to be a suitable material for certain severe wear areas, especially
where spalling i s a factor.
It has even been proposed to improve the strength and further
lower the porosity of prereacted magnesite chrome grain refractory product
by adding from 3 to 6% chromic oxide to the refractory batch prior to form-
ing and firin~. The method oi manufacture and improved properties of such
prereacted basic refractory brick is described in U. S. Patent No. 3, 594,199
While strength is markedly increased an~ porosity further reduced b~ this
process, brick made therefrom remain less resistant to spalling or therma
shock than conventional direct bonded brick. Moreover, with the prefiring
of nearly all the raw materials, and refiring in brick shape form, plus the
addition of pure chromic oxide, the selling price re~uired to cover costs and
reasonable profit is necessarily high, as high as rebonded fused grain basic
brick. Such brick incorporating chromic oxide has previously been suitabl
for use in only selected limited applications, where furnace shutdowns or
~esrere thermal shock is absent. -
It is further known in the art that chromic oxide added to a
refractory batch acts as a pressing aid or lubricant to increase the "as
- 25 pressed" density and density after firing of many classes of refractories,
" . ' .
r~
.~, . ~ 5~
' 1~, . .
`~,
'''
.
I ~ `~ ~
l -
1 1~59S36
¦ including ma~nesia, magnesia-chrome, alumin~, and 2irconia. A port~on of
I the increased density is due to the substitution of chromic oxide material for
¦ ingredients of lower specific gravity. The specific gravity of Cr2O3 is 5.1
¦ to 5. 2 while that of periclase is 3. 5 to 3. 6. In any event, ccrresponding
¦ porosity improvements are cited for such brick in U. S. Patent No. 3,192, 058
which discloses the method of rnanufacture. However, the use of chromic
l oxide must be confined to very specific compositional fields in order to
¦ realize impr~ved service results commensurate with the increased manu-
facturing costs incurred with its use. Benefits beyond the effects of higher
density and reduced porosity must be achieved in order to make a significant
improvement in service life i.n the lining of an industrial furnace. Therefore,
chromic oxide has not been used comme, cially as extensively as might be
inferred from the literature.
A primary object of this invention is to provide an economical,
improved h gh fired direct konded basic: refractory shape combining both in-
creased resistance to slag penetration and resistance to thermal shock,
spalling or "slabbing", equal to or better $han conventional direct bonded
; shapes.
An additional object of this invention is to provide a method
of manufacture for an improved high fired direct bonded basic refractory
shape.
A further object of this invention is to improve the micro-
structural features of direct bonded basic refractory shapes which in turn
improve slag resistance and physical properties.
Another important object of ~his invention is to provide an
improved high fired direct bonded basic refractory brick which gives im-
. ' -6-
.,
., , .
~
1~59536
proved service life in linin~s of industrial furnaces, a~ld a method of manu-
facture for such a brick.
These and other objects will become apparent from the
specification and clalms.
SUMMARY OF THE INVENTION
.. .~
This invention is directed to improved, commercially useful,
high fired direct bonded basic magnesite-chrome refractory shapes, such
as brick and the like, and eliminates many of the disadvantages of prior, co _
ventional direct bonded basic refractory shapes. More particularly, the
present invention provides a high fired direct bonded basic magnesite-chroIr ,
refractory shape characterized by the presence of chromium enriched spine
structures distributed in the matrix which bond the individual periclase
crystals. The refractory shape is made by adding chromic oxide po ~der to
magnesite periclase and chrome ore to form a mixture which is pressed int
the desired refractory shape, and fired to a temperature of at least 1700C.
The mixture comprises (1) from about 40% to about 75% by
weight magnesite periclase containing at least about 94% MgO, (2) from abo t
25% to about 60% by welght chrome ore and (3) from about 0. 5% to about 10%
by weight chromic oxide powder.
Preferably, the mixture comprises (1) from about 55% to
about 65% tjy weight (magnesite) periclase containing at least about g4% by
weight MgO, and especially from about 96% to about 99% by weight MgO,
(2) from about 35~0 to about 45% by weight chrome ore and (3) irom about 2%
to about 7% chromic oxide powder consisting essentially (90+70) of -325 mes
lOS953~;
particles, the mixture having a lime to silica ratio no greater than 1:1,
preferably no greater than 0. 5:1, and a total silica content of less than
about 3%, and especially less than about 2%.
In accordance with one preferred embodiment of the present
invention, a high fired direct bonded basic magnesite-chrome refractory
brick is made by firing the brick to a temperature of about 1760C for at
least about 4 hours.
According to the method of the present invention, an improvec
direct bonded basic magnesite-chrome refractory shape is made by adding
from about D. 5% to about 10% by weight chromic oxide powder to a conven-
tional high purity magnesite-chrome ore mix to provide a mixture as de-
scribed ab~ve which is pressed into a refractory shape, and high fir~d to a
temperature of at least 1700~C.
Preferably, the shape is fired to a temperature of at ;east
IS 1700C for at least about 2 hours. It - presently preferred to fire to a
temperature of about 1760C for at least about 4 hours.
ln accordance with one preferred embodiment, this invention
; provides a method of making an improved direct bonded basic refractory
brick from a mixture of (1) from about 55% to about 65% by weight periclase
having an MgO content of from about 9~% to about 99% by weight, (2) from
;~ ¦ about 35% to about 45% by weight chrome ore and (3) from about 2% to about
¦ 7% chromic oxide powder consisting essentially of -325 mesh particles,
¦ the mixture having a lime to silica ratio of no greater than 1:1 and a total¦ silica content of less than about 3%. The mixture is prcss.ed into the shape1 of a refractory brick, and fired to a temperature of about 1760C ~>r at
. -8-
. , .
_
105953~ l
¦ least about 4 hours.
l The refractory products of thls ;nvcntion fired to a temper-
- ¦ ature of at least 1700C exhibit an overall c:ombination of properties which
I is superior to that of conventional direct bondec1 basic magnesite-chrome
¦ shapcs similarly produced from perlclase and virgin or unfired cnrome ore
I and free of significant quantities of fused or prereacted magensite-chrome
¦ materials, but without the addition of chromic o~cide povwder. Thc higll fired
I products of this invention have superior resistarlce to s'ag penetration and
¦ erosion, equally good or better resist:ance to thermal shock and spalling or
l0 ¦ "slabbing", enjoy improved service life in industrial furnace linings, and ar
¦ economically and functionally suitable for use under the severe conditions of
high temperature industrial refractory service.
It is to be understood that the foregoing general description
and the following detailed description are only illustrative and exemlilary
lS and there will be obvious modifications therein without departing from the
spirit or sc:ope of the present invention.
','' ' .
~ DETAILED DESCRIPTION OF THE INVENTION
: ', ' . - .
This invention is suitable for the production of a variety of
high fired ~irect bonded basic magnesite-chrome refractory shapes, particu
larly brick. Preferred embodiments of the in~Tention will now be described
in detail with reference to the manufacture of high fired direct bonded basic
magnesite-chrome refractory brick.
In accordance with the present invention;, a high fired direct
bonded basic magnesite-chrome refractory brick may be prepared by adding
from about 0. S% to about 10% by weight chromic oxidc powder to a conven-
,;' ,'~ . ' .
~ , _9_
l ~
~05953ti
tional peric]ase-chrome ore mix to rorm a mixture which is thcn pressed
into the shape of a refractory brick, and fired to a ternr)erature of at least
. l1700C.
I The lime to silica ratio of the conventional brick is usuall~
I 1:1 or less than 1:1, preferably less than 0. 5:1~ so that the native silicates
¦ in the refracto~y bricls are predominately forsterite, magnesium silicate,
¦ and monticellite. It has been found that the lime to silica ratio of convention-
al periclase-chrome ore brick is particularly suitable for the brick of the
present invention.
¦ Moreover, In order to facilitate firing at 1700C and above,
the combin~tion of periclase and chrome ore should resu]t in an SiO2 content
¦ of less than 3%, preferably less than 2%.
I In accordance with the present invention, direct bonded basic
¦ magnesite-chrome refractory brick can vary widely in the magnesite- chrome
¦ ratio. To maintain a desirable balance of economics, adequate strength,
I and spall resistance, the range commonly practiced is 40 to 75% periclase
¦ with 25 to 60% chrome ore by weight. The preferred range optimizing all
aspects is 55 to 65% periclase and 35 to 45% chrome ore. Compositions in
accordance with this invention contain essentially from 20 to 50% chrome ore
By USi'lg high purity periclase of from ~ 6 to 99% purity and sm~ll quantities
of chromic oxide, a final chemical composition in the 45 to 75% MgO range
is achieved.
All materials are size graded for brick-making including
particles from -3 mesh, or preferably 6 mesh, to -325 r~iesh, Tyler.
- 25 The periclase can be dead burned rnagnesite but is nearly
always selccted from available synthetic rnagnesite or periclase which is a
. --10--
. . ..
0 ~ l.
` ~l
1059S3~
dead burned dense aggregate of grains of MgO crystals with~ accessory phase
and im,t~urities. Typically, the periclase is used in the composition as
coarses 6 x 48 mesh, intermediates 48 x 200 mesh, and fines -325 mesh,
Tyler screens.
A suitable periclase useful in this in~tention may have a
- chemic~l composition of from about 94% to about 99% and above by weight
MgO, up to 2% by weight SiO, up to 1% by weight Fe2O3, up to 1% by weight
Al2O3, up to 1. 5% by weight CaO, and up to 0. 3% by weight B2O3. An
example of a specific periclase found to be useful according to the present
invention has a composition of 97. 4% MgO, 0. 9% SiO, 0. 3% Fe O3, 0. 3%
Al2O, 0. 8% CaO, and 0. 2% B O .
The chrome ores are obtained from natural deposits. Re-
fractory grade chrome ore is essentially a solid solution of spinel minerals
conl;ainine oxides of chromium, magnesi~n, aluminum and iron, accompanie
by a siliceous mineral gangue. The chemical composition varies depending
on the location of the deposit and the particle size of ore selected from
crushing operations. The SiO2 content can vary from 2 to 7%, Cr2O3 from
30 to over 50%, and the remaining FeO, MgOJ and Al2O vary depending
upon the nature of the ore and its country of origin. Concentrated ore, con
sisting OI sized and washed particles containing from 1 to 2. 5% silica are
commonly used in conventional direct bonded brick production. The chrom
ore is crushed, if necessary, to provide sizes typically below 8 and 10 mesh
sized graded for the batch. -
In accordance with the concept of this invention, the chromic
oxide powder is added in an arnount of from about 0. 5% to about 10% by
;~ weight. A particularly preferred range of chromic oxide powder addition
~ ,,...... . -11-
,' . . . ..
.
l .
1 1~)595;~
¦ is fron about ~% to about 7% by wcigllt, and about 4% h~s been found to be
¦ especially suitable,
¦ The chromic oxide powder useful in this inventioll is a fincly
di~rided powder WhiCIl consists essentially (90+~0) of -325 mesh partic~es.
I In one finely divided form which is commercially available for use as a
l pigment, the average particle size of individual particles of cllromic oxide
¦ powder is no greater than about 10 microns in diameter. This very fine
¦ form is suitable for use in this invention. The chromic oxide powder,
which usual~y possesses a rich green color, is water insoluble.
Mosl; available grades of chromic oxide are high purity, i. e.
above 97% Cr2O3, and this degree of purity is desirab~e for use in this in-
vention.
This invention therefore provides an improved high fired
direct bon1ed basic magnesite-chrome refractory shape, particular~y brick,
formed from a mixture of:
(1) from about 40% to about 75%, preferably from about 55%
to about 65%, and especially about 60%, by weight, periclase containing at
least about 94%, and preferably from about 96% to about 99%, by weight
MgO,
(2) from about 25% to about 60%J preferably from about 35%
to about 45%, by weight, chrome ore; and
(3) from about 0. 5% to about 10%, preferably from about 2%
to about 7%J and particularly about 4%, by weightJ chromic oxide powder
consisting essentially oE -325 mesh particle sizeJ the chromic oxide powder
.
t' -12-
' _ ' " . _
1059S3~j
¦ prcferably of high purity, such as at least abou~ 97% by ~tei~ht Cr203, the
¦ periclase-chrome ore - chromic o~icle powder mixture preferably having
¦ a lime to silica ratio of no greater than 1:1 and especially 0. 5:1 or less,
¦ and a total silica content of less than 3%, particularly less than 2%.
¦ In accordance with the n~ethod of the present invention, the
refractory hatch essentially comprised of the chrome ore, periclase and
chromic oxide as described above is thoroughly mixed to disperse the fine.s.
A temporary or ~green~ binder and tempering liquid, typic:ally lignosulfonat~
and water, are mixed into the batch until the mass is uniform and pressable.
~- 10 l Pressing iS carried out using the standard ecluipment and in accordance with
the standard methods commonly employed in the refractories industry.
The shapes can be dried, but in any event are fired in a kiln
.; to a minimum temperature of 1700C (3100F) for at least about 2 hours and
cooled at a rate usually not exceeding 1t)0C/hr. to avc~id microCrac]:s. A
firing temperature of at least 1700C is an important process requiremcnt
for the achievement of the desired microstructure. A particularly superior
product is attained when the refractory shape is fired to a temperature of
~, about 1760C. It is advantageous to fire to about 1760C for at least 4
' hours.
Compared to a conventional direct bonded brick composition
wi,thout the chromic oxide addition, the ~rick of this inVention has increa~ed
density, ultra-high temperature (1600C) bending strength and improved
compressive load bearing resistance at 1701)C while the porosity is low.
Physical properties of direct bonded compositions with and without the
chromic oxide additions are shown in the examples below.
.. '`,-- :-- .. . - . .
~ ~ -13- `
i ,, .
'"''' ''''
:~,
. .. '.
I ~ ~
. I
1 105~3~
I An important feature of the improved direct bonded ~rick of
¦ this invention is that cven though thc compositions containing thc chromic
oxide are of increased density and reduced porosity after firing at high
l tempera~ures, the refractory retains its spall or thermal shock resistance.
The feature believed responsible for the superior slag resist-
ance and actual improved performance in service, is the formation of widely
distributed chromium rich spinel crystals in the periclase microstructure
which inhibit slag or foreign silicate penetration. Slag and foreign silicates
penetrate deeper into the microstructure of conventional direct bonded brick
not having this additional, improved bonding feature. As a result, individual
periclase crystals in conventional direct bonded brick are floated out or
eroded away in a high liquids zone. Deep slag penetration is also detrin~ental
to service life because the altered zone has different thermal-mecrJanical
properties. The boundary between altered and original structure is the loca-
tion of stress concentration and subsequent fracture, crack propagatic~i, and
eventual slabbing, a common wear mechanism of refrac$ories.
Many facets of this invention are further illustrated by the
following examples which are not to be construed as limitations thereof.
Various other con positions, embodiments, modifications, and equivalents of
these examples will readily suggest thernselves to those skilled in the art wit _
out departing from the spirit of the present invention or the scope of the
appended claims. All percentages and parts referred to herein are by weigh
unless otherwise indicated~ Ali particle sizes or sieve classifications are
Tyler series.
~
, ' ~S~S3~; ,
EX~M rLE
A series of experimental high fired direct bonded magnesitc--
chrome compositions of approximately 60% MgO were prepared in the lab-
oratory. The prepared batches were pressed at 15, 000 psi into 6" x 1'! x 1'
bars in a hydraulic press. ~fter drying, the test shapes were fired to
- 3200F (1760~C) in a high firmg kiln and held at that ter~perature for at leas
4 hours before the cooling cycle was started.
The analysis of the starting raw materials is as follows:
,, .'
High Purity Periclase
MgO 97. 6%
-- Si2 0~ 7%
- -- F O 0. 2%
2 3 %
CaO 1. 0%
B O 0. 2%
2 3
Chrome Ore
. ,
-10 Mesh -65 Mesh
'.'' . .
' SiO2 2.8% - 2, 2%
2 3 35. 5% 36.1%
,~ 21) FeO 15. 2% 15. 9%
Al23 29. 8% 30. 2%
; CaO 0.3% 0.2%
MgO 16. 4% - lS. 4~o
,,- ~ .'
~ ~ -15-
'' , . ,
. , . . _. .
105~1536
¦ The an~lysis of the mix compositiolls (% by weight) is shown
¦ in Tahle I.
- ¦ . TABI,E I
.~ I
. I ~ Sieve Classi- .
¦ fication (Mix~ A B N D E
l .__ _ _ _ _
Periclase -8 ~ 28 mesh 26 26 26 26 26
. Ball Mill Fines 22 22 22 22 22
(60% minimum .
- 325 mesh) -
Chrome Ore -10 mesh . 32 32 32 32 32
;. -65 mesh 9 9 9 9 g
:~ High Fired -8 + 28 mesh 5. 5 5.5 5. 5 5. 5 5. ~
13rick Rejects, -~18 mesh 5. 5 5.5 5.5 5.5 5.5
. 60~o MgO .
l 5 Chromic
Oxid e,
ADDED _ -325 mesh 0 0.5 1. 5 5,0 10.0
,.~
Lignosulfonate .
: Binder, ADDED 2.0 2.-0 2.0 2.0 2.0
BuLk Density
. After Drying
; at 300~F . g/cc 3.01 3.06 3.05 3.15 3.25
:: ' :
. ~ ~ . _ .
;.~ Based on the chemical analysis of the raw materials the
I chemical composition of the control Mix A was as follows:
.
1,' . '
~ ~ -16- . - `
.1
;. . Il ' ' .
1 1~59~3~
~ SiO2 1.7% ,
¦ 2 3 7.2%
¦ 2 3 13,8% ,
¦ CaO 0.7%
¦ MgO 60.3% .
¦ Cr23 16.3%
Mixes B, N, D, and E increased in Cr2O3 content commen-
surate with the additions of chromic oxide powder listed in Table I.
Each sample was measured b~fore and after firing to deter-
mine dimensional changes due o reactions between the raw materials.
Properties were evaluated by testing for open or apparent porosity, bulk
; density, hot modulus of rupture at 2700F and hot crushing resistance at
2800F. Physical properties are sho~;n in Table II: .
TABLE II
: :~15 Mix Mix Mix Mix Mix
A B N D E
: ~ Linear Firing
Change -% ~0.07 -0.02 -0.07 -0.05~0.03
:j Fired Bulk
Density g/cc 2.96 3.01 3.01 3.103.13
Apparent
: . Porosity-% ~17.8 - .~17.8 16.7 16.315.1
Hot Modulus of
Rupture at
2700F-psi 505 395 .415 520 640
: 25 Hot Crushing
Strength at
2800F-psi 780 745 875 9301280
r-~- ~ ' ~~~
~ i
' ` '' ' 1~
`
,.` ~ ' '~ _
' I lU5y53~
It can be observed that as the chro~nic oxi~e powder ~ddition
. l is increased, the bulk density and crushin~ resis~ance are in~leased while
the porositS~ is reduced.
EXA MP LE 2
. 5 Additional experimental 60% MgO class direct bonded composi
tions were processed in the laboratory usinglar~r batches than were used in
Example 1. 9" x 4 1/2" x 2 1/2" rectangular brick were pressed under form
ing pressures of 12, 000 psi, dried at 22QF, and fired in a commercial high
firing tunncl kiln to a peak temperature of about 3200F (1760C) for a minl
murn of 6 hours. - .
. . The chemical composition of the periclase used in these
experiments is as follows:
: High Burity Pcriclase
.. .
. MgO 97. 8%
l 15 SiO2 0. 6%
: 2 3 %
A123 O. % , .
CaO 0. ~i%
B O ~ 0.2%
`: . 23 ..
The cl-rome ore used was the same material shown in
Example 1 as -10 mesh.
- The analysis of the mix compositions is shown in Table~,. III. ` .
-18- .
.- _.,
.. .. .. ... .. ~ . ... ~ _ .. . .
l ~
I lOS9S3~i
.. I . .
TABLE III
.. . , .. ~ .. .
I Sieve Classi-
¦ fica_ion (Mix) F G H I
l Periclase -4 + 8 mesh 19. 5 19~ 5 19. 5 19. 5
: 5 ¦ -8~ 28mesh 6.5 6.5 6.5 6.5
I Ball Mill Fines
- . ¦ (60% minimum
I -325 mesh) 30.0 30. 0 30.~ 30.0
.: I Chrome Ore -10 mesh + 28 mesh21. 521.5 21; 5 21. 5
1 -28 mesh 12.0 12.0 12.0 12.0
Ball Milled
: -150 mesh 5.0 5.0 5.0 5.0
High Fired
Brick Rejects -48 mesh 5. 5 5. 5 5. 5 5. 5
Chromic Oxide,
. ADDED _ -325 mesh 0.0 2.0 3.~ 5.0 .
. Lignosulfonai;e (solut;on)
.~ ADDED 3. 5 3. 5 3. ;~ 3. 5
Bulk Density,
Dried at 220F g/cc 3.10 3.15 3.17 3. 24 .
'',; . .
,-, , . .~
. : The testing was the same as was used to evaluate the .
. ~ . materials in Example 1, with the addition of higher temperature modulus of
~ rupture tests and the addition of a compressive load test. Improvements in
; ~ fired density, porosity, were noted as in Example 1. In addition, the ex-
tremely high temperature properties at 290ûF and 3100F show marked im-
11 ~ ~ I
' ~.' ' , ' ' ' ' ' ' " '' .
' .... .: . ' ' .
--1 9
7~
~' ' ,
,
., ~_ _. _... . .--..... .... .... ~__ _ . ~ .
I ~(~5~S3ti - .
¦ provement due to the additional bonding caused by the formation of magne
si~n chromite spinel in the matrix Test results are shown in Table IV:
¦ TABLE IV
' I
I Mix Mix Mix Mix
I F G H I
¦ F`ired Bulk Density - g/cc 3.03 3.11 3715 3. 18
¦ Apparent Porosity -% 18.0 15.5 14.0 13. 5
I Hot Modulus of Rupture - psi
¦ at 2700F 615 510 630 750
¦ at 2800F 555 355 455 460
¦ at 2~30F 145 155 210 210
Hot Crushing Strength - psi
at 280CF 640 890 1080 780
Compressive Load
Resistance, 50 psi
at 3100F
Hrs. to failure 0.1 0.75 0. 75 No failure
after 2 hrs.
..
EXAMPLE 3
Improvements in high temperature properties shown in
Example 2 can also be achieved in high ~ired direct bonded brick where a
low porosity is achieved by the use of high density periclase having a high
lime to silica ratioO This periclase when combined with the lowest silica
chrome ore yields a direct bonded composition with an overall lime/silica
ratio of above 1. 3:1 and can be as high as 2:1 or adjusted with various lime
source additions. Such brick can be pressed to a high density. tend to
expand less on ~irin~, or in many cases actually shrinks. This product is
avail~ble commercially and well known in the art.
~ , `.
y -20-
..L
: j
~ ` 1()5~S3t;
E~perimental batches were processcd into brick shapes,
fired ~nd evaluatcc3 using the same procedures used in Example 2. The
¦ chemical composition of the starting raw materials is as follows:
I High Lime Periclase
¦ MgO 96.2%
SiO2 1. 2%
¦ Fe23 0. 2%
2 30. 2% .
. CaO 2. 2%
. B O 0. 02%
'. 23 . .
.. . Low Silica Chrome Ore
., `. . , sio2 0-9% , . .
. 2 3 26. 0%
2 3 lS. 9%
. 15 Cr2346. 5%
. ~ MgO 10. 5%
CaO 0. 2%
Experimental compositions and test properties are given in
I Fable V
''' , . . ' ' ,.
,:~ .'
-, ~ ' ,
. -21-
.' '' ". ` .
~Q5953ti
TABL~ V
Sieve Classi-
fication (Mix) J K
er;clase -~ + 8 mesh 5. 5 S. 5
-8 + 28 mesh 37.0 37.0
Ball Mill ~ines 15. 0 15. 0
(G0% -325 mesh)
Chrome Ore - -28 mesh 42. 5 42. 5 .
- Chromic Oxide,
ADDED _ -325 mesh - 4.0
- Lignosulfonate (solutionj
ADDED _ 3. 5 3.5
Fired Bulk
Density- g/cc 3.29 3.36
Apparent Porosity
-% (kerosene method) 14.0 13. 6
Hot Modulus of
Rupture
at 2700"F - psi 565 550 .
at 290C~F - psi 260 325
CaO/Silica ratio 1.4:1 1.4:1
, - - .
.
From the test results, the addition of 4, 0% chromic oxide
to the high lime/silica ratio, low porosi~y direct bonded type refractory
produces only moderate improvements in the highest temperature strengths.
- 25 Further evaluation of this compositional field revealed that the chromium
. rich spinels do retain a role in bonding areas ~f the hot face in service, but .
.'' .;. ' . . ,, .
,~, , . ' . ~ . ~
~ ~ ~ -22~ -
if''-
, ,. - I '
-.
I . lOSgS36
¦ are less effective in inhibiting slag penctration in these high lime/silica
- ¦ ratio compositions. An improvement in penetration resistance could not be
.: . I detec~ed compared to the same base mix without the chromic oxide addition.
On this basis, compositions with low CaO/SiO2 ratios are preferred.
. . ¦ l:)irect bonded compositions, Mixes J and K, were compared
¦ with a composition, Mix C, in which the chromic oxide addition was 4. 0%, in
. ¦ a rotary slag test referred to in the industry as the Valley Dolomite Slag Test
. furnace. The base composition of Mix C was the same as that of Mix A showr
: . in Table I. Siliceous electric furnace (EF) and Argon Oxygen Decarburizatior .
... (AOD) vessel slags were selected for the comparison of Mix C to Mixes J and
. . K because of their reactive nature with basic refractories and because they .
are thought to be involved in determining the wear rate of refractory linings
. - . in such vessels.
. The slag compositions are as follows:
. Chemical Cs:~mpos. i.tion - Synthetic Slag (%) .
. . L-5 (AOD) M-1 (EF)
CaO 27 33 . .
Si2 - 54 33 .
:: . Fe2O3 ~ 5 20 .
A12O3 ~ 4 -
., . MnO 5 5
. l. . MgO 5 5
. The original inner diameter (opening) of the refractory lined
-. ~ ~lag furnace is about 3". Ten pounds of sgnthetic slag pelletized to about 1"
.; .
diameter pellets were fed to the furnace at a rate of 4 lbs, durin6 the first
. r - - ~ hour and 2 lbs. /hr. dur;ng the îollowing three hours. The refractory
~',J~ hot f3cc was maintaincd at 3050 - 3150F. ' After cooling, the crodcd
~'''~`'' ' ' ~' ,. , '
~...................................... -23- .
I . ~.
; l ~
~ ~ ; ~
.- I 1(~5~S36
; I test shal)es were removed and c~lt to expose the inlerior structure from hot
to col~ face. The depth of slag penetration was estimated h~ the physical
¦ dellsification and discoloration and later confirmed ~y petrographic examina-
I tion. Mix C having 4.0% chromic oxide vras arbitrarily rated as unity or 1.
! I Ratir~gs greater than 1 indicate greater or deeper penetration. l~esults are
l given in Table VI below:
.' l .
TAB~.E VI
Relative Slag Penetratio
,: . I .
~ l Slag Mix J Mix K Mix C Conve}ltional (1)
':' I
: 10 L-5 1. 49 1, 63 1. 00 1, 18
M;l 1.09 1.11 1,00 1.02
; 60~o MgO class direct bonded without chromic oxide - similar to M~x A.
For the composition tested in Table VI, Mixes J and K, al-
though possessing desirable characteristics of low porosity and high density,
15 resis~ed the penetration of the siliceous slags to a lesser extent than Mix C.
Even more surprising is the poor relative results oE Mix K em~lGying chromi
oxide compared to Mix J or Mix C. These unexpected results can be ex-
plained by the fact that the native silicates of Mix C, controlled by the lime/
silica ratio; are more compatible and less reactive to the silicates of the slag
20 whose phases are determined by the lime/silica ratio, Considering Mix C
~ and slags L-5 and M-1, the lime¦silica ratios are about 1:1 orless than
- ~ . 1:1. Thc chromium cnriched spinels of Mix K are not as stable in the high
lime cnvironment of thc basic brick composition and are there~ore not as
: ` .
~ 24-
~ _
, ~ .
.
~ l ~ ~
, I
` I lOS~S3f~ .
¦ available to act as slag inhibitors as the chromite spincls in Mix C which
¦ has a lower lime/silica ratio. This again indicatcs the preferability of¦ lower CaO/SiO2 ratios.
¦ The importance of the low porosity and chromium rich spinel
¦ bonding, in combination with the preEerred low lime/silica ratio or lowbasicity of the native sil~cates, in resisting slag penetration is even more
I apparent from examination of the microstructures of the refractories în-
¦ volved in the following examples.
¦ EXAMPLE 4
l Equally important to increased slag resistance is the optimi-
¦ zation of mechanical properties, particularly thermal spall resistance.Direct bonded compositions were prepared similar to Mix A in Example 1,
one without chromic oxide and one modified by addition of 4% chrol~ic oxide.
After firin~,s in a commercial tunnel kiln at temperatures above 1700' C
- 15 (3100F) for at least four hours, the bricl; was tested for spall resis'ance
along with other commercially available classes of magnesite-chrome
; ~ refractory brick.
The test used for evaluation of thermal shock resistance is
called the prism spalling test. Prisms, 3 x 2 x 2", are cut from each brick
sample. The prisms are placed into an electrically preheated test furnace
until they reach 1205C (2200F). After holding at that temperature for 20
minutes, the samples are removed and cooled in still air for 10 minutes.
-; This procedure~ is repeated for up to 40 cycles or discontinued for those
~- ~;amples iractured by extensive crack propagation. Specirnens able to with-
~25 &tand a hi~her nurnber of cycles are more thermal shock resistant.
~'~ ~ ~ -25_
'. ~ ~ ! '
~ L
. `., .
` I 1059536
¦ The following classes of refractory bricks were compared:
. I Class Basic
I Refractory Chemical Co~nposition (~0) Cycl ~s
; l . S ~ _ e"O3 Al2S Cr~3 CaO ~
¦ Conventional .
. ¦ Direct Bonded 2. 0 7. 0 13, 7 15. 9 0. 9 60. 5 35-4
~.~
Direct Bonded
. with 4%
Chromic Oxide
Added 1. 6 6. 8 13.1 19. 0 0. 9 58. 6 38+
" _ ._ , .
; Rebonded, High
Fired, Fused
Grain 1.5 11.2 6.9 17.8 0.7 62.0 23
: . ' .
. High Fired
. 15 Direct Bonded
~lased on Pre-
. reacted Mag-
. nesia-Chrome
Sinter __ 2.9 9.5 6.6 22.7 1. a 57.1 15
,': . ~ .
Fused and - . -
. Cast Block 2.5 10.5 8.0 20.-0 0.5 56.5 2
(FeO)
:, ~ ,
. Prism spalling results indicate that while the brick of this
invention, high fired direct bonded wi~h chromic oxide added, has low por~
. osity and good slag resistance, the ther.rnal spalling resistance is as good a
. 25 or better than conventional direct bonded brick. The high fired brick with
. chromic oxide has better shock resistance than rebonded fused grain, pre-
. .~: rcacted grain base, or îused cast 60% MgO basic refractory produ .ts.
. ~ ~
"~.~ :. , .
. ,~ ,,..... .~
: ~ ' ~ 2~-
'~ ' _
____
,, ''~:
I .
l ~
~ l ~
1(~53~
¦ EX~M J'LE: 5
I Conventional direct bonded brick, Mix A of Examl~le 1, was
¦ modified by addition of 4% cl-romic o~ide to form Mix C and after firing was
¦ subjected to sla~ging tests and petrographic examination.
¦ The dynamic rotary slag test was again used. Slagging was
¦ carried out at 3050-3100F with 10 ll~s. of synthetic slag applied over a
period of four hours. The slag composition was as follows:
Slag Composition ~%)
CaO 37, 5
SiO2 37. 5
FeO 18. 0
A123 2. 0
, MnO 3.C'
P205 1. 0
~aF2 1. 0
:
Erosion and penetration factors were chosen as 1. 0 for
the composition including chromic oxide. Values over unity indicate more
erosion loss and penetration. Products were those tested in Example 4.
The te~t ults were as follows:
27-
''`,,' ~ .~ . .
.' ' ~ . ,
L ~
i2 ' ~ _
l()S~53~:i
Product Class, Slag Slag
Fired Erosion_ Penetration
Conventional (Mix A)1. 3 1.1
Mix ~ plus 4% Chromic
Oxide (Mix C) 1. 0 1~ 0
Rebonded Fused
Grain Brick . 6 1. 0
Pre-~ eacted
Grain Brick . 9 1. 0
Fused Cast . 5 . 7
Reference will now be rnade to the drawings which are for
illustrative purposes only.
., ' . . . 1, .
BRIEF DESCRIPT~ON OF THE DRA~INGS
FIG. 1 is a photomicrograph at magnification of 35X showing
the microstructure of slagged brick formed from Mix C in accordance with
the present invention.
FIG. 2 is a photomicrograph at magnification of 35X showing
the Inicrostructure of slagged brick formed from conventional Mix A.
- FIG. 3 is a photomicro~raph at magnification of 130X showing
~0 the microstructure of slagged brick formed from Mix C in accordance with
the present invention behind the hct face region of the brick.
FIG. 4 is a photomicrograph at magnification of 130X showing
the microstructure of slagged brick formed from Mix C in accordance with
the present invention in the hot face region of the brick.
- ~ . ' ,' ' . ':
~' ' . ~.
~ -28- ~
11159536
¦ FIG. 5 is a photornicrograph at magnificati~n of 130X showing
I the microstructure of slagged brick formed from convcntional Mix A in the
¦ hot face region of the brick.
¦ FIG. 6 is a photomicrograph at magnification of 130X showing
¦ the microstructure OI a conventional 50% MgO clirect bonded brick after in-
¦ use service. - -
FIG. 7 is a photomicrograph at magnification of 130X showing
¦ the micrc)structure of ~rick formed from Mix C in accordance with the pre-
¦ sent invention in the hot face region of the brick after in-use service.
¦ FIG. 8 is a photomicrograph at magnification of 140X showing
the microstructure of brick formed from Mix C in accordance with the pre-
sent invention ~ 1-2 mm behind the hot face after in-use service~
DET~ILEI~ DESCRIPTJC!N OF THE DRAW~GS
` . .'- ~ .
Figs. 1-5 show the microstructural differences between
is slagged direct bonded brick made in accordance with the present invention
and slagged conventional direct bonded brick r30t rnade in accordance with
the present invention. Fig. 1 shows the microstructure at low magnification
of brick îormed from Mix C in accordance with the present invention and
; slagged as described in Examples 4 and 5. Fig. 2 shows the microstructure
at the sam- magnification of brick formed from conventional Mv~ A and
slagged in the same manner. Slag appears at the top of both photomicro-
graphs.
The mechanism of slag penetration in both the brick of tl-e
prescnt invention and the convcntional b~ ick includes penetration through
~, '
i ; -29-
~ . _
o ~ ~ -
1 1055~5;~;
boundaries betwccn individual periclase and chromite crystals. An out-
standing feature of this invention, visil)le in the microstructure of Mix C,
I even at the low magnification of Figs. 1 and 2, is the s~lperior amount of
¦ periclase-chromite bonding of the Mix C brick as compareci to the Mix A
1 standard direct bonded brick.
¦ The amount of slag liquid as a function of depth behind the
hot face was found to more rapidly decrease for the Mix C brick than for the
conventional Mix A brick. The Mix C brick also exhibits more secondary
spinel at smaller depth behind the hot face. As shown in Fig. 3, "anchors",
or recrystallized protrusions of spinel extend from chromite crystals
(marked "Cr") into the brick matrix in the brick of Mix C. These "anchors"
of spinel are believed to significantly contribute to the strength of the brick.
The microstructural cietails of the hot face regions of the
Mix C al!d Mix A brick are shown in Figs. 4 and 5 respectively. As noted
above, the amount of periclase (P)-chroAnite (Cr) bonding is one of the out-
standing features of the brick of this invention. The microstructure of the
hot face re~,ion of the conventional direct bonded brick made from Mix A,
shown in Fig. 5, exhibits a greater amount of intragranular slag penetration
than that of the Mix C brick, The sponge-like appearance of the chromite
particles ~fter slag attack appears to be characteristic of attack by a slag
of the chemical composition used in Exarnple 5.
The photomicrographs of the drawinas show that d brick mad
in accordance wlth the present invention retains spinel bonding closer to the
hot face regi~n than conventional direct bonded brick, That accounts for
the superior slaE~ ero~lon and penetration resistance, as well ns the improv d
1 105953~;;
,-- ¦higl~ lem~erature properties of bricli made by this invention as compared to
¦ conventional direct bonded 1 riclc. I\Ioreover, brick made in accordance
¦ with the present invention is more useful than other slag resistant products
¦ available, such as rebonded fused grain brick, prereacted grain brick, and
¦ fused cast brick of Examples 4 and 5 due to its su~,erior thermal shock
re si stance .
Figs. 6-8 show the miclostructural details of brick after
service in a 100 ton AOD vessel. Fig~ 6 shows the microstructure of a 507
MgO direct bonded brick at the region of the hot face from above the metal
line. In this conventional used brick, silicates (S) have disrupted much of
the bonding bctween rounded periclase crystals (P). Silicate phases at the hot
face are merwinite and monticellite which are also present in the reduction
cycle slag composition.
Samples of improved direct bonded brick made in acccrdance
with the present invention in the 60% MgO class were examined after a_hieving
excellent re.sults in AOD test panels compared with other direct bonded pro-
ducts. Fig. 7 shows the microstructural details of the improved direct bonded
. brick of this invention at the region of the hot face. Intercrystalline silicate
penetration is exhibited only at the immediate hot face.
The photomicrograph of Fig. 8 taken at 1-2 mm behind the hot
face shows the presence of spinel structures ~Sp) which apparently play a
~ major role in the retention of brick integrity near the working surface. This
- 1~ ond retention accounts for the improved service life achieved by the direct
~; bonded brick of the present invcntion. ~ .
Tl e ~tc,tural integrlty resultin~ from the bondin~ by thermally
a~ ~ tab1e 8plncl~ clo~o ~o thC hot facc, in coml)ination ~th the good thcrmal
~ ~ " J ' ~* ~
~ ~ ~ '' ' _
~ l ~ ~
I ~OS~S36
¦ shock resistance, accounts for the superior performance of this improved
¦ direct bonded brick in oxygen blown open hearth roofs and back walls as
well as AOD vessels and electric furnace side walls.
I The invention in its broader a.spects is not limited to the
¦ specific details shown in the examples and drawings and described in the
¦ specification.- Departures may be made from such details wit'nout depart-
¦ ing from the scope or spirit of the invention as set forth in 'he appended
Icl~is.
I 1 . I
: ' ' : , '
. , ~ . . .
~ ' i '
