See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/292887249

Development of a Zirconia-Graphite Refractory for a SEN Application

Article in InterCeram: International Ceramic Review · June 2014

DOI: 10.1007/bf03401049

CITATIONS READS

0 98

4 authors, including:

Gaurav Gugliani
Indian Institute of Technology (Banaras Hindu University) Varanasi
11 PUBLICATIONS 27 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Statistical analysis of wind data View project

All content following this page was uploaded by Gaurav Gugliani on 11 March 2016.

The user has requested enhancement of the downloaded file.

 136 INTERCERAM – REFRACTORIES MANUAL I/2014

Special TechnologieS

ied. F
bon p
80 %
mater
fractio
stabili
amou
tents w
G. Gugliani*, S.K. Pal*, A. Sen**, B. Prasad** We al

Development of a Zirconia-Graphite
certai
carbo

Refractory for a SEN Application

99 %
were
(ACB

The auThor absTracT Keywords 2 Exp

2.1 Ra
The corresponding author, Dr. Zirconia-graphite refractories are generally used in the slag-like zone of zirconia, graphite,
graph
Sumit Kumar Pal, received his the subentry nozzle in continuous casting. A zirconia-graphite refractory fixed carbon, spalling
Ph.D. from Jadavpur University has superior thermal shock resistance and corrosion resistance compared resistance Ca-sta
and his B.Tech. in Ceramic to MgO−C and Al2O3−C refractories. The effect of graphite on the zirco- INTERCERAM 63 (2014) [3] tions,
from University of Calcutta nia-graphite refractory is not well documented. Therefore, in the present silicon
(India). He is currently an work an attempt has been made to evaluate the effect of graphite on the carbo
Assistant Professor at the properties of the zirconia refractory. Graphite can be classified in terms
and a
Department of Ceramic En- of fixed carbon content and in this work three different types of graphite
gineering, National Institute of Technology, with varying fixed carbon content were used (99, 94, and 80 %). The raw m
Rourkela. E-Mail: [email protected] amount of zirconia was maintained at 82 % and that of the graphite at the zi
18 %. Other processing parameters and ingredient concentrations were Table
constant throughout the work. Refractory samples were fabricated using comp
the cold isostatic method and refractory samples were coked at 1000 °C
comp
in a reducing atmosphere. The effect of fixed carbon percentage on ap-
parent porosity, bulk density, spalling resistance, and oxidation resist-
ance were evaluated. 2.2 Sa
The C
fractio
carbo
1 Introduction The corrosion resistance of a zirconia- may be corroded by the action of the two unifor
Submerged entry nozzle/subentry nozzles graphite refractory increases with an in- mechanisms noted below. mixer
(SEN) play a vital role during the flow of crease in the zirconia content. But, as zirco- 1. Corrosion of the binding material and resin
molten steel from the tundish to the mold. nia has low thermal conductivity, the high impurities present at grain joints by alka- mixed
The nozzle is used to prevent molten steel percentage of zirconia leads to substantial line ions (Na+) and by fluor (F–) with the then p
from coming into direct contact with at- thermal expansion thereby increasing the formation of liquid phases at the temper- ture a
mospheric oxygen and to avoid molten steel possibility of spalling cracks in the slag ature of the steel. The f
splash during the flow of steel from the band of the subentry nozzle. Although an 2. Attack of the ZrO2 grain peripheries and trolled
tundish to the mold, thus ensuring that increase in the graphite content improves dissociation into numerous small parti- zircon
good quality steel is produced during the the thermal spalling resistance parameter it cles within a vitreous “slag” with a low then
casting process. The lifetime of a subentry results in poor refractoriness. Researchers fusion point. Consequently corroded at 200 m
nozzle depends upon its capability to with- have carried out experiments to optimize its periphery, the zirconia which was 1000
stand the attack of molten slag which leads zirconia and graphite compositions. They stabilized and crystallized in the form of specim
to a high risk of corrosion. Therefore, an found that compositions with more than cubic crystals (c. ZrO2) becomes destabi- a redu
improvement in the corrosion resistance of 80 mass-% zirconia will be prone to lized and transforms into small particles furna
refractory materials used at the slag line lev- spalling cracks in the standard preheating of monoclinic zirconia (m. ZrO2). acteri
el helps in achieving longer sequence casting practice for SEN. However, zirconia con- In addition in relation to its rheological
of steel. As aluminous or magnesian refrac- tents of up to 85 mass-% can resist spalling properties (fluidity, surface tension), this 2.3 Ch
tories cannot cope with the corrosion when using a strict and controlled preheat- liquid slag layer can easily wet the nozzle 2.3.1 S
caused by mold powders, zirconia-graphite ing practice for SEN. surface and reach parts that are not protect- Slag c
materials are commonly applied at the slag ed by inserts (especially the orifices). The re- the ro
line level due to their excellent corrosion re- 1.1 Corrosion mechanism sult of this is a corrosion of the lower part of 150 ×
sistance as compared to other refractory The bound carbon and graphite present in the nozzle which can bring about premature The s
materials. the refractory are oxidized by the oxygen rupture and stoppage of the casting process tempe
present in the air and by the oxides present [1]. dryin
in the slag, resulting in the formation of a Herein, the effect of different fixed carbon start o
decarburized layer. This decarburized layer (FC) percentages in graphite on the physi- at 100
* National Institute of Technology, Depart. of Ceramic
Engineering, Rourkela, Odisha (India) results in a loosening of the bond between cal, thermal, and mechanical properties of and 5
** OCL India Limited, Rajgangpur, Odisha (India) zirconia grains and as a result zirconia grains zirconia-graphite refractories has been stud- slag w

136-140_ST_Kumar_Pal.indd 136 23.05.14 13:58

 INTERCERAM – REFRACTORIES MANUAL I/2014 137

Special TechnologieS

ied. Flake graphite with varying fixed car- Fig. 1 • Thermal

expansion behavior
1 0.007
bon percentages (99 % FC, 94 % FC, and
80 % FC, respectively) was used as a filler as a function of 0.006
fixed carbon
99% FC
material. Different zirconia particle size 0.005 94% FC
fractions were chosen. The amount of Ca- 80% FC
stabilized zirconia was 82 mass-% and the 0.004 94% FC +ACB

∆L/L
amount of graphite with different FC con- 0.003
tents was fixed at 18 mass-%.
0.002
We also studied the possibility of replacing a
certain amount of graphite with amorphous 0.001
carbon black. In this regard, 2 mass-% of 0.000
99 % FC graphite and 94 % FC graphite
-0.001
were replaced by amorphous carbon black
(ACB). 0 200 400 600 800 1000 1200 1400 1600

s
Temperature / °C
2 Experimental
2.1 Raw materials for the zirconia-
graphite refractory
ling
Ca-stabilized zirconia of different size frac- Simultaneously the temperature was in- 3 Results and discussion
14) [3] tions, flaky graphite (purity 99, 94, 80 %), creased to 1550 °C by increasing the gas 3.1 Dilatometry test
silicon metal as an antioxidant, amorphous pressure and fuel supply. The temperature Generally, the strength of a zirconia-graph-
carbon black, resin powder, resin liquid, was kept constant at 1550 °C for 30 min. ite refractory comes from carbon interlock-
and an antioxidant were used as starting After 30 min of testing, the molten metal ing due to polymerization of the resin. Sin-
raw materials. The batch compositions of and slag was tapped out from the machine tering of zirconia particles is purposefully
the zirconia-graphite mixes are shown in and fresh slag and metal again added to it avoided due to the requirement of high tem-
Table 1. In the Table 1 the word C denotes for further testing. In this way 2 tapping was perature, which would lead to loss of carbon
composition. For the present work five done. The testing was performed on the ba- from the refractory. A dilatometry study was
compositions had been chosen. sis of ASTMC-874. The slag has a C/S ratio carried out for the samples to determine the
of 3 (mass percentage ratio of CaO/SiO2). maximum possible temperature up to which
2.2 Sample preparation a specimen could be fired without sintering.
The Ca-stabilized zirconia of different size 2.3.2 Spalling resistance It can be observed from Fig. 1 that sintering
fractions, flaky graphite of different fixed Specimens of size 50 ± 2 mm in the form of initiates at around 1100 °C for refractories
carbon percentages, and antioxidant were a square with 75 mm height were first coat- containing 80 % FC graphite, 94 % FC
e two uniformly heated and mixed in a Sigma ed with an antioxidant material to protect graphite, 99 % FC graphite, and 94 % FC
mixer for 10 min after which the powder them from oxidation. Then the specimens graphite with ACB. Sintering initiates at the
l and resin was added and the mixture was then were dried for 10 h followed by air cooling. same temperature, that is around 1100 °C.
alka- mixed for 5 min. Liquid phenolic resin was The specimens were then kept in a furnace One observes therefore that carbon content,
th the then poured into the zirconia-graphite mix- for 10 min maintained at 1200 °C followed whether in the form of graphite or in an
mper- ture and it was mixed for a further 30 min. by water quenching for 10 min. This cycle amorphous form, has no particular effect on
The final mixture was then kept in a con- was repeated for 15 cycles. Also Young’s the sintering temperature. Therefore, the
es and trolled atmosphere for aging over 16 h. The modulus of the specimen was calculated firing temperature for the samples was fixed
parti- zirconia-graphite refractory mixture was prior to the Spalling test so that comparative to 1000 °C, which is lower than the sintering
a low then shaped (diameter 150 mm, length data can be generated. initiation temperature; this was to avoid any
ded at 200 mm) using a cold isostatic press at kind of carbon loss from the refractories.
h was 1000 bar. After curing at 270 °C for 5 h the 2.3.3 Oxidation test
rm of specimens were fired at 1000 °C for 10 h in Specimens of height 50 mm were kept at 3.2 Phase analysis (XRD)
stabi- a reducing atmosphere using a Bell crank 900 °C for 2 h. Before and after the samples The XRD pattern of the zirconia-graphite
rticles furnace. The specimens were cut for char- were placed in the furnace an initial and refractory samples after firing is shown in
acterization. final sample weight was noted and in this Fig. 2. A cubic zirconia and tetragonal zirco-
ogical way the weight loss was calculated. nia phase was identified as the major phase
, this 2.3 Characterization
nozzle 2.3.1 Slag corrosion test
Table 1 • Batch formulation of the zirconia-graphite refractory with varying fixed
otect- Slag corrosion resistance was evaluated with carbon percentage
he re- the rotary drum test. Samples of dimension
Chemical composition C-1 C-2 C-3 C-4 C-5
art of 150 × 40 × 40 mm were cut from each batch.
CaO-stabilized zirconia 82 82 82 82 82
mature The samples were then coated with a high-
rocess temperature coating material followed by Graphite 99 % FC 18 16 – – –
drying inside a dryer for 10 h. Before the 94 % FC – – 18 16 –
80 %FC – – – – 18
arbon start of testing the samples were preheated
physi- at 1000 °C for 1 h. Then 50 mass-% metal Resin binder 12 12 12 12 12
ies of and 50 mass-%-LD (Linz and Donawitz) Antioxidants 0.5 0.5 0.5 0.5 0.5
stud- slag was placed inside the sample holder. Amorphous carbon black – 2 – 2 –

136-140_ST_Kumar_Pal.indd 137 23.05.14 13:58

 138 INTERCERAM – REFRACTORIES MANUAL I/2014

Special TechnologieS

2 30 60 3 4
3000
80C 22
2000

Apparent porosity / %

Bulk density / g/cm³

1000 C T,Cu
0 20
3000
94C+ACB
Intensity / a. u.

2000

1000 18
T,Cu
C
0
3000 94C
2000 16
C
1000
T,Cu 80%FC 94%FC 94%FC +ACB 99% FC 99% FC +ACB
0
3000 C 99C Carbon / %
2000
Fig. 3 • Apparent porosity as a function of fixed carbon
Fig. 4 •
1000 T,Cu
Table 3 • Properties evaluated with amorphous carbon black
0 6
30 60 Composition C2 C4
Apparent porosity / % 21.3 22.2
2Θ/° Bulk density / g/cm3 3.13 3.14
Modulus of rupture / kg/cm2 66 66

Width loss / mm
Fig. 2 • XRD pattern of the zirconia-graphite refractory with 80 % FC graphite,
94 % FC graphite, 99 % FC graphite, 94 % FC graphite + ACB Slag corrosion resistance index 19.12 24.19

with small quantities of a monoclinic phase. of the ACB-containing refractory is com- which 2 mass-% ACB replaced some of the
It can be seen from the XRD pattern that as paratively higher for every fixed carbon per- graphite has higher apparent porosity than a
the purity of graphite decreases the intensity centage of graphite. The apparent porosity refractory without ACB.
of the carbon peak decreases and the lowest of the samples decreases from 21.1 to 18 % It can be observed from Fig. 4, that bulk
intensity peak was observed for 94 % fixed as the purity of graphite increases from 80 density increased as the percentage of fixed
carbon graphite with ACB which reveals to 99 %. This could be because 99 % FC carbon present in the graphite increased.
that the percentage of fixed carbon present graphite has lower amounts of volatile mat- Moreover, the BD of the ACB-containing
in the zirconia-graphite refractory has de- ter content and 80 % FC graphite has higher refractory is comparatively less than that of
creased. amounts of volatile matter content. The samples without ACB for every fixed carbon
high volatile content in the 80 % FC-con- percentage of graphite. Again, this could be
Fig. 6 •
3.3 Apparent porosity (AP) and bulk taining graphite vaporizes during coking, attributed to the fact that 99 % FC-contain-
density (BD) leaving behind pores, hence the apparent ing graphite has less volatile content com-
The different property such as apparent porosity increases as the percentage of fixed pared to the 94 and 80 % FC-containing
porosity (AP), Bulk density (BD), Modu- carbon present in the graphite decreases. graphite samples. Additionally, ACB has 68 to
lous of Rupture(MOR) and corrosion Re- Moreover, the apparent porosity of the higher free energy. Therefore, the volatile crease
sistance values for different Fixed carbon ACB-containing refractory is 21.3 and content and free energy are responsible for of the
graphite containing Zirconia refactories has 22.2 %. This could be due to the fact that the leaving behind the pores after vaporization parati
been reported in Table 2. In Table 3, differ- refractory in which ACB replaced some of during coking. Thus, BD is higher for the out A
ent properties of ACB contaiing refractories the graphite had a lower carbon content af- 99 % FC-containing graphite refractory. of gr
has been reported for composition C2 and ter coking, which was observed from the appar
C4. It can be observed from Fig. 3, that there carbon intensity peak in the XRD analysis. 3.4 Cold modulus of rupture crease
was a gradual decrease in apparent porosity Amorphous carbon black has a higher free It can be observed from Fig. 5 that the cold the cr
as the fixed carbon percentage of flake energy compared to graphite, therefore, it modulus of rupture (CMOR) gradually in- is app
graphite increased from 80 % FC to 99 % sublimates at higher temperatures leaving creases as the FC content of the graphite centra
FC. Moreover, the apparent porosity value behind the porosity. Thus, the refractory in increases. The CMOR value increased from of the
Hence
Table 2 • Variation in AP, BD, MOR and Corrosion resistance for different percentages of fixed carbon result
99 % FC graphite 94 % FC graphite 80 % FC graphite ulus o
Trial T-1 T-2 T-3 T-1 T-2 T-3 T-1 T-2 T-3
3.5 Sl
Apparent porosity / % 21.5 17.9 18 20.2 23.1 21.1 22.8 23.3 21.1
It can
Bulk density / g/cm3 3.15 3.25 3.26 3.22 3.09 3.14 3.13 3.11 3.17
2
sistan
MOR / kg/cm 76 73 74 70 65 70 65 68 68
increa
Corrosion resistance*, width loss / mm 15.7 20.36 17.81 17.18 20.74 19.11 21.81 24.19 20.55 in the
* The corrosion resistance test was carried out with the Rotary Drum Method with metal and slag [basicity (C/S) = 3], at a tempera- tainin
ture of 1550 °C ± 50 °C, metal and slag were replaced twice over one run. resista

136-140_ST_Kumar_Pal.indd 138 23.05.14 13:58

 INTERCERAM – REFRACTORIES MANUAL I/2014 139

Special TechnologieS

4 3.30 5 80

Modulus of rupture / kg/cm²

3.25 75
Bulk density / g/cm³

3.20 70

3.15
65

3.10
60

3.05
55

3.00
80%FC 94%FC 94%FC +ACB 99% FC 99% FC +ACB 50
ACB 80% FC 94%FC 94%FC +ACB 99%FC 99%FC+ACB

Carbon / % Carbon / %
Fig. 4 • Bulk density as a function of fixed carbon Fig. 5 • Cold modulus of rupture as a function of fixed carbon

ck
6 7
24
2
4 22
Width loss / mm

20
9
18
of the
than a 16

14
bulk
fixed 12
eased.
aining 10
80% FC 94%FC 94%FC +ACB 99%FC 99%FC+ACB
hat of
arbon Carbon / %
uld be
Fig. 6 • Slag corrosion resistance as a function of fixed carbon Fig. 7 • Scanning electron microscopic image of (a) 99 % FC graphite, (b) 94 %
ntain-
FC graphite, and (c) 80 % FC graphite in the zirconia-graphite refractory
com-
aining
B has 68 to 76 kg/cm2 as the percentage of FC in- of graphite. The slag resistance increases as trate the zirconia refractory resulting in
olatile creased from 80 to 99 %. The CMOR value the fixed carbon percentage increases. This more width loss compared to a zirconia-
le for of the ACB-containing refractory is com- is caused by the covalent nature of the bond graphite refractory without ACB.
zation paratively less than that of refractories with- along the basal planes of the graphite. This
or the out ACB for every fixed carbon percentage covalency characteristic increases as the 3.6 Spalling resistance
y. of graphite. This could be related to the purity of graphite increases resulting in low All samples were tested for 15 cycles. No
apparent porosity of the material as an in- free energy over the entire surface area. Ac- crack was observed until after 15 cycles. This
crease in the number of open pores reduces cordingly, both reactivity and wettability by could be related to the thermal conductivity
e cold the cross-sectional area across which a load the liquid slags and metals are very low [3]. of the graphite. As the percentage of fixed
ly in- is applied. The pores also act as stress con- As 99 % fixed carbon graphite may have carbon present in the graphite increases, the
aphite centrators, thereby resulting in a reduction very low free energy, its reactivity and wetta- carbon promotes higher thermal conductiv-
from of the flexural strength of the specimen [2]. bility in the molten metal and liquid slags is ity and as a result the graphite easily pro-
Hence, a decrease in the apparent porosity very low. Therefore, a refractory with graph- motes the transfer of heat from the hotter
results in an improvement in the cold mod- ite containing 99 % FC shows better slag side to the cooler side, thereby reducing the
ulus of rupture of a specimen. corrosion resistance compared to graphite temperature gradient present inside the
-3
with 94 and 80 % FC. Moreover, the ACB- refractory wall. Thus, as this gradient is
3.5 Slag corrosion resistance index containing refractory has a lower slag resist- reduced the thermal shock resistance im-
1.1
It can be observed from Fig. 6 that slag re- ance due to its higher free energy compared proves.
17
sistance of the zirconia-graphite refractory to graphite. This high energy increases car-
68
increases as the percentage of fixed carbon bon reactivity at high temperatures, thereby, 3.7 Scanning electron microscopy
.55 in the graphite increases and the ACB-con- causing it to become oxidized resulting in a The microstructure of the zirconia-graphite
era- taining refractory has a lower slag corrosion loosening of the zirconia-graphite bond. refractories with graphite having 99 % FC,
resistance for every fixed carbon percentage Therefore, it was easy for the slag to pene- 94 % FC, and 80 % FC are shown Fig. 7. The

136-140_ST_Kumar_Pal.indd 139 23.05.14 13:58

 140 INTERCERAM – REFRACTORIES MANUAL I/2014

Special TechnologieS

samples were fired at 1000 °C in a reducing 4 Conclusions ACB. This could be because the ACB has
atmosphere. It can be observed that in the Our studies on zirconia-graphite refrac- higher free energy therefore it vaporizes
case of 99 % FC-containing graphite, the tories with variation in the fixed carbon during coking resulting in a deterioration of
flaky nature of the graphite is retained in the content of the graphite, suggested that re- the thermo-mechanical properties as well as
refractory. This could be the reason for its fractories prepared with 99 % FC showed corrosion behavior of the refractory. There-
better thermal spalling resistance. Pores enhanced physical, mechanical, and thermal fore, ACB is not a potential candidate to
were also observed in the samples. For 94 % behavior as well as high corrosion resistance enhance thermal spalling resistance as well
FC-containing graphite, it can be observed when compared to 94 % FC- and 80 % FC- as corrosion resistance of refractories.
that the flaky nature of the graphite was re- containing graphite. This could be due to
tained in the sample. For 80 % FC-contain- 99 % FC having the lowest volatile content
ing graphite, the flaky nature of the graphite thereby retaining a large amount of carbon References
is not as well maintained. An increase in the after coking. This carbon helps in improv- [1] James, P.: Natural flakes graphite and carbon fiber
number of pores can also be observed in the ing the corrosion as well as spalling resist- addition to dolomite carbon refractory. Proc. UNIT-
CER 2 (1989) 451−456
samples from 99 % FC to 80 % FC. The poor ance of the refractory. [2] Callister, W.: Material science and engineering. 7th
spalling resistance as well as the poor corro- There was no remarkable improvement in Edition. John Wiley & Sons (2007), Minnesota,
sion resistance of 80 % FC-containing the thermo-mechanical properties or cor- USA, ISBN 978-0470419977
[3] Aneziris, C.G.: Magnesia-carbon bricks: A high-duty
graphite may be due to the destruction of rosion behavior of the refractory where refractory material. Interceram Refractories Manu-
the graphitic structure in the refractory. 2 mass-% of the graphite was replaced by al (2003) 22−27

Received: 02.12.2013

I N D E x o f ADv E Rt I S E RS

ECRef European Centre for Refractories GmbH D-Höhr-Grenzhausen 116

Euromoney Institutional Investor (Jersey) Ltd. HKG-Wan Chai IFC
Maschinenfabrik Gustav Eirich GmbH & Co KG D-Hardheim 81
FORM+TEST Seidner&Co. GmbH D-Riedlingen 91
JUMO GmbH & Co. KG D-Fulda 85
Linn High Therm GmbH D-Eschenfelden 87
LUM GmbH D-Berlin 89
MunichExpo GmbH D-München 86
Nürnberg Messe GmbH D-Nürnberg 88
Rimini Fiera S.p.A. I-Rimini IBC
Tosoh Europe B.V. NL-Amsterdam 83
Unifair Exhibition Service Co. Ltd. CN-Guangzhou BC

136-140_ST_Kumar_Pal.indd 140 23.05.14 13:58

View publication stats