-
Impact of activated nut coke on energy efficiency in the blast
furnace
Lena Sundqvist kvist 1, Carina Brandell
2, Maria Lundgren
1
1Swerea MEFOS AB
Box 812, SE- 971 25 Lule, Sweden
+46 920 20 19 00
[email protected]
[email protected]
2 Luossavaara-Kiirunavaara AB (LKAB)
Box 952, SE-971 28 Lule, Sweden
+46 920 38 000
[email protected]
(Note: Please add more authors in the same format if
necessary)
Keywords: Blast furnace nut coke, activation, energy efficiency,
thermal reserve zone
INTRODUCTION
When the temperature in the thermal reserve zone is reduced, the
equilibrium between CO/CO2 and Fe/FeO is shifted towards
metallic
Fe at a specific reducing power of the gas as can be seen in
Figure 1. The use of highly reactive nut coke mixed ferrous
materials is
one possible method for reducing the temperature.
Figure 1 Fe-O-C equilibrium diagram including the Boudouard
curve as well as the degree of reduction in the presence of CO
and
CO21
In studies conducted by Nomura et al. 2, 3
it has been shown that the reactivity of coke can be increased
either by adding a catalyst to
the coal blend prior coking or to treat the produced coke.
Reactive coke was successfully produced by using a coal type with
Ca rich
ash in the coal blend for coke making2. In laboratory testing
the coke showed high chemical reaction rate with CO2 at 950 C, but
the
reactivity index (CRI) increased less and the strength after
reaction (CSR) could be more or less kept unchanged if the addition
of the
Ca-rich coal was less than 10%. In full scale coke production
5-7% of the Ca rich coal was added. Adiabatic BF simulator trials
with
sinter showed decreased thermal reserve zone temperature and
higher reduction degree of sinter for the Ca-rich coke compared
to
ordinary coke. The total reductant rate of the operational BF
decreased when changing from ordinary coke to the Ca-rich coke.
Results
were also reported from the addition of iron ore or iron oxide
prior to the production of a reactive coke 4. The iron oxide
caused
decreased strength after reaction and increased reactivity of
the coke but the strength measured as drum index was proper. The
iron
oxide has a tendency to react with the silica brick in the coke
oven at temperatures of 1200C but not at a temperature of
1100C.
A catalyst can be added post coke making and deteriorating
effects on strength can be avoided3. In laboratory testing it was
estimated
that 70% of the addition made as a solution or slurry with a
desirable composition should sustain raw material handling. The
reaction
rate and the weight loss are increased with the addition of Fe
and Ca containing compounds and the results indicate that the
thermal
-
reserve zone temperature may decrease and improved reduction of
ferrous material possibly be achieved. The authors think that
more
studies are needed to explore and confirm the effect
further.
By using a high ratio of nut coke mixed with the ferrous
material the high temperature properties of the iron ore in terms
of reduction
degree was improved up to 10% addition5. Improved shaft
efficiency decreased reductant rate and the permeability of the
cohesive
zone were stated in industrial trials charging approximately 100
kg nut coke/tHM. The reactivity of the coke is not stated but the
coke
has a CaO content of 0.25 wt%.
Aiming for reduced consumption of reducing agents in the BF, the
most appropriate method for activation of nut coke were
analyzed
in the laboratory and later up-scaled for use in pilot scale
treatment at LKAB in Malmberget. To explore the effect on the
process
conditions and the energy efficiency operational trials in the
LKAB experimental BF in Lule were planned and conducted. The
results from the trials are presented and discussed.
EXPERIMENTAL
The LKAB Experimental Blast Furnace (Experimental BF)
The first campaign in the Experimental BF was conducted in 1997
and these trials were conducted as one part of the 29th
campaign. It
is usually operated twice every year and the length of each
campaign is approximately 6 to 8 weeks. The operation and control
of the
Experimental BF are comparable with the industrial BF but the
response time is shorter.
Figure 2 Schematic layout of Experimental BF plant
The Experimental BF, of which the schematic layout is shown in
Figure 2, is equipped for flexibility in type of in injection
and
process settings. The plant is compact and all equipment is
installed in a single building. The raw material handling allows
the usage
of up to four different ores simultaneously. Injection of
pulverized coal, gas or oil is possible, as well as injection of
fluxes or other
oxides.
Table I Positions for installed measurements used in the
evaluation of data
Type of equipment Position below stockline Position on
radius
Blast furnace burden level 0
Upper skinflow -0.44 30 , 90 ,150 ,210 , 270 , 330
Upper shaft probe -0.90 From 30 to 150 , 9 positions
Lower skinflow -1.05 30 , 90 ,150 ,210 , 270 , 330
Lower shaft probe -3.46 From 30 to 150 , 9 positions
Inclined shaft probe -4.21 From 210 to 0
Tuyeres -5.7 30 , 150 , 270
-
The Experimental BF is also well-equipped for measuring of
temperature and pressures. The shaft has no water cooling,
thermocouples are installed inside the refractory and pressure
measurements over the height of the shaft. In order to study
the
atmosphere inside the shaft and collect material samples during
operation, material and gas/temperature probes are fitted to the
blast
furnace, as shown in Figure 3.
Figure 3 The Experimental BF including shaft probes, inclined
probe and vertical probe
The vertical temperature profile of the Experimental BF can be
measured by inserting a thermocouple from the top allowing it
to
descend with the burden. This measurement normally withstands
temperatures up to approximately 1250C before the thermocouple
is
damaged. The levels where thermocouples and other important
blast furnace equipment have been installed are listed in Table I
and
the radial sampling and measurements with probes are shown in
Figure 4.
Figure 4 Radial position of lower shaft and inclined probes
-
Raw Materials
The ferrous burden for all periods was LKAB olivine pellets from
Malmberget. Quartzite was added to reach the desired slag
volume
and a limestone and BOF slag was used for basicity adjustment of
the final slag. The chemical compositions of raw materials,
including the activated coke, are given in Table II.
Nut coke and coke used in the coke layers were crushed and
screened into the fraction 15-30 mm. Three types of treatments were
used
for nut coke to be added to ferrous layers; a solution with
Zr-tracer, a slurry of magnetite with Zr-tracer and a slurry of
burnt lime with
Zr-tracer. The solutions and slurries were sprayed on the coke
under rotation in a drum (left photo in Figure 5) and finally the
coke
was dried (right photo in Figure 5) and samples for analyses
taken before transportation to the Experimental BF plant.
Figure 5 Processes of preparation of activated nut coke by
coating and drying of coated coke afterwards
The original nut coke used as well as the Zr-traced one has
similar chemical composition as the coke used in the coke layers.
As the
activated coke was coated with slurries of magnetite or hydrated
lime the chemical composition was changed somewhat. The content
of C was lowered at the same time as either the content of Fe or
CaO increased. The slurries also contained zirconium sulphate
for
tracing the activated nut coke in samples collected during the
trials.
Figure 6 Moisture content in coke and nut coke used during the
Experimental BF tests
Samples of raw materials were collected every time the raw
material bins were filled up. Moisture content of coke and ferrous
material
as well as the particle size distribution of pellets were
analyzed for each sample. In case of changed moisture content to a
new stable
level the new values were introduced in the system and used in
the recipe calculations. The variation in measured moisture content
of
pellets was minor and the variation in coke moisture is shown in
Figure 6. To minimize the variations in coke moisture one
single
delivery of coke was used during the trial and in general the
moisture was in the range of 4-6%. However, a few significant
changes in
moisture was experienced although this precaution. The lime
coated coke had lower moisture content compared to the other
coke
types due to longer time for drying after coating.
0
2
4
6
8
10
12
14
13-05-01 00:00 13-05-06 00:00 13-05-11 00:00 13-05-16 00:00
13-05-21 00:00 13-05-26 00:00
% m
ois
ture
Large sample Small sample
Zr Large Zr small
Lime large Lime small
Magnetite large Magnetite small
Lime coated coke
-
Table II Raw materials used during the trial and their
composition, given in wt.-%
MPBO Limestone Quartzite
BOF
slag
Nut coke
Coke PC
Lime
activated
Magnetite
activated
Original &
Zr-traced
CaO 0.43 58.0 0.199 40.5 1.30 0.03 0.03 0.03 0.37
MgO 1.31 0.93 0.23 11.8 0.09 0.09 0.09 0.09 0.16
SiO2 1.76 1.06 97.9 8.75 7.03 6.49 6.62 6.62 3.24
Al2O3 0.35 0.48 0.75 1.71 3.12 3.03 3.09 3.09 1.80
TiO2 0.30 0.02 0.03 1.49 0.16 0.18 0.18 0.18 0.06
V2O5 0.21 0.01 0.004 4.57 0.00 0.19 0.00 0.00 0.00
Na2O 0.07 0.13 0.061 0.022 0.022 0.022 0.04
K2O 0.01 0.09 0.247 0.02 0.098 0.069 0.07 0.07 0.1
S 0 0.069 0.029 0.063 0.57 0.55 0.55 0.55 0.340
P 0.01 0.002 0.005 0.29 0.020 0.007 0.007 0.007 0.017
Mn 0.04 0.01 0.01 3.13 0.00 0.00 0.00 0.00 0.02
Fe 66.8 0.058 0.149 19.067 0.41 2.02 0.53 0.53 0.609
C 85.4 85.4 87.1 87.1 82.9
H2 0.087 0.087 0.089 0.089 4.10
O2 0.20 0.20 0.20 0.20 3.77
N2 1.16 1.16 1.188 1.19 2.18
Test Conditions
The test periods operated and evaluated consisted of four main
parts
Reference period without nut coke in ferrous layers
Reference periods with 150 kg/tHM of original or Zr-traced nut
coke charged in ferrous layers
Test period with 150 kg/tHM of Zr-traced, magnetite activated
nut coke charged in ferrous layers
Test period with 150 kg/tHM of Zr-traced, lime activated nut
coke charged in ferrous layers
From the operational tests, stable data periods stated in Table
III were selected for the evaluation of in-furnace conditions,
operational conditions as well as energy efficiency of the set-ups.
The two periods for lime coated nut coke has the same start but the
long period
extends longer in time including some slightly more varied
conditions.
Table III Evaluation periods, 150 kg/tHM of nut coke was used in
the ferrous layers in all test periods except for reference
Start End Recipe, type of nut coke Short name of period
Duration
2013-05-02 21:00 2013-05-04 10:00 Reference without nut coke
Ref. 37.0 h
2013-05-15 14:01 2013-05-15 23:54 Original nut coke Orig. 9.9
h
2013-05-15 23:56 2013-05-17 01:53 Zr-traced coke Zr-traced 26.0
h
2013-05-17 14:00 2013-05-19 12:30 Magnetite coated Zr-traced nut
coke Magnetite 46.5 h
2013-05-20 14:00 2013-05-21 15:00 Lime coated Zr-traced nut coke
Lime 25.0 h
2013-05-20 14:00 2013-05-22 03:30 Lime coated Zr-traced nut coke
long Lime long 37.5 h
Recipe changes were made when starting to use a new type of nut
coke, when the moisture content of coke changed, for heat level
control and in order to adjust the slag basicity for sufficient
alkali removal. The amount of coke in the ferrous layers was 150
kg/tHM
when nut coke was added, except for during a ramping-up period
in the beginning of the reference period with nut coke. As the
efficient C content varied between the nut coke types the amount
of coke in the coke layers was influenced correspondingly.
During the all periods the pulverized coal injection (PCI) was
kept fairly constant and changes in the amount of reducing agents
were
adjusted by changing the coke rate. Also other basic operational
parameters were successfully kept more or less constant during
the
tests as can be seen in Table IV.
-
Table IV Indicative and actual operational parameters during
trial periods for evaluation
Indicative Ref. Orig. Zr traced Magnetite Lime Lime long
Prod. Rate* 1.5 1.50 1.45 1.48 1.53 1.52 1.51 tonne/h
Tot. blast flow 1587 1584 1585 1580 1582 1582 Nm3/h
Blast air flow 1500 1500 1500 1500 1500 1500 1500 Nm3/h
O2 enrichment 110 111 110 110 110 110 110 Nm3/h
O2 in blast 50 50.6 50.3 50.4 50.1 50.2 50.2 Nm3/h
O2 to lances 60 60.5 59.3 59.3 59.7 59.4 59.3 Nm3/h
Tot. O2 in blast 26.4 26.6 26.6 26.6 26.8 26.7 26.7 Vol%
PCR 150 (225) 150 (226) 153 (221) 148 (219) 148 (226) 150 (228)
151 (228) kg/tHM (kg/h)
Moisture in blast 20 20 20 20 20 20 20 g/Nm3
Blast temp. 1200 1196 1196 1196 1196 1196 1196 C
Ave. blast temp. at tuyeres** 1120 1117 1114 1114 1114 1113 1113
C
Flame temp. ~ 2150 2187 2187 2190 2184 2180 2180 C
* from material balance, **average close to tuyeres
RESULTS AND DISCUSSION
C consumption in the BF during trials
The C consumption for each evaluation period was analyzed in
heat and mass balance model based on the principles of RIST
diagram6. The calculations are using data converted to units on
a per tonne hot metal (tHM) basis. In principle quite long
stable
operational periods are required for getting reliable data for
the heat and mass balance. On short-term e.g. the coke reserve can
be
either built up or consumed and this will create errors if not
taken into account.
Figure 7 Heat losses during each evaluation period Figure 8
Charged and by the process consumed C for each
evaluation period
Initially the collected Experimental BF data for each period was
inserted in the model assuming that data as e.g. weights and
analyses
for charged and tapped material and gas analyses are correct.
The model iteratively calculates the blast volume and total heat
losses.
The achieved values for blast volume are compared with the
measured ones. Actual heat losses for the Experimental BF are based
on
measured heat losses for water cooled part, estimated heat
losses based on temperature measurements on the steel shell and
calculated
heat losses related to the top gas. A large deviation between
totally estimated heat losses based on data and the one calculated
in the
model indicate changed amount of coke in the coke reserve. As
can be seen in Figure 7, the un-known heat losses varies and
e.g
during the period with Zr-traced nut coke the heat losses
estimated in the model based on the charged C is much lower than
those
stated based on the consumed C. During the first period with
original nut coke the starting point was a higher fuel rate in
order to
avoid disturbances caused by changed temperature and gas
profiles. The coke reserve is accumulated and during the following
period
it is then possible to reduce the fuel substantially without
coming into low heat level.
-200
0
200
400
600
800
1000
1200
1400
Ref. Orig Zr Magnetite Lime Lime long
He
at lo
ss M
j/tH
M
Calc. tot heat losses charg. Calc. tot heat losses consump.Not
known heat losses charg. Not known heat losses consump.
466
481
457459
453 453
463
474
471
458459 459
1410
1420
1430
1440
1450
1460
1470
450
455
460
465
470
475
480
485
Ref. Orig Zr Magnetite Lime Lime longTe
mp
era
ture
, C
C, k
g/t
HM
Charged C C consumed in process HM Temp
-
To overcome the impact on heat and mass balance from
accumulation or consumption of raw materials the amount of consumed
C
(coke and coal) is calculated based on the gas analyses and
under the use of measured and calculated heat losses. In Figure 8
the
amount of charged and by the process consumed C, respectively,
are stated. As can be seen, coke accumulation occurs during the
first
period with nut coke when using original coke in the ferrous
layers. During the next period when using Zr-traced nut coke
the
consumption of some accumulated coke reserve is coke occurs.
The heat level is significantly higher during all test periods
compared to during the reference period as shown in Figure 9. This
results
in an increased consumption of C for periods when original or
Zr-traced nut coke is used. For periods with activated nut coke the
C
consumption is lower although higher heat level compared to
during the reference. As can be seen the average hot metal
temperature
during the reference is ~1429 C compared to 1446-1465 C during
trial periods. Additionally, the Si content in hot metal differs
as
well, se Figure 9. To be able to compare the C consumption for
the different evaluation periods, model calculations for
normalized
conditions were conducted. For normalized heat level a hot metal
quality with 1.35 wt.-% Si and hot metal temperature of 1430 C
were selected, quite similar as for the reference period. In
Figure 10 the difference in estimated C consumption for trial
periods
relative reference period is shown for normalized conditions. It
can be concluded that the C consumption is reduced with ~ 4-6
kg/tHM (Figure 10) if differences in heat level are not taken
into account. For normalized data the reduction in C consumption is
~6-8
kg/tHM (Figure 11).
Figure 9 Average hot metal quality for each evaluation period
Figure 10 Differences in by the process consumed C for each
relative the reference period for each trial period
During periods with higher heat level and higher Si content in
hot metal the slag basicity is increased and the slag volume
reduced.
This results in reduced alkali recovery to the slag. In Figure
12 the average K2O yield to the slag is shown. When the yield is
low,
accumulation, recirculation and scaffolding may occur.
Recirculation of alkali results in increased C consumption in the
lower part of
the furnace due to the reduction and gasification of alkali
compounds. In the upper part these are oxidized again. The thermal
and
chemical energy returned in the upper part is evened out in the
heat and mass balance but in the actual BF it will result in
increased
energy consumption. Moreover, during this period the high Si
content in hot metal requires an increasing amount to counteract
the
accumulation of alkalis.
Figure 11 C consumption for normalized hot metal quality Figure
12 K2O yield and slag basicity
1429
1458
1465
1446
1456
1448
1.35
1.69 1.611.43 1.38 1.35
1
1.5
2
2.5
3
3.5
4
4.5
1410
1420
1430
1440
1450
1460
1470
Ref- Orig. Zr traced Magn. Lime Lime long
Wt-
% C
an
d S
i
Tem
pe
ratu
re,
C
HMT (C) C (%) Si (%)
0.0
11.2
8.2
-5.8
-4.3 -4.1
1410
1420
1430
1440
1450
1460
1470
-6
-4
-2
0
2
4
6
8
10
12
Ref. Orig Zr Magnetite Lime Lime long
Tem
pe
ratu
re,
C
Dif
fere
nce
in C
co
msu
mp
tio
n, k
g/tH
M
Change in C consumption relative ref. HM Temp
0.0
4.9
2.1
-7.8
-6.4-5.7
-10
-8
-6
-4
-2
0
2
4
6
8
10
430
435
440
445
450
455
460
465
470
475
480
Ref. Orig Zr Magnetite Lime Lime long
Dif
f. in
C c
on
sum
p. r
ela
tive
ref
., k
g/tH
M
Esti
mat
ed
C c
on
sum
pti
on
, kg/
yHM
C consumed in process Normalised C cons. in process Diff.
Normalised cons.
0.80
0.85
0.90
0.95
1.00
1.05
1.10
0%
20%
40%
60%
80%
100%
120%
140%
160%
Ref Orig Zr traced Magn. Lime Lime long
Slag
bas
icit
y, B
2
K2O
yie
ld, %
K2O yield Slag basicity, B2
-
Impact on gas efficiency
The gas efficiency was calculated from top gas analyses and from
in furnace gas analyses measured during shaft probing. The gas
efficiency calculated from top gas analyzes increases with
approximately 1.5 % when activated nut coke is used as can be seen
in
Figure 13. Periods during which original and Zr-traced coke is
charged in the ferrous layers are operated under quite high heat
level
and high coke rate. In spite of this, the gas efficiency is in
the same range as for the reference period that has significantly
lower heat
level. This indicates that the addition of nut coke in general
is beneficial for the gas distribution. It could also be noted that
although
that the hot metal indicated high heat level the top gas
temperatures and amount of water dosage in the BF top could be kept
at desired
levels, Figure 14.
Figure 13 Average burden descent, top gas efficiency, vol.-%
of
H2 per evaluation period calculated from top gas analyses
Figure 14 Average top gas temperature and water dosage at
top per evaluation period
At the upper probe level the gas efficiency is higher during
periods with activated coke, see Figure 15. At the lower probe
level the gas
efficiency is as seen in Figure 16 highest for periods with
magnetite-activated nut coke. In between the tuyeres the gas
efficiency is
high also for lime-activated coke. Above the raceway the gas
efficiency is quite similar for all periods except for during the
magnetite-
activated ones.
Figure 15 Gas efficiency per evaluation period calculated
from
gas analyses during probing at upper probe level
Figure 16 Gas efficiency per evaluation period calculated
from
gas analyses during probing at lower probe level
Impact on thermal reserve zone temperature
The temperatures measured within the shaft are influenced both
by the chemical reactions and the heat transfer from ascending
gas.
When the general heat level in the furnace is increased the
temperature isotherms is moved upward in the BF. It is difficult to
compare
the temperature at one vertical position in the shaft for
periods with different heat level without knowing that this
position is within the
thermal reserve zone. As indicated by the vertical temperature
profiles Figure 19 the thermal reserve zone starts for several of
the
measurements approximately at the position of upper shaft probe
(0.90 m below stockline) and ends before or at the position of
the
lower probe level (3.46 m below stockline). Therefore the
temperatures at upper and lower probe, as shown in Figure 17 and
Figure
18, cannot in general be used for estimations on the thermal
balance controlling the thermal reserve zone temperature. These
temperatures seem to be influenced by the thermal level of the
process. Additionally, the probes for temperature measurements
are
47.747.5
47.7
49.2 49.2
49.0
46.5
47.0
47.5
48.0
48.5
49.0
49.5
0
1
2
3
4
5
Ref. Orig. Zr traced Magnetite Lime Lime longG
asef
fici
en
cy, %
Bu
rde
n d
esc
en
t, c
m/m
in, v
ol%
H2
Burden descent vol%H2 Gas efficiency, %
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
160
165
170
175
180
185
190
Ref. Orig. Zr traced Magnetite Lime Lime long
Wat
er
do
sage
, lit
res/
min
Top
gas
te
mp
era
ture
, C
Top T Water dosage
0
10
20
30
40
50
60
1 2 3 4 5 6 7 8 9
Gas
eff
icie
ncy
, %
Position
Ref. Orig. Zr traced Magn. 1 Magn. 2 Lime and lime long
0
5
10
15
20
25
30
1 2 3 4 5 6 7 8 9
Gas
eff
icie
ncy
, %
Position
Ref. Orig. Zr traced Magn. 1 Magn. 2 Lime and lime long
-
water cooled. However, it is indicated that the use of 150
kg/tHM of nut coke results in a flatter temperature profile with
less extensive
central gas flow. In the vertical temperature measurement a
lowered thermal reserve zone temperature is measured for the period
with
lime-activated coke as seen in Figure 19. The period with
magnetite-activated nut coke shows similar temperature profile as
for non-
activated nut coke.
Figure 17 Horizontal temperature profile measured during
stated
evaluation period at upper probe level
Figure 18 Horizontal temperature profile measured during
stated evaluation periods at lower probe level
Based on the horizontal and vertical measurements conducted
during the evaluation periods but also in connection to them a
measured
temperature was deduced and used in the heat and mass balance
calculations. If the measured temperature is used in the heat and
mass
balance calculations the shaft efficiency reached more than 100%
for Zr-traced and magnetite-activated periods. Assuming a
similar
shaft efficiency of 99% for all periods, a lower temperature is
estimated for all evaluation periods when using activated nut coke,
Figure 20.
Figure 19 Vertical temperature profiles measured within
evaluation periods
Figure 20 Measured and calculated thermal reserve zone
temperature
450
500
550
600
650
700
750
800
1 2 3 4 5 6 7 8 9
Tem
pe
ratu
re
c
Position
Ref. Orig. Zr traced Magn. 1 Magn. 2 Lime and lime long
450
550
650
750
850
950
1050
1 2 3 4 5 6 7 8 9
Tem
pe
ratu
re
c
Position
Orig. Zr traced Magn. 1 Magn. 2 Lime and lime long
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
0 200 400 600 800 1000 1200
Dis
tan
ce f
rom
sto
cklin
e, m
m
Temperature C
Ref. Zr tracedMagn. 1 Magn. 2Lime and lime long
1000 1000
1060 1060
960 960
1023
9941002
965 961
974
900
920
940
960
980
1000
1020
1040
1060
1080
Ref. Orig Zr Magnetite Lime Lime long
Tem
pe
ratu
re,
C
Approximate measured Calculated for 99% shaft efficiency
-
CONCLUDING REMARKS
During the Experimental BF trials stable operation could be
achieved when using 150 kg/tHM of nut coke in the ferrous layers.
The
gas efficiency was in general high for all periods and increased
from the reference the to the activated nut coke periods.
The charging of high amounts of nut coke is believed to improve
the horizontal gas distribution.
High gas efficiency could be achieved although high coke rate
during periods with original and Zr-traced nut coke
The horizontal gas profile at upper shaft probe level became
flatter with less significant central gas flow
The top gas temperature and water dosage was not significantly
increased when high heat levels were reached
Further increase in gas efficiency during periods with activated
nut coke is likely caused by lower thermal reserve zone
temperature
that improves the indirect reduction of ferrous material and
decreases the direct reduction. Lower thermal reserve zone
temperatures
were indicated by measured data and heat and mass balance
calculation results. Lowering of the thermal reserve zone
temperature
moves the equilibrium for Femet/FeO to CO/CO2 towards Femet
during the periods with activated nut coke.
The C consumption calculated for all periods using normalized
hot metal heat level in terms of Si content and tap temperature
states a
reduction of C consumption with 6-8 kg/tHM when activated nut
coke is charged. The highest savings are achieved during the
period
with magnetite-activated nut coke.
For industrial implementation the method for activation has to
be further developed. In the method used the coke reaches a
moisture
content corresponding to saturation and cannot be charged
without previous drying.
ACKNOWLEDGEMENTS
The research work presented in this paper has been carried out
within the project of Innocarb, RFSR-CT-2010-00001, that is co-
funded by the Research Fund for Coal and Steel (RFCS). Swedish
Energy Agency is greatly acknowledged for additional financial
contribution. The paper is a contribution from CAMM, Centre of
Advanced Mining and Metallurgy, at Lule University of
Technology that supported the research scientifically and
economically.
REFERENCES
1. A. K. Biswas, Principles of Blast Furnace ironmaking, Theory
and Practice, Cootha Publishing House, Brisbane, Australia(1981),
pp 122-132, p 74
2. S. Nomura, H. Ayukawa, H. Kitaguchi, T. Tahara, S. Matsumaki,
M. Naito, S. Koizumi, Y. Ogata, T. Nakayama, T. Abe, Improvement in
Blast Furnace Reaction Efficiency through the Use of Highly
Reactive Calcium Rich Coke, ISIJ Int. Vol 45
(2005), No. 3 pp 316-324
3. S. Nomura, H. Kitaguchi, K. Yamaguchi, M. Naito, The
characteristics of Catalyst-coated Highly Reactive Coke, ISIJ Int.
Vol. 47 (2007), No. 2 pp 245-253
4. S Nomura, H Terashima, E Sato, M Naito, Some Fundamental
Aspects of Highly Reactive Iron Coke Production, ISIJ Int. Vol. 47
(2007), No. 6 pp 823-830
5. S. Watakabe, K. Takeda, H. Nishimura, S. Goto, N. Nishimura,
T. Uchida, M. Kiguchi, Development of High Coke Mixed Charging
Techniqiue to the Blast Furnace, ISIJ Int. Vol. 46 (2006), No. 4 pp
513-522
6. Patrick Lawrence Hooey, Axel Bodn, Chuan Wamg, Carl-Erik Grip
and Bjrn Jansson, Design and Application of a Spreadsheet-based
Model of the Blast Furnace Factory, ISIJ International, Vol. 50
(2010), No. 7, pp. 924930