NIPPON STEEL TECHNICAL REPORT No.

123 March 2020

UDC 669 . 162

Technical Report

Progression of Experimental Blast Furnace

in Hasaki R&D Center
Takuya NATSUI* Kohei SUNAHARA
Shinichi SUYAMA Kaoru NAKANO
Yoshinori MATSUKURA Yutaka UJISAWA
Takanobu INADA

Abstract
The experimental blast furnace in Hasaki R&D Center (Kamisu City in Ibaraki Prefec-
ture), constructed as a melting furnace in 1982, carried out a total of 50 test operations over
27 years up to 2008. This article reviews the history of the technological development and
knowledge obtained from the experimental blast furnace.

1. Introduction versity one-ton/day furnace”) as “the experimental furnace with

Employing an experimental blast furnace is one of the R&D which qualitative information pertaining to the state of the progress
methods used to investigate ironmaking technology that is utilized of various processes in a blast-furnace-type reactor is obtained, and
at the early stage of the development of a new process technology, appropriate problem presentation is provided thereby to the funda-
or at the stage where the solutions to problems arising from various mental R&D on an experimental basis and/or to the survey on
conditions are required. However, such furnaces always suffer from acommercial blast furnace basis”.
the inevitable problems of their dissimilarity to a commercial blast In Nippon Steel Corporation, after the prewar R&D based on the
furnace, or in other terms, discrepancies such as in the thermal level, 1.2 m3 experimental blast furnace in the Kamaishi Works (ex-Mitsui
reducing agent rate, in-furnace flow time, and in-furnace load. Such Mining Co., Ltd.), 7) a one-ton furnace (0.596 m3) was constructed in
dissimilarity or discrepancies are inevitably caused by the relatively Higashida in Yawata Works in 1934, and a three-ton furnace (4.6
large thermal loss due to the small scale of the experimental blast m3) was constructed in Tobata in 1944, both belonging then to cor-
furnace. Furthermore, relatively large-scale R&D resources are re- porate technology research and development laboratories. In these
quired, and therefore, not only the test cost, but also other issues in- furnaces, a total of thirty-eight test operations over fifteen years
cluding securing sufficient operator resources are common problems were conducted. From the test operations, various information was
across countries and periods of time. obtained about the usability evaluation results of raw materials such
Figure 1 shows the transitions of the iron and crude steel pro- as ore powder, reduced iron sand, and anthracite coal, and informa-
duction and experimental blast furnaces. 1–4) Systematic tests on an tion pertaining to flux injection, low Si operation, mixed charging,
experimental blast furnace were conducted after 1916 in the US un- and ferroalloy production. 8) Such information was transferred to the
der the control of the US Bureau of Mines and later by a research abovementioned Tokyo University one-ton/day furnace upon its
association that consisted of US and Canadian steel companies, and construction. In addition, domestically, in the experimental blast
in Europe, by the alliance of Belgium and France after 1957. 5) How- furnace of the ex-Nippon Kokan K.K. (furnace volume 0.63 m3, later
ever, both in the US and Europe, the experimental projects became expanded to 3.2 m3) built in 1967, various R&D studies on the de-
inactive in the latter half of the 1960s. Under such circumstances, velopment of various new processes were conducted, and it was
the research based on an experimental blast furnace conducted by confirmed that the reduction of the reducing agent rate by 30 kg/t
the Institute of Industrial Science (IIS), University of Tokyo, from (5%) per 100 Nm3/ton of the injected reductant gas, and lowering of
1955 to 1981 produced a lot of information on ironmaking technol- the direct reduction ratio to about 10% are possible. 9–11)
ogy in the early stage of the postwar production-growth period. The experimental blast furnace (12 t/d, last stage furnace volume
Tate, 6) being aware of its problems, defined the abovementioned ex- 4.0 m3) in the Hasaki R&D Center of the ex-Sumitomo Metal Indus-
perimental blast furnace (hereinafter referred to as the “Tokyo Uni- tries, Ltd. (Kamisu City in Ibaraki Prefecture) was built as a melting

* Senior Researcher, Ironmaking Research Lab., Process Research Laboratories

20-1 Shintomi, Futtsu City, Chiba Pref. 293-8511

- 126 -
 NIPPON STEEL TECHNICAL REPORT No. 123 March 2020

Fig. 1 Transitions of iron and crude steel production and experimental blast furnaces

furnace with a furnace volume of 1.3 m3 in 1982 at the same time as furnace). After 1996, the furnace was equipped with measuring de-
the termination of the operation of Tokyo University’s one-ton/day vices such as a sampling device and the like, and the furnace height
furnace. At this time when the domestic crude steel production ex- was extended and an additional material hopper was installed on the
ceeded 100 million tons a year, overtaking the US, the Japanese furnace top. Thus, as a blast furnace simulator, the test operations
steel industry was suffering from the sharp rise of the crude oil continued for the evaluation of raw material quality, etc., and in
price, and was forced to change its energy source. Then, the follow- 2008, after the completion of fourteen test operations as an experi-
ing R&D was conducted: development of coke rate reduction tech- mental blast furnace, it terminated its service after a total of fifty test
nology, the shift from heavy oil injection to pulverized coal injec- operations during the past twenty-seven years. Currently, there are
tion, and development of new processes as substitutes for the blast only two experimental blast furnaces in operation: one from the
furnace method. In the early period, the furnace was mainly used for COURSE50 Project in Japan, 4) and the other in LKAB (Luos-
studies on the maximum amount of pulverized coal injection, and savaara-Kiirunavaara Aktiebolag), a mining company in Sweden.
the feasibility of producing ferroalloys with the blast furnace proc­ In this article, the history of the experimental blast furnace of the
ess to cope with the electric furnace process. By hot-connecting the Nippon Steel Hasaki R&D Center (hereinafter referred to as Hasaki)
melting furnace and the shaft-type reducing furnace (8 t/d, furnace from 1982 to 2008 (Table 1), and the information obtained there-
volume 1.3 m3) constructed adjacently in 1984, and by separating from are reviewed.
the function of iron ore reduction by a gas reductant and the func-
tion of melting the reduced iron, R&D was promoted for a new 2. Test Operations of the Experimental Furnace as
ironmaking process, the SC Process (Sumitomo/Shaft-Cupola), tar- Melting Furnace
geting the alleviation of raw material quality specifications, drastic 2.1 Pulverized coal injection and combined injection, and fer-
energy saving, and cost reduction. 12) Furthermore, by using the roalloy production development period
melting furnace independently, test operations were conducted to After the oil crisis in the 1970s, as a substitute for the heavy oil
develop new ironmaking process technologies such as oxygen blast injection that was started in the 1960s, the study on pulverized coal
furnace and scrap melting, and a total of thirty-six test operations injection was started in the early 1980s. An actual size model of the
using the furnace for melting were conducted until 1988. lower part of the Kokura No. 1 Blast Furnace with a 48 degree-
After 1989, with the construction of a ceramic heat-exchange- fanned section (after the third repair and improvement, furnace vol-
type hot stove and with the alteration of lance-type blasting to ume 750 m3) was constructed in the Amagasaki R&D Center in
tuyere-type blasting, the furnace was used as an experimental blast 1972 (transferred to Hasaki in 1978, furnace volume expanded to 44
furnace (Sumitomo/Small Test Blast Furnace: STBF) for the devel- m3). Based on the information obtained from the actual size tuyere
opment of the combined blasting of high-rate pulverized coal and combustion experiment conducted on the model, 13) study on the
powder ore (hereinafter referred to as ultra-combined blasting for maximum amount of pulverized coal injection using the melting

- 127 -
NIPPON STEEL TECHNICAL REPORT No. 123 March 2020
Table 1 Chronology of the experimental blast furnace

Campaign
Schedule Operation subject Equipment transition
No.
SC Evaluation of coke properties and maximum amount of
1982/3-11 1982 Melting Furnace (MF) established (10 t/d, 2.2 m3, 3 mH).
1st-5th PCI
6-12th 1982/12-1983/10 Ferroalloy production
1984 Shaft Furnace (SF) established (8 t/d, 1.3 m3, 3 mH).
13-15th 1984/4-1984/7 Development of SC method Reduction ore hot conveyor, Hot cyclone, and Sampling
sonde were installed.
16-17th 1984/9-1984/11 Ferroalloy production
18-22th 1985/2-1985/10 Development of SC method
Development of oxygen blasting and ultra combined
23-32th 1985/12-1987/10 1987 Hot stove established.
blasting for furnace
1988 Furnace height extension (SL: TY+3.0 mH→3.5 mH)
33-36th 1988/3-1988/12 Development packed bed type scrap melting process
2nd tuyere installation in shaft (TY+0.6 m, 1.2 m)
STBF 1989 Tuyere/Browpipe and hot blast control valve system
1989/5-1991/4 Development of ultra combined blasting for furnace
1st-7th 1990 Furnace height extension (SL: TY+3.5 mH→5.0 mH, 3 m3)
1995 Installation of measurement systems
8th 1996/3/11-3/17 Large amount of PC injection and low slag rate tests (dripping and cohesive zone samplers, liquid level detector,
stock level detector)
HBI charging, reduced iron and ore powder injection
9th 1997/4/14-4/18
tests
Evaluation of effect slag rate and low slag sinter
10th 1997/10/27-10/31
properties on permeability
11th 2000/3/25-3/29 Evaluation of effect of high Al2O3 slag on the operation. 1999 Ground flare stack was installed.
12th 2001/1/29-2/2 Wasted plastic powder injection test
2003 Furnace height extension (SL: TY+5.0 mH→6.0 mH, 4.0 m3)
13th 2003/11/7-11/13 Evaluation of sinter reducibility and coke reactivity
Vertical prove was installed.
14th 2008/11/16-11/21 Evaluation of effect of mixed charge on permeability 2007 Hopper for mixed charging was installed.

priced ferroalloys, development of the production of ferroalloys in

the blast furnace method was studied. The in-furnace high tempera-
ture refining region expanded by the high-amount pulverized coal
injection was targeted, and was considered as appropriate for the
production of ferromanganese and ferrochromium. Until 1984, nine
test operations were conducted, and high carbon manganese of [Mn]
= 75%, and high ferrochromium of [Cr] = 60% were produced. 15, 16)
2.2 New ironmaking technology (SC Process) development period
The SC Process that was developed independently in the 1980s
by the ex-Sumitomo Metal 12) is a new ironmaking process to cope
Fig. 2 Results of the coal combustion test with the future scarcities of high-quality coal and high-quality iron
ore, wherein the function of a blast furnace is divided into the func-
furnace was conducted. tions of a melting furnace and a shaft reducing furnace to improve
As shown in Fig. 2, by oxygen-enriched ambient temperature air productivity (Fig. 3). Figures 4 and 5 show the appearance of the
blasting, even under the condition of the high pulverized coal injec- melting furnace at the time of its construction in 1982, and its sche-
tion rate of pulverized coal/oxygen (PC/O2) = 1.4 kg/Nm3, a combus- matic diagram, respectively. The reduced iron reduced to the degree
tibility of coal as high as 87.5% could be maintained. Furthermore, of 80–90% in the shaft furnace is carried to the melting furnace by
study on the combined injection of pulverized coal with ore powder the hot conveyer, and melted by coke. In the melting furnace, it was
and flux materials was conducted. In the actual trial operation of the confirmed that, since the coke solution-loss reaction does not occur,
Wakayama No. 4 Blast Furnace (after the third repair and improve- the coke strength is sufficiently maintained even at the bottom of the
ment, furnace volume 2 700 m3), reduction of [Si] of the molten pig furnace as shown in Fig. 6. 17)
iron (hereinafter referred to as hot metal) by 0.1% was confirmed 2.3 Oxygen blast furnace, scrap melting development period
per injection of ore powder of 30 kg/t. 14) During the period from 1985 to 1987, test operations were con-
In addition, to solve the problem of the electricity cost of ferroal- ducted to confirm whether the oxygen blasting accompanying the
loys that used to be produced in the conventional electrical method, injection of a large amount of pulverized coal is practically feasible
and to secure the competitiveness against the then imported low- as an ironmaking process. 18) As a result of the test operation, under

- 128 -
 NIPPON STEEL TECHNICAL REPORT No. 123 March 2020

the conditions of blasting the mixture of the ambient temperature

oxygen and nitrogen with a mixing ratio of O2 = 60% and of PC/
O2 = 1.2 kg/Nm3, a pulverized coal rate of 407 kg/t, coke rate of 258
kg/t and productivity of 7.35 t/d/m3 were achieved. Figure 7 shows
the result of the operation (plotted) and the result of the calculation
by using one-dimensional blast furnace models (full line, broken
line). Since the calculation result of this model agrees with the pro-
duction operation result, through the simulation under the condition
of PC/O2 = 1.2 kg/Nm3 assumed for a large commercial blast furnace
(3 680 m3), it was estimated that the operation result of a productivi-
Fig. 3 Concept of the SC Process
ty of 3.30 t/d/m3, pulverized coal rate of 375 kg/t, coke rate of 180
kg/t, and reducing agent rate of 555 kg/t is thermally achievable.
In 1988, under the condition of the blasting of oxygen-enriched
ambient temperature air under the normal furnace top pressure, and
by using coke and pulverized coal, a test operation was conducted
for melting 100% steel scrap. As opposed to the ordinary cupola
process that uses the low-reactivity, high-quality, large-size lump
coke exclusively, since the coke generally used for the blast furnace
is used in this operation, in the furnace, the coke for blast furnace
use is combusted evenly. Consequently, the coke becomes highly
reducible as compared with the weak reducibility in the case of the
cupola, and desulphurization and carburization are promoted there-
by. As a result, as shown in Table 2, the reducing agent rate became
275–290 kg/t, and further improved to 240 kg/t due to the addition
of air through the second tuyere installed in the shaft section. 19) The
productivity remained at 15 t/d/m3 due to the equipment restriction.
However, it was estimated that, on the condition of a bosh gas ve-
Fig. 4 Appearance of the melting furnace (1982) locity of 0.5 Nm/s, a productivity of 30 t/d/m3 is possible, and fur-
thermore, that the reducing agent rate can be reduced to 150 kg/t
under the condition of a gasification degree of 50%. 20) Further, in
the extended type of the packed-bed-type scrap melting process us-
ing a converter 21) where iron ore was added, heat efficiency as high
as that of the all-scrap melting process was confirmed. 22)

Fig. 5 Schematic diagram of the SC pilot plant

Fig. 6 Distributions of coke properties and gasification degree in melt- Fig. 7 Comparison of operation results of oxygen injection and calcula-
ing furnace tion results

- 129 -
NIPPON STEEL TECHNICAL REPORT No. 123 March 2020
Table 2 Results on all scrap operations of the experiment blast furnace

Indices Case No. 1 2 3

Bosh gas volume (Nm3/h) 800 714 614
Flame temperature (°C) 2 500 2 700 2 700
Productivity (t/d∙m3) 11.7 14.7 14.7
Coke rate (kg/t) 275 150 125
Coal rate (kg/t) 0 140 115
Fuel rate (kg/t) 275 290 240
Oxygen rate (Nm3/t) 104 147 122
Additional air (Nm3/t) 0 0 74
By product gas (Mcal/t) 981 1255 881
Hot metal temperature (°C) 1 508 1 486 1 480
C (%) 4.70 4.63 4.35
Si (%) 0.29 0.35 0.21
S (%) 0.032 0.041 0.036
Fig. 8 Appearance of the experimental blast furnace (2008)
Slag CaO/SiO2 (-) 1.33 1.15 1.13
MgO (-) 15.8 22.2 22.1 (capacity 800 kg) and two hoppers for the iron source materials of
(%S) / [%S] (-) 58.3 72.8 35.1 ore family such as sintered ore, pellet and lump ore (capacity of 700
kg each) are installed. By a cut gate valve, the volume of a charge of
a coke bed is divided into five or six units, and each unit is charged
3. Test Operation as Experimental Blast Furnace intermittently (inching charging). The ore family iron source materi-
3.1 Outline of equipment al in the hopper is charged uniformly in the radial direction by a ta-
With the alteration of the lance-type blasting system to the ble feeder. The gas cleaning system consists of a dust catcher and a
tuyere/blowpipe-type system and the installation of a hot stove con- venturi scrubber, and the furnace top gas is combusted and dis-
ducted in 1987–1989, the experimental furnace was modified from charged by the flare stack after the water sealing equipment.
an SC-type melting furnace to a small experimental blast furnace The furnace is equipped with one sampling device for the cohe-
(STBF) that resembled a commercial blast furnace. After 1989, a to- sive zone and three sampling devices for the dripping zone (upper,
tal of fourteen blast-furnace-type test operations were conducted, in- middle, lower). These sampling devices analyze the in-furnace gas
corporating the evaluations obtained from the actual blast furnace compositions and measure temperatures during operation, and addi-
operation. During this period, the furnace height was extended three tionally, are able to take samples of the in-furnace material during
times, the stock level of the burden was heightened from 3.0 m to operation by exchanging the probe of the device. Figure 11 (a)
6.0 m, and the inner volume was expanded from 1.3 m3 to 4.0 m3. shows an example of sampling the in-furnace material with the sam-
After 1995, measurement systems and charging hoppers were addi- pling device. With the front end tip of the cohesive zone sampler,
tionally installed. The appearance of the experimental blast furnace and with the dripping zone samplers, samples of the cohesive mate-
in 2008 is shown in Fig. 8, and the equipment flow and the sche- rial could be frequently taken. Occasionally, the insertion of the
matic diagram of the experimental blast furnace are shown in Figs. probe into the bottom part of the furnace had to be abandoned due
9 and 10, respectively. to the high resistance force to thrusting.
The hot stove is of the direct heat exchanging type, consisting of The liquid level detector dips the probe directly into the liquid of
a two-stage heat exchanger of a metallic type and ceramic type (em- the molten slag and the hot metal produced and stays at the hearth
ployment of SiC heat transmission tube), and the blast temperature bottom during the operation, and enables the detection of the surface
at the exit of the hot stove is about 1 050°C. However, the blast tem- levels of the molten slag and the molten slag/hot metal boundary
perature into the furnace is about 800°C due to the temperature drop (Fig. 11 (b)). The stock level detector detects the level of the burden
before the blast reaches the tuyere tip. The hot stove system has a surface and measures its descending rate. The rigid vertical-type
maximum air blast rate of 900 Nm3/h with two 500 Nm3/h air com- probe follows the descent of the burden, and the vertical distribu-
pressors, and supply capacities of 400 Nm3/h of oxygen, 600 Nm3/h tions of the in-furnace temperature and the gas compositions are ob-
of nitrogen, and 30 Nm3/h of LPG. The furnace has three tuyeres (35 tained.
mm in diameter) and the blast volume at each tuyere is controlled Figure 12 shows the results of the measurement by the liquid
by the hot blast air control valve installed at each branch pipe based level detector and the stock level detector. The in-furnace vertical
on the blast volume measured by the flowmeter equipped to each stresses of the experimental blast furnace are about 5–10 kPa or
branch pipe. There is one tap hole, and a tap hole opening machine about 7 kPa on average, and estimated to be about 7% of a commer-
and a mud gun. The hot metal and the molten slag are tapped to a cial blast furnace (about 100 kPa) even in the neighborhood of the
rectangular ladle with a capacity of about one ton for about five cohesive zone (Fig. 12 (a)). From the mechanical balance between
minutes with a tap to tap time of about two hours. The injection sys- the vertical stresses and the buoyancy from the hot metal and the
tems are as follows: pulverized coal (300 kg/h), ore powder (150 kg/ molten slag phase, the floating state of the deadman coke in the ex-
h) and the flux materials powder (30 kg/h). The system is designed perimental blast furnace can be estimated. Figure 12 (b) shows the
so that each powder is injected into the furnace via a distributing transition of the burden stock level during tapping, and the liquid
equipment using a carrier gas of nitrogen through the injection lance levels identified by the amount of drainage of the hot metal and the
installed below the tuyere. Atop the furnace, one hopper for coke liquid level detector. It is considered that the packed coke bed sinks

- 130 -
 NIPPON STEEL TECHNICAL REPORT No. 123 March 2020

Fig. 9 Experimental blast furnace and peripheral facilities

Fig. 10 Schematic diagram of the experimental blast furnace

- 131 -
NIPPON STEEL TECHNICAL REPORT No. 123 March 2020

Fig. 11 State of collected samples by samplers and liquid level detector

Fig. 12 Vertical stress, variation of stock level with drainage and state of hearth (eleventh test operation)

in the slag phase and floats on the hot metal phase, and it was con- ing lower basicity in the dripping zone has good wettability with
firmed that the state of the deadman coke agrees with the assump- coke.
tion of the abovementioned mechanical balance. Figure 14 shows an example of the results of the in-furnace
After the test operation, the furnace was cooled by nitrogen for sample analysis by the dissection survey and the measurement dur-
two weeks, maintaining the designated stock level, and a dissection ing the operation. The carburization of the hot metal dripping from
survey was conducted. From the top of the furnace, a sample was the upper part of the dripping zone progresses (Fig. 14 (e)), and from
taken at each coke bed and ore bed at predetermined positions in the right after the hot metal/molten slag separation, the slag starts to as-
radial direction. Figure 13 shows an example of the in-furnace bed similate with ash and the basicity decreases. The slag FeO right be-
structure. It was confirmed that the slag formed at above 1 400°C in fore dripping is about 1.0%, which drops to the tapped molten slag
the middle of the cohesive zone has larger angles of contact with level in the upper part of the dripping zone. From the data, the slag
coke, and has not got wet as compared with the slag formed in the physical property can be estimated. In addition, since the analysis
upper cohesive zone right before separation of the molten slag from result of the in-furnace content obtained in the dissection survey and
the hot metal. However, it was confirmed that the molten slag hav- that of the sample taken by the sampling devices during operation

- 132 -
 NIPPON STEEL TECHNICAL REPORT No. 123 March 2020

agree with each other in general, the in-furnace state under any test group that controls the hot stove operation in the instrument pulpit,
conditions can be directly grasped by the sampling device and the the charging group that conducts manual operation in the instrument
dissection survey result. pulpit, mounting work of the material hoppers with a hoist, and the
The one test blast furnace operation continued for about five blending work of raw materials, the powder injection group that
days, and was conducted by about fifteen shift operators on a three- controls the powder material transportation quantity such as pulver-
shift basis. Each shift consisted of the following groups: the blasting ized coal powder and charging to powder material hoppers, the in-
strumentation group that handles various sampling devices, the tap-
ping group that operates the compressed-air-driven tap hole opening
machine, conducts tapping operation using the oxygen opening
method and tap-hole-closing operation by a hydraulic mud gun and
the handling of the molten slag/hot metal ladle, the utility group that
controls the working conditions of equipment and controls the sup-
ply of utilities, and an overall operation control supervisor and as-
sistant operation control supervisor.
3.2 Ultra-combined blasting development period
In the first to the seventh test operations during the period from
1989 to 1991, from the viewpoints of extending the coke oven fur-
nace life, direct use of ore powder, the blast furnace productivity en-
hancement, and the hot metal composition control, the ultra-com-
bined blasting technology in which large amounts of pulverized
coal, ore powder, and slag formers are simultaneously injected in
through the blast furnace tuyeres was developed. 23, 24) Its concept is
shown in Fig. 15.
In 1988, a new coke packed-bed-type combustion furnace (1.5
m long × 1.0 m deep × 2.35 m high with one 65 mm in diameter
tuyere) was built adjacent to the experimental blast furnace across
the hot stove. After fully grasping the combustion state within the
raceway and its periphery based on the detailed study using the fur-
nace, a continuous six-day operation was conducted in the experi-
mental blast furnace under the conditions of 300 kg/t + 100 kg/t and
200 kg/t + 200kg/t of the pulverized coal rate and ore powder rate.
The reducing agent rate was maintained at 600 kg/t, and no slag
FeO increase and no insufficiency in the reduction of the ore powder
were recognized. In the total of eight tests using only one tuyere
conducted during the same period at the Wakayama No. 3 Blast Fur-
nace (after the third repair and improvement, furnace volume 2 150
Fig. 13 Dissection of experimental blast furnace (thirteenth test opera-
tion) m3) and the Wakayama No. 5 Blast Furnace (after the third repair

Fig. 14 Vertical distribution of sample analysis results by dissection and sampling sonde (eleventh test operation)

- 133 -
NIPPON STEEL TECHNICAL REPORT No. 123 March 2020

rate. 25) In the softening and melting test conducted prior to the ex-
perimental blast furnace operation, 26) the high permeability of HBI
at high temperature was confirmed, and upon application of HBI to
the blast furnace, not only improvements in the productivity and the
reduction in the reducing agent rate, but also a great contribution to
the improvement in permeability were expected.
In the test blast furnace operation, to supply HBI, two methods
were used: charging lump HBI from the furnace top and the injec-
tion of the powder reduced iron (PRI) through the tuyeres. The test
operations were conducted for six cases: the four cases of the HBI
mixing ratio of 0, 25, 50 and 100%, one case of PRI (powder re-
duced iron) injection of 200 kg/t, and one case of powder ore injec-
tion (ore injection: OI) of 200 kg/t for comparison purposes. To
Fig. 15 Concept of ultra combined blasting for blast furnace
compare and verify the improvements in permeability and produc-
and improvement, furnace volume 2 700 m3), in the test under the tivity, all blasting conditions were fixed.
condition of 200 kg/t + 200 kg/t of the pulverized coal rate and ore Throughout the test operation, the furnace condition remained
powder rate, from the fiber scope observation, 24) melting of the ore stable. According to Fig. 16 wherein the representative operation
powder was observed and it was confirmed that the ore powder in- data are prepared, the 5.5% production rate improving effect and the
jection was functioning sufficiently within the raceway. In addition, 4.3% reducing agent reduction effect were obtained per HBI 100 kg/
although the effect of decreasing [Si] by FeO was not as remarkable t, and above HBI 50%, a significant decrease in the resistance to the
as that of a commercial blast furnace test, 14) hot metal chemical in-furnace permeability (permeability resistance) was confirmed. In
compositions almost equal to those of the commercial blast furnace the case of PRI, no changes were observed in the production rate
were obtained. Additionally, to prevent the piping abrasion, a plug- and the reducing agent rate. However, in the case of OI, as the pro-
type pneumatic conveyor powder transportation technology was es- duction rate deteriorated, accompanied by the worsening in the re-
tablished. ducing agent rate, it was concluded that, for the improvement of
After the extension of the furnace height in 1990, a stock level blast furnace productivity using HBI, top furnace charging is more
change test was conducted under identical operation specifications. advantageous than the injection through tuyeres.
As a result, by changing the stock level from 3.5 m to 5.0 m, de- In addition, the effect of using HBI in the experimental blast fur-
crease in unburnt char trapped in the furnace top was confirmed, and nace was analyzed from the viewpoint of kinetics by using the blast
the furnace top temperature dropped by 100–150°C, the gas utiliza- furnace mathematical models. 27) As shown in Fig. 17, the analysis
tion efficiency increased by 5–6%, and the reducing agent rate de- clarified that the productivity improvement, the reduction of the re-
creased by about 100 kg/t. Further, according to the dissection sur- ducing agent rate, and the decrease in gas utilization efficiency due
vey, in the case of a stock level of 3.5 m, the cohesive layer was to the increase in the HBI charge ratio are thoroughly elucidated.
found in the upper part of the furnace. Therefore, a thermal reserve And additionally, based on the results of this test operation and the
zone was not formed. kinetics analysis, the effect of the employment of HBI in the com-
3.3 Blast furnace raw material evaluation technology develop- mercial blast furnace was estimated. Later on, demonstration tests
ment period up to HBI 100 kg/t were conducted in the Wakayama No. 5 Blast
3.3.1 High pulverized coal rate, low slag rate test Furnace (after the third repair and improvement, furnace volume
In the eighth test operation conducted in March 1996, under the 2 700 m3), and the effects of the productivity improvement and the
ordinary blast furnace raw material condition, namely a sintered ore reducing agent rate reduction were confirmed as predicted by the
ratio of 75% and a lump ore ratio of 25%, the change in the raw ma- theoretical analysis. 28)
terial properties in the furnace due to the difference between the 3.3.3 Evaluation of low slag sintered ore
high pulverized coal rate operation (PCR > 200 kg/t, equal to or The tenth test operation was conducted in October 1997, six
higher than the one in the commercial blast furnace operation) and months after the ninth test operation. The test operation was con-
the entire coke operation was studied. However, as stated earlier, as ducted for quantitative analysis of the effects of the high tempera-
the blast air temperature of the experimental blast furnace is lower ture character, reducibility, and the slag rate of the low SiO2 sintered
than that of the commercial blast furnace, to simulate the tuyere ore on the blast furnace permeability. 29) Table 3 shows the test con-
front combustion condition of the commercial blast furnace, the de- ditions. The test operations were conducted for five cases of four
gree of oxygen enrichment was increased to fit the oxygen excess different types of sintered ore and one case in which the material
coefficient. As a result of the test operation, in the high pulverized property and the blast furnace slag rate were changed by adjusting
coal rate operation, the soot generated at the furnace top increased, the amount of the flux materials. The blasting condition, charged O/
and an increase in the strength of the coke at the furnace bottom C, and the target slag compositions are fixed.
considered to be the effect of the unburnt char generated in the fur- As a result of the test, in Fig. 18, the effects of the blast furnace
nace was observed. In addition, it was confirmed that these phenom- slag rate and the high temperature permeability resistance index of
ena become more remarkable in the low slag rate operation where the sintered ore (KS) 26) measured by the softening and melting test
the sintered ore ratio is lowered to 25%. on the blast furnace permeability resistance index (KR) 30) are
3.3.2 Evaluation of reduced iron shown. Focusing on Case 1 in Table 3 as the basis of the compari-
The ninth test operation was conducted in April 1997 to quanti- son, in Case 2, with the low SiO2 of the sintered ore, KS decreases
tatively grasp the effects of hot briquette iron (HBI) on the blast fur- and the blast furnace slag rate decreases along with the decrease of
nace productivity improvement and reduction of the reducing agent SiO2. In Case 3, the decrease in the blast furnace slag rate is covered

- 134 -
 NIPPON STEEL TECHNICAL REPORT No. 123 March 2020

Fig. 16 Relative production rate, reducing agent rate and permeability through the reduced iron melting tests using experimental blast furnace

Table 3 Experimental conditions

Case No. 1 2 3 4 5
Sinter Sample A B B C D
SiO2 (%) 5.01 3.89 3.89 4.63 4.60
CaO (%) 9.99 9.09 9.09 9.44 9.86
MgO (%) 1.09 1.11 1.11 1.19 0.94
Al2O3 (%) 2.09 1.87 1.87 1.95 1.84
FeO (%) 7.47 5.88 5.88 6.93 5.86
RI (%) 64.6 68.0 68.0 68.1 66.9
KS × 105 1 500 648 648 1 085 1 195
Fig. 17 Calculated results of experimental blast furnace operation RDI (%) 41.7 42.8 42.8 45.3 44.2
TI (%) 76.9 81.3 81.3 75.1 67.2
Fluxes (kg/t-HM) 33.5 13.3 74.1 5.5 19.3
by the blast furnace flux materials. The figure shows that the effect Slag rate (kg/t-HM) 302 258 302 271 271
of KS on the blast furnace permeability is significant, and the effect Ore/Coke (-) 3.64 3.58 3.60 3.59 3.61
of the blast furnace slag rate is considerably small as compared with
that of KS. Further, under these test conditions, there were no effects
of RI and RDI of the sintered ore on the permeability. er, in the case of the Kashima No. 2 Blast Furnace, the blast furnace
Based on the result of this test operation, to grasp the effect of slag rate changed with the change of KS. However, despite that, in
KS on the permeability in the commercial blast furnace, the soften- both blast furnaces, since the behavior of the change of the values
ing and melting test result was formulated and was incorporated into calculated by the model into which the KS evaluation model was in-
the blast furnace mathematical models, and the commercial blast corporated without taking into consideration the influence of the
furnace permeability based on this model 27) was evaluated. The cal- blast furnace slag agrees with the behavior of the change of the ac-
culation results agree very well with respect to not only the static tual values, it was concluded that, even in the commercial blast fur-
pressure distribution on the experimental furnace wall when high/ naces, similarly to the result of the test blast furnace operation, the
low SiO2 sintered ore is used, but also, as Fig. 19 shows, very well effect of the blast furnace slag rate is small compared with that of
with respect to the actual KR values obtained from the Kokura No. 2 KS.
Blast Furnace (after the second repair and improvement, furnace 3.3.4 Evaluation of high Al2O3 slag
volume 1 850 m3) and the Kashima No. 2 Blast Furnace (after the The eleventh test operation was conducted in March 2000. The
third repair and improvement, furnace volume 4 800 m3). In the case objectives of the test operation were to lower SiO2 of the sintered
of the Kokura No. 2 Blast Furnace, during the period of the subject ore aimed at blast furnace permeability improvement, and to clarify
operations, the blast furnace slag rate was almost constant. Howev- the in-furnace phenomena focusing on the permeability in the lower

- 135 -
NIPPON STEEL TECHNICAL REPORT No. 123 March 2020
Table 4 Results of experimental blast furnace operation

Case 1 Case 2 Case 3 Case 4 Case 5

Pig output (kg/tap) 784 695 734 873 666
Slag output (kg/tap) 225 225 225 225 225
Melting time (min) 97.7 96.6 96.8 100.5 112.5
RAR (kg/pt) 742 783 823 773 848
Top gas temperature (°C) 360 347 406 410 397
Top gas ηCO (%) 40.5 40.5 37.1 40.5 41.9
KR (1/m) 3 080 3 366 3 323 3 190 2 815
Pig temperature (°C) 1 457 1 446 1 428 1 418 1 390
[C] (%) 4.75 4.73 4.56 4.48 4.44
[Si] (%) 0.77 0.7 1.35 0.63 0.72
[S] (%) 0.025 0.023 0.056 0.05 0.057
Slag temperature (°C) 1 533 1 541 1 503 1 500 1 498
(Al2O3) (%) 13.4 18.6 20.4 19.2 16.5
(MgO) (%) 5.36 8.5 4.75 4.63 10.6
(CaO/SiO2) (%) 1.49 1.46 1.21 1.44 1.18
Viscosity (poise) 2.35 2.5 5.93 6.03 1.45
Fig. 18 Comparison of effect of KS on KR with that of slag rate Viscosity 1 500°C (poise) 3.07 3.51 6.04 5.64 1.43
Crystallization temp. (°C) 1 354 1 430 1 376 1 404 1 366
ΔTc (°C) 179 111 127 96 132
Drainage rate (kg/s) 4.93 5.17 3.24 3.15 5.99
ΔTc = Slag temperature − Slag crystallization temperature

Fig. 19 Effect of KS on KR in the commercial blast furnace

part of the furnace and the slag drainage characteristics on the fur-
nace working floor, to cope with the rise of the slag Al2O3 concen-
tration coming from the blast furnace slag reducing movement pro-
moted from the viewpoints of reducing the amount of slag disposal
and the environmental protection. 31) The test operations were con-
ducted, under the condition of fixed per tap slag output, for five cases Fig. 20 Effect of the slag content on drainage rate
wherein two types of sintered ore were used and the slag Al2O3, slag
MgO, and the CaO/SiO2 levels were differentiated by the addition to the effect of the wettability via the slag static hold-up, and the ef-
of flux materials. Table 4 shows the results. fects of the slag viscosity and the crystallization temperature are
Figure 20 shows the effect of the slag contents on the slag drain- small. Furthermore, as Fig. 22 shows, the permeability resistance
age rate on the 1 500°C corrected base, eliminating the influence of index of the cohesive zone (KRM) is confirmed to depend on the sin-
temperature. Along with the rise of slag Al2O3, the slag viscosity rises tered ore KS. Namely, even in the high Al2O3 sintered ore, the rise
and the slag drainage rate lowers, and as MgO rises, the slag viscos- of the KS value is suppressed by increasing MgO, and the permea-
ity lowers and the drainage rate rises. On the other hand, no effect of bility of the cohesive zone is maintained thereby.
the slag crystallization temperature on the slag drainage rate was 3.3.5 Evaluation of material quality
observed. Namely, the slag drainage phenomenon is a fluidity-domi- In the twelfth test operation in January 2001, waste plastic pow-
nant phenomenon, and since the high MgO slag content in the high der was injected through the tuyere. However, since a sturdy hang-
Al2O3 content slag, despite raising the crystallization temperature, ing was built right after the start of the injection of the waste plas-
lowers the slag viscosity, high MgO was found to be effective in en- tics, the test operation was terminated.
hancing the slag drainage rate. The thirteenth test operation was conducted in November 2003
As Fig. 21 shows, a positive correlation is recognized between with the objectives of evaluating the quality of the high-reducibility
the calculated static hold-up in the dripping zone and the permeabil- sintered ore and inspecting the effectiveness of the high-reactivity
ity resistance index in the dripping zone (KRL). The pressure drop in coke under the joint use state with the high-reducibility sintered
the dripping zone of the experimental blast furnace rises with the in- ore. 32) In the experimental blast furnace, the reducing agent rate was
crease of the slag Al2O3, slag CaO/SiO2, which is mainly attributed 700–800 kg/t, higher than that of the commercial blast furnace due

- 136 -
 NIPPON STEEL TECHNICAL REPORT No. 123 March 2020
Table 5 Test cases of the experimental blast furnace

Coke (CRI)
Coke A (25.6) Coke B (42.2)
Sinter A (65.3) Case 1 -
Sinter (RI) Sinter B (72.3) Case 2 Case 3
Sinter B-fine (86*) Case 4 Case 5
*
Apparent RI

Fig. 21 Effect of calculated static hold-up on permeable resistance in-

dex of dripping zone KRL of experimental blast furnace

Fig. 23 Vertical distribution of temperature and gas composition

achieved.
As shown in Table 5, the test operations were conducted for the
four cases wherein the combination of each of the two types of coke
and each of the two types of sintered ore arranged as below was
Fig. 22 Effect of sinter high temperature property KS on permeable changed. Two types of coke were produced by changing the coal
resistance index of cohesive zone KRM of experimental blast blending ratio so that the coke reactivity index (CRI) of the coke is
furnace
differentiated by 10%. The two types of sintered ore have different
levels of reducibility index (JIS-RI). One case in which the apparent
to: the large per ton hot metal heat diffusion from the furnace be- JIS-RI of the high-reducibility sintered ore is further improved by
cause of the restricted furnace volume, higher furnace top gas tem- grain-refining was additionally arranged.
perature because of the restricted furnace height and the restricted In the case where the coke reactivity is fixed and the sintered ore
blast air temperature. Therefore, upon evaluating the sintered ore reducibility is changed, in addition to the rise of the hot metal form-
quality and the coke quality, various countermeasures were taken to ing rate, increase of the top gas utilization and improvement in the
conduct the operation under the in-furnace reducing condition as reducing agent rate by the enhanced JIS-RI, the improvement in
closely as possible to the actual blast furnace operation conditions. permeability was confirmed. In the case where the sintered ore re-
As for the modifications of the equipment, aiming at heat loss ducibility is fixed and the coke reactivity is changed, improvement
improvement, the furnace height was extended, and the stock level in the top gas utilization efficiency and lowering of the top gas tem-
was set at the height of 6.0 m from the tuyere level, heightened from perature by the use of high-reactivity coke were confirmed, while
the past 4.5 m level. By the extension of the furnace height, the fur- the amount of solution loss carbon decreased. As shown in Fig. 23,
nace volume was expanded to 4.0 m3 from the past 3.0 m3. Further- according to the measurement by the vertical probe at the same
more, as new sensors to grasp the in-furnace state of the reaction time, with the high-reactivity coke, the thermal reserve zone tem-
further, a rigid-type vertical probe that follows the descent of the perature dropped to 920°C from 980°C, and the improvement in re-
burden was installed, and the gas sampling devices that enable si- action efficiency was confirmed. Simultaneously, to grasp the in-
multaneous gas-sampling in the height direction were installed on furnace phenomena when the high-reactivity coke is used, evalua-
the furnace wall (Fig. 10). The modifications in the blasting condi- tion of the strength of the in-furnace coke sampled by the sampling
tion are: increase of the bosh-gas volume anticipating improvement device during operation was also conducted.
in the heat loss by the increased production, the rise of the tempera- According to Fig. 24, when the high-reactivity coke was used
ture set in front of the tuyere anticipating increase in the hot gas (Case 3), the amount of solution loss carbon decreased as compared
flow rate, decrease in pulverized coal rate, and the dehumidified with Case 2, and as a result thereof, the degradation of coke was
blasting by the application of nitrogen injection. Additionally, as a considered to be suppressed. However, when KR was maintained at
result of thinning the charged material layer aimed at improving the the same level as that of Case 2, the improvement of the in-furnace
reactivity efficiency, and lowering the targeted hot metal tempera- permeability was not confirmed, which is considered to be attributed
ture, an operation with a reducing agent rate of below 600 kg/t was to the extremely small load in the experimental furnace as compared

- 137 -
NIPPON STEEL TECHNICAL REPORT No. 123 March 2020
Table 6 Layer structures of test cases

Fig. 24 Comparison of effect of CRI on coke degradation, reaction ra-

tio and KR

Fig. 26 Influence of layer structure on reducing agent rate, gas utiliza-

tion and pressure drop

rials of ore and coke which had been previously blended according
to the predetermined weight ratio (Fig. 10).
Table 6 shows the layer structure for each test case. The conven-
tional layer structure of ore and coke charging is taken as the basis
for comparison, and the following three cases of mixed layer ar-
rangement were prepared: Case 1: half of the entire charging materi-
als in terms of weight are mixed (partly mixed condition); Case 2:
the coke of the mixed layer of Case 1 is grain-refined (partly refined-
Fig. 25 Effect of JIS-RI and CRI on reducing agent rate coke mixed condition); and Case 3: fully mixed case of Case 1
(completely mixed condition).
with that of a commercial blast furnace. Figure 26 shows the result of each case of the test operation
Figure 25 shows the effects of the sintered ore JIS-RI and the with respect to the reducing agent rate, top gas utilization efficiency,
coke CRI on the reducing agent rate. With a fixed CRI, along with and the in-furnace pressure drop. Together with the order of increase
the rise of JIS-RI, the top gas utilization rises and the reducing agent in the mixing ratio, or the charging layer structure change from the
rate is lowered. When JIS-RI is fixed, since the abovementioned conventional layer arrangement to the partial mixing and the com-
trend is remarkable when the high-reactivity coke is used, it was plete mixing arrangement, the top gas utilization is improved and
confirmed that the operation-improving effect of the high-reactivity the reducing agent rate is enhanced. From the result of the measure-
coke works effectively under the joint use with the high-reducibility ment by the vertical probe, as compared with the base case of the
sintered ore. Further, JIS-RI of the grain-refined high-reducibility conventional layer structure, in Case 3 of complete mixing, the ther-
sintered ore is equivalent to 86%. However, its effect on the gas uti- mal reserve zone temperature and the top gas temperature are lower,
lization efficiency and the reducing agent rate was small. The rise of and improvement in the top gas utilization was confirmed. Further-
the top gas temperature is considered to offset the reducibility im- more, in the result of the dissection survey of the experimental blast
proving effect by grain-refining. furnace, in the mixed layer of coke and ore, the bedrock-like fu-
3.3.6 Evaluation of the effect of mixed layer sioned ore layer normally observed in the conventional ore/coke
The fourteenth test operation, the last test operation, was con- layer structure was not confirmed. This is considered to be attributed
ducted in November 2008, five years after the implementation of the to the aggregate effect of the mixed coke. In the experimental blast
thirteenth test operation. To realize the high productivity and low re- furnace wherein the load exerted by the burden is smaller than that
ducing agent rate, the effect of the mixed layer of ore and coke of a commercial blast furnace, the lowering of the pressure drop was
(proximate arrangement of ore and carbonaceous material) on reac- not confirmed. In the cases of partial mixing, the pressure drop is
tivity and permeability was evaluated. 33) For the implementation of higher than that of the base case, the phenomenon of which is con-
this test, in addition to the then existing hoppers for coke and ore sidered to be attributed to the decrease in the void fraction of the
use, an additional hopper was installed for charging the mixed mate- packed bed, and the increase in the slag holdup caused by the coex-

- 138 -
 NIPPON STEEL TECHNICAL REPORT No. 123 March 2020

nace, together with the fundamental experiments and the utilization

of mathematical models and the investigation on commercial blast
furnaces, continues to be relevant.
The Hasaki experimental blast furnace, built in 1982 and used
for the development and study of technologies including those of
new processes for a quarter of a century, terminated its service in
2008, and was dismantled and removed with the hot stove included
in 2018 after the completion of the technology transfer to the Kimi­
tsu experimental blast furnace (furnace volume 12 m3, completed in
2015) that was newly constructed for the COURSE50 Project. 4) To
close this article, we wish to express herein our sincere gratitude to
all who participated in the development and operation of Nippon
Steel’s experimental blast furnace.

References
1) The Bureau of Mines, The Ministry of Agriculture and Commerce: Ref-
erence document pertaining to iron and steel industry. 1919
2) The Bureau of Mines, The Ministry of Commerce and Industry: Refer-
ence document pertaining to iron and steel industry. 1931
Fig. 27 Relationship between reducing agent rate and gas utilization of
3) The Japan Iron and Steel Federation: Handbook for Iron and Steel Sta-
experimental blast furnaces
tistics. 1961–2018
4) The Japan Iron and Steel Federation HP: http://www.jisf.or.jp
5) Tate, M.: Met. Technol. (Jpn.). 399, 36 (1963)
istence of the mixed coke of remarkably decreased grain size due to 6) Tate, M.: Tetsu-to-Hagané. 70 (11), 1501 (1984)
prior solution loss reaction and the relatively sound slit coke in the 7) Nakata, Y. et al.: Tetsu-to-Hagané. 16 (11), 1205 (1930)
dripping zone. 8) Yasumoto, T.: Blast Furnace Ironmaking Method. Sangyotosho, 1954
9) Miyashita, T. et al.: Tetsu-to-Hagané. 57 (11), S351 (1971)
10) Miyashita, T. et al.: Tetsu-to-Hagané. 58 (5), 608 (1972)
4. Conclusion 11) Ando, R. et al.: Nippon Kokan Tech. Rep. 54, 371 (1971)
Much information was obtained from the experimental blast fur- 12) Miyazaki, T. et al.: Tetsu-to-Hagané. 73 (15), 2122 (1987)
nace of Nippon Steel constructed in Hasaki after those in Higashida, 13) Hatano, M. et al.: Tetsu-to-Hagané. 62 (5), 505 (1976)
Tobata, and that of the University of Tokyo which have supported 14) Mizuno, Y. et al.: Tetsu-to-Hagané. 70 (4), S35 (1984)
the development of blast furnace operation technologies, and the in- 15) Kamei, Y. et al.: Tetsu-to-Hagané. 79 (4), 449 (1993)
16) Kamei, Y. et al.: Tetsu-to-Hagané. 79 (4), 456 (1993)
formation obtained continues to be used up to the present day. As 17) Miyazaki, T. et al.: Tetsu-to-Hagané. 72 (4), S120 (1986)
for the problem of dissimilarity which is inevitable in a small-scale 18) Yamaoka, H. et al.: Tetsu-to-Hagané. 77 (12), 2099 (1991)
experimental blast furnace, thanks to the development of equipment 19) Miyazaki, T. et al.: Tetsu-to-Hagané. 73 (4), S129 (1987)
and operation technologies, improvements in the reducing agent rate 20) Kamei, Y. et al.: Tetsu-to-Hagané. 79 (2), 139 (1993)
21) Yamamoto, T. et al.: CAMP-ISIJ. 6 (4), 1012 (1993)
and the gas utilization have been achieved, approaching the com-
22) Ishida, H. et al.: CAMP-ISIJ. 10 (1), 197 (1997)
mercial blast furnace levels as shown in Fig. 27. 23) Yamagata, C.: et al.: CAMP-ISIJ. 4 (1), 143 (1991)
Furthermore, owing to today’s highly accurate mathematical 24) Yamagata, C.: et al.: CAMP-ISIJ. 4 (4), 1020 (1991)
models, the issue of dissimilarity has been considerably comple- 25) Ujisawa, Y. et al.: Tetsu-to-Hagané. 92 (10), 591 (2006)
mented, and the precise estimation of the operation of commercial 26) Mochizuki, K. et al.: Tetsu-to-Hagané. 72 (14), 1855 (1986)
27) Takatani, K. et al.: ISIJ Int. 39 (1), 15 (1999)
blast furnaces based on the quantitative analysis of the operation re- 28) Ujisawa, Y. et al.: CAMP-ISIJ. 22 (1), 282 (2009)
sult obtained from the experimental blast furnace has become possi- 29) Matsukura, Y. et al.: Tetsu-to-Hagané. 87 (5), 350 (2001)
ble. However, there is still room for further study on the effects of 30) Matoba, Y. et al.: Tetsu-to-Hagané. 60 (5), S354 (1974)
the in-furnace load on the difference in the ore-softening and shrink- 31) Sunahara, K. et al.: Tetsu-to-Hagané. 92 (12), 875 (2006)
32) Natsui, T. et al.: Tetsu-to-Hagané. 99 (4), 267 (2013)
ing behaviors of iron ore, and on the permeability. Therefore, the
33) Natsui, T. et al.: CAMP-ISIJ. 25 (2), 958 (2012)
significance of the R&D method using an experimental blast fur-

- 139 -
NIPPON STEEL TECHNICAL REPORT No. 123 March 2020
Takuya NATSUI Yoshinori MATSUKURA
Senior Researcher Senior Manager
Ironmaking Research Lab. Experimental Blast Furnace Project Div.
Process Research Laboratories Process Research Laboratories
20-1 Shintomi, Futtsu City, Chiba Pref. 293-8511

Kohei SUNAHARA Yutaka UJISAWA

Senior Manager, Head of Research Plant General Manager, Ph.D. (Environmental Studies)
Ph.D. (Environmental Studies) R & D Planning Div.
Experimental Blast Furnace Project Div.
Process Research Laboratories

Shinichi SUYAMA Takanobu INADA

Production & Technical Control Dept. Dr. Eng.
Production & Technical Control Div. R & D Planning Div.
Kashima Works

Kaoru NAKANO
Chief Researcher
Ironmaking Research Lab.
Process Research Laboratories

- 140 -