SMELTING REDUCTION FOR IRONMAKING, Bhubaneswar (18 - 19 December 2002) 1

HIsmelt - The Alternative Ironmaking Technology

P D Burke, S Gull
HIsmelt Corporation
PO Box 455 Kwinana 6966 Australia

Peter Burke – Manager Technical, HIsmelt Corporation

Steve Gull – Executive Manager Marketing, HIsmelt Corporation

Synopsis furnace has reached a level where it becomes

increasingly difficult to make further significant
The Global Steel Industry needs an alternative to improvements. The complexity of modern
conventional ironmaking technology to meet the furnaces and their associated infrastructure has
challenge of increasing environmental and cost raised capital requirements, reduced operating
pressures. The extensive and very successful flexibility and the impacts of downtime (relines),
testing of the HIsmelt Process provides the basis has become a major logistical hurdle.
to commercialise this technology to meet these
challenges. The technology addresses key issues The driving force behind the interest in alternative
that currently confront the industry, such as, high ironmaking is:
capital and operating costs, environmental
constraints and lack of production flexibility. • The ability to utilise cheaper and more
abundant raw materials such as non-coking
In the final phase of demonstrating the commercial coals and non-agglomerated ores,
viability of the technology, the first commercial • Smaller economic plant sizes,
scale HIsmelt plant will be constructed at • Competitive capital and operating costs,
Kwinana in Western Australia. The plant will • Reduced environmental problems through the
produce around 820,000 tpa of high-grade pig iron elimination of coke ovens and sinter/pellet
that will be marketed as a high quality feed for the plants, and
global EAF sector. Construction will commence in • Flexibility of operation.
early 2003 with first production planned for last
quarter of 2004. With a growing domestic and regional steel
demand, India has the potential to become a major
Importantly for India, the HIsmelt Process has the player in the steel industry. Large reserves of high-
potential to deliver a low cost, quality metal feed grade iron ore, low labour costs and high quality
utilising fine iron ore and domestic non-coking management provides India with a competitive
coals in a process that has significantly reduced advantage.(1) Impasse to the expansion of the
capital expenditure and is more environmentally Indian steel industry include (2):
friendly than traditional ironmaking routes.
• The lack of good quality domestic coking
Introduction coals forcing a dependence on costly
coking coal imports to meet blast furnace
Even without the prevailing worldwide economic coke requirements,
climate, the global nature of the steel industry • High Al2O3 content of the ore negatively
creates a fiercely competitive environment. The impacting on sinter quality and tapping
pressure to make incremental improvements in practice,
operating costs and product quality continually • High silicon in the hot metal due to
increases, as does the pressure for significant unfavourable cohesive zone formation as a
improvements in environmental performance. result of sinter properties,
• High cost of capital.
In the field of ironmaking this pressure has led to
improvements in blast furnace technology and The alternative ironmaking technologies offer the
productivity - much of which has revolved around Indian iron and steel industry the opportunity to
control and conditioning of the material inputs to overcome some of these issues.
the furnace and add-on equipment for control of
emissions. The technology surrounding the blast
SMELTING REDUCTION FOR IRONMAKING, Bhubaneswar (18 - 19 December 2002) 2

History Table 1: Vertical Vessel Availability

The HIsmelt Process has its origins in the early Having successfully demonstrated the commercial
1980’s when Rio Tinto Limited identified the viability of the technology, the HRDF was put
potential to adapt the high scrap rate steelmaking under care and maintenance in mid 1999. The plant
and iron bath coal gasification processes developed has been partially demolished in preparation for the
by Klockner Werke to the direct smelting of iron construction of the Kwinana Commercial HIsmelt
ore.(3) Development of the HIsmelt Process Plant. The same location will be used to take
commenced with a 60 tonne K-OBM converter. advantage of some of the existing infrastructure.
After successfully proving the concept, a small
scale pilot plant was constructed and operated at The HIsmelt Process
Maxhutte Works in Germany. The plant design was
based on an enclosed horizontal vessel to overcome HIsmelt is an air based direct smelting technology
process containment issues experienced in the 60 which is simple yet innovative. The HIsmelt
tonne converter trials. process is illustrated in Figure 1. The process
occurs within a vertical Smelt Reduction Vessel
Following the successful conclusion of these trials (SRV) under pressure. The vessel is refractory
the HIsmelt Research and Development Facility lined within the hearth and water cooled in the top-
(HRDF) was established in Kwinana, Western space. The refractory hearth contains the molten
Australia. The plant had a rated capacity of iron bath. A thick slag layer is situated above the
100,000 tpa and was a direct scale-up of the small metal bath. Iron ore fines, coal and fluxes are
scale pilot plant. The principle objective of the injected directly into the melt in the SRV. Upon
HRDF was to demonstrate scale-up of the core contact with the iron bath, dissolution of the carbon
plant (process and engineering) and provide in the coal occurs, which reacts with the oxides in
operating data for commercial evaluation of the the iron bearing feeds, forming carbon monoxide.
technology. Rapid heating of the coal also results in cracking of
the coal volatiles releasing hydrogen.
Whilst the process scale-up was successfully HOT AIR BLAST
demonstrated, the complexity of the engineering, 1200 C
(OXYGEN ENRICHED)
poor plant availability and difficult operational
requirements brought into question the commercial
viability of the plant design. After three years of
operation the plant was substantially reconfigured
WATER-COOLED
to address these issues. The horizontal Smelt PANELS IN TOPSPACE OFFGAS

Reduction Vessel (SRV) was replaced with a

COAL & ORE COAL & ORE
vertical vessel. This allowed extensive use of water
cooled elements in the process top space to address
availability issues. Injection of solids was through
water cooled top injection lances as opposed to
submerged tuyeres thereby simplifying the METAL
SLAG

engineering and overcoming operational issues

encountered with the submerged bottom tuyere
used in the horizontal vessel. REFRACTORY LINED HEARTH

Figure 1: HIsmelt Vertical Vessel

Installation of the vertical vessel proved successful.
The plant achieved very high plant availability as A fountain of molten material, consisting largely of
shown in Table 1 whilst still maintaining high slag, erupts into the top space by the rapid
process performance despite the increased vessel expulsion of the carbon monoxide, hydrogen and
heat losses due to the added water cooling. nitrogen carrier gas from the molten bath. Hot air at
Duration Availability Metal
1200°C is blasted into the top space through a
Production water-cooled lance. The carbon monoxide and
[t] hydrogen is post combusted with oxygen (from the
Campaign 7 52 days 98.9 % 8450 hot air blast). The heated metal and slag fall back
(Feb - Dec 1997) (8 Periods) into the bath providing the energy for direct
Campaign 8 32 days 98.9 % 5390
(Mar - Sep 1998) (4 Periods)
smelting of the iron ore. Ejected slag coats the
Campaign 9 48 days 99.6 % 8270 water-cooled panels, which reduces energy loss.
(Dec 98 - Jun 99) (3 Periods) The offgas from the process is partially cooled in a
Cumulative 132 days 99.2 % 22 100 membrane tubed hood. The sensible and chemical
energy in the offgas can then be used to effect
SMELTING REDUCTION FOR IRONMAKING, Bhubaneswar (18 - 19 December 2002) 3

some preheating, pre-reduction and/or calcination operating campaigns. The longest of these
of the metallic feed and fluxes. Then the offgas is was a 38 day run during which 99.8%
cleaned in a scrubber and used as fuel for the hot plant availability was achieved.
blast stoves or in a co-generation plant.
(2) Production capabilities have been assessed
The vessel is equipped with a forehearth for across the full range of ferrous feed
continuous tapping of hot metal. This maintains a reduction levels from hematite ores
optimum bath level within the SRV and provides a through to DRI feed. Intermediate levels
clean product stream without the need for external of pre-reduction were tested by blending
slag and metal separation. Slag is tapped ore fines with DRI. Oxygen enrichment
periodically through a water cooled notch. of the hot air blast (up to 30% oxygen
content), has been successfully utilised to
HIsmelt has a number of unique features that sets increase the operating intensity of the
it apart from the other direct smelting processes. vessel, resulting in the expected increases
Firstly, HIsmelt uses the metal bath as the primary in productivity.
reaction medium. Other direct smelting processes
generally top-feed ore and coal, with smelting via (3) Extended pilot plant operation with high
char (plus a small amount of metal) in the slag post combustion (PC) and low coal
layer. Dissolved carbon in metal is a more readily consumption has been demonstrated
available reductant than char in slag, since the latter repeatedly. PC in the range of 60-80%
requires a gas-phase intermediate (CO). In other has been achieved routinely and, with cold
words, HIsmelt achieves significantly faster iron ore fines as ferrous feed, coal rates of
smelting rates by using carbon in a more active (i.e. around 880 kg per tonne of hot metal
dissolved) form. (kg/thm) have been recorded. This
translates to a coal rate of 600-620 kg/thm
A second differentiating factor is the degree of for a scaled-up plant with an ore
mixing in the melt. Injecting feed materials preheater.
directly into metal generates a large volume of
“deep” gas. This creates a strong buoyancy-driven (4) A significant portion of the pilot plant
upward plume that in turn causes rapid turnover of operation utilised ore fines with a high
liquid. It has been calculated that this turnover is phosphorus content (0.12% P). The
in the order of tonnes per second. Under these phosphorus removal capability of the
conditions there is very little potential for process has been readily demonstrated
establishing significant temperature gradients with an average of 85-95% of the
(greater than 20-30 oC) in the liquid phase and the phosphorus input reporting to the slag
system operates with an (essentially) isothermal phase.
melt. The rapid mass turnover promotes good heat
transfer from the top space to the bath without (5) Typical steelplant waste materials have
significant over-heating of individual liquid also been successfully utilised. A mixture
droplets. Implications are significant for hearth of blast furnace and LD dusts and sludges,
refractories in the slag-line region, since good millscale, pellet fines and casthouse dusts
mixing leads to the bricks being exposed to low were blended with ore fines and injected
FeO and uniform (low) temperature. without any need for feed agglomeration.
These trials resulted in the expected
HRDF Operations increase in productivity and reduction in
coal consumption. Dust losses were not
The major achievements of the Kwinana HIsmelt increased, achieving iron unit yields of
Research and Development Facility during the greater than 97% which was similar to the
vertical vessel phase of the development were: operation using ore fines. The zinc and
lead components of the waste materials
(1) The Kwinana plant was run cumulatively were found to partition to the dust,
for over 3000 hours (132 days). Total suggesting that reasonably high levels of
metal production was around 22100 zinc and lead can be accommodated and
tonnes, together with 8600 tonnes of slag. potentially recovered. In a commercial
A total of 63000 tonnes of solids facility, these waste materials have a high
(primarily ore, coal and flux) was injected value, reducing coal consumption to less
through the two injection lances. Plant than 650kg/thm without the need for
availability in excess of 99% was preheat.
demonstrated in several continuous
SMELTING REDUCTION FOR IRONMAKING, Bhubaneswar (18 - 19 December 2002) 4

(6) The range of coal types used has allowed implying hearth refractory life well in
their impact on process performance to be excess of 1 year.
quantified. Coals with higher volatile
contents have the expected negative (8) No water leaks have occurred and no
impact on production as a result of the repairs have been required on any of the
energy loss due to gasification and water cooled elements in the Smelt
cracking of the volatile components. Reduction Vessel.
Other potential production impacts from
coal, oxygen, moisture and ash content (9) Handling of hot process offgas in a
have been examined and the findings membrane-type hood has been
accounted for in commercial flowsheeting. demonstrated over a period of 1100
All coal used have exhibited an ability to operating hours. Results show that,
be used in the process and coal selection despite the steady-state nature of the
remains an economic decision. (Table 2). HIsmelt process, the hood readily comes
Throughout these trials, the plant to equilibrium in terms of accretion
produced a consistently high quality hot behaviour. Operation is simple and self-
metal (see Table 3). regulating, requiring little or no operator
intervention.
Ferrous Fe (T) Fe (Met) C Silica +
Materials Alumina Kwinana Commercial HIsmelt Plant
Hamersley – 61.0 % 7.5 %
DSO Fines
(0.075% P)
The outstanding success of the HRDF operation
Hamersley - 62.4 % 5.6 % convinced Rio Tinto Limited to pursue the
Brockman Fines commercial realisation of the HIsmelt Process. In
(0.12%P) 2002 an unincorporated joint venture was formed
DRI 90.5 % 84.2 % 5.5 % between subsidiaries of:
Steelplant Wastes 53.3 % 10.1 % 5.5 %
Coals Fixed Volatiles Ash CV
C • Rio Tinto Limited;
Coal A 73.2 % 9.8 % 12.0 % 28.3 GJ/t • Nucor Corporation;
Coal B 69.3 % 16.8 % 6.9 % 30.5 GJ/t • Mitsubishi Corporation; and
Coal C 68.8 % 25.7 % 4.8 % 32.9 GJ/t • Shougang Corporation;
Coal D 49.9 % 38.5 % 9.4 % 30.2 GJ/t

for the purpose of constructing and operating a

Table 2 Range of Raw Materials Used commercial HIsmelt facility. The Kwinana
Commercial HIsmelt Plant will produce around
Hot Metal Analysis 820,000 tpa of high quality pig iron. The new
Typical Range Comment
Analysis Possible
plant will be built on the existing HRDF site in
Carbon 4.0 ± 0.2% 3.5 - 4.5% Easily Kwinana, Western Australia.
controllable
Silicon - - No silica
reduction in the
Construction of the new plant will commence from
Process the beginning of 2003 with expected completion by
Manganese 0.1 % 0 - 0.2% Depends on Mn the end of 2004. Figures 2 and 3 illustrate a
in ore
Sulphur 0.1 ± 0.02% 0.08 – 0.15% Desulphurised simplified flowsheet and a computer generated
prior to use image of the commercial plant to be built at
Phosphorus* 0.03 ± 0.01% 0.02 – 0.05% Based on ore Kwinana.
with 0.12% P
Temperature 1480± 15°C 1450 -1550°C Easily
controllable

Table 3 Typical Analysis of HIsmelt Hot Metal

(7) Plant operating parameters have been

developed to allow close control of hot
metal carbon and temperature. Slag
chemistry and inventory have been
optimised to achieve hot metal quality,
productivity and minimise refractory
wear. Refractory wear rates less than 1 kg
per tonne of hot metal were achieved,
SMELTING REDUCTION FOR IRONMAKING, Bhubaneswar (18 - 19 December 2002) 5

• reduced flux consumption, and

Oxygen Plant
Iron Ore Feed • reduced refractory consumption.
880 t/day

The ease with which the HIsmelt process can be

Hot Blast
Stoves
started, stopped and idled provides flexibility that
ORE
PREHEATER is fundamental to the EAF producer. This allows
1200 °C
the effective coupling of iron and steelmaking
Hot Blast
Hot
operations without the need for expensive and
Ore under-utilised facilities for holding and disposing
20MW
Power Plant
of, surplus hot metal production.
Dryer Fuel
Gas Power Plant=
Power Generated
Coal Power Consumed
Internally
Value to Integrated Mills
SMELTER
Slag Metal
By far, the major value to the integrated producers
is the fact that the HIsmelt process produces hot
Figure 2: Simplified Kwinana Plant Flowsheet metal without the need for coke ovens and sinter
plants. The direct use of low value grades of iron
ore fines without pre-treatment increases the
flexibility of raw material supply for steel
producers, assuring a greater competitiveness in
costs. The use of steaming coals also brings
substantial costs advantages. The flexibility to vary
feedstock with productivity demands, allows the
operator freedom to continuously optimise
operating costs.

A HIsmelt facility uses much of the same

equipment as a blast furnace. Hence, a HIsmelt
Figure 3: Future Kwinana Plant plant can be easily ‘retro-fitted’ into an integrated
site.
The Future of HIsmelt
Raw material selection along with critical operating
The successful operation of the 6M unit proposed
parameters (e.g. hot blast rate and level of oxygen
for Kwinana will open the way for general
enrichment) can be readily adjusted to maximise
applications of this type of smelter. The 6M
the efficient use of the process in line with
HIsmelt facilities are ideally sized to provide high steelmaking requirements. Again process flexibility
quality metallics to the EAF producer. From the is important.
proposed Kwinana plant the scale-up from 6M to
8M is only two-fold, and should not pose any The economic performance of the process at
significant additional development hurdles. The relatively low production capacities permits the use
8M HIsmelt unit, or a combination of 8M units, of multiple units to replace large blast furnaces.
would be able to replace a blast furnace in situ. This, coupled with the potential for significant
increases in productivity of each unit through raw
Value to Mini Mills material selection, means that the site wide impacts
of downtime and relines can be more easily
Stable supply of high-quality virgin iron units is a accommodated.
major issue for EAF producers, especially those
with plans to penetrate further into flat products. The metal can be mixed directly with blast furnace
EAF operators can feed up to 30-50% hot metal as iron or alloyed to provide precise low silicon
a feed. Hot metal is a high quality substitute for content for the LD charge. Methods of “slag-less”
cold pig iron, DRI and premium quality scrap, steelmaking are widely practised throughout Japan.
providing high value-in-use in the charge mix due Blast furnace iron is being desiliconised and
to the following: dephosphorised before it is charged to the LD. (4)
Thus compared to a blast furnace hot metal,
• increased productivity, HIsmelt hot metal has the following potential
• reduced tap to tap time, advantages:
• lower energy consumption per tonne,
• lower residual contents, • reduced flux consumption,
• consistent high quality products, • increased liquid steel yield,
SMELTING REDUCTION FOR IRONMAKING, Bhubaneswar (18 - 19 December 2002) 6

• increased productivity due to a reduced

blowing time,
• production of higher quality (low
phosphorus) grades,
• reduced re-blows,
• reduced refractory consumption,
• decreased consumption of ferro-alloys,
and
• production of ultra-clean steel.

Conclusion

The HIsmelt process has moved from pilot plant

status to a position where it is ready for commercial
application. The Kwinana Commercial HIsmelt
Plant will be the final step in demonstrating the
commercial viability of the HIsmelt process. The
HIsmelt Process addresses all of the key
requirements of the industry for a successful
alternative ironmaking technology, combining a
high level of technical achievement with simple
engineering concepts and existing proven plant
technology. The flexibility offered by the process,
results in a wide range of opportunities for
steelmakers in both the integrated and EAF sectors,
to add value to their operations in a way that suits
their particular circumstances.

In particular for India, the HIsmelt Process will

deliver a low cost, quality metal feed to either a LD
plant or EAF facility utilising domestic fine iron
ore and non coking coals, in a process that has
significantly reduced capital expenditure and is
more environmentally friendly, compared to the
traditional ironmaking technologies.

References

1. A. Chatterjee, “Steel Industry in South East

Asia with Particular Reference to India”, 1998
ICSTI/IRONMAKING CONFERENCE
PROCEEDINGS, 1999, p93.
2. S. Dewan, “Techno-economic Evaluation of
Smelting Reduction and Other Alternative Iron
Making Processes in India”, SEAISI Quarterly
April 2000.
3. K. Brotzmann, Howe Memorial Lecture, “New
Concepts and Methods for Iron and Steel
Production”, ISS AIME 70th Steelmaking
Conference, 1987, pp 3-12.
4. Development of New Slag-less Steelmaking
Process at NKK Fukuyama Works.