Today, about 20-25 kg of fine iron oxide-rich dust per tonne of raw steel arises in the off-gas cleaning systems
of integrated steelworks in Germany.
Because of their zinc, lead and alkaline content these materials cannot
be recycled back into the production process and have to be recycled externally or dumped - at some expense.
Limited landfill capacities, rising costs, tighter laws and governmental regulations, and the public
environmental discussion, make it increasingly difficult to continue the current practice. Therefore, a new technology
had to be developed to recycle these residuals back into the production process and prevent dumping, for environmental
and cost reasons.
The resultant technology, named OxiCup, is based on a modified cupola furnace which turns cold-bonded
carbon-containing agglomerates into hot metal. The main advantages of this process for a steel mill are:
- The process is very similar to a blast furnace, hence minimal operator training is necessary
- It supplements BF-grade hot metal for the steel mill and achieves higher utilisation of iron units
- Zinc-enriched sludge can be sold for further recovery
- All kinds of metallic revert metals, like skulls, desulphurisation metals and metals from slag processing
can also be treated
Development of the process
The process consists of the physical combination of fine grained waste oxides with a coke breeze reducing
agent into bricks (called C-bricks) which are charged to a shaft furnace, together with other iron-bearing by-products,
coke and fluxes (see Figure 1
). While the charge is descending in the furnace the material is heated, and at about
1,000°C the coke breeze inside the bricks is converted to CO-gas which directly reduces the iron oxide grains.
The high temperature and the high surface area of the fine waste material make the reaction rate very fast.
In laboratory tests the samples were reduced and metallised within a few minutes, whereas in the shaft furnace
the material spends about 20 minutes in the required temperature range between 1,000°C and 1,400°C. The brick
turns to direct reduced iron (DRI), which further descends to the melting zone of the furnace where it melts
together with other metallic by-products.
Carburisation of the hot metal takes place predominantly in the hearth of the cupola. Hot metal and slag are
tapped continuously and are passed through an iron and slag separator; then the hot metal is sent for further
treatment, and the slag is sent to a slag granulation system. The zinc content of the residues is volatilised
and leaves the furnace with the top gas, and as strongly enriched shaft furnace dust.
After extensive laboratory testing the first industrial trial was made in a foundry cupola. The 10 t/h
furnace was operated for more than 10 hours with up to 30 per cent of C-bricks. There were no major changes
in the composition of hot metal and the production rate decreased, as expected, to 8.5 t/h. There was a slight
increase of iron oxide in the slag from 1 per cent to 2.5 per cent, which equates to 1 kg Fe/t HM.
The CO content in the top gas increased and there was also a rise in dust due to the relatively low strength of
The main conclusion of the trial was that the Fe-oxides were reduced and converted to hot metal.
Thyssen Hamborn shaft furnace
To prove the process over longer periods of time, ThyssenKrupp Stahl, together with Mannesmann,
Küttner, B.U.S and Messer Griesheim, decided to build a pilot shaft furnace in Duisburg as a feasibility
demonstration plant for a hot metal production of 15 t/h at a cost of DM 15 million (€7.7 million).
Peripheral equipment such as a recuperator and a wet disintegrator-gas-cleaning-system, were bought used from a foundry
nearby in order to minimise costs.
The furnace itself was built in the supporting steel structure of former blast furnace No.3 in Hamborn.
shows a schematic of the plant. The inner diameter of the hearth is 2.4 m and of the shaft 2.6 m.
A total of 17,000 m³/h of cold blast is heated in a recuperator up to 520°C and the furnace
is equipped with six cast copper tuyeres and a system to inject oxygen via supersonic lances. The advantage
of this technology, compared to oxygen-enrichment to the blast, is more effective penetration, which is very important
for low coke and refractory consumption rates. The top gas leaves the furnace at about 300°C via a gas off-take
to the disintegrator gas cleaning system. The furnace-shaft itself is more than 5 m longer and the charge
material in this part of the shaft seals the process against the atmosphere. The top gas is cooled, washed and
then burned, either in the recuperator for heating up the cold blast, or in a torch, because for this stage of
the project there was no connection to the gas mains of the steel mill.
The furnace is lined with cast refractory material and operates as a typical wet bottom cupola.
Iron and slag flows continuously through the taphole. Slag and iron are separated in a conventional front
spout and the process retains the normal cupola flexibility. It is possible to shut down the furnace in one minute.
Following much laboratory testing,
the optimised C-brick was a cement-bonded 110 mm hexagonal brick,
as shown in Figure 3
, as used for some alloying materials. Together with the waste oxides, the bricks contain
LD-fine dust, BF-sludge and mill scale, 15 per cent of coke breeze and cement, which is intensively mixed in a special
The bricks were then formed in a vibrating press and cured by storing for five days before they were handled
as a bulky material. The Fe-content of the bricks is about 50 per cent.
Scrap and metallic revert materials
In the first six weeks after start up
of the furnace only common scrap was used. From then on the charged materials
consisted of 100 per cent skulls from the steelplant as shown in Figure 4
and the magnetic fraction
of desulphurisation slag. Material in the size range of 10-600 mm can be treated without any problems, but larger
pieces have to be charged in a limited amount. The Fe-content of these metallic revent materials is about 70-80 per
cent because of adherent slag.
Operation of the plant
After solving the common technical start up problems, the focus of optimisation
was on refractory campaign life.
Due to the slag rate of about 350-450 kg/t hot metal, which is very high compared to a foundry cupola
(60 kg/t HM), the tap hole wore very quickly. By redesigning the geometry of the front spout and making
a change in the composition of the refractory material, the taphole campaign life was extended to 14 days:
the same value as for scrapbased cupolas. It was possible to tap up to 10,000 t of hot metal
and 4,000 t of slag per campaign.
The development of the charged material is shown in Figure 5
. The proportion of C-bricks was gradually
increased to a maximum 70 per cent. The production of hot metal decreased as expected due to the lower Fe-input.
In the pilot stage the plant produced almost 50,000 t of hot metal and processed more than 5,000 t
of C-bricks. As shown in Figure 6
, the amount of bricks in the charge had no major effect on the hot metal
composition, whereas the use of metals from desulphurisation slag results in fairly high sulphur contents.
The influence of sulphur on the solubility of carbon in hot metal can also be seen.
The carbon content decreases
to 4 per cent as the sulphur content rises to 0.3 per cent, at Si-levels of about 1.5 per cent. The variation
of the Fe-content in slag is only a matter of refractory wear at the slag notch. In contrast to a foundry operation
Si was reduced from the slag.
More significant is the correlation of the percentage of C-bricks in the charge materials and the permeability
of the furnace, measured as back pressure in the bustle pipe (see Figure 7
The increase of back pressure
is caused by the increase of shaft gas and there is no indication of a quasi-cohesive zone and no clogging
or hang-ups were observed.
The void-factor rises with increasing amounts of C-bricks because the bricks have all the same size, and
the conditions in the packed bed move in the direction of the ideal mono-grain structure which decreases the back
pressure. The increase of the rate of C-bricks from 55 per cent to 70 per cent of the charge material showed no
further increase of the back pressure. It can be assumed that both effects eliminate each other.
The furnace conditions were stable in every stage of the trials, and even one hour off-blast caused
no problem when returning to normal operation.
Unfortunately, the second-hand blowers and the off-gas cleaning
system were at the limit of their capacity and so a further increase of the C-bricks content was not possible.
Conclusion and outlook
The main conclusion of the operation of the demonstration plant is that the process to reduce
(C-bricks) to hot metal in a shaft furnace is stable and reliable. A total of 22.5 t/h of C-bricks were charged,
in addition to 9 t/h of skulls. The use of skulls up to 600 mm in size in the shaft furnace instead of in
the converter, increases the performance of the BOF plant, allows higher direct tapping rates and lowers the cost
Clean Zn coated scrap can now be used in the BOF shop in large quantities. The Zn rich sludge of the OxiCup
furnace can be sold for further recovery. Undesirable fine materials can be taken out of the sinter plant,
which increases performance and lowers stack emissions.
|Figure 8. Pilot OxiCup plant at Thyssen Krupp Stahl|
The highlights of the achieved results are:
- 740 t/d maximum production - 100 per cent skulls
- 70 per cent maximum portion of bricks in charge
- 22.5 t/h maximum consumption of C-bricks
- 450 t/d maximum production in 70 per cent bricks - 30 per cent skulls mode
- 14 day taphole refractory campaign life
- 6 week hearth refractory campaign life
- Use of China-coke without problems
- Use of 20 per cent blast furnace coke
The results of the 70 per cent brick trials are equivalent to a specific productivity per m³ of working volume
of about 11 t/m³ per 24 h, and more than 110 t/m² per 24 h per m² of hearth area.
The plant is now rebuilt as a commercial operating installation, while a brick-making facility is to be
built in the area of the furnace.