THE PRODUCTION OF IRON-RICH IRON-COKE FOR BLAST FURNACE OPERATION

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP80-00926A007900040001-2
• Release Decision: RIPPUB
• Original Classification: K
• Document Page Count: 27
• Document Creation Date: December 14, 2016
• Document Release Date: April 25, 2001
• Sequence Number: 1
• Publication Date: January 1, 1955
• Content Type: REPORT

. P130DUCTI8VRSFv o-ftea1fkM% {p7F flMRF8%RR2F2 A89jTFWdV001-2
THE
By Dr.-Ing. C. Eymann of Walsum
In the last few years the "iron-coke" process has been developed by
the Thyssen Gas- and Wasserwerke G.m.b.H. and the Bergwerksgesellschaft
Walsum G.m.b.H. and up to the present time more than 12,000 tons of
iron-coke has been prepared mainly at the Prosper coke ovens in Bottrop.
This iron-coke has been used partly in a slagging gas producer at the
Thyssen works, it has been intermittently tested in a large blast furnace at
the August-Thyssen iron works and it has been finally given a continuous
trial over a prolonged period and as a 100% charge at the Friedrichshutte
A.G. works at Herdorf (Sieg). Recent developments have shown that a
highly bituminous coal, i.e. a gas coal which cannot be made into a
satisfactory blast furnace coke by itself, can be used successfully as
a binder when mixed with 40% of finely ground iron ore and suitably coked.
When mixed with a suitable iron ore this coal can be used to produce a
solid large-sized iron-coke with an Fe content after coking of 35%. At
the same time, due to the reaction between the oxygen of the iron ore
and the coal a greater quantity of gas is produced which entails the
consumption of coking coal during the coking operation.
The recovery of valuable by-products, such as tar and benzol is no
lower when a fine ore of difficult reducibility is used the the production
of iron-coke, the recovery of ammonia is less. The content of hydrocyanic
acid in the gas produced during the distillation process is markedly less.
The recovery of valuable by-products, such as tar and benzol is no lower
when coke and the consumption of fuel (gas) during this operation are
later offset in the smelting of the iron-coke in the blast furnace when
both ore and coke are saved. The equivalent recovery of ordinary coke
in the iron-coke smelting process for the production of pig iron is better
than in the normal operation of coking the usual grades of coal and then
using this coke in the usual manner. This is because although there is
a loss of efficiency in the coking chamber when iron-coke is produced,
the intimate contact between ore and carbon in the smelting operation
increases the metallurgical efficiency here. The larger proportion of
iron-coke can be charged in the form of pieces of over 15 mm size, the
remainder can be used as coke breeze in the sintering process. Here
also the heat of reduction of the iron can to a greater or less extent
replace the heat of combustion. The larger size of the fine ore in the
coke breeze is advantageous here, and in order to take full advantage of
the coke-equivalent in i.roz-coke care must be taken to see that no
oxidation of the reduced iron, i.e. no roasting of the iron-coke takes
place during sintering.
In the mixing of fine coal and fine ore the ore to some extent forces
its way into the spaces in the agitated fine coal, thus the weight per
unit volume of coking coal in the coal-ore mixture is increased due to
the entrainment of the heavier ore. The quality of the coke is thus
improved in a manner similar to stamping or compressing. In the production
of iron-coke therefore the reduction in the coal capacity of the coke oven
does not correspond exactly with the ore content in the mixture.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 r F e se 0 /09/ 7: Cel P80-00926~Q079t00040~01-2
If a. Coax~~ g1%q oea coi9~ie e~ploy ~~ is, p4ssi e to produce
a very hard grade of iron-coke. This, however, aan lose its strength
at a higher temperature. Such an iron-coke contains clusters of ore
(Fig. 1), which on being heated to the higher temperatures ruling in
in the blast furnace cause the localised production of concentrates
of oxygen which destroy the coke skeleton. Local concentrations of
ore weaken the structure of the coke and favour the collapse of the coke
particles, and iron-coke of this sort can undergo collapse during its
passage down the blast furnace shaft. The preparation of iron-bearing
coke has been undertaken experimentally in several places in the last
ten years using medium bituminous coal. In cases where the work has
not been fully successful the difficulty has usually been traced to the
formation of "nests" of ore in the mixture. The strength of iron-coke
as determined at room temperature gives no indication of its behaviour
in the blast furnace, for good iron-coke must maintain its strength as
it passes down the furnace stack. This can only be obtained by the
proper fine milling of the coal, the use of finely-milled ore and
uniformity of mixing the two (Fig. 2).
The mixing of fine ore or flue dust with finely-ground coal is facilitated
by the addition of a little oil or something similar. At the same time
the production of gas is enhanced and also the weight of the mixture which
can be poured into a coke oven and settled therein satisfactorily is in-
creased. The addition of oil produces a stable ore-coal mixture in which
the two constituents remain uniformly mixed in spite of repeated unloading,
re-loading and general agitation. Therefore, there is no tendency to form
local concentrations of ore in the mixture, even with a high percentage
of ore present, and the effects which this can cause are therefore avoided.
In an iron-coke made with fine-ground coal and fine ore there are no ore
"nests". Oxidation can only take place on the surface for in the interior
of each coke particle metallic iron is combined and therefore protected
against further oxidation, i.e. against a reduction in the coke-equivalent
(Fig. 3). If these conditions are not fulfilled however then small
concentrations of oxidation can be set up when the iron-coke is in storage
and these "nests" of oxidation can have the same effect as ore "nests",
when the iron-coke passes down the blast furnace shaft the coke skeleton
is disrupted.
A condition for the production of good iron-coke is the use of a coal
with the right bitumen content. Coals which are low in bitumen content
can only be combined with a little fine ore, conversely highly bituminous
coals can carry a greater quantity of ore. In the latter case the strength
of the coke is increased so that a smeltable iron-coke can be made from
a mixture of finely-ground coal and a suitable ore without the addition
of a "fat" coal. To make iron-coke in this way has been the main object
of our work.
For the production of iron-coke we utilised the Walsum coal which contains
between 33 and 34% volatiles (dry analysis) and a magnetite dust. The
coke. prepared from this coal is normally of a small size and a brittle
character. By the addition of 5 - 7% by weight of coke breeze, which is
10 - 14% by volume the coke made from Walsum coal becomes larger-sized
and its Abrasion Index rises by about 7 points. If in place of the coke
breeze the same volume of magnetite dust is added, i.e. 20 - 30% of
magnetite by weight then the coke loses its fragility, it becomes coarse
in lump size and in strength and structure it resembles blast furnace coke
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 (Fig. 4a hprp)ipd
of the compressed ore-coal mixture is enhanced. This is one of the reasons
why a hard iron-coke can be made from a highly volatile coal (Fig. 5). The
impregnation of coal with ore is responsible both for the increased bulk
density of the ore-coal mixture and for the improved hardness and strength
of the resulting coke.
The type of iron ore used plays a decisive part in the production of iron-
coke. Easily-reduced ores quickly lose their combined oxygen during coking
and thus neutralise the bitumen content of the coal. These ores thus lower
the recovery of the by-products, the yields of tar and benzole are both
reduced. Ores which are difficult to reduce on the other hand have little
influence upon by-product recovery, they simply act as inert substances
during coking. They can only be added to coal in appreciable quantities
if the coal used is sufficiently bituminous.
Fig. 6 shows the relationship between the gas yield and the length of
coking time during the coking of Walsum coal with and without the addition
of iron ore. The reduction of the ore) as the gas yield.*indicates, sets
in early and in the positions in which the coke quickly reaches a high
temperature, for example in the layers of coke next to the kiln walls.
The figures on the curves give the calorific value of the gases generated
in kilocalories per kilogram of coal. When Walsum coal is coked by itself
from the 18th hour onwards the production of gas falls sharply, whereas
when iron ore is included in the charge the production of gas proceeds
at the same rate as at the beginning and is not complete until after
20 hours of coking. A few tests were carried out in which coking was
continued for a further 4 hours to examine the further gas yield, the
gas yield continued practically undiminished in these cases. In the
production of ordinary coke the reaction was practically complete after
20 hours. The calorific value and volumes of gas produced up to this
point have been compared with the same values for the production of gas
during the preparation of iron-coke. The difference in these figures can
be ascribed to the reduction gas. The actual volume and analysis of this
reduction gas cannot be accurately assessed as it reacts with the other
gases generated, which reaction is considerably influenced by the catalytic
action of the iron. The reduction gas which is generated in the 20th to
24th hours of coking iron-coke also contains the gas from the final after-
gassing of coke as ordinarily produced. In general however ordinary coke
is not over-coked in this manner. In the presence of iron ore however the
release of gas and with it the reduction of the ore proceeds so vigorously
that an extension of the normal coking time is not necessarily uneconomic.
The loss of combustion heat in the release of reduction gas into the coke
oven gas is offset by the enhancement of the reduction reaction. During
the extra coking time the heat input to the coke oven amounted to 100
kilocalories per kg of dry coal. The production of the reduction gas gave
a credit of 145 kilocalories per kgm of coal, i.e. during this time 45
kilocalories per kg represent the heat-giving value of the reduction gas.
If the heat balance of the coking operation is drawn up it is apparent that
the firing of the oven is dependent upon the time of coking, i.e. on the
degree of ore reduction. The sensible heat of the reduction-coke, which
is lost when the oven is pushed, must be regarded as heat input. The
sensible heat and calorific value of the coked material is higher than
calorific value and sensible heat before the reaction has taken place.
In this case a heat displacement has taken place from the coal side to the
ore side. There are various reasons which can be advanced to account for
Approved For Release 2001/09/07 : CIA-RDP80-00926A007900040001-2
 this. Fo'ApagfOeF4 ' itari&TOc1~R,WEptFA- gRRZ6tyg7oq99P4 9Oac- ermined
for the production of ordinary coke and then under the same conditions for
the production of iron coke the displacement can be calculated by difference.
However, the specific fuel consumption for the coking of coal is not a
fixed quantity but is dependent upon the way in which the process is carried
out. If a lower specific firing rate be used then the heat displacement
drops from 585 kilocalories down to 575 kilocalories if the coking time
be short and down to 540 kilocalories if the coking time be longer. Under
those.conditions, therefore the fuel consumption for the production of iron
coke increases. Results of this kind are not unusual. In individual tests
specific fuel consumptions of 540 kilocals/ton of coal have been recorded,
and even lower values than this. In ordinary practice results of this
kind can be obtained only if a continuation of the firing is avoided and
the coking operation brought to an end the moment the reaction is complete.
If the coke is heated for longer than this then one part of the unnecessarily
applied heat is lost in the chamber walls and in the regenerator while
the other part raises the final temperature of the coke unnecessarily high
and is lost by radiation and in the chimney gases when the oven is pushed
and the coke quenched. In practice it is generally impossible to bring
the cycle time so well into line with the actual coking time. The coal
is slightly damp and the charged weight varies within certain limits so a
certain degree of overheating of the coke cannot generally be avoided. In
addition the temperature of the oven walls is not fully uniform, therefore
to make certain that all the charge is coked some of the coke has to be
overheated and it is well-known that for these reasons a not inconsiderable
proportion of the applied heat has to be wasted.
With the coupling together of the reactions of coking the coal and reducing
the iron ore however it is possible to bring about a considerable improvement
in the efficiency of heat utilisation. Any heat applied to the ore-coal
charge after the coking reaction and the formation of a strong iron-coke will
in part, as shown above, be utilised in the formation of gas and the re-
duction of iron oxide. The reduction of the ore acts as a brake on the
temperature as this reaction starts when the temperature is between 700
and 8000 C. In the production of iron-coke therefore an improved heat
utilisation is possible and a reduction in external fuel consumption, and
this explains the improved thermal conditions when coal is coked in
presence of magnetite ore. Unlike the normal coking procedure, a standover
in the oven is not necessarily a source of loss when iron-coke is being
made.
Heat losses in the production of iron-coke are compounded of the fuel
consumption in the firing chamber, and in the losses involved in quenching
and cooling. In four sets of tests this amounted to between 320 and 410
kilocalories per ton of moist ore. In this process the coking action is
combined with that of sintering. The thermal waste is therefore in the
region of but a little lower than that found in the case of ordinary
sintering under suction. It is not therefore out of the way to ascribe
these losses to the sintering reaction. The reduction process therefore
may be regarded as proceeding without any thermal consumption.
The economy of the iron-coke process is thus assured by the increased heat
utilisation, the increased gas yield, the saving in sinter costs and the
noticeable degree of pre-reduction of the charge which entails a saving
of coke in the smelting process.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 W corresARphqY6qcg26,d%jHM ;tQO/economy of he Os 2n ~R ng0 process in
the iron-coke production method has been published by H. Barking and
C. Eymann in the periodical "Stahl and Eisen". The sintering costs are
lower than that in the normal suction sintering process. Finally it is
of importance to note that iron-coke can be prepared from a pure gas-coal
and this increases the range of fuels which can be utilised for the smelting
of iron.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
THE DEMAG-HUMBOLDT LOW SHAFT FURNACE PROCESS
By: Prof. Dr. E. Hofmann of Berlin-Charlottenburg
Since the turn of the century the pig iron production of the world has
increased from an annual 40 million tons to more than the quadruple amount.
The previous double production up to about 80 million tons until 1913
before the first Great Warp consisting of the production of the most
industrial countries of that time like USA, Germany, England, France,
Belgium and Luxembourg, is mainly due to the improved gas cleaning plants
as well as to the improved heating technics and operating economic
conditions (Figure 1). In consequence hereof it was possible to increase
the furnace measures as well as the number of furnaces continuously in the
above mentioned countries. The further increase up to 167.5 million tons
per year during the second period until 1953 is a consequence of the
metallurgical progressive development of the blast furnace process and a
more economic utilization by a suitable preparation of the ore. In
addition thereto', there were the efforts of many other countries like
Russia, Japan, Canada, India, South Africa and Australia for an independent
industry.
With reference hereto, a further increase of the production for meeting the
still high demand of the world for steel can only be obtained - as there is
a high raw material consumption - by using more and more fine ore as an
additional raw material for the smelting processes. With regard to the
fact, that in many countries there are no deposits of cokeable coal at all,
whilst in other countries the deposits of cokeable coal - which are of
high importance for the production of suitable, resistable blast furnace
coke until now - become more and more less, another process should be
started which renders possible the production of a technically valuable pig
iron in an economic way, for countries having non-cokeable coal at their
disposal. Although the fuel consumption for the agglomeration of fine ore
can be delivered by the sintering and pelletizing plants - being developed
to a considerable throughput capacity during the last both decades - with
waste cdke and fuel of low value, in all cases the fuel consumption of the
blast fi 'naces themselves is to be covered by a good blast furnace coke.
Even.the use of electro low shaft furnaces with electric power for the
production of pig iron in countries having cheap current at their disposal
will not be a real solution. As to R Durrer the erection of the above
mentioned electro low shaft furnaces will be favorable at those places only,
where the costs for the production per 6 KWH are not higher than for 1
Kg coal.
Already more than 10 years ago 0 Diettrich of Klockner-Humboldt Deutz
AG, Kolb-Deutz, has recommended the single stage low temperature car-
bonization of so-called combined ore-coal briquettes in a short, rectangular
low. shaft furnace of about 3- 5m shaft height only. After a test seris
carried through in a successful progression since 1952 in co-operation with
the Demag AG, Duisburg, by the Demag-Humboldt Niederschachtofen GmbH,
Duisburg, in the meantime the so-called D H N process has been tested so
far in some long time tests in a stepwise completed, half-technical pilot
plant in Koln-Kalk with a daily pig iron capacity of 12 - 15 tons, that
the erection of big plants up to about 200 tons daily capacity does not
mean a risk any longer.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 inhe fo~da~oladee2n0;d0~i9~ IB3P ~7~~~ential
characteristics of the DHN process.
Special care has to be taken for the grain structure of the single raw
material components, as the Demag Humboldt low shaft furnace process is
based on the charging of those raw briquettes only, containing all amounts
of ore, fluxes and fuel, necessary for smelting and ready for charging,
and tuned to a certain slag figure, depending on the sort of pig iron to
be produced.
On one hand, the reducibility depending on the just prevailing stage of iron
oxidation and on the other hand the slag ratio are important and decisive
for the ores. Consequently, difficulty reducible ores have to be binded in
a fore fine-grained form than slightly reducibles ores. Morever the grain-
sizes as well as the sorts and amounts of the various fluxes like limestone
dolomite lime-hydrate, bauxite or such like necessary for a suitable slag
production, have carefully to be examined with regard to their binding
strength in the briquettes in accordance with the prevailing slag ratio of
the ores.
Special care has to be taken for the coal, as after the carbonization the
coal will have to act as a binder in order to keep together all further raw
material components in the combined ore-coal briquettes.
During extensive investigations lasting for many years, the conditions were
tested, on which this result could surely be obtained with the various sorts
of coal.
With reference to the favorable results of the in-tamping of coal mixtures
of smaller baking capacity in coke ovens for the production of well-carrying
blast furnace coke, there was the change-over to the briquetting of more
slightly baking gas and open burning coal in order to obtain a briquette -
by the use of high super-charging pressures at the presses - resulting
sufficiently strong carbonization coke after carbonization in the Fischer-
retort.
During these tests it was clearly found out, that in addition to the super-
charging first of all the period of carbonization has a decisive influence
on the resistibility conditions of the produced carbonization coke. During
short periods of smelting of less than 30 minutes there were obtained not
only the required values of resistibility but also an increased tar output.
In case of highly sweeling and baking coal, which show the unfavorable
condition of sticking together after the carbonization and disturbing the
equal flowing of gas through the furnace square section during the
throughput of the briquettes in the furnace, the baking capacity can be
reduced by the addition of blast top furnace dust, very fine ore and coking
duff resp. non bituminous coal. The addition of low shaft,top furnace gas
will be most economical, as this contains carbonization coke already beside
of fine ore components; by its equal distribution in the briquet this
carbonization coke will be the support of the structure and will have a good
influence on the coming carbonization in the furnace.
Consequently it is possible to use all normal pit coal - from the slightly
baking coal up to the highly baking coal - by a short period of carbonization
for the DUN process.
Approved For Release 2001/09/07: CIA-RDP80-00926AO07900040001-2
 r Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
Now it is quite obvious, that just with regard to the necessary short
period of carbonization of about 25-30 minutes only, the flushing gas
process will be the only suitable one of all well-known carbonization
processes, and it is carried through most easily and effectively in the
low shaft furnace itself where high amounts of gas with high contents of
perceptale heat of the respective temperatures are at disposal, -- and --
as has been proved by the smelting tests -- the carbonization period will
last for about 25--30 minutes only in the low shaft furnace. Such a high
carbonization speed could only be reached in a special carbonization
furnace with extraordinary high costs for equipment.
Finally, the binder is of certain importance; in first line it is used
as pitch, tar-pitch-smelting, bitumen, sulphite liquor and suchlike, for
securing the necessary storing-stabilitiy and resistibility of biquettes as to our experiences of about 30--40 kg resistibility in raw briquettes --
until the real carbonization process becomes effective in the furnace.
Moreover, the cold raw briquettes, charged into the furnace at about 360--
4000 C top furnace gas temperature in order to avoid the discharge of
carbon, have to stand this temperature interval without bursting, crushing
or sticking together.
Only in this way there will be the guarantee, that after the throughput in
the furnace the briquettes will arrive before the tuyeres in a good condi-
tion of shape, porosity and reactivity, in order to reach the full
efficiency of their well-increased surface -- won by the intimate mixture
of all fine-grained raw material components -- with the still existing
porosity and the highly accellerated metallurgical reactions.
As the former tuning of raw briquettes to certain values of resistibility
does not give sufficient information on their real behaviour in the
furnace, they are successfully being tested in an briquettes checking
apparatus, expecially developed by the DHN, in which the briquettes --
produced as to various views and with various binders -- will be exposed
as far as possible to the real reactions occuring during the throughput in
the upper carbonization section of the low shaft furnace.
This is a checking apparatus, which mainly consists of a well-insulated,
small electrically heated furnace, through the inside of which top furnace
gas can be transferred if wanted.
By means of a movable weight on a weighing bar with a heatresisting pressure
stamp any pressure on the briquettes can be tested and the behaviour of the
samples with regard to compression can be obtained exactly from a notofying
writer.
As to our experiences, at an original height of 32 mm of the raw briquettes
the compression of 4 mm is possible under a loading of 4 kg without pausing
any disturbances in the operation of the furnace; the more, as after .
completion of carbonization in the furnaci, in a depth of 0,8 m below the
charging surface the resistibility has reached more than the double value --
compared with the raw briquette -- whilst the maximum loading in the
furnace is less than 2 kg. In correspondence with the conditions of a big
plant the cold raw briquette is charged at 4000 C and heated up to 8000 C
within about 25 minutes.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 Pitches 9,PPf 2 col; t!@ A~4 Q~ w @-~0~~6 ~ t t~xmnre,
a specially developed tar-pitch-smelting of the Chemical Plant "Plutn" of
the Rheinelbe Bergbau A.G. in Gelsenkirchen, as well as a sulphite li=luor.
The results of these checks under 4 kg loading are stated on fig. 2. The
different behaviour of the various binders is clearly recognizable.
The most favourable values - for the raw material used in this case - of
about 2-1/2 mm compression at 4 kg loading and about 30 - 35 kg cold com-
pression strength of the raw briquettes were reached by the use 6% Pluto-
tar-pitch-smelting and a mixture moisture of about 6%. After completion
of carbonization the values of resistibility of these briquettes increased
up to 65 - 80 kg. Despite the fact, that the sulphite-liquor briquettes
did prove a good behaviour even under a loading of 4 kg and showed a
compression of about 2,9 mm only, their use should be restricted to those
cases, where the tar-pitch-production from the used coal is insufficient
(for example anthrazite), the more, as under normal conditions those
briquettes are not water-proof and consequently they cannot be stored in
open weather.
Fig. 3 shows the schematic arrangement of the pilot low shaft furnace
plant in Koln-Kalk. The left upper part represents the furnace and the
charging arrangement. The furnace has a slight oval square section of
1,04 m2 and an effective hearth area of 0,82 m2 between the 4 tuyeres.
The furnace has an effective charging height of 3,2 m from the tuyeres
level to the top stock line level, although during the continuous
operation the furnace had been operated on the average of about 2,4 to
2,8 m effective height only, and the top furnace gas temperature had been
kept constantly on about 380 - 400?C for reason of a sure avoiding of
C-discharging out of the tar. The charging of raw briquettes is carried
through via the conveyor belts, the weighing machine, trap bucket and
slide, whilst for building reasons a filling bucket with valve, Parry
cone and cover serves as top closing. The gas cleaning plant consists
mainly of a dust catcher, for a most possible separation of the dry dust.
Then there is a pre-cooler, driven by direct current, with warm water for
gas cooling to about 90 - 1000 C and a mud basin thereunder, a desinte-
grater for a pressure increase of about 350 mm water column with a tar-
discharge and -transfer, a drop catcher with Raschig-rings and a tube
cooler for oil catching.
The cyclone group intended for the dry cleaning had proved to be super-
fluous during the operation, consequently the saved pressures were for
the benefit of the whole cleaning system.
During the whole test period part of the low shaft furnace gas, cleaned
in this plant to less that 0,09 g/1kn3, with a calorific value on the
average of 1400 - 1450 kcal/Nn , heated the recuperator for the pre-heating
of the blast to the average of 450? C at the tuyeres, whilst the greater
part was burned at the surplus gas burner and discharged into the open
weather without being utilized.
Finally, the lower part of.this fig. shows the blower, the recuperator,
the slag utilization, the cast pig bed as well as the weighing of pig
iron and dust resp. their transport.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 In. order aWr %"s&2% 4bn C*_ P~4 * %7 00%0 q% I?
t
as a sample of the continuous 3-weeks-test-serie that part of the test,
comprising 100 work hours, carried through for the production of Bessemer
pig iron with 1,5% Si and 0,05% S.
As can be seen from fig. 4 the uniformity of the operation of the furnace
and the products won during the last 3 days will catch the eyes at once.
Although during the first 2 days after the furnace was changed on another
burden, some of the S-values increased to more than 0,1%, and in one single
case even to 0,198%, there was reached yet a total average of 0,057%, and
during the last 3 days of even less than 0,04%. Beside of the other small
alterations of the Mn-, Si- and P-contents, the relatively high C-contents
of the pig iron, amounting to 4,3 to 4,5% C, are very remarkable, although
they have caused very small N2-contents of 0,003 to 0,004%.
Furthermore, the slag composition was a rather uniform one, consequently
the slag rations of
P1 = (._n) = 1,30 - 1,32
(Si02 )
or p2 = (CaO) + (MgO) = 1,14 - 1,17
2 (Si02) + (A1203)
showed no important alterations.
The net calorific value of the top furnace gas amounted to an average
of 1425 kcal/Itn at an average gas constitution of 30,6 vol. % CO;
3,4 vol. % CH4; 0,1 vol. % CmHn; 3,4 vol. % C02; 7.6 vol. % H2 and 54,9
vol. % N2. Single higher variations from the average value were caused
by the fact, that the terms of regular taking of samples, that means
about 5 minutes after charging, for controlling the influence of progression
of carbonization, had been postponed in case of these samples.
The Fe-contents of the dust amounted to 14 - 20%, and the fix. C-contents
amounted to 48 - 56%.
Special care had been taken for a systematic digging-out of the furnace,
cold-blown with nitrogen after the last cast in order to render possible
the conclusions concerning the sectional loadings on the single layers
as well as to the progression of the metallurgical processes with in-
creasing depth.
The digging-out has been carried through in 16 different layers of about
200 mm heigh each, and photos were taken of the respective surfaces of
the furnace square section. Furthermore, the whole sample material taken
from each layer had been subdivided in 17 lots as to wall-, middle- and
centre zone and was examined. This further examination consisted of the
completion of petrographic briquette-cuts as well as of a chemical
examination with regard to oxygen separation in the different zones and
layers. Unfortunately, the various, explaining photos about the petro-
graphic and photographic considerations cannot be shown in this report.
On the other hand fig. 5 shows the progressive reduction of the iron-
oxygen-compositions in the briquettes during their throughput in the
furnace.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 F~~IeAo o ouItaAo DP80-00926A0079ta001400Q111t
Mist thApgBy
g - gave an essen is co ibution
by their various considerations and almost agreeing results to the
scientific investigation of the metallurgical processes in the low shaft
furnace.
In consideration of the burden basis and raw material analyses of the
charged material as well as of the analyses and amounts of the output
material, detailed balances for iron, manganese, sulphur, carbon, blast,
gas, dust and tar were made, the results of which are summarized in the
total balance of material as to fig. 6. The fuel consumption on the
average of 2191 kg coal/t pig iron, obtained from the balances, can be
considered as a favourable one, since on one hand it includes the amount
of fuel required in the conventional blast furnace process for sintering
and coking, and, moreover, a further reduction of fuel consumption can
be expected for a big plant as there is the possibility to use a higher
blast temperature at a simultaneous reduction of the losses through the
wall. Consequently, the fuel consumption for the only pig iron production
in a big low shaft furnace - with the raw material at disposal and the
pig iron required -will amount to about 1000 kg/t pig iron, with relation
to coke of 85% C.
As finally has been proved by the agreement of the balances with the
equilibrium examinations, based on the meditations of Th. Kootz and
W. Oelsen, the D H N - process seems to be of good influence on the
conditions of equilibrium, with a good manganese reduction and favourable
desulphurization, at a some higher degree of saturation of pig iron for
carbon as a consequence of the porosity, caused by the carbonization,
as well as of the almost general equal distribution of the fine grains
of ore, fluxes and fuel with their bigger surface.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 pamagnpow,
WITH SPECIAL REFERENCE r TO DEVEL' NTS IN
GERM&NY
By: K Meyer of Frankfurt am Main
Sintering and pelletizing, being processes for the lumpification of
fine-grained iron ores, differ as to the grain sizes of raw materials
used, the kind of fuel required, and the appearance of finished products.
(I) REASONS FOR, AND IMPORTANCE OF2 PELLETIZING
Since the beginning of this century, fine-grained iron ores have been given
a lumpy shape suitable for blast furnace treatment. There is only a decade
between the beginnings of sintering and pelletizing as lumpifying methods.
While sintering took but a few years to attain vital importance,
approximately three decades elapsed before Swedish and American engineers
resumed the works of Andersen (1912) and Brackelsberg (1913) and developed
pelletizing for commercial use. This difference in time is chiefly due to
the materials being far better suited for sintering than for pelletizing.
Along with the ever increasing quantities of very fine-grained ores
appearing in the market, the prospects for pelletizing have been improving,
the more so as the greatest part of these ores contain a high percentage
of iron and their simultaneous treatment in a sinter plant greatly lessens
its capacity.
Pelletizing will develop most favorably in countries where large quantities
of fine-grained ores are available, principally Sweden and the USA. For
practically all the other industrial countries, the importance of sintering
remains undiminished. This is especially true for Germany, as in this
country the blast furnaces are fed with ores of the most different origin.
(II) RAW MATERIALS AND SITES SUITABLE FOR PELLETIZING AND SINTERING PLANTS
The fields of application for sintering and pelletizing, respectively, are
limited by the absolutely different fine-ore screen analyses required for
each of the two processes. This is the reason why the two methods are by
no means competitive but supplementary.
While fine ores up to 15 mm are suitable for sintering, the maximum grain
size for pelletizing is 0,2 mm. TABLE 1 shows typical screen analyses of
fine ores. The limits between the two grain sizes are very sharply drawn.
If it were intended to prepare ores of coarser grain size for pelletizing,
it would become necessary to give them first the desired degree of fineness.
This, however, would incur considerable costs for grinding. On the other
hand, the advantages that may be expected when smelting pellets in blast
furnaces, are not yet great enough to justify such additional costs. It
is not usual for smelters to grind raw materials (except ground basic slag
and cement). While, in general coarser fine ores will accumulate at
screening and chrushing plants in the vicinity of blast furnaces, fine-
grained ores will mostly be found in dressing plants. This will result in
a fixation of the sites for sintering and pelletizing plants, respectively:
the former at the smelter, the latter in the immediate vicinity of dressing
plants. These locations are desirable for other reasons as well. For
instance, coke breeze and blast furnace gas, the fuels required for
sintering plants, will best be obtained at smelters.
Approved For Release 2 /- : ' WpffWO07900040001-2
 -~ Approved For Release 2001/09/07 : CIA-RDP80-00926A007900040001-2
The low mechanical strength of sinter calls for short distances between
the places of production and consumption.
Matters are different for the location of pelletizing plants. The required
fuel, chiefly oil, is independent from the smelter. Transport of dry,
dusty material means a nuisance alongside the traveling route and entails
dust losses. Wetted material, however is sensitive to frost, particularly
when the respective dressing plant is located in the extreme North. Hard-
burnt pellets have a high mechanical strength, and, besides, are easily
transportable, from which it results that their production absolutely
belongs to dressing plants. H B Wendeborn has especially outlined these
special conditions.
III PELLETIZING PROCESS
On having the sufficiently fine-grained stage, two further steps are
necessary before a pellet may become a blast furnace charge. The next step
is to give the green pellets the required diameter. They are to be
thermically hardened. For the production of wet balls (green pellets),
several factors are of importance, as there are surface tension of the
wetting agent, wettability of the ore, its swelling property, grain size,
and grain surface, as pointed out by M Tigerschiold, H F Reich and the
author. Therefore, micaceous ores and ores that had been annealed to about
100000 once before (pyrite cinders) will not easily form balls, whereas
raw materials of plastic properties will. The existence of the green
pellets is only an intermediate stage in the process of manufacturing
finished pellets. For further treatment in shaft furnaces or on sinter
strands they need a certain minimum of strength, which will be 2 kg (~ 3)
chrushing strength for shaft furnaces and 1 kg for sinter strands. The
higher strength can only be attained by additives or by a more thorough
fine-grinding, which means higher operating costs.
For the production of green pellets, two basic ways are possible, viz.
(a) Forming the balls from smallest particles on the snow-balling principle
in drums or on discs;
(b) Rough pressing of cylinders and rerolling them in drums
For the snow-balling principle, cylindrical drums are mostly used in Sweden
and the USA; on their dimensions and effectiveness M Tigerschiold has
already reported in detail. According to this report, the length of the
drum is to be 2 to 3 times its diameter, its slop about 2 to 5%.
Production will fluctuate, depending on the size of the pellets to be
produced. Variation of rotation speed will change the diameter of the
pellets. To ensure good functioning it is important that the inside of the
drum shell has a certain gripping capacity, which will best be produced by
a layer of the wet ore itself. Excessive growth of this layer will be
prevented by a scraper. An advantage of the drum lies in its relatively
high production capacity, whereas it is a drawback that the pellets, that
are at the same time leaving the drum, are however, not of a uniform
diameter. For this reason, it is imperative to have a screen follow the
drum, by means of which the required grain size will be separated for
further treatment, whilst the undersize will be returned to the drum,
sometimes up to 400% of the quantity produced.
It has first of all been in Germany where the granulating disc has chiefly
been used for the production of green pellets. Compared to the drum, it
shows some substantial advantages, such as lower weight as well as grea?~qr
variation with respect to rotation speed, slope and height of disc wall.
Approved For Release 20MJQfiW : CIA-RDP80-00926AO07900040001-2
Mon W.-
 It is thepddi c that will as2001/09/07 ba1 s R disch09 n Othem over O the disc
~ gi g
wall within narrow grain-size limits. This is a function of the motion of
the material. FIGURES 1 AND 2 show the path of motion in a drum and in a
disc, respectively. FIGURE 3 represents the front view of a granulating
disc, namely a special variant. A channel is fastened to the outside
border of the disc. Balls coming from the disc will first roll into this
channel, where they can be covered with another material, which is another
essestial advantage, as compared to the drum. The sizes of the green
pellets vary according to slope and rotation speed. FIGURE 4 shows pellets
produced on discs.
For iron ores the productive capacity will fluctuate between 3 and 25 tons
per sq. mo per 24 hours. Discs of 2, 3, 4, 4.5, a and 5.5 m diameter are
built. Capacity, however, does not increase proportionally to the enlarged
surface of the disc. The productive capacity per square metre of a 5-metre
disc will only reach 70% of that of a 2 -metre disc. This results in setting
a limit to the enlargement of the surface. FIGURE 5 is a section of a
5 -metre disc in operation for the pelletizing of flotation cinder;FIGURE 6
shows the removal of green pellets. This disc was developed by Lure.., in
collaboration with Maschinenfabrik B Beumer. Several discs are now in
operation. Experiments for an improvement of efficiency of the disc surface
were various, intensive development is still going on. It seems untimely
however, to publish promising partial results at this early date.' Up to
the present, practically all of the fine ores suitable for pelletizing could
be formed on discs.
When looking for machines offering both a high throughput capacity and a
fairly uniform size of pellets, the vacuum press (known from ceramic
industry) was also minutely tested for a longer period. Reference is made
to the article by A Stirling and to that by G Sengfelder. The results of
these tests are approximately as follows: only fine ores of suffiently
plastic properties can be formed on presses; the porosity required for
reaching a high degree of oxidation during the hardening process is lower
than is the case with balls formed on discs; this results in an increase in
bivalent Fe content and in a decrease in strength; the vacuum press is much
more sensitive towards fluctuations in moisture content than are the drum
and the disc; moreover, it is extremely sensitive to hard foreign particles.
The last step in the production of finished pellets is the hardening
process applied to attain the necessary mechanical strength. This is done
by cautious heating up to the point where sintering is beginning. Details
on hardening possibilities have been given by M Tigerschiold.
Until lately, pellets were exclusively produced in shaft furnaces which are
round for smaller capacities and rectangular for larger ones. The
dimensions commonly used in the USA are 7 ft. 14 ft. 34 ft. in Sweden 6 ft.
10 ft. section.
The green pellets are charged into the top and discharged at the bottom.
Heat supply is done by strongly heated combustion gases. Up to now, the
capacities reached were about 350 to 400 tons per shaft furnace that is
often composed of several of the above units. Recently, however, mention
was made of remarkable increase in capacities reached by an increase in
blowing rates, stepping up to 800 - 900 tons in American furnaces. More
details on shaft furnaces have been given by M Tigerschiold.
Shaft furnaces have proved advantageous for pelletizing magnetites, because
a considerable part of the required heat is supplied by oxidation to
hematite. But shaft furnaces are badly or not at all, suited for hematites.
Approved For Release -~ /07 : CIA-RD~PW-00626-A,007 0040001-2
 /07 IA-RDP 0-00 6400790 040001; 2
rwu- Annrfly
or such o
u ~n a of empera ure a weep t eir sintering
and melting points. There are other disadvantages, too, such as the
difficulty of maintaining a uniform gas passage within the furnace, and
the danger of disintegration or agglomeration of the charge. It will,
therefore, be understood that tests have also been made for the purpose
of using the sinter strand for the hardening of pellets. Developing tests
were simultaneously being made in the USA and in Germany. Up to now, one
largescale plant for a capacity of 1000 tons per 24 hours was built by the
firm of McKee at Babbit. Lurgi has been carrying through such tests since
1949 and is now erecting a 1100-ton plant. All results were at first
established in a laboratory plant and then controlled in a pilot plant.
All these tests have been guided by the desire to pelletize practically
all materials of finest grain size on the sinter machine, making it a
principle to exclude all additives of salts or bentonite, as are needed
for shaft furnaces. Neither has the material been mixed with solid fuel.
The test results gained up to now clearly show that it is possible to
produce pellets up to 30 mm (1-1/4").
Special arrangements in the design of the burning machine as well as the
feeding and at the operational control have prevented direct contact of
the hot gases with the side walls and grate bars of the pallets. The use
of suitable alloys means an additional safety factor for trouble-free
operation. The application of the sinter machine as a pelletizing apparatus
is based on the devise: safety of the machine first. Shockless feeding
of the pellets onto the machine is very important, for which purpose special
devices have been created. Drying and burning of the pellets is done by
sucking; cooling is done by pressing.
FIGURE 7 is a scheme of a Lurgi pelletizing plant.
As compared to the shaft furnace, the sinter machine has several remarkable
advantages: It can be built in one unit for capacities up to 1500 tons;
temperature control is easy, its adjustment precise and adaptable to the
properties of the respective ores; the low depth of layer and relative
repose of the pellets do not require such high initial strengths; badly
burnt material can easily be screened off and recirculated to the process;
heat losses can be reduced to a minimum by circulating part of the gases;
the range of application is practically unlimited; much less is required
of the range of the grain sizes than is the case with shaft furnaces.
(IV) A PELLETIZING PROCESS FOR PYRITE CINDERS, SIMULTANEOUSLY CLEANING
THEM OF COPPER, LEAD, AND ZINC. - VUOKSENNISCA-IMATRA PROCESS
There is a possibility of carrying through chemical reactions simultaneously
with a thermal treatment of green pellets. This proves important wherever
the value of high-prized raw materials is lessened by impurities. This is
especially true for pyrite cinders poor in phosphorus, but contaminated
by Cu, Pb, and Zn. By the classical process, cleaning the materials of
nonferrous metals is possible where contents are high, but uneconomical
where they are low. In its Works of Imatra, the firm of Vuokesenniska AB
developed a thermal process for the utilization of such cinders. This
process was taken over by Lurgi for sale.
FIGURE 8 is a scheme of this process. It has been made possible to
pelletize and clean flotation pyrite cinders by one step.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 ..&_ Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
TABLE 2 gives some values before and after pelletizing, respectively.
Intensive developmental work has helped to reduce operational costs to
such a degree that this process appears to be absolutely economical. The
plant at Imatra now treats about 130 - 140 tons of flotation pyrite cinders
per day.. It was taken in operation in 1952 already.
(V) APPLICATION OF KNOWLEDGE GAINED AT PELLETIZING TO GENERAL DOWN-DRAUGHT
SINTERING
The question that arises is: do sintering and pelletizing contrast or is
there a chance of usefully applying any of the knowledge gained at the
development of one process to the other process? The latter question
should be answered in the affirmative, as doubtlessly, there are, apart
from the optimum grain distributions for sintering or pelletizing,
respectively, many other grain distributions that will suit neither group.
This will principally be the case whenever different kinds of ore, whose
properties are varying, have to be treated.
With various ores of a high percentage of fines it has proved possible
substantially to increase the efficiency of sinter grates, when the
thoroughly mixed mass-of components was, for a longer time, subjected to
rerolling in a drum. But it is also possible to separate finest ores from
coarser lots, for instance by screening or air sifting, and to sinter the
coarser section effectively, and to pelletize the fines fraction.
At the sintering process the required fuel, that is of a solid form, is
usually mixed with the ore and ignited. At the pelletizing process the
heat supply is done, by the greatest part and sometimes altogether, by
means of heated gases from the outside no matter whether it is question
of shaft furnaces or sinter strands.
Since 1949/50 Lurgi has experimenting on combining both burning techniques
and FIGURE 9 is the respective scheme. An extended ignition furnace covers
part of the sinter grate, so that the hot gases are sucked through the
sinter mixture for a longer time than is necessary for ignition alone.
The result obtained is surprising in many respects:
1. Depending on the effect exercised by the hot gases, the
total fuel requirement will decrease to 75 to 80% of the
requirement of fuel when using coke breeze alone.
2. The degree of oxidation of sinter increases from 90 to 92%
to 95 to 98% and approaches that of the black burnt sinter.
3. The capacity of the sintering machine decreases when prolonging
the effect of the hot gases.
4. The sinter is not more slagged to such a high degree and yet the
quantity of return fines is smaller.
5. Applying this process means a better adaptation of the sintering
to the respective fuel situation. The first sinter plant of this
kind was built by Lurgi in co-operation with the firm of Arbed
at Saarbrucken-Burbach. The sinter grate of an effective suction
area of 50 sq.m. and a daily capacity of 1000 tons has been in
operation since August 1952. In 1954 a further machine was
switched over to mixed firing at the same Works. Since May 1954 a
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
sinter grate working by this process is also in operation at Huttenwerk
Phoenix, Duisburg. Several other Works in France an Italy have decided
on changing over to mixed firing.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 Approved Fof&e 2%% 9/9b~ ,-RDP80-00926A007900040001-2
By Dr.-Ing. D. Fastje of Watenstedt-Salzgitter
The metallurgical principles of the Krupp-Renn process and its operation
have already been reported in "Stahl and Eisen" 69 (1949) pp 319 to 325 so
that only thq more important features of this process need be recapitulated
before the new developments are considered.
The Renn process A s a reduction process for iron ore which uses a revolving
kiln furnace, sud in which the iron oxides are first reduced by means of
the fuel mixed .n with the charge to an iron sponge at a temperature of
between 600?C a rd 1000?C. This sponge is then converted in a higher
temperature zone to larger bodies of low-carbon iron which are.known as
"Nodules" (Lt;py en), caused by the sponge being welded. At the same time
the gangue mip.erals in the ore form a semi-liquid slag in which the nodules
remain embec:dpd. The separation of metal and slag is completed by mechanical
methods type:-axing in conjunction with the kiln. The final products are
gangue-free nodules which contain between 93 and 95% Fe and which will
contain varying quantities of sulphur and phosphorus according to the
analyses of the ore and fuel being utilised., but which on the other hand
contain no manganese or silicon as These elements are not reduced from
their oxides it the relatively low operating temperature of the Krupp-Renn
process. For the same reason the carbon content of the nodular iron
produced is atso relatively low.
Fig. 1 shows vchematically the method of operation of the process. The
ore which according to its reducibility must be crushed to a particle size
of 2 to 20 me diameter, is mixed with finely crushed fuel and fed to the
inlet end of the kiln where it passes song in the opposite direction to
the combust:Lon gases which are generated by a flame located at the outlet
end of the kiln. In the first zone of th kiln, which is known as the
pre-heat zce, a gradual warming-up of the charge takes place and moisture
and water 4# crystallisation are driven off. In the second zone of the
furnace.. Jere the temperature is between 100?C and 1000?C the iron oxides
are reduc(td to iron sponge by the reducing action of the carbon present
in the fuJ'i. The carbon monoxide generated inside the charge by this
reaction #urns in the free space above End t%us supplies enough heat for
reductior+ to proceed continuously. In the final zone of the furnace, in
which th temperature is raised to about 1250 to 1300?C by the help of
a pulverised coal- or ore flame, there occurs 4 welding of the finely
divided 4.ron sponge into nodules due to the comhtned effects of the oxidising
combust`ivn gases and the turning action of the _uln. A heavy and viscous
slag is,formed from the gangue constituents of t1k,: ore, which slag has only
a low 4 on content, and the nodules are carried -% the outlet of the kiln
in this slag. The material issuing from the outl.nt of the kiln is quenched
with water and cooled and is then passed through suitable crushing machinery
in whi,eh the nodules remain unaffected, and they exe then removed from the
finer-ground slag by means of screens and magnetic separators.
One tj'rvantage of the Krupp-Renn process is that it is largely independent
of t}.: quality of the fuel used as a reducing agent: that is to say
prat ically all the so-called "waste" fuels can be used for this purpose
suck] as coke breeze, brown coal semiaeoke (Braunkohlenschwelkoks), fine
ant', :^acite, low-grade coal slack etc. The ash content of the fuel
Approved For Release 2001/09/07 : CIA-RDP80-00926A007900040001-2
 used play s` o9 ec~j l5 ff% 441JA2/017eq~&? *902jgM i7 pff.40~8 t nt
of the fuel must be low as these gases cannot be"sefully employed inside
a Renn kiln. The sulphur content of the fuel is of considerable importance
as a part of this sulphur is transferred to the nodules and the low
operating temperature and the very low basicity of the slag offers no
hindrance to sulphur pick-up by the iron.
The most important metallurgical features of the Renn process are as
follows:
1. The process operates almost exclusively with slags which, in the
range of temperature at which iron exists as a tacky, low-carbon
mass, i.e. iron can form nodules, is sufficiently heavy and viscous
as to be able to hold these nodules in suspension. As a general
rule only slags with a silica content of 55 to 70% possess the
necessary degree of viscosity.
2. The slag volume should not fall below 800 to 1000 kg per ton of
nodules, because otherwise the iron will coagulate into balls to
too great a degree and this will cause difficulties in the operation
of the kiln.
Thus the most important application of the Renn process is already
indicated: It is particularly suited for the treatment of low-grade
highly siliceous ores which are uneconomic to smelt directly by present
day techniques in the blast furnace and which are unsuited for beneficiati.on
or any of the usual ore-preparation processes prior to blast furnace
treatment.
From this there arises the question whether the Renn process should be
considered only as a method of pre-treatment of low grade ores, i.e. as
a first stage in the production of iron in the blast furnace or whether
it can be considered as being already a member of the real iron-producing
processes. This question must be answered in one way or the other according
to circumstances. Nodules with a high phosphorus and sulphur content,
which are produced when a phosphoric ore is reduced with a sulphur-containing
fuel, must be first charged to the blast furnace if an iron is to be
produced which can be converted to steel by the normal known methods. This
is economical particularly in cases where a blast furnace plant is already
in existence. In effect this raises the capacity of the blast furnace since
the production of pig iron from Renn nodules entails no reduction or slag
formation and only 200 kg of blast furnace coke are required per tonne of
pig iron produced.
In other cases, where the quality of the ore and the low sulphur content
of the fuel are such that the production of nodules low in phosphorus and
sulphur is feasible, then these nodules can be processed directly into
steel in the electric arc furnace or in the open hearth furnace, without
the cost increase involved in first producing a pig iron. In this instance
the Renn process is not a pre-treatment method, but is an iron-producing
process, in fact a method of generating a form of "artificial" scrap.
Unlike the ordinary grades of commercial scrap this Renn nodule "scrap"
possesses the advantages of easy chargeability and freedom from the
unwanted "tramp" elements such as Copper, Nickel, Chromium, Molybdenum.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
At the`pzesent time developments are taking place tbich will tend in the
future toake the Renn process more important as an iron-producing
method. The technical development of the are furnace and its increasing
size, together with the increasing efficiency in the generation of electric
power make it likely that the large electric arc furnace will be found to
be the most suitable metallurgical apparatus for the direct conversion of
Renn nodules into steel. The advantages of supremely easy charging and
melting which are obtained when Renn nodules are used may well offset the
fact that owing to the sulphur contained in this material a higher slag
volume than normal must be carried. With the use of the oxygen lance the
electric arc furnace is becoming more able to undertake refining operations,
and thus it is not impossible that ways may also be found to process high-
phosphorus Renn nodules direct to high quality steel in the arc furnace.
These technical developments, which are already well under way, in any
case mean that it is advisable when considering the construction of new
iron and steel plants to take into consideration more and more the
possibilities of the Renn kiln large electric furnace combination
which both technically and economically in the saving of first cost possesses
some definite advantages when compared with the orthodox blast furnace and
open hearth resp. basic converter system..
The industrial development of the Renn process began rather more than 20
years ago with the erection of a large experimental installation at Essen,
Germany, in which up to the outbreak of the second World War low grade
German ores were treated. In the former Japanese areas of Manchuria and
Korea there were 4 Renn kilns installed by the end of the 1930's and 2
such kilns were in operation on the Japanese mainland. During the war all
these plants were developed and extended and new plants were put down in
other localities so that according to information received there were 24
Renn kilns operating under Japanese supervision by the time the war ended.
During the war also one Renn furnace came into operation in Czechoslovakia
and in Germany the big Renn plant at Watenstedt was erected.
Since the end of the war the Czech plant has been expanded to two kilns
and a Renn plant installed in North Spain which will come into operation
in a few months. A plant comprising two kilns has been constructed and
put into operation at Unterwellenborn in the East Zone of Germany.
The plant at Borbeck, and the largest and most important Renn installation
in Germany at Salzgitter-Watenstedt both fell victims to the post-war
dismantling programmes. A few weeks ago, however, the decision was taken
to re-construct the Watenstedt plant in a more up-to-date form and with
increased capacity. A comparison between the arrangement of the kilns
before dismantling and the newly planned system is shown in Fig. 2.
The original plant before dismantling consisted of 3 furnaces of 4.2 metres
diameter and 70 metres length with an annual capacity of about 450,000 tons
of ore. At that time a raw ore containing between 26 to 28% Fe was treated,
the annual production of nodules being between 110,000 and 120,000 tons.
The kiln charge consisted of ore, fuel, an intermediate product arising
from the slag milling plant and flue dust, and this charge was extracted
from bunkers by plate conveyors and weighed out in scale lorrys. The
constituents of the charge were mixed in revolving feeders and the prepared
charge then fed into the inlet end of the kiln by means of a bucket elevator.
This intermittent charging system is to be abandoned in the new project,
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 rimnril q~ roov, ed ,F r Rlease 2001/09/07 : CIA-RDP80,-00926A007900040001-2
A y c~ ~o ispense with the arrangement of weighers, feeders
and bucket elevators which require a great deal of maintenance. In the
new installation the preparation of the charge will be undertaken in a
manner similar to that employed in sinter plants, rotary table feeders and
automatic weigh-feed belts being employed, while the prepared charge will
be taken to the mouth of the kiln by means of belt conveyors. The kilns
themselves will be enlarged, 2 kilns of 4.2 metres diameter and 95 metres
length and one kiln 4.6 metres diameter and 110 metres length being
installed. These three now furnaces will together be able to treat some
600,000 tons of ore per annum.
It is to be noted that particular consideration has been given to the
question of the type of ore which will be processed in the new Renn plant.
A wet concentration plant has been erected which can produce a concentrate
from the ores of the Salzgitter area. The local lean, siliceous ores
contain between 25 to 28% Fe and 22 to 27% Si02, and as a result of
technical improvements in the concentration process and the introduction
of new methods the concentrate contains 39% Fe and only 14% Si02. In
practice however this concentrate has to be diluted with "middlings"
containing only 30 to 38% Fe, but it is possible that in the future these
"middlings" may be treated directly in the Renn plant. This will have a
double advantage, on the one hand owing to the increased iron content of
the charge the net output of Renn nodules will be increased to about
180,000 to 200,000 tons p.a., which will raise the economic efficiency
of this process, while on the other hand the wet concentrators attached
to sinter plants and blast furnaces in this area will be able to produce
a concentrate containing more iron, i.e. with a Fe content of 40.5 to
41%, with a correspondingly lower silica content. In this manner the new
Renn plant will become an important factor in the entire economy of the
Salzgitter orefield. It is not unlikely for example that much of the
fine ore now produced here in the ore crushing and screening plants, and
which at present must be either sintered or pelletized before charging
to the blast furnace may in future be processed into Renn nodules before
blast furnace treatment. By the choice of this system there is also the
possibility that at the same time the dust and the mud from dust-catching
plants, which are not altogether very suitable for sinter plants, may be
treated and their lead and zinc contents recovered from the flue gases
issuing from the Renn kiln,
The Renn process should also be tested for the treatment of the American
taconite ores, which in the future must be further exploited in order to
make up for the depletion of reserves in the Lake Superior deposits.
These are very low-grade ores with a high silica content and they can
only be made suitable for blast furnace treatment after a costly operation
of fine milling, often with a poor recovery. Beneficiation is only
possible in a taconite in which the iron is combined as magnetite, which
can therefore be enriched by magnetic separation. In the case of the
non-magnetic taconites, in which the iron is held as carbonate, haematite
or in a silicate complex, no suitable pretreatment process has yet been
devised.
Some years ago the Renn process was tried in America on a poor ore containing
27 to 32% Fe. In the processing of a taconite, however, there is one grave
disadvantage to be contended with, namely that because of the very high
content of silica it is necessary to add substantial quantities of lime
or alumina to the charge in order to produce a wor1able Renn slag. This
Approved For Release 2001/09/07: CIA-IJP80-00926A007900040001-2
 ? entails a p erloatpivegy "large volume 90 7 slalg R P.8 -OAn 079q p( OQ OOCloW
iron content can account for a loss of 3 to 5% in recovery. In addition
the capacity of the Renn kiln is reduced owing to the large volume of
slag carried.
A higher degree of efficiency due to the lower slag bulk and greater
throughput may be obtained in the case of processing Salzgitter ores, if
instead of charging the raw ore the concentrates and some of the middlings
from a magnetic or similar beneficiation plant be charged to the Renn
kiln. It is for example thus possible that in the case of a magnetic
taconite the fine grinding should not be taken so far as normal, but in-
stead a relatively coarse-mesh product should be sufficient and therefore
along with the concentrates and tailings a lower-grade "middlings" could
be recovered with an iron content of perhaps 35%.Fe. This intermediate
product might then be processed into iron nodules in a Renn kiln. This
would reduce the total crushing costs and it may also be expected to
improve the overall iron recovery.
In non-magnetic taconite deposits there are bodies of ore which possess
a certain magnetite content and which can be enriched by magnetic
separation, but not sufficiently for blast furnace smelting. If, however,
the Renn furnace is proposed instead of the blast furnace then an enrich-
ment to 35 to 40% Fe suffices, which is possible in a number of cases.
The wholly non-magnetic taconites, which are today untreatable can probably
be made suitable for Renn furnace smelting if they are mixed with concen-
trates or high-grade middlings from a magnetic taconite concentrator before
being charged to the kiln. Provided this mixture contains about 40% Fe it
should be smeltable. A thorough investigation of these possibilities is
clearly called for in order to understand the fundamental technicalities.
To assess the economics of the Renn process the following data may be used
as a general guide
Under present-day German conditions the first cost of the construction of
a Renn plant for the annual production of 400,000 tons of nodules from
about 1.2 million tons of ore with 35% Fe content amounts to about 90 million
Deutschriarks (8 million F). This plant would consist of 5 kilns of 4.6
metres diameter and 110 metres length. Allowing for depreciation and
interest together at the rate of 15% this amounts to a capital charge of
roughly 34 124 (F 3) per tonne of nodules produced.
Taking the price of raw ore, "middlings" ore concentrates at 12 to per
ton (about 22/ -- per ton) then the ore cost works out at about 36 DK
(F 3.3) per ton of nodules. A consumption of "waste" fuel at the rate
of 1000 kg per ton of nodules (20 cwts per ton) is to be expected, at
a price of about 40 EM (F 3.7) per ton.
The operating costs, i.e. power, maintenance, wages etc. will run at the
rate of about 30 EM (F 2.75) per ton of nodules.
The total costs therefore for the production of Renn nodules from a low-
grade acid ore in a large modern plant will amount to about 140 TAI per
ton, i.e. about F 13 per ton.
_410010P.-
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 ~O
TFiis costAl5jugwred tMaasd'2MM//d: 6L? P iQ9iZ O 79Q0811Q001-2
potentialities of the Renn process under various conditions and in
different places, particularly in those cases where a further processing
of the nodules in the large electric are furnace can be envisaged. If
the conversion cost in such a furnace be assessed at 80 DM (F 7.3) per
ton of ingots produced, then the production of electric steel at a cost
of 220 DM per ton of ingots (F 21 per ton) becomes a practical possibility,
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 Approo 2M/ R. , Rh-?8R26AO07900040001-2
$y Dr.-Ing. F. Wesemann of Dusseldorf
I should like to report on investigations on the testing of the mechanical
strength of metallurgical coke conducted recently at a number of blast-
furnace plants. The mechanical strength of coke is measured by the drum
test developed some thirty years ago by Simmersbach. Before the last war,
a German standard (DIN) specification was laid down for this test. The
method of evaluation in this specification, however, underwent several
modifications of the so-called Ilseder index although these have recently
become controversial.
In the past, random determination of the drum index was usually regarded
as sufficient, but in recent years more and more blast-furnace plants have
changed over to continuous testing of the coke on delivery. Continuous
testing, however, requires mechanisation of the preliminary screening.
Preliminary screening of the coke with a fork with prongs 50 mm. (2. in.)
apart, according to the DIN specification, as well as the screening of the
coke after the test to 40 mm. (1 -9/16 in.) was previously done by hand.
W.-Wolf, at the blast-furnace plant of the Westfalenhutte, recently
developed a mechanical screening method which is now being employed at
several blast-furnace plants. The device consists of a number of screening
boxes stacked one above the other. These boxes 1 sq.m. (1.2 sq. yards)
in area have plates with circular holes ranging from 100 to 10 mm.
(4 inches to 3/8 in.) in diameters The stack stands on an electrically
driven jolting table which is vibrated at a frequency of 1000 cycles per
min. and an amplitude of 3.67 mn. (0.144 inch). The electric drive is
time-controlled and switches itself off after 40 seconds. A great number
of tests showed that for a given coke the percentage of the various
fractions is the same with mechanical screening as when it was done by
hand. When using mechanical screening in combination with the drum test
the instructions in DIN 51,712 were followed with the sole exception that
the fraction larger than 60 mm. (2-3/8 in.j was used instead of that
obtained with the hand fork with prongs 50 mm. (2 in.) apart.
The fact that mechanical screening reduces the total time needed for the
drum test and permits the screen analysis of metallurgical coke to be
made without additional manual labour has led to remarkable progress in
determining the indices characterizing the mechanical strength of coke
(Fig. 1).
The screen analysis of coke as found by mechanical screening before drum-
testing is given on the left in Fig. 1 where all grades above 60 mm. are
designated "D". On the right are given the letters denoting the size ranges
used to represent the strength indices of the drum-tested coke, viz.
A = percentage above 60 mm.
B = percentage from 40 to 60 mm.
C = total percentage below 40 mm., and
L = percentage below 10 mm.
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 -RDP Giese letters Y S9 u?sxeadetoa cnar0acter/~ze the mecna Q0 a7.6 s?07902R4QQOOjhe
upper and lower part of Fig. 2. The former strength index in the upper
part of the diagram is expressed by A plus B, i.e. it represents the
total quantity of the sizes above 60 mm. In the next line, the original
Ilseder index A minus C, i.e. sizes above 60 mm. less the amount found
below 40 mm., and finally in line 3 the new Ilseder index which consists
of the size A above 60 mm. after drum-testing, divided by D the percentage
above 60 mm. before drum-testing; from the quotient is subtracted the
amount C below 40 mm. obtained after testing. All three indices, however,
do not take into account the fractions below 10 mm., which should not be
omitted since this is an important factor in blast-furnace operation.
Furthermore, the mathematical expression of the new Ilseder index is in
many respects not satisfactory.
A suitable characteristic for the strength of coke should contain three
main components, viz.
(1) D = percentage above 60 mm. before the drum test
(2) A = percentage above 60 mm. after the drum test, and
(3) L = percentage below 10 mm. after the drum test.
The three components D, A, and L represent a selection and summation of
the original values obtained in mechanical screening and drum-testing.
It is obvious that if all the actual values obtained in the whole test,
i.e. all the screenings rising by 10 mm. stages were incorporated in a
formula, better information could be obtained than if only three factors
are employed, but graphical or numerical evaluation of such a formula
would be too elaborate for practical use. The new index, k, consists
therefore of only the three components A, D, and L, in the form shown
in the lower part of Fig. 2.
A. . D
100 , L
The product (A . D)/100 representing the percentages of the grades above
60 mm. before and after drum-testing, should be large, whereas L, the
grade below 10 mm. after testing, should be as low as possible. As L
is in the denominator, the value of k increases with decreasing L.
(A . D)/100 is plotted against L in a coordinate system; i.e. k is
represented by a point in the diagram. The value of the index k can
be read directly from the diagram without having to use a slide rule.
A number of tests, e.g. taken over a month, gives a "cluster" of points;
two of these "clusters" for different cokes are shown in Fig. 3. It
can clearly be seen from the diagram that the two clusters are well
separated and experience has shown that this type of diagram discriminates
sharply between the strength properties of different cokes.
Fig. 4 shows the results obtained for five types of coke, but for the sake
of clarity only the cluster boundaries are plotted. It must, however, be
stressed that the position of the cluster in the diagram does not in any
way appraise the coke or its suitability for use in the blast furnace.
The particular position of the cluster in the diagram is not governed
solely by the mechanical strength of the coke but mainly by the screened
fraction of the coke as-delivered. If, however, a record is kept over a
long period (e.g. a month) of the test results of the coke from one
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 s`fzpplior Ac ygdcFo clus~aO2% 9 IPZg: IA- RDP80g-OOqualQPy9and40 1 rmity
of the coke can be drawn. In these long-period records, not only the
position but also the area of the cluster are regarded as the criteria
for quality. Evaluation of a large number of test results has shown
that changes in the coking practice of one and the same supplier, such
as variation in composition and grainsize of the charge, coking time or
temperature, are clearly reflected by the displacement of the point in
the diagram.
Further research is, however, necessary in order to establish in what way
blast-furnace operation is affected by movement of the points and by
changes in the area of a cluster in the diagram, or, to put it another
way, in which part of the diagram must the cluster representing a certain
coke lie in order to be regarded as good material for the blast furnace.
As my time is limited, further advantages offered by this method of testing
cannot be discussed. I will only mention the correlation between the new
coke index k and the results of the shatter test or Wolf's abrasion index;
further, the influence of the sampling place and of transport and loading
and unloading on the index k as well as the brittleness of the coke, which
will be taken into account by modifying the index k.
The Coke Properties Sub-Committee of the Committee on Coke Production
cooperates closely with the Steinkohlenbergbauverein (Coal Mining Companies
Association) and we trust that this close cooperation will be beneficial
to both parties in their effort to improve the quality of metallurgical
coke which in turn will ease the operation of the blast furnace.
U
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2
 Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2 25X1A
Approved For Release 2001/09/07 : CIA-RDP80-00926AO07900040001-2