Reaction Zones in the Iron O r e

Sintering Process
by R. D. Burlingame, Gust Bitsianes, a n d T. L. Joseph

ESPITE almost fifty years of commercial pracD

tice, the sintering of iron ore has received little
fundamental study. Much of the theoretical work1-'
has dealt with the constitution of sinter produced
under widely varying conditions. While these studies
have broadened our knowledge of the changes that
occur in the sintering zone and in the freshly formed
sinter during the early stages of cooling, they provide little insight into the changes that precede the
formation of sinter. These preliminary changes
merit study as a part of the overall process.
Hessle.' working with beds of Swedish magnetite
concentrates, was one of the first investigators to
study the sintering process in its entirety. On the
basis of temperatures observed at various levels of
the bed during sintering, he postulated a number of
distinct reaction zones to account for the chemical
changes leading to the formation of sinter.
A more direct method of attack is that of arresting the sintering zone after it has progressed part
way through the bed. A study of a vertical cross
section through such a quenched bed provides direct
information on the changes taking place at various
levels. This method was used by McBriar et al.' to
show that several well-defined zones of chemical
change existed within beds that were typical of
British sintering practice. The same general method
of attack was developed independently in the present
investigation to study partially sintered beds typical
of American practice.
Experimental Sintering Equipment

The sintering operation was carried out on an

experimental scale with the equipment shown in
Fig. 1. The refractory-walled sintering chamber A
was 11 in. deep and averaged 9 in. in diameter. Air
was introduced through a tapered flow section B,
which contained the orifice C for accurate metering
of the incoming air. This section was located directly above the square ignition housing D, which
in turn rested upon the sintering chamber A. The
bed was ignited with burner E.
R. D. BURLINGAME, Student Associate AIME, and G. BITSIANES,
Member AIME, are Weirton Steel Co. Fellow and Associate Professor, respectively, School of Mines and Metallurgy, University of
Minnesota, Minneapolis. T . L. JOSEPH, Member AIME, is Assistant
Dean. Minnesota School of Mines and Metallurgy, Minnesota Institute of Technology, Minneapolis.
T P 4256C. Manuscript, Dec. 5, 1955. Blast Furnace, Coke Oven,
and Raw Materials Conference, Cincinnati, April 1956.
-----

TRANSACTIONS A l M E

The required suction for the operation was f u r nished by a fan F, which had an air capacity of 500
cfm (stp). Hot exhaust gases from the sintering
chamber were cleaned in the dustcatcher G before
entering the exhaust fan.
In the study of partially sintered beds, it was
essential to find some technique for removing the
entire charge from the sintering pot without disarranging the unsintered bottom portion. This problem was finally solved by sintering the charge in a
removable basket, which snugly fitted the sintering
chamber. This basket was constructed of two thicknesses of window screen and was lined with a 3/16in. layer of asbestos paper. The bottom of the basket
consisted of two thicknesses of wire screen, which
were fastened to the basket wall. For high fuel mixtures, additional insulation was provided by a somewhat thicker layer of asbestos cement.
Preparation of Partially Sintered Mixtures

The moist feed was carefully placed in the sintering basket, to prevent segregation of the particles,
which varied widely in size and composition. A
thermocouple was placed in the center of the basket
with the hot junction halfway down, and the mixture was evenly distributed around it.
During ignition and throughout the sintering of
the upper half of the bed, the hot junction temperature increased very little. When the sintering zone
reached the halfway point, as indicated by the sudden increase in the hot junction temperature, the
charge was quenched. During quenching the suction
was turned off and the orifice was tightly stoppered
to prevent further influx of air. At the same time,
nitrogen was admitted to the sintering chamber
through the orifice tap. As soon as the nitrogen had
displaced the air and products of combustion, the
charge was removed from the sintering pot for immediate dissection.
It is impossible to preserve the exact zone structure of the bed at the instant that combustion is
arrested unless the downward transmission of heat
is also immediately stopped. Fortunately, heat transfer is very slow in beds containing a stationary
fluid, especially if the particle size is small. It follows that the minimum quantity of nitrogen should
be used to displace the air and that static conditions
be established as soon as possible. A very steep
temperature gradient across the combustion zone
for some time after the quench was evidence of inJULY 1956, JOURNAL O F METALS-853

This page o f Metals Transactzons AIME follows p. S O O A . The inter?ening non-Transactions pages appeared i n the Journal of Metals.

significant heat transfer. After completion of the

dissection, the bright orange color of the hot sinter
was found to grade sharply into a black zone of
unfused particles of magnetite over a distance of
only l/z in. The zone samples and zone dimensions,
therefore, represent the conditions as they existed
during sintering.
Dissection Procedures
Fig. 2 is a pictorial explanation of the two methods
of dissection. One scheme proceeded according to
sketches 1-2-3-4, while the other one followed the
sequence 1-2A-3A-4A-4. The former method was
the simpler, but the latter procedure also provided
an opportunity to photograph the zone structure.
Sketch 1 of Fig. 2 represents the basket containing the hot, partially sintered charge that has just
been lifted out of the sintering pot. The charge is
inverted and the basket is then cut along the dotted
line, sketch 2. Successive horizontal layers of bed
are removed by alternately cutting the basket and
striking off the samples or layers, sketch 3. Each
horizontal section or sample is placed in a Pyrex
flask, sketch 4.
In the second method of dissection, the asquenched charge is laid on its side and the wire
basket is cut along the dotted line as shown in
sketch 2A. This exposes a 120" sector of the bed
below the combustion zone. Starting at the bottom
of the charge, successive slices of the unsintered bed
are obtained by driving sampling blades into the
material as indicated by the dotted lines of sketch
3A. Each sample is caught in a sampling trough,
sketch 4A, and then transferred to a storage flask,
sketch 4. The unexcavated bed clearly reveals the
zone structure. To avoid oxidation, the hot particles
of magnetite adjacent to the combustion zone are
swept with nitrogen when removed from the bed.
Resin Impregnation and Sectioning
Samples from various levels of the unsintered
portion of the charge revealed the changes occurring
ahead of the combustion zone, but the conversion
of the loose granular material into sinter in the very
combustion 'One
not be studied by
this technique alone. In order to prepare polished
specimens from the sintering zone for microscopy,
the granular bottom portion had to be bound coherently to the sinter cake. This was accomplished by

Fig. 1-Experimental sintering machine: A, sintering chamber;

B, flow section; C, orifice for metering incoming air; D,
ignition housing; E, burner; F, fan; and G, dustcatcher.
854-JOURNAL

O F METALS, JULY 1956

Fig. 2-Dissection

procedures for partially sintered charges.

Fig. 3-Sampling
procedures for resin-impregnated charges:
1) Bakelite-filled charge sawed vertically, 2 ) center slab of
sketch 1 sawed into columns, 3) reimpregnated columns
sawed into slices 1/16 in. thick, and 4) 1-in. disks for work
a t higher magnifications.

impregnating the entire charge with hard-setting

resins, by a specialized procedure.
After a partially sintered charge had been cooled
and dried in nitrogen, it was transferred to a sheet
metal impregnating vessel of slightly larger dimensions. A hose connected the bottom of this vessel to
a reservoir filled with a solution of 70 pct Bakelite
and 30 pct acetone by weight. By slowly raising the
reservoir, the liquid thoroughly filled the charge of
sinter by upward percolation. After a period of
soaking, the excess solution was drained off. A tough
resin bond was produced by curing the charge at
150F for a t least 8 hr. However, reimpregnation
was always necessary before sufficient strength was
developed for good sectioning.
Sectioning was performed on a large diamond
saw. Sketch 1 of Fig. 3 shows the Bakelite-filled
charge after being sawed into three vertical sections. The porous sinter is on top and the impregnated granular or unsintered portion of the charge
is on the bottom. The center slab was in turn sectioned into short columns, which contain the combustion zone, sketch 2, Fig. 3.
TRANSACTIONS A l M E

--

- -

Table I . Screen Analyses of Raw Materials

T y p i c a l Sinter

W t Pct of Size Fraction

Size
Fraction

Ore
Fines

Sinter
Returns

Flue
Dust

Roll
Scale

Coke
Breeze

Mixture.* Pct

Dry
Stale

After
W e t Miring

Inch

Mesh

* Ore

fines, 44.8 pct; flue dust, 20.0; sinter returns, 20.0; roll scale, 4.0; coke breeze, 1.7; water, 9.5 ( w e t basis). Total carbon, 5

At this point, the short columns were impregnated

once more, using methyl methacrylate monomer.
The extremely tough lucite matrix provided by this
technique was essential in the preparation of the
specimens for final sectioning and polishing. The
columns could then be sectioned into slices only
1/16 in. thick, as in sketch 3. After a light grinding,
these slices were ready for macrographic study. For
work a t higher magnifications, 1-in. disks containing the sintering zone were cut from these slices and
mounted in conventional briquettes for ease in
polishing, sketch 4, Fig. 3.
Raw Materials

Raw materials were used in proportions typical

of American sintering practice. Charges were proportioned on a wet basis as follows: Labrador ore
fines, 46.5 to 43.3 wt pct; flue dust, 20.0; sinter returns, 20.0; roll scale, 4.0; coke breeze, 0.0 to 3.2;
moisture, 9.5. The screen analyses and chemical
compositions of each component and of a typical
charge
in Tables I and 11.
- are given
Microscopic examination of the Labrador ore fines
showed that the mineralogical composition varied
with particle size. The coarser particles were either
limonitic, hematitic, or banded combinations of these
minerals. Because the finer limonitic particles
coated the coarse pieces, the ore appeared to be
entirely limonitic.

pct.

The typical mixture balled readily when moistened and subjected to a mixing and rolling action.
A comparison of the size distribution in the dry
and in the wet state is shown in Table I. Pelletization occurred by the formation of thick coatings of
fine particles around the coarse pieces. Since the
major proportion of the fines originated in the ore
and in the flue dust, the external color changes in
the sintering bed were the result of chemical changes
in these fines.
Presentation of Results

Vertical sections of partially sintered beds revealed a number of definite zones or layers in which
various physical and chemical changes occurred
during the sintering operation. Fig. 4 shows dissection of a partially sintered bed according to the
second technique shown in Fig. 2. Although it is not
readily discernible in the picture, a very narrow
band, called the zone of combustion and sinter formation, is recognized as the region of transition
between the coherent sinter cake and the under-

Table I I . Chemical Analyses of Raw Materials

W t Pct of Conslituent

Constituent

Ore
Fines

Flue
Dust

Sinter
Returns

Roll

Scale

Typical
Slnler
Coke Mixture,*
Breeze D r y Slate

Total
Fe++
SiOz
A1?03
CaO
MgO
P
S
Mn

C
Volatile matter
Ash
Combined H.0
Combined C 0 2

* Ore fines, 44.8 pct; flue dust, 20.0; sinter returns, 20.0; roll scale,
4.0; coke breeze, 1.7; water, 9.5 ( w e t basis). Total carbon, 5 pct.
Fig. 4-Zone

TRANSACTIONS A l M E

structure o f partially sintered bed.

JULY 1956, JOURNAL OF METALS-855

Fig. 5-Variations
in chemical analysis in relation to zone
structure, low fuel mixture.

lying, granular portion of the bed. The first distinct

region of unsintered material is the blue-black zone
of calcination and reduction. Directly beneath it is
the deep red zone of dehydration. In sharp contrast
to this is the light yellow dry zone. Finally, the
dark brown wet zone forms the bottom layer of the
bed. The white spot a t the center of the charge is
the thermocouple sheath, and the white areas enclosing the charge are sections of the asbestos lining
of the sintering basket.
A series of charges of varying composition, particularly with respect to the fuel, were sintered at

Fig. 7-Variations
in chemical analysis in relation to zone
structure, high fuel mixture.

various rates of air flow, quenched, and then dissected. Because the zone relationships were qualitatively the same as those just outlined, the pattern
was considered to be general. At this point, it
seemed necessary to correlate the more important
variables of the process with the zone structures.
The most important variables in any sintering
operation are those associated with combustion, the
quantity of fuel in the mixture, and the rate of air
flow through the bed. This investigation was primarily concerned with a study of the effects of
variations in the amount of fuel at a constant rate
of air flow typical of commercial practice. No direct
information has been published on the specific rate
of air flow through industrial beds (cubic feet of
air per minute per square foot of wind-box a r e a ) ;
moreover, estimation is difficult because of the unassessable effect of leakage. The rate of advance of
the sintering zone is, however, a direct function of
the specific rate of air flow. Expressed in this way,
sintering velocities of 1/2 to 3/4 in. per min are typical
of plant practices involving 11 to 13-in. beds and
suctions of 20 to 25 in. water gage. Throughout the
present study, sintering rates of about l/z ipm were
obtained with a pressure drop of about 25 in. water
gage across the 11-in. bed. It was concluded, therefore, that the specific rate of air flow of 55 cu ft per
min per sq ft (stp) was typical of commercial practice. With the air flow held constant at this value,
changes in the sintering process were studied as the
fuel content was increased from 3 to 6 pct C (dry
basis). Using the first method of dissection, Fig. 2,
samples were removed for chemical analysis. Plotting the data as a function of zone location gave a
chemical profile of the partially sintered bed.
Variations in Chemical Analysis Through
Partially Sintered Beds

Fig. &Variations
in chemical analysis in relation to zone
structure, low fuel mixture.
8 5 6 J O U R N A L O F METALS, JULY 1956

Figs. 5 and 6 summarize the data from the low

fuel mixture and Figs. 7 and 8a present the findings
TRANSACTIONS AlME

of the high fuel charge. Variations in moisture,

combined water, combined CO,, and combustible
carbon are shown in Figs. 5 and 7, and changes in
total iron, ferrous iron, sulfur, and phosphorus are
plotted in Figs. 6 and 8a. Zone widths observed from
the color changes are also indicated on the graphs.
The Wet Zone-As the hot gases from the combustion zone flow through the unsintered portion of the
bed, they are cooled in supporting endothermic reactions and in providing the charge with sensible
heat. At the level of the wet zone, the gases are
effective only in the removal of moisture from the
charge. Figs. 5 and 7 show that the percentage of
moisture varied from zero at the junction of the wet
and dry zones to almost 1 2 pct a t the grate level for
both low carbon and high carbon mixtures. The
moisture values are somewhat lower than those
existing a t the instant combustion was stopped, because of unavoidable losses during quenching and
dissection.
Because the sinter mixture contained 9.5 pct moisture, i t is apparent that some condensation occurred
a t the very bottom of the charge. In the 3 pct C
charge, the gases became saturated with water after
flowing only one fourth of the way through the wet
zone. The dew point was not reached in the bed
containing 6 pct C, however, until the gas stream
had traversed about three fourths of the wet zone.
Since the rate of air flow was the same in both
cases. this difference was attributed to the much
higher temperature of the gas stream in the high
fuel charge.
The effects of condensation can be rather serious.
If the moisture is high enough to convert the pelletized structure of the feed into a sludge, the permeability of the bed will be sharply decreased.
While it is essential to add enough water to pelletize
the feed satisfactorily, the moisture content should
be minimized to prevent excessive condensation in
the wet zone. Sludge formation did not occur in the
charges investigated, although aggregates in the bottom layer of the bed were covered with a water film.

Fig. 80-(Original)
Variations in chemical analysis in rela
tion to zone structure, high fuel mixture.
TRANSACTIONS A l M E

DISTANCE FROM BOTTOM OF BED- INCHES

Fig. B b ( M o d i f i e d ) Variations in iron analyses with zone

structure, high fuel mixture.

The Dry Zone-As shown by Figs. 5 to 8 a inclusive,

little chemical change takes place within the dry
zone, for the materials are not hot enough to initiate
the chemical reactions that characterize the overlying zones. This region may be regarded as a zone
of preparation, in which the charge is preheated to
reaction temperature.
Very little difference was noted between the state
of aggregation of the particles in the wet zone and
in the dry zone. Heat and chemical effects may cause
pellet disintegration above the wet zone. However,
the materials used in this study formed tough aggregates, which suffered very little breakdown in the
lower zones of the bed. The small amount of dusting observed may have accompanied the evolution
of water vapor during drying. During the sintering
of some charges, the drop in permeability may be
due partly to excessive breakdown of the pellets as
they are heated.
Zone of Dehydration-After
the charge has been
heated to the temperature range of this zone, the
hydrated iron oxides begin to dissociate thermally.
In both the high fuel and low fuel runs, the elimination of combined water across this zone, Figs. 5
and 7, changes the color from yellow to red.
This zone may also be termed the zone of partial
sulfur elimination, for a t both fuel levels the sulfur
content of this portion of' the charge decreased
markedly. The removal of sulfur from the fuel, the
major source of this element, is usually attributed
to simple oxidation. Since hot water vapor is evolved
within this layer, some sulfur may be steamed out,
as suggested by the work of Powell7 and Thompson."
The increases in other constituents are probably
caused by concentration as a result of the loss of
combined water from this zone.
Zone of Calcination and Reduction-At this level
of the bed, the gas stream is hot enough to calcine
the carbonates, and it also contains a sufficient conJULY 1956, JOURNAL OF M E T A L S 8 5 7

centration of carbon monoxide to reduce a part of

the hematite to magnetite. The curves for combined
COI in Figs. 5 and 7 indicate that decomposition of
the carbonates is practically complete within this
zone for both high fuel and low fuel mixtures. Reduction of hematite to magnetite is shown by the
increase in ferrous iron across this zone, Figs. 6 and
8a. Further evidence of reduction at this level is the
familiar change in color from red to blue-black.
Moreover, the particles of this zone were all strongly
magnetic. The degree of conversion to magnetite
was greater for the high fuel charge than it was for
the low fuel mixture because of the higher concentration of reducing agents. Further changes within
this region are the continued removal of sulfur and
the sharp increase in total iron by concentration.
In agreement with other studies,' the elimination of
sulfur is favored by lower percentages of carbon in
the mixture.
I t was not until the arrival of the zone of calcination and reduction that the aggregates of particles
began to break down. The coatings of fine particles
that were built up around the coarse pieces during
mixing were observed to spa11 away from their
central kernels and to disintegrate. Most of the
coarse pieces were badly cracked. The extent of
breakdown increased rapidly as the particles were
heated to peak temperature. Actually, at the junction of this zone with the overlying sinter cake, the
material was little more than an incoherent powder
in many areas.
Zone of Combustion and Sinter Formation- Figs.
5 and 7 show that the percentage of carbon remains
fairly constant throughout all the zones of unsintered material. The abrupt decrease from the amount
in the charge to a very low value marks the zone
of combustion. Within a very narrow region, a large
quantity of heat is released, and much of it is absorbed by the small, contiguous particles. The particles then fuse, flux, or dissolve in the adjacent
viscous mass. Therefore, this region is also referred
to as the zone of sinter formation. The hot products
of combustion are drawn out of this zone of maximum temperature and alter the lower, unsintered
portion of the bed in accordance with the trends.
Table I l l . Zone Dimensions
Zone Width, I n c h e s
1 . n
-. ~

Fuel
Mixture,
3 Pct C

Region
\Vet zone material above the dew point level
D r v zone
zone of dehydration
Zone of calcrnation and reduction
Total of a l l rntermediate zones

1.25

n 5n

H
irh
-----

Fuel
Mixture,
6Pct C
2.75

n 75

The sulfur is practically burned out at this point

in the low carbon charge, but a substantial amount
still remains in the high fuel mixture. Further reduction of hematite raises the ferrous iron to the
maximum percentages found within the sinter cake,
and increases the total iron by concentration.
Zone of Sinter-The zone of sinter is a broad zone
of the finished product. Chemical change within the
sinter seems to be restricted to a partial reoxidation
of the magnetite by the stream of air in the early
stages of cooling. In the high fuel charge, the ferrous
iron was decreased from a maximum of over 40 pct
to about 26 pct by reoxidation of the upper part of
858-JOURNAL

OF

METALS, J U L Y 1956

the sinter, Fig. 8a. Less reduction occurred in the

low carbon charge and reoxidation decreased the
ferrous iron to a value no greater than that of the
original mixture, Fig. 6. Visual evidence of reoxidation was found in the more reddish cast of the
upper, cooler sinter, and in the abundance of microscopic crystals of hematite around the edges of the
pores in this region. The sinter closer to the combustion zone was blacker and contained few grains
of hematite.
In the high carbon mixture, the final sulfur content was somewhat more than twice that of the
residual sulfur in the low fuel charge. Considering
the overall reduction in weight during sintering, the
horizontal phosphorus curves indicate that only a
very small amount, if any, of this element was lost
in the operation.
A substantial difference was found in the strengths
of the top and bottom halves of the sinter cake. The
upper half was noticeably weaker than the lower
portion, and the difference was greatest for the low
carbon mixture. This weakness may be due to the
relatively poor preheat and low peak temperatures
attained in the top portion of the charge.
Zone Dimensions of the High Fuel and Low Fuel
Mixtures-Because the specific rate of air flow was
the same for both cases, the peak temperature in the
combustion zone was considerably higher in the high
fuel mixture than in the low carbon charge. The
widths of the intermediate zones, therefore, were
greater in the former case than in the latter.
The term intermediate zones refers to all portions
of the bed between the sinter cake and the dew
point level of the wet zone. As the width of the
sinter cake increased, the width of the wet zone decreased by about the same amount. The dimensions
of the intermediate zones, however, remained fairly
constant through most of the run. The widths of
these zones, tabulated for easy comparison, are
shown in Table 111.
The intermediate zones constitute the entire region
of transition from the initial moist charge of 9.5 pct
water to the final sinter cake. Although their total
breadth was twice as great in the 6 pct C mixture
as it was in the 3 pct C charge, the change from the
wet charge to the immature sinter occurs over a
surprisingly short distance.
Sintering Zone-The thin slices sectioned from the
lucite-impregnated bed, as in sketch 3 of Fig. 3,
were suitable for photographing the sintering zone.
Macrographs were obtained from several specimens
and a composite print was made, which reproduced
a wide section of the bed at actual size, Fig. 9. The
granular, unsintered portion of the bed is in sharp

Fig. 9-Micrograph
of zone of combustion and sinter formation. Area reduced approximately 50 pct for reproduction.
TRANSACTIONS A l M E

Fig. 10-Microgrophs of zone of combustion and sinter formation.^,

reduced approximately 50 pct for reproduction.

contrast to the vitreous, cellular appearance of the

sinter cake. The sintering zone is, however, rather
irregular and diffuse. This lack of definition is unavoidably associated with the raw materials used.
Chemically, the system was highly heterogeneous,
and the various components were all present in a
wide range of particle sizes. As a result, local differences in the conditions of heat transfer and in
the fusibility of t h e materials produced a broad,
undulating zone of sintering. It was impossible to
locate a clearly representative sintering front in a
highly magnified field. Typical views of the sintering zone are shown in Figs. IOU, b, c, and d.
Fig. 10a shows the sintering front advancing upon
a cluster of particles ranging in size from fines to
aggregates of medium size. The sinter a t the top is
easily recognized by its gray color and impervious
structure. The tendency of the fines to form shells
around the coarse pieces has produced pellets whose
diameters are about twice those of the kernels. The
plastic sinter has engulfed a large piece of ore, A,
and solution is progressing a t the edges of the particle. Internal cracking is evident.
Fig. lob shows two iarge pieces of ore in varying
stages of ingestion by the igneous solution. The
larger piece of ore, A, has just been contacted by
the fused mass and, because of its size, it is probable
that complete solution would not have occurred had
sintering continued. An outer shell of fines is clearly
discernible. The smaller piece of ore, B, has been
TRANSACTIONS AlME

6, and d; A

and B, ore: c; A, return sinter; B, ore. X10. Area

enveloped by the liquid phase. Reduction to magnetite seems to have occurred over its entire surface
before the sintering zone arrived. The white center
is hematite; the gray peripheral zone, magnetite.
In Fig. 10c, a large piece of return sinter, A, and
an adjacent fragment of ore, B, are being gradually
absorbed by the plastic or semimolten mass. The
coating of fines on the lump of return sinter is the
only means of distinguishing it from the new sinter.
The ore particle suffered extensive cracking during
heating and partial reduction.
A wide variety of particle sizes are shown in Fig.
10d, including two large pieces of ore that dominate
the field. Reduction of the smaller piece of ore, A,
almost entirely to gray magnetite was probably
accelerated by the extensive cracking. In the larger
ore lump, B, the progress of reduction is revealed
by the distinct layer of gray magnetite around the
core of white hematite. The difficulty of converting
large pieces of ore into a coherent mass of sinter is
clearly evident from these photographs.
Ferrous lron Gradient in the lron Ore Sintering Process

The variations noted in iron analyses across the

partially sintered charges were significant and deserve further comment, particularly in the high fuel
mixture. In this respect, Fig. 8 b reveals that the
sinter zone was running as high as 40 pct ferrous
iron and 62 pct total iron. A simple stoichiometric
calculation shows that these analytical results would
JULY 1956, JOURNAL O F METALS--859

Table I V . X-Ray Examination of Reaction Zones in a Partially Sintered lron Ore Bed.
High Fuel Mixture: lron Radiation
Total
Iron
Anal ysls,
Average
Pet

Zone
No.
3

Designation
- -Dehydration

p
p
p
p
p
-

Ferrous
Iron
Analysis.
Average
Pet
12.19

Reduction

56.40

18.48

Hot sinter

62.76

40.39

Cool sinter

62.85

26.53

860-JOURNAL

OF METALS, JULY 1956

Lines Found

--

55.52

require 45 pct of the iron to be in the wiistite state

with the remainder as magnetite. Such a high wustite
content was unexpected and indicated that reducing
conditions may be quite high within iron ore sintering beds, and that the wiistite phase may be important in the overall mechanisms of the process.
Since the results of chemical analyses in themselves are not always reliable in predicting phase
relationships, the presence of the wiistite phase was
recently checked by means of the X-ray diffraction
method. In these experiments portions of the same
samples as had been used for the chemical analyses
were X-rayed by the powder method and the resulting patterns compared with standard iron oxide
patterns for positive identification. The pattern from
the hot sinter zone was also carefully measured,
and the resulting d,, values for the various lines
were checked against the standard data as given in
the literature. The final results for the dehydration,
reduction, and sinter zones are summarized in Table
IV, along with an estimate of their phase content
from visual comparison with phase standards. The
high percentage of wiistite phase in the hot sinter
zone is definitely confirmed and the estimated
amount checks well with the results calculated from
the chemical analyses.
Inspection of the original figure, Fig. 8a, reveals
that the ferrous iron relationship could have been
more judiciously drawn as shown in the modified
figure, Fig. 8 b , accompanying this discussion. A
number of reasons support such a change. For instance, most high carbon charges, when sintered to
completion, yield sintered products composed mainly
of magnetite, and thus running about 20 pct ferrous
iron. This value of ferrous iron could then well
serve as an anchoring point, A, at the extreme top
of the charge in plotting the overall ferrous iron
gradient. Furthermore, the two ferrous iron points
shown in the sinter layer of the original, Fig. 8a,
represent the averuge composition of the top and
bottom halves of the sinter zone, respectively. The
original plot thus shows the average point in the
bottom half of the sinter zone at 40.19 pct ferrous
iron, and as a maxlmum in the ferrous iron relationship. Obviously, this condition is unrealistic.
Points of higher ferrous iron content must exist in
this layer, and it is logical to assume that they are
located at the zone of combustion and sintering
where the strongest reducing conditions are generated. All of these ideas have been used in redrawing the ferrous iron plot as shown in Fig. 8b.
The sketch as now shown indicates that extremely
high ferrous iron coritents may exist at the zone of
combustion and sintering, B-possibly
running over

X-Ray Results
Phase Content,
Pet
60 hematite
40 magnetite
Trace foreign
75 magnetite
25 hematite
Trace foreign
60 magnetite
40 wiistite
Trace foreign
95 magnetite
Trace hematite
Trace wiistite

d,,values checked
for a wiistite and
14 magnetlte lines.

50 pct ferrous iron. More realistically, the phase

relationships under these conditions would dictate
that over 70 pct of the iron be present in the form
of the wiistite phase. These conclusions strengthen
the contention that the wiistite phase plays an important role in the sintering of high fuel charges.
The mechanisms must involve a reduction of much
of the iron to the ferrous state at the sintering interface, with a consequent reoxidation of the sintered
material by the incoming air stream.
Summary

Dissection procedures were developed for immediate and accurate sampling of all the zones in
partially sintered beds. Six zones were clearly
marked by changes in color. From the top to the
bottom of the bed, these zones included: 1 ) the cake
of sinter, 2) the narrow zone of combustion and
sinter formation, 3 ) the zone of calcination and reduction, 4) the zone of dehydration, 5) the dry zone,
and 6 ) the wet zone. With 3 pct and 6 pct C in the
mixture, dry granular materials were converted into
a coherent mass of sinter within distances of 1.5
and 2.75 in., respectively.
The chemical composition changed gradually
across the various zones. The ferrous iron increased
in the zone of calcination and reduction and attained
a maximum in the freshly formed sinter. Further
out in the sinter cake, the percentage of ferrous iron
decreased as a result of air oxidation. The extent of
increase in ferrous iron and the final amount in the
sinter appears to depend upon the amount of fuel
used. A substantial amount of sulfur disappeared
from the zone of dehydration, presumably by reaction of steam with the coke. In accordance with
general practice, sulfur was eliminated more completely in the low fuel charge.
Methods were developed for impregnating and
sectioning partially sintered beds of iron ore. This
permitted a study of the very narrow sintering zone,
which advances as portions of the charge soften,
fuse, and merge into the cake of sinter. Typical
vertical sections extending across the sintering zone
are included to show how the zone advances. Gross
discontinuities were observed in the relatively coarse
mixture investigated. Charges of finer materials
should be studied to relate changes across the sintering zone with changes in raw materials, and in important variables of operation.
Acknowledgments

The investigation described herein was made possible by a generous fellowship program supported
by the Weirton Steel Co. Div., National Steel Corp.
TRANSACTIONS AlME

In particular, the writers gratefully acknowledge

the interest, support, and encouragement of Julius H.
Strassburger.
Vernon Bye, of the Minnesota Mines Experiment
Station, is credited with a portion of the analytical
work.
References
1 G. M . Schwartz: l r o n Ore Slnter. AIME Trans., 1929, vol. 84,
PP: 39-69.
R. Hay and J. McLeod: Some Aspects of Sintering Iron Ores.
J o i ~ r n a lWest Scotland Iron and Steel Institute, 1942-1943, vol. 50,
pt. 6, pp. 55-64.
:'E. Cohen: Radiographic Studies of the Process of Sintering lron
Ores. J o u ~ n a llron and Steel Institute, London, 1953, vol. 175, PP.
160-166.

I R. Wild: The Chemlcal Constitution of Sinters. Journal Iron and

Steel Institute London 1953, vol. 174, pp. 131-135.
.-, B. Hessle: ' ~ e w~ e i a l l u r g i c a lExperience in Sintering. Jernkuntorets Annaler, 1945 vol. 129, no. 8, pp. 383-446.
('E. M. McBriar, W. Johnson, K. W . Andrews, and W. Davirs:
Natolre nf
I r o n s t o n e Sinter.
Journal Iron and Steel Institute. Lon-- -.-..-.....
- - ~
don. 1954 vol. 177. PP. 316-323.
7 A. R.'powell and J. H. Thompson: Study of the Desulphurlzation of Coke by Steam. Coal Mining Inoestigntions Bulletin No. 7,
Carnegie lnstitute of Technology. Pittsburgh, 1923.
' J . H. Thompson: Forms of Sulphur in Steamed Coke and Their
Action in the Blast Furnace. Report of 1nz;cstigations No. 2518, U. S.
Bureau of Mlnes Washington D. C., 1923.
: ' R . A. Elllot: ' s u l p h u r ~ l i k i n a t i o nand the Use of Sulphur as
Fuel i n the Sintering of Iron Ores. Canadian Mining Journal, 1952,
vol. 73. no. 7, pp. 51-56.

p
p
p
p
p
-

Discussion of this paper sent 12 copies1 to AlME by Sept. 1, 1956

will appear in AIME T~ansactlonsVol. 2011, 1957. and in JOURNAL
OF METALS,
October 1957.
-

Technical Note

Metallographic Identification and Crystal Symmetry of Titanium Hydride

by L. D. J a f f e
N previous metallographic work on titanium and
Itinguishing
its alloys, difficulty has been encountered in disspheroidal particles of titanium hydride,
dispersed in a-titanium, from other phases that may
be present, such as p-titanium. This problem is
common in examination of commercial unalloyed
titanium.'
In the course of an investigation of the Ti-H
phase diagram, it was noted that titanium hydride,
dispersed in a matrix of a-titanium, showed strong
optical anisotropy when examined under polarized
light. With sensitive tint illumination, particles of
the hydride, when the stage is rotated, changed from
bright yellow to bright blue in color. Since Ptitanium is cubic, it showed no color change on rotation, remaining dark gray or blue-gray. Although
a-titanium is hexagonal, it showed only a minor
color change, going from pale pinkish-blue to pale
purplish-blue. These colors could be modified somewhat by adjustment of the illumination. They were
observed in specimens mechanically polished, both
without etching and after etching with hydrofluoric
acid-nitric acid mixtures in either water or glycerine. The strong color change of the hydride seems
to offer a simple method for its metallographic
identification.
Optical anisotropy of titanium hydride is inconsistent with its accepted face-centered-cubic structure.' Samples of iodide titanium containing 6 to 40
atomic pct deuterium were prepared for diffraction
studies. After deuterating and slow cooling' the
samples were held 21Y2 hr at 400C, water
quenched, held 61 hr at 100C and quenched, then
held 64% hr at 255OC and quenched. Debye exposures with MoKa X-rays and 1.111A neutrons
were made at room temperature of samples asdeuterated and after the 100" and 255C treatments.
Table I shows the lines observed, in addition to
those attributable to a-titanium. These data show
that the hydride phase was not cubic, but probably
tetragonal. The titanium atoms appeared to form a
body-centered-tetragonal lattice complex, two titanium atoms per cell, with approximate dimensions:
-

------

---

--

p~

L. D. JAFFE, Member AIME, formerly with Watertown Arsenal,

Watertown, Mass., is now Chief, Materials Section, Jet Propulsion
Laboratory, California Institute of Technology, Pasadena, Calif.
T N 333E. Manuscript, Mar. 20, 1956.

--

--.
-

---

TRANSACTIONS A l M E

--

- - --

.
-

--

Table I. Diffraction Data

Relative Integrated Intensity

---

--

Interplanar
Spacing, A

X-Ray*

Neutron*

Lattice Complex
Indexing
BCTt

FCTt

* < indicates lines overlapped by lines of a-titanium, so only

maximum limit to intensity of hydride line could be set.
t BCT and FCT indicate body-centered-tetragonal and face-centered-tetragonal, respectively.
t $ indicates neutron lines not fitting indexing of lattice complex
of titanium atoms, and presumably arising from scattering by
deuterium atoms.
a = 3.12A, c = 4.18A, and c / a =1.34A.
This is
equivalent to a face-centered-tetragonal complex,
four titanium atoms per cell, with approximate dimensions: a = 4.42A, c = 4.18A, and c / a
0.946A.
The presence of the 2.98 and 1.81A neutron lines
indicates that the true Bravais lattice is not identical
with this lattice complex, but may be primitive
tetragonal with dimensions some multiple of those
of the complex.
The relative intensities of the hydride and atitanium X-ray lines as functions of overall composition indicated that the hydride, presumably in
equilibrium with the a at 100" to 255"C, contained
more than 62 atomic pct H.

Acknowledgments
The experimental work covered in this report was
done at Brookhaven National I,aboratory, through
the courtesy of D. Gurinsky. The cooperation of
J. Sadofsky, 0 . Kammerer, A. McReynolds, and
A. Cendrowski is gratefully acknowledged.
References
1 G. A. Lenning, C. M . Craighead, and R . I. Jaffee: Constitution
and Mechanical Properties of Titanium-Hydrogen Alloys. AIME
Trans.. 1954, vol. 200, pp. 367-376: JOURNAL
OF METALS,
March 1954.
2 A. Chretien, W. Freundlich, and M. Bichara: Etude du systeme
titane-hydrogene; preparation d'un hydrure de titane, TiH2. Comptcs
Rendus. 1954, vol. 238, pp. 1423-1424.

JULY 1956, JOURNAL O F METALS-861