BREAKTHROUGH TECHNOLOGIES

FOR THE NEW MILLENNIUM

Winston L. Tennies
Gary E. Metius
John T. Kopfle
Midrex Technologies, Inc.
Charlotte, NC, USA
ABSTRACT
Continued development of the MIDREX Direct
Reduction Process has resulted in technological
innovations providing for enhanced productivity
and economics. With the incorporation of lump ore
use, iron oxide coating, oxygen injection, and
OXY+, the productivity of a MIDREX Shaft
Furnace can be increased up to 50 percent versus
the original practices in the early 1970s. Midrexs
HOTLINK System provides a simple, reliable way
to charge hot DRI directly from the shaft furnace to
the EAF, resulting in higher meltshop productivity
and lower production cost. Given the importance
of environmental considerations today, the use of
DRI or HBI produced in MIDREX Direct
Reduction Plants allows integrated steelmakers to
reduce coke consumption and increase hot metal
output.
KEYWORDS
Direct Reduction, Ironmaking, Environment
INTRODUCTION
The MIDREX Process was developed during the
1960s by the Midland-Ross Corporation, which had
pioneered shaft furnace technology and
stoichiometric gas reforming for use in processing
minerals. In 1969, the first MIDREX Plant was built
in Portland, Oregon, USA. Since then, 49 additional
modules have been sold in 18 countries. During the
quarter century since commercialization, many
technical improvements have been made, including
larger shaft furnaces, in-situ reforming, greater heat
recovery, improved catalysts, and hot briquetting.
Midrex continues to advance the state-of-the-art in
shaft furnace direct reduction technology in the areas
of raw material flexibility, shaft furnace productivity,
product characteristics, flowsheet options, and energy
efficiency.
This paper describes two areas of technology
development that Midrex is presently pursuing:
increasing shaft furnace productivity (specifically via
oxygen use) and hot charging of DRI to the EAF. It
also discusses environmental considerations with
regard to ironmaking processes.
SHAFT FURNACE PRODUCTIVITY
The single most important factor contributing to the
success of the MIDREX Process has been increasing
shaft furnace productivity. Historically, the major
challenge has been to improve the rate and degree to
which CO and H
2
are consumed in the shaft furnace,
which strongly influences productivity and energy
efficiency. Over the last 30 years, the utilization of
these gases in MIDREX Shaft Furnaces has increased
by more than 25 percent. This has been achieved by
improving the uniformity of solid/gas contact and
increasing the reducing gas temperature entering the
shaft furnace.
Original Practice (1970s)
The original practices followed in the early 1970s
utilized 100 percent pellet feed and low reducing gas
temperatures due to the sintering tendencies of the
pellet burden materials. This practice will be referred
to as Case 1 and will form the starting base for
comparison of the process development. Fig. 1 shows
the general arrangement of the reducing gas cooler and
natural gas addition used in both Cases 1 and 2. The
summary data for all six cases and the comparison can
be seen in Table 1 and Fig. 2, respectively.
FIG. 1.
Original Practice (1970s)
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1 2 3 4 5 6
prod
cost
oxyg
ng
elec
FIG. 2.
Production Comparison:
Increasing Shaft Furnace Productivity
prod cost oxyg ng elec
1 1.000 1.000 0.000 1.000 1.000
2 1.130 0.957 0.000 0.976 0.885
3 1.241 0.927 0.000 0.960 0.806
4 1.368 0.960 0.425 0.969 0.731
5 1.455 0.997 0.733 0.990 0.687
6 1.504 1.015 1.000 0.985 0.665
prod Production ratio from a base of Case 1
cost Cost ratio of oxygen plus natural gas from a base of Case 1
(Oxygen - $42/ton, Natural Gas - $1.8/MMBtu, Electricity - $0.04/kWh)
oxyg Consumption ratio of oxygen/t from a base of Case 6
ng Consumption ratio of natural gas from a base of Case 1
elec Consumption ratio of electrical power from a base of Case 1
Reformer
Cooler
Natural
Gas
Shaft
Furnace
Hot Reducing Gas
TABLE 1
Production Data
case enr% bgtm brtm prod oxypt ngpt elec cost
1 4.5 780 789 88.8 0.0 268.6 135 $19.65
2 3.5 850 814 100.3 0.0 262.3 120 $19.19
3 3.5 918 834 110.2 0.0 257.9 109 $18.87
4 4.5 1050 860 121.5 17.5 260.3 99 $20.10
5 3.0 961 857 129.2 30.2 265.8 93 $21.26
6 3.0 1050 898 133.6 41.2 264.6 90 $21.83
enr% - Percentage of natural gas in reducing gas at furnace entrance
bgtm Temperature (C) of reducing gas at furnace entrance
brtm Temperature (C) of shaft furnace burden
prod Tons per hour of production from MIDREX Plant
oxypt Flow (Nm
3
/t) of oxygen consumed
ngpt Flow (Nm
3
/t) of natural gas consumed
cost - Cost of oxygen plus natural gas
(Oxygen - $42/ton, Natural Gas - $1.8/MMBtu)
Case 1 Original practice 1970s
Case 2 Practice using lump ore 1980s
Case 3 Practice using coating of oxide feed materials 1990s
Case 4 Oxygen Injection practice late 1990s
Case 5 OXY+ practice - 2000
Case 6 Combination practice with oxygen injection and OXY+ - future
Note: All production percentage variations are commented on from a reference of Case 1.
Lump Ore Practice (1980s)
In the mid-1970s, lump ore was first used and the
practice was widely adopted in the 1980s. This
provided an additional benefit of preventing the
sintering of the shaft furnace burden as reducing gas
temperatures increased and added about 13 percent to
the system productivity. Case 2 data shows that the
reducing gas temperatures increased from 780C to
850C.
Iron Oxide Coating Practice (1990s)
Development work continued in the 1980s, and led
to the introduction of in-plant coating of iron oxide
feed materials with CaO or CaO/MgO in many plants
by the mid-1990s. As more interest in this technique
grew, the practice of pre-coating oxide pellets was
adopted by international pellet suppliers and has
permitted the nearly universal application of higher
reducing gas temperatures, where these pellets are
available. This further improved productivity by an
additional 11 percent. As Case 3 indicates, the
reducing gas temperatures increased to slightly more
than 900C. Fig. 3 shows the changes in general
arrangement that resulted.
Through Cases 1 to 3, the temperature of the reducing
gas had increased by more than 100C, while the
temperature of the shaft furnace burden had increased
by only 40C. All of these improvements were
accomplished with no major capital investment or
major modifications to the typical MIDREX Direct
Reduction Plant equipment.
FIG. 3.
Iron Oxide Coating Practice (1990s)
Oxygen Injection (late 1990s)
Up to this point, the quality of the reducing gas was
held nearly constant, while the temperature of the
reducing gas at the shaft furnace was allowed to
increase. From this point, the development efforts led
us to increase reducing gas temperatures at the cost of
reducing gas quality. The results that followed
showed a clear production advantage for the higher
reducing gas temperature versus the loss in reducing
gas quality that resulted from the oxygen combustion.
The introduction of oxygen injection, combustion of a
portion of the reducing gas CO and H
2
by O
2
,
achieved this effect with great success.
Oxygen injection, shown in Fig. 4, as currently
designed consists of the introduction of high purity
oxygen into the flowing hot reducing gas stream
through a multiple nozzle arrangement. This practice,
as shown by Case 4 data, has resulted in reducing gas
temperatures in excess of 1000C, and burden
temperature increases of up to 70C, over the Case 3
conditions. The investment costs, assuming that
oxygen will be available from an across-the-fence
supply, for a typical oxygen injection system are
primarily the oxygen supply piping and the flow
control and safety equipment. The only operating
cost increase is due to the consumption of oxygen at a
typical rate of 12 to 15 Nm
3
/t. The payback is an
increase in shaft furnace productivity by up to an
additional 12 percent, as compared to the previous
case. Today, many MIDREX Plants are using oxygen
injection as a means of increasing production at
minimum investment cost.
Since 1969, there has been a major increase in
MIDREX Direct Reduction Plant productivity. Today,
the state-of-the-art in operations, including use of
lump ore, iron oxide coating, and oxygen injection,
results in a productivity increase of about 37 percent,
versus the first generation plants.
OXY+ (2000)
OXY+ generates a reducing gas by reacting oxygen
and natural gas at about a 0.5 stoichiometric ratio.
The burner is designed to be installed directly in the
reducing gas duct after the reformer, as indicated in
Fig. 5. The addition of in-situ partial oxidation of
natural gas, to generate additional CO and H
2
as well
as sensible heat, offers new possibilities to production
augmentation of existing plants and capital savings in
new facilities.
Unlike the oxygen injection systems, OXY+ closely
controls the combustion mixing of the oxygen and
natural gas to maintain consistent gas quality and
temperature. This control minimizes the temperature
increase of the gas entering the shaft furnace and
provides for additional opportunity to extend plant
production capacity, as can be observed in the Case 5
data.
Reformer
Natural
Gas
Shaft
Furnace
Hot Reducing Gas
FIG. 4.
Oxygen Injection (late 1990s)
FIG. 5.
OXY+ (2000)
Reformer
Oxygen
Shaft
Furnace
Hot Reducing Gas
Natural
Gas
Reformer
Natural
Gas
Shaft
Furnace
Hot Reducing Gas
OXY+
Natural
Gas
Oxygen
The application of OXY+ results in a potential
increase in shaft furnace capacity of approximately 21
percent, over Case 3 conditions. It can be seen by
comparing the data presented for Cases 4 and 5, the
benefit of OXY+ over oxygen injection. A
commercial OXY+ installation is underway and is
expected to be operational in mid-2000.
Combination Practice with Oxygen Injection and
OXY+ (future)
The optimum productivity is achieved by maximizing
the reducing temperature of the burden and the
quality of the reducing gas entering the shaft furnace.
These two factors are the keys to optimizing the
production of any direct reduction furnace and its
related gas generating equipment. By utilizing a
combination of the two operating practices, discussed
previously, oxygen injection and OXY+, shown on
Fig. 6, as well as maintaining the natural gas in the
reducing gas stream, it is possible to independently
control the shaft furnace burden temperature and the
reducing gas temperature. This permits the operator to
maximize the performance of the shaft furnace by
maximizing the utilization of the reducing gases within
the furnace. This practice offers the potential for a
production increase of approximately 5 percent over
Case 5 conditions, the production level achieved by
the plant operating only with OXY+. Midrex expects
to commercialize the combination approach in late
2000 or early 2001.
FIG. 6.
Oxygen Injection and OXY+
Summary of Shaft Furnace Productivity
Improvements
Given that the single most important factor in the
technical and economic success of an ironmaking
process is productivity, the enhancements to the
MIDREX Process since 1969 have been largely
focused on that aim. Much of the productivity gains
have been made by allowing higher reducing gas
temperatures via lump ore use, iron oxide coating, and
use of oxygen. As Fig. 2 shows, the use of all the
enhancements described above results in a total
productivity increase of 50 percent, compared to the
original design and practices. Midrex will continue to
push the envelope by developing new technologies
and practices that enable higher productivity.
Reformer
Natural
Gas
Shaft
Furnace
Hot Reducing Gas
OXY+
Natural
Gas
Oxygen
Oxygen
HOT DRI CHARGING - HOTLINK
Optimization of DR/EAF production and energy
efficiency have taken a new step with the close
coupling of a MIDREX Shaft Furnace and an EAF to
achieve increased productivity and energy savings in
the production of high quality steel. The concept,
shown in Fig. 7, under study for several projects
places a MIDREX Hot Discharge Furnace just
outside and above the wall of the meltshop. This
provides the opportunity to discharge directly from
the shaft furnace to a hot DRI surge bin and then from
the surge bin directly to the EAF by gravity.
This type of arrangement has been used on all
MIDREX HBI plants, in which the hot DRI reaches
the briquetting machines in excess of 700C. The
HOTLINK Modules are equipped to handle any upset
problem and are fully capable of discharging at full
production through a DRI cooler located next to the
hot DRI surge gin. The primary goal of the
arrangement shown below is to supply hot DRI to the
EAF as the primary discharge method from the DR
plant, but be fully capable of switching to the cold
DRI mode and back to the hot DRI mode with no
delay in operation.
We expect the typical hot DRI temperature to average
700C during each heat. This will result in an
electrical energy savings of 120-140 kWh/t of liquid
steel, when 95% of the metallic charge material is hot
DRI. HOTLINK is currently the basis of two detailed
feasibility studies, and we expect it to be part of one of
our next new plant contracts.
.
FIGURE 7 HOTLINK FLOW DIAGRAM
FIG. 7.
HOTLINK System
THE MIDREX PROCESS A GREEN
ALTERNATIVE
The continued technical developments of the
MIDREX Process not only enhance direct reduction
economics, but provide an environmental benefit,
since they result in more efficient use of gas and
electricity. In addition, the supplementing or
replacement of blast furnace hot metal with Midrex
DRI or HBI is an environmentally friendly approach.
Following is a description of the possibilities.
In the United States, many sinter plants and coke
ovens will be closed in the future to reduce emissions
of carbon dioxide and pollutants. Total US coke
consumption has averaged 23.6 million tons during
the last five years, while production was just 17.6
million tons, as shown in Fig. 8. The six million ton
shortfall was met largely by imports from China and
Japan. China is closing a number of small coke ovens
as part of its iron and steel industry restructuring
effort. Though there is a plan to replace these small
facilities with larger, new ones, the changeover may
not be orderly. Once Chinese coke exports decrease,
the impact will be felt not only in the US, but
worldwide.
Worldwide, integrated mills will continue to spend
large amounts for maintenance of old sinter plants and
coke ovens, since new installations are unlikely due to
tightening environmental controls and the significant
capital requirements. An excellent option for
steelmakers to manage this situation is to use a
metallic charge such as HBI. In the US, hot briquetted
iron has been used in many blast furnaces to boost hot
metal production continuously, or to increase output
temporarily while relining other blast furnaces or
repairing coke ovens.
In addition, HBI use in blast furnaces reduces coke
consumption and total carbon dioxide emissions. DRI
and HBI are also being used in basic oxygen furnaces
to supplement scrap. In Japan and Korea, integrated
producers have started using HBI in their blast
furnaces and BOFs. Since the blast furnace is a very
efficient ironmaking unit, it will remain the
predominant method for the foreseeable future.
However, due to the need to ameliorate the
environmental problems of sinter plants and coke
ovens, the blast furnace will become more of a melting
furnace in the future.
FIG. 8.
US Coke Production and Consumption
Source : AISI
1998
1997
1996
1995
1994
Consumption
Production
0
5
10
15
20
25
M
i
l
l
i
o
n

n
e
t

t
o
n
s

/

y
e
a
r
FIG. 9.
Greenhouse Gas Emissions for
Selected Ironmaking/Steelmaking Processes
Fig. 9 shows greenhouse gas emissions for various
ironmaking/steelmaking processes, including blast
furnace/basic oxygen furnace, pig iron/EAF, coal-
based DRI/EAF, natural gas-based DRI/EAF, and
scrap/EAF. The figures show total carbon emissions
in the form of carbon dioxide per ton of liquid steel.
The calculations take into account excess energy that
is converted to electricity. Note that the figures
assume use of 100 percent pig iron or DRI in the
EAF. Since this is not usually the case, and scrap is
used along with virgin iron, emissions in those cases
will usually be lower than shown in the graph.
As Fig. 9 shows, scrap-based steelmaking has much
lower carbon emissions than integrated steelmaking.
Also, the use of recycled scrap steel makes efficient
use of an otherwise unwanted waste product.
However, it is not generally feasible to use 100
percent scrap to produce high quality steels. Charging
of alternate iron to the EAF allows the best of both
worlds: the environmental benefits of scrap along with
the quality benefits of the use of virgin iron.
Production of DRI or HBI via gas-based or coal-based
processes provides for lower carbon emissions than
production of iron by the blast furnace/BOF route.
0
100
200
300
400
500
600
700
Canada Japan US
BF-BOF
EAF-Pig
EAF-Coal DR
EAF-NG DR
EAF-Scrap
K
g
-
C
/
M
t
-
s
t
e
e
l