Utilizing
simulations
to
guide
furnace
designs
for
the
E-
™Nugget
process
Introduction
Need Carbontec
E-Nugget
Process
•
Carbontec
Energy
Corporationhas
developed
a
novel
iron
smelting
proces s
that,
as
proven
in
an
extended
pilot
operation,
can
produce
high
quality
pig
iron
nuggets
on
a
continuous
basis
–
Can
convert
both
iron
ore
and
steel
mill
wastes
into
low
sulfur
iron
nuggets
–
Utilizes
biomass
reductants
in
place
of
coke
•
Carbontec
is
moving
forward
to
construct
a
plant
that
will
produce
100,000
tonnes/yr
of
pig
iron
grade
nuggets
–
We
are
modeling
the
core
nugget
process
to
provide
a
tool
to
assist
in
the
final
design
of
the
production
plant
by
avoiding
pitfalls
in
the
joining
of
the
core
process
with
production
infrastructure
Aaron
Fisher,
Christopher
Walton
(LLNL)
Chenn
Zhou,
Armin
Silaen,
Haibo
Ma
(CIVS,
Purdue
University
Northwest)
John
Simmons
(Carbontec)
Approach
•
Iron
ore
and
biomass
are
blended
and
pressed
into
briquettes
•
Briquettes
are
fed
through
multiple
temperature
zones
of
a
linear
furnace
•
The
iron
ore
and
biomass
run
through
a
series
of
chemical
reactions
and
the
iron
ore
is
reduced
to
iron
•
The
metallic
iron
then
flows
from
the
briquettes
and
pools
into
nugget
sized
pieces
•
The
resulting
material
is
cooled
tumbled
and
the
iron
nuggets
are
magnetically
separated
Benefits
Approach
(continued)
Results
(continued)
•
Process
replaces
coke
in
iron
smelting
–
First
plant
will
displace
90M
lbs
coke/yr*
–
Converting
coal
to
coke
requires
1.75-2.5
MJ
/
lb**
–
If
5%
of
US
iron
production
were
produced
with
this
approach,
2
PJ/yr
of
energy
could
be
saved
by
reducing
coke
needs
by
1
billion
lbs
•
E-Iron™nuggets
are
good
feedstock
for
electric
arc
and
BOF
furnaces
–
96.5%
Fe,
2.9%
C,
0.017%
S
*2014
AIST
average
coke
rate
is
~900
lbs/NTHM
**2007
IEA
Tracking
industrial
efficiency…
p110
E-Nugget
Process
Simulations
•
Star
CCM
model
of
experimental
process
–
Models
include
multiphase
solid/gas
flow,
radiative
and
convective
heat
transfer,
and
16
reaction
chemistry
model
–
2D
and
3D
simulations
on
up
to
512
CPUs
–
Temperature
varied
for
different
furnace
zones
•
Model
validated
against
experimental
data
Results
Physical
model
of
briquettes
on
a
feed
boat
Wood pyrolysis
blast
furnace
r
methane combustion
Lumped wood model Chemical
reaction
model
Temperature
zones
the
feed
boat
moves
through
Time
(s)
•
Simulations
were
run
on
4
experimental
cases
–
M-11
was
run
at
higher
temperatures
(1600K)
and
had
a
full
yield
of
Fe
–
S-02
was
run
at
higher
temperatures
(1600K)
and
utilized
different
ore
and
flux
conditions
and
yielded
~50%
–
M-12
and
S-04
were
run
at
lower
temperatures
(1300K)
failed
to
yield
Fe
did
not
yield
metallic
Fe
Experiment
M-11
Experiment
M-12
Temperatures
at
the
end
of
the
heatingzone
in
2
experimental
cases.
The
highertemperatures
in
M-11
led
to
a
successful
experiment.
Time
(s)
Averag
temperatures
of
the
briquettes
throughoutthe
experiments.
We
believe
the
Fe
must
reach
T~1500k
to
succeed
in
coalescingthe
metallic
Fe.
Experiment
M-11
Experiment
M-12
Time
(s)
Time
(s)
Normalized
mass
fractionsof
the
various
iron
oxides
in
the
successful
M-11
experiment
and
the
unsuccessful
M-12
experiment.
In
both
cases
nearly
all
of
the
oxides
are
reduced
indicatingthat
the
Fe
coalescence
after
reduction
is
the
dominant
factor
differentiating
success
from
failure.
•
All
4
simulated
cases
showed
most
of
the
Fe
oxides
reducing
before
the
end
of
the
experiment
•
After
reduction
the
Fe
must
reach
a
high
enough
temperature
to
flow
and
coalesce
•
M-
and
S-
experiments
utilized
different
iron
ore
sources
Material
components
(s)
(Cellulose,
Hemicellulose,
Lignin)
Active material (s)
Tar
vapor
(
Char
(s)
+
(1-
)
Gas
(g)
Gas
(g)
k 1
k 2 k 3
k 4
This
work
was
supported
by
the
U.S.
Department
of
Energy
(DOE),
Office
of
Efficiency
and
Renewable
Energy
(EERE),
Advanced
Manufacturing
Office
(AMO)
under
contract
no.
DE-AC52-07NA27344.
POST-724662
APPENDIX
A
6
Temperature (K)
Temperature (K)
Normalized ma ss fraction
Normalized ma ss fraction