United States
Environmental Protection
Agency
Water Engineering
Research Laboratory-
Cincinnati OH 45268
Research and Development
EPA/600/S2-87/048 Sept. 1987
x>EPA Project Summary
Reactivation of Granular Carbon in
an Infrared Traveling Belt Furnace
Ramin Nur and Robert W. Horvath
An all-electrical Shirco* carbon regen-
eration furnace and its air pollution
control system were evaluated for cost
and process effectiveness in carbon
reactivation at the Pomona Advanced
Wastewater Treatment Research Facil-
ity. The pilot scale Shirco furnace was
operated within the range of 102 kg/d
(224 Ib/d) to 196 kg/d (434 Ib/d).
The Shirco carbon regeneration sys-
tem was as effective as the multiple
hearth and rotary kiln furnaces in re-
activating the exhausted granular
activated carbon. The Shirco furnace
required less operational skill but more
maintenance labor than the multiple
hearth or rotary kiln furnaces. The high
maintenance requirement of the Shirco
furnace was caused mainly by prema-
ture deterioration and breakdown of
the heating elements and conveyor belt
mistraction inside the furnace.
A cost estimate based on a typical
regeneration capacity of 182 kg/h (400
Ib/h) has been made for the Shirco
furnace regeneration system. Compari-
son of this cost estimate to those that
were reported for the multiple hearth
and rotary kiln furnaces indicates that
capital cost for the Shirco furnace is
lower than that for the multiple hearth
furnace and higher than that for the
rotary kiln regeneration unit. The opera-
tion and maintenance cost for the Shirco
furnace was, however, higher than those
for both the multiple hearth and the
rotary kiln furnaces. The overall process
cost for the Shirco furnace system based
on the operation and maintenance of
the pilot unit was estimated to be 61.8
Mention of trade names or commercial products
does not constitute endorsement or recommenda-
tion for use.
ct/kg (30.9 ct/lb) for the carbon
regenerated.
Tills Pro/act Summary was developed
by EPA't Watar Engineering Research
Laboratory, ClndnnaU, OH, to announce
toy finding* of the rssearcn pro/act that
la fully documented In a separate report
of the same Wt/e fsee Pro/act Report
ordering Information at back).
Introduction
This study was conducted to evaluate
and determine the cost-effectiveness of
the Shirco infrared traveling belt carbon
regeneration system. It was a portion of a
much larger investigation into the removal
of potentially hazardous trace organics
from wastewater.
An extensive pilot plant study on the
granular activated carbon adsorption
process for wastewater treatment has
been jointly conducted since 1965 by the
Sanitation Districts of Los Angeles County
and the U.S. Environmental Protection
Agency at the Sanitation Districts'
Pomona Advanced Wastewater Treat-
ment Research Facility in Pomona, CA.
Initially, the Pomona carbon study utilized
a multiple hearth furnace system for
carbon regeneration during the first 10 yr
of pilot plant operations. During this
period, the study concerned itself with
the evaluation of the various treatment
process parameters, mainly pretreatment
requirements, carbon characteristics,
hydraulic loading rates, adsorption ca-
pacity, backwash requirement, and mode
of regeneration. Different types of carbon
regeneration furnaces were not evaluated,
since during that initial period the multiple
hearth furnace system was very effective
and reliable in regenerating the spent
activated carbon in wastewater treatment
processing.
-------
However, because carbon regeneration
is a major factor in the cost of activated
carbon treatment, investigation of other
regeneration systems was deemed de-
sirable. A study at this site with a rotary
kiln (a less capital intensive process) was
reported previously. An investigation of
the Shirco furnace was conducted be-
cause it uses a different method of con-
tacting the heated regenerant gases with
the carbon, which could yield lower loss
of carbon during the regeneration cycle.
In addition, the use of electricity for heat
production by the Shirco furnace may be
advantageous in certain locations.
Granular activated carbon (Filtrasorb
300) was exhausted by exposure to
secondary effluent in a downflow carbon
contactor and then regenerated in the
Shirco regeneration furnace. The per-
formance of the carbon after regeneration
was evaluated by its adsorption capacity
and efficiency in the next adsorption cycle.
This process was repeated for several
cycles, and after each regeneration,
carbon quality was compared to that of
its virgin state. The performance of the
electric Shirco furnace was compared to
that of the multiple hearth and the rotary
kiln furnaces that were used in the past
to regenerate the same type of carbon
from the same carbon contactors and
exposed to the same activated sludge
plant effluent.
The ability to regenerate carbon to its
pre-exhausted conditions, the ease of
control and operation, the extent of carbon
losses, and energy consumption were
the basis for this evaluation.
Carbon Regeneration System
The infrared traveling belt carbon
regeneration furnace system was manu-
factured by Shirco, Inc. The regeneration
furnace is a rectangular, horizontal sy
tem consisting of an insulated enclosu
through which the carbon is transport*
on a continuous woven wire conveyi
belt. This system was assembled from
series of three modules that were boltc
together before conveyor belt installatic
(Figure 1) with the final dimensions >
0.81 m (32 in.) wide by 3.4 m (135 ir
long and a height of 0.86 m (34 in.). Th
stainless steel modules were factory line
with a thermal-shock-resistant, ceram
fiber blanket insulation system and wei
equipped with support rollers for tr
conveyor belt. The Shirco furnace systei
had a rated total regeneration capaci
range of 327 kg (720 Ib) to 381 kg (840II
of granular activated carbon/24 h of coi
tinuous operation.
Spent carbon is automatically release
into the furnace feed hopper where it
mixed with water by a small variabl
Exhaust Blower
_
d
PJ
me
ff
I'liffl
ri
On
W
F-V
tr
i- View f
k rq
Scrubber
• Afterburner
Heating Element,
Control Centers
Electrical Cover
Plain View
Exhaust Ducting
Afterburner
Exhaust
Blower
Blower
Drying/Activation Module
Feed Module
Drive System
Heating Element
Control Centers
Figure 1. Shirco furnace regeneration system.
2
Scrubber
Water Inlet
Elevation View
Outlet for Carbon
Slurry Transport
-------
speed vibrator. The mixture of water and
carbon is fed into the regenerator by a
variable speed, spiral type, hollow, stain-
less steel screw conveyor. Inside the
furnace, the wet carbon on the conveyor
belt is leveled by an internal roller into a
layer approximately 1.9 cm (3/4 in.) thick
spanning 2/3 of the width of the belt.
The carbon then moves through the
various heating zones to accomplish the
regeneration process. The required re-
generation temperature is provided by 24
silicon carbide heating elements located
approximately 15 cm (6 in.) above the
carbon layer. The regenerated carbon is
then discharged at the exit end into a
quench tank for cooling before transport
to the adsorption contactors.
The Shirco furnace is equipped with an
afterburner and venturi scrubber unit for
air pollution control of the exhaust gases.
The required heat energy in this unit is
provided by a cluster of 12 shorter silicon
carbide heating elements on the top of
the unit. After being cooled down to
approximately 32°C (90°F) by the venturi
scrubber, the burned air-exhaust mixture
is dispersed into the atmosphere through
a stainless steel stack by an exhaust
blower.
Carbon Adsorption
During the Shirco carbon regeneration
study, unchlorinated and unfiltered
secondary effluent from the Pomona
Water Reclamation Plant was treated
directly by the carbon adsorption system.
The Pomona Water Reclamation Plant is
a 0.44 mVs (10 MGD) activated sludge
plant, and is located adjacent to the
research facility where this study was
conducted. The three contacting carbon
columns were operated in series in a
downflow mode at a constant rate of 6.3
L/s (199 gpm) thereby providing a
hydraulic loading rate of 2.4 L/s/m2 (3.5
gpm/ft2) and an empty-bed contact time
of approximately 10 min for each column.
The first carbon bed was backwashed
daily with unchlorinated secondary ef-
fluent to maintain good hydraulic condi-
tions for operation. A portion of the carbon
treated water was stored and provided a
sufficient amount of water for back-
washing the second and third columns in
the series every 2 and 4 wk, respectively.
The carbon contacting columns were
taken off stream after they had treated a
total volume of 32,200 m3 (8.5 MG) of
Pomona Water Reclamation Plant second-
ary effluent during the first adsorption
cycle and approximately 41,600 m3 (11.0
MG) during the three cycles that followed.
The breakthrough of trace organics under
study occurred during this 3 mo of opera-
tion, and the COD removal efficiency of
the carbon columns usually leveled off at
this point. The spent carbon was
thoroughly backwashed before being
hydraulically transferred to the elevated
dewatering chamber.
Carbon Regeneration
Carbon retention time in the furnace
was set at 18 to 35 min, depending on
the extent of carbon exhaustion and
operating temperature of the Shirco fur-
nace. The regeneration temperature was
maintained at 760° to 815°C (1400° to
1500°F) in the first (drying) zone and
899° to 955°C (1600° to 1750°F) in the
second zone. This particular temperature
range was chosen since the previous
regeneration furnaces tested at this site
operated in a similar mode. These tem-
peratures were also recommended by
Shirco, Inc., who uses a similar unit at
their own research facilities. In general,
during the first and second regeneration
cycle the carbon in the final column was
regenerated under lower temperatures
than the carbon in the first column. The
carbon in the first column, however, was
regenerated under higher temperature
since it was spent to a higher degree
than the other two columns in the series.
A control led oxidizing atmosphere neces-
sary for the carbon activation process
was obtained by the steam generated in
the drying zone and flow co-current with
the carbon through the length of the
furnace. The regenerated carbon was
discharged from the Shirco furnace into
a quench tank and was continuously
educted back into the contacting columns.
Four adsorption and three regeneration
cycles were conducted during the study,
and appropriate samples were taken for
evaluating the carbon adsorption and
carbon reactivation efficiencies. In the
course of carbon regeneration, a numbr
of control tests measuring such parame-
ters as apparent density, iodine number,
molasses number, and methylene blue
number, were performed to regulate the
regeneration process and monitor the
quality of the regenerated carbon. In
general, approximately 550 h was re-
quired for each cycle resulting in the
regeneration of 4,756 kg (10,500 Ib) of
carbon. Laboratory analysis on regen-
erated carbon was performed on an hourly
basis for apparent density, every 2 h for
molasses and iodine numbers, and once
every 8 h for methylene blue number.
Grab samples of spent carbon collected
during carbon transfer and the hourly
samples of regenerated carbon were
composited over the regeneration period.
These composited carbon samples were
analyzed for apparent density, iodine,
methylene blue and molasses numbers,
and ash content.
Performance Of Regeneration
System
During the initial shakedown operation
of the Shirco carbon regeneration system,
a number of mechanical difficulties were
encountered. These problems were traced
to inadequate design of portions of the
regeneration system. The problems were
generally in the areas of the carbon feed
system, leveling roller, carbon movement
through the furnace (sudden stopping of
the belt), faulty tracking of the belt inside
the furnace, furnace and afterburner
temperature control, and temperature
monitoring and recording instrumenta-
tion. A number of system modifications
were performed, and most of the problems
were corrected before the first carbon
regeneration cycle. Some problems, how-
ever, were major design problems that
could not be corrected at the research
site.
In the course of thermal regeneration,
the organic pollutants on the surfaces of
the external and pore areas of carbon are
oxidized and removed. This oxidation
process, however, does not completely
remove the adsorbed organics from the
carbon pores. Therefore, a certain amount
of the capacity is normally lost in every
thermal regeneration cycle. In addition,
the change of pore size distribution during
the regeneration process may also con-
tribute to the reduction of carbon adsorp-
tion capacity. The carbon adsorption
capacity recovery was monitored by the
determination of the iodine number,
molasses number, and methylene blue
number of both spent and regenerated
carbons.
Iodine and molasses numbers are
related to the surface area of the pores
with a diameter larger than 10 and 28
angstroms, respectively. A continuing
decrease in the iodine number with
respect to regeneration cycle was ap-
parent though the cyclic thermal regen-
eration was basically effective in restoring
the operational adsorption capacity. The
molasses number was found to gradually
increase with each successive regenera-
tion cycle. Since the molasses number is
related to the surface area of the pores
with a diameter larger than 28 angstroms,
the increase in molasses number in-
-------
dicated an enlargement of micropore
structures to macropore structures in the
carbon during the repeated thermal
regeneration process. This shift in pore
size distribution also caused a reduction
of total surface area of the carbon, which
was indicated by the reduction of the
iodine number. Methylene blue number
is related to surface area of carbon pores
with diameters larger than 15 angstroms.
The methylene blue number of the carbon
was not affected to the same extent as
iodine and molasses number during the
thermal regeneration.
The effects of the Shirco furnace carbon
regeneration on the various carbon
characteristic numbers as discussed
above are similar to those reported for
the multiple hearth study. Apparently, all
three regeneration systems, the multiple
hearth, rotary kiln, and Shirco furnace
could repeatedly restore the carbon
adsorption capacity equally well following
each adsorption cycle.
Because of stretching and corrosive
damage during regeneration, the con-
veyor belt required replacement after
approximately 1,800 h of operation. Some
of the heating elements had to be replaced
after 550 h of operation. The entire 24
units of the heating elements in the
furnace were replaced before the final
regeneration. Since several of the new
heating elements broke in half and be-
came inoperative by the last days of final
regeneration after operating for approxi-
mately 600 h, the average life of this
component is estimated to be no longer
than 750 h.
Cost Estimates
The cost estimates for the Shirco fur-
nace system have been divided into two
subcategories; namely, capital cost and
operation and maintenance costs. The
equipment cost consists of the carbon
feed system, furnace system, and the air
pollution control system, which consisted
of an afterburner and venturi wet scrub-
ber. The capital cost also includes the
initial engineering cost, equipment ship-
ping and installation cost, and contin-
gency. The operation and maintenance
costs include the utilities, operation and
maintenance labor, carbon makeup, and
maintenance materials consisting of
heating elements and conveyor belt re-
placement. Costs are summarized in Table
1. Total estimated process cost is 61.8
ct/kg (30.9 ct/lb) of carbon regenerated.
Conclusions
The following conclusions can be drawn
from the pilot plant study of the Shirco
regeneration furnace and its comparison
to the multiple hearth and/or rotary kiln
furnaces:
• The all-electric Shirco furnace was
found to be as effective as the
multiple hearth and rotary kiln
furnaces in reactivating granular
activated carbon that had been ex-
hausted by an activated sludge plant
effluent.
• The Shirco furnace is insulated with
a thermal-shock-resistant ceramic
fiber blanket. Unlike the multiple
hearth furnace's refractory lining,
this type of insulation was able to
withstand rapid startup and shut-
down.
• The steam generated in the drying
zone of the Shirco furnace moves
co-currently with the carbon to ai
the activation process; thus th
auxiliary process steam used for th
multiple hearth and rotary kiln fui
naces was not required.
• The Shirco furnace system require
less operational skill but more main
tenance labor than the multipl
hearth and rotary kiln furnaces.
• The usable life span of the Shire
furnace components such as heatin
elements and the conveyor belt wa
much shorter than expected.
• The energy cost per pound of carbo
regenerated by the Shirco furnac
was higher than for the two othe
furnaces studied previously. This i
because of the higher cost of electri
city to generate the same amount c
heat energy produced by fossil fue
for the multiple hearth furnace.
• All three regeneration furnace sys
Table 1. Coat Estimates for the Shirco Furnace Based on Operation of the Pomona Pilot Plar
Category $1K> $(K)
Ct/lb
carbon
Capital
Equipment
Carbon feed system
Furnace system
Air pollution control system
Total Equipment Cost
Shipping and Installation
Engineering
Contingency
Total Capital Cost
Capital Amortization
Operation and Maintenance
Utilities
Power
Water
Labor
Carbon Makeup
Maintenance Material
Heating elements
Conveyor belt
Other
Total Process Cost
25
345
125
495
124
49.5
49.5
718
4.87
11.7
0.135
3.75
4.9
4.16
1.187
0.208
30.91
-------
terns required an afterburner and a
venturi wet scrubber for effective
emission control of air pollutants.
However, since no fossil fuel is used
in the Shirco furnace, the total ex-
haust gas volume was less.
• The average carbon loss for the
Shirco furnace was slightly higher
than 7%, which was reported for the
multiple hearth and rotary kiln
furnaces.
• The total capital cost of the Shirco
furnace is estimated to be lower
than that for the multiple hearth
furnace and higher than that for the
rotary kiln system. The total process
and operation and maintenance costs
for regeneration of carbon by the
Shirco furnace were, however, high-
er than costs for the other regenera-
tion systems studied previously.
The full report was submitted in ful-
fillment of Contract No. 68-03-2745 by
the County Sanitation Districts of Los
Angeles County under the sponsorship of
the U.S. Environmental Protection
Agency.
Ramin Nur and Robert W. Horvath are with County Sanitation Districts of Los
Angeles County.
Irw/n J. Kugelman was the EPA Project Officer (see below).
The complete report entitled "Reactivation of Granular Carbon in an Infrared
Traveling Belt Furnace," (Order No. PB 87-209 466/AS; Cost: $ 13.95, subject
to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
For further information, Sidney Hannah, can be contacted at:
Water Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S2-87/048
CHICAGO
------- |