EPA/600/2-86/015
January 1986
DESIGN AND CONSTRUCTION OF A MOBILE
ACTIVATED CARBON REGENERATOR SYSTEM
R. H. Hiltz
MSA Research Corporation
Evans City, Pennsylvania 16033
Contract No. 68-03-2110
Project Officer
John E. Brugger
Hazardous Waste Engineering Research Laboratory
Releases Control Branch
Edison, New Jersey 08837 ^
HAZARDOUS WASTE ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45258
-------
-------
TECHNICAL REPORT QATA
(Please read Instructions on the reverse before completing)
. REPORT NO.
EPA/600/2-86/015
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
DESIGN AND CONSTRUCTION OF A MOBILE ACTIVATED
CARBON REGENERATOR SYSTEM
5. REPORT DATE
January 1986
6. PERFORMING ORGANIZATION CODE
A-918
7. AUTHORtSI
R.H. Hiltz
S. PERFORMING ORGANIZATION REPORT NO.
MSAR 81-107
9. PERFORMING ORGANIZATION NAME AND AOORESS
MSA Research Corporation
Division of Mine Safety Appliances Co.
Evans City, Pennsylvania 16033
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-03-2110
12. SPONSORING AGENCY NAME AND ADDRESS
Hazardous Waste Engineering Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final-8/6/74-2/8/79
14. SPONSORING AGENCY CODE
EPA/600/12
IS. SUPPLEMENTARY NOTES
Project Officer: John E. Brugger (201)321-6634
vK
16. ABSTRACT
''Activated carbon adsorption has become a standard procedure for the cleanup of
contaminated water streams. To facilitate such cleanup at hazardous waste and spill
sites, mobile carbon, adsorption units have been constructed and *re now in use. Their
primary drawback is the logistics associated with the disposal of spent (contaminated)
carbon and its replenishment with fresh, active carbon.
This program was undertaken to assess the feasibility of designing and building
a mobile carbon regeneration unit - including an incinerator/scrubber to destroy the
offgases - for field use in conjunction with mobile carbon adsorption systems. A
system was designed and built based on technology developed in the earlier fabri-
cation of a laboratory-sized regenerator and on an in-depth evaluation of factors
affecting system design and size. Housed in a standard van-type of trailer, the
system met all weight and size limitations for over-the-road transportation. The
system includes a direct fired, rotating barrel kiln to thermally regenerate the car-
bon, an incinerator and scrubber to destroy the desorbed materials and treat the off-
gases, 2nd a separator to reclaim the reactivated carbon granules.
Test runs using spent carbon from an on-site treatment of a 'spill were quite
successful. The carbon was returned to essentially 100% activity with an 88% volume
recovery. The unit has been delivered to the US EPA for their use...
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lOENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
54
20. SECURITY CLASS (Thisparti
UNCLASSIFIED
22.
IpT.
EPA P»na 2220-1 (R«». 4-77) PRIVIOUS COITION i* OOSOUETE
t^5^if^v^y^'ik.ig^«i,^^-f..&^^^
-------
DISCLAIMER
The information in this document has been funded wholly or in. part by the
United States Environmental Protection Agency under Contract No. 68-03-2110
to MSA Research Corporation. It has been subject to the Agency's peer and
administrative review, and it has been approved for publication as an EPA
document. Mention of trade names or commercial products dees not constitute
an endorsement or recommendation for use.
ii
-------
FORWARD
Today's rapidly developing and changing technologies and Industrial
products and practices frequently carry with them the Increased generation
of solid and hazardous wastes. These materials, if improperly dealt with,
can threaten both public health and the environment. Abandoned waste sites
and accidental releases of toxic and hazardous substances to the environment
also have important environmental and public health implications. The
Hazardous Waste Engineering Research Laboratory assists in providing an
authoritative and defensible engineering basis for assessing and solving
these problems. Its products support, the policies, programs, and regula-
tions of the Environmental Protection Agency, the permitting and other
responsiblitles of State and local governments and the needs of both large
and small businesses in handling their wastes responsibly and economically.
Activated carbon is proving to be an invaluable tool in the cleanup of
hazardous chemical storage sites and in the response to Industrial and
transportation disasters. The addition of a mobile regeneration system
such as designed and constructed in this project will further facilitate
the efforts of field workers and allow onsite destruction of the pollutants
from the environment.
David G. Stephan, Director
Hazardous Waste Engineering
Research Laboratory
111
-------
ABSTRACT
Activated carbon adsorption has become a standard procedure for the
cleanup of contaminated water streams. To facilitate such cleanup at
hazardous waste and spill sites, mobile carbon adsorption units have been
constructed and are now in use. Their primary drawback is the logistics
associated with the disposal of spent (contaminated) carbon and its
replenishment with fresh, active carbon.
This program was undertaken to assess the feasibility of designing and
building a mobile carbon regeneration unit - including an incinerator/
scrubber to destroy the offgases - for field use in conjunction with mobile
carbon adsorption systems. A system was designed and built based on tech-
nology developed in the earlier fabrication of a laboratory-sized regener-
ator and on an in-depth evaluation of factors affecting system design and
size. Housed in a standard van-type of trailer, the system meets all weight
and size limitations for over-the-road transportion. The system includes a
direct fired, rotating barrel kiln to thermally regenerate the carbon, an
incinerator and scrubber to destroy the desorbed materials and treat the
offgases, and a separator to reclaim the reactivated carbon granules.
Test runs using spent carbon from an onsite treatment of a spill were
quite successful. The carbon was returned to essentially 100% activity with
an 88% volume recovery. The unit has been delivered to the US EPA for their
use.
This report was submitted in fulfillment of Contract No. 68-03-2110 by
MSA Research Corporation under the sponsorship of the U.S. Environmental
Protection Agency. This report covers a period from August 6, 1974 to February
8, 1979, and work was completed as of October 15, 1981.
jf.
*?'
*s~
I
K
IV
-------
•—prTr—r--*-^*- i**i-ar.-T "
CONTENTS
Foreword ill
Abstract iv
Figures vi
Tables vi
1. INTRODUCTION 1
2. RESULTS AND CONCLUSIONS 3
3. PHASE I - SYSTEM DESIGN 5
TECHNOLOGY ASSESSMENT 5
Classification of Spill Chemical 5
Review of Potential Regeneration Systems 10
Direct and Indirect Heating 11
System Functions 12
Carbon Feed 12
Regeneration 13
Indirect Fired Rotary Kiln 18
Direct Fired Rotary Kiln 19
Incinerator 19
Scrubber 20
PRELIMINARY SYSTEM DESIGN 20
BENCH SCALE STUDIES 25
Bench Scale System ...... 27
Chemical Classification and Selection 27
Regeneration/Incineration Tests 28
Materials of Construction 30
4. PHASE II - COMPONENT SPECIFICATION DEVELOPMENT 31
KILN 31
INCINERATOR/AFTERBURNER 34
SCRUBBER 35
REGENERATOR POWER AND CONTROLS 37
TRAILER DESIGN 38
GENERAL REQUIREMENTS 39
5. PHASE III - PROCUREMENT, ASSEMBLY AND TEST 40
PROCUREMENT 40
ASSEMBLY 41
ASSEMBLY REVIEW 41
. SYSTEM CHECKOUT 41
-------
FIGURES
Number
1
2
. 3
4
5
6
7
8
9
10
11
Flow Diagram of Proposed Carbon Regeneration System 6
Mobile Regeneration Unit 21
Trailer Design 26
Enclosed Trailer 44
Trailer Open for System Operation 44
Profile of Kiln and Incinerator 45
Head End of Kiln 45
Scrubber arid Filtration Equipment 46
Tanks and Pumps Beneath Trailer Bed 46
Control Panel and Product Screener 47
Trailer in Transit 47
TABLES
1 Classification of Potential Spill Chemicals Relative to
Their Ease of Desorption From Activated Carbons During
Regeneration 8
2 Material and Heat Balance for Granular Carbon Regeneration
and Adsorbate Recovery 14
3 Effect of Direct and Indirect Heating on Gas Flows for
Thermal Des'orbtion . . . 15
4 Material and Heat Balance for Granular Carbon Reactivation . 16
5 Effect of Direct and Indirect Heating on Gas Flows During
Pyrolytic Regeneration 17
6 Effluent Gases for Regenerators Operated Under Different
Conditions 18
7 Incinerator Size Estimates for 0.5 Sec Residence Time at
760"C 20
8 Process Calculation Summary .... 23
9 Estimated Weight for Mobile Carbon Regeneration Unit for
45.4 kg/hr Direct Fired Rotary Kiln 24
10 Typical Concentration of Gases Exiting From the Laboratory
Scale Regenerator and Scrubber . 29
11 Kiln Design Operating Parameters 33
12 Scrubber Design Parameters 36
13 Operating Conditions of System 42
14 Iodine Numbers ' 43
VI
m
-------
SECTION 1
INTRODUCTION
One of the principal techniques now in wide use to decontaminate water is
granular activated carbon, which has the ability to adsorb selected classes of
chemicals preferentially from aqueous solutions or dispersions. Activated
carbon is characterized by a highly developed network of small pores and
interstices which contribute to its high sur'ace-to-volume ratio and,
consequently, to its high adsorptive capacity.
In spill situations where hazardous chemicals have entered the water
ecosystem, where materials are leaching from landfills or waste storage sites,
,., —— i" other similar situations, activated carbon treatment becomes a standard
! practice. To facilitate the cleanup at such sites, portable carbon adsorption
| systems have been developed in recent years. These are trailer or skid mounted
•! units which can be moved over the road or by air to the place where the cleanup
! is to be conducted.
These portable systems have proven quite versatile and effective. However,
they have a significant drawback—the logistic and administrative barriers
associated with the disposal of spent (contaminated) carbon and its replace-
ment. Large quantities of carbon are usually required for cleanup operations.
And though the carbon can technically be regenerated commercially it must
be moved to a facility having the necessary processing equipment, and
appropriate permits for carbon reactivation. Also, fresh carbon must be
moved to the field site to keep the cleanup operation active. This approach
to regeneration can be complicated If hazardous materials are involved
(such as PCB's or TCDD) for which there are strict regulations or prohibitions
affecting commercial reactivation. Commercial regeneration facility operators
are unwilling to operate their equipment routinely at the temperatures needed
to destroy PCB's and TCDD because of economic reasons. Further, the regeneration
market for carbon contaminated with these substances is not large enough to
justify the facility operator's incurring the public concern that would result if
the facility were to accept such substances. Because of these problems with
commercial reactivation, carbon used to treat PCB's or TCDD is currently being
disposed of in chemical landfills rather than being regenerated. Thus, the
toxics accumulated on the carbon are not destroyed, only stored, possibly to
enter the environment in the future.
Clearly, the utility of portable activated carbon systems (particularly
when used on toxic chemical contaminated carbon not commercially regenerable)
could be enhanced if regeneration of the carbon could be achieved at the
cleanup site. Since the adsorption system has been adapted to a mobile base,
it was not difficult to conceive of a regeneration process that could be
fitted onto a skid or mobile trailer bed.
-------
BVK-a^a
Host materials that can be adsorbed by carbon can be stripped by means of
cteam or other thermal treatments to regenerate the adsorptive capacity of the
carbon. The residue from this stripping then contains the adsorbed materials
(in an aqueous carrier) at a much higher concentration but in a iduch smaller
volume than in the original contaminated water. This residue must be disposed
of, however, which still may require transport to an acceptable facility. To
be most effective, the regeneration system should strip the adsorbed material
and convert it to a form suitable for disposal at the spill site.
In a prior Federally sponsored program (Juhola, A.J., "Laboratory
Investigation oC the Regeneration of Spent Activated Carbon", U.S. Public
Health Service, Contract No. 14-12-469, Final Report MSAR 70-184),
a laboratory-sized carbon regeneration unit was developed that used thermal
treatment to strip the adsorbed material from the carbon. This unit was
coupled to a gas-fired incinerator to provide high temperature thermal
destruction of the stripped material. The laboratory unit served as a start-
ing point for the design of a portable unit to provide onsite carbon regenera-
tion and waste material incineration. ,
The goal of the current program was to design, construct, and test a full
size mobile system for field regeneration of carbon. Several tasks were pur-
sued toward this goal. These included:
(a) an in-depth assessment of existing carbon regeneration technology;
(b) selection of representative pollutants for which such a system might
find use;
(c) preparation and evaluation of a preliminary design;
(d) laboratory-scale studies of the proposed design;
(e) modifications of the design into final specifications for fabr.i-
cation;
(f) fabrication of a mobile system; and
(g) shakedown testing of the unit prior to transfer to the US EPA for
further testing and field use.
The studies undertaken for each of these tasks are reported In the vari-
ous sections of this report.
-------
SECTION 2
RESULTS AND CONCLUSIONS
The program was successful in designing and fabricating & trailers-mounted
carbon regeneration system based on design calculations and testing of a
laboratory-scale system. The system consisted of a rotary kiln thermal regen-
erator, an incinerator or after burner to degrade the material stripped from
the carbon, and a scrubbing tower to remove undesirable materials from the
offgases.
The unit is self-contained for operation at a spill or hazardous waste
site when deployed near a source of fresh water and provided with fuel. As
constructed, the system is capable of regenerating 45.4 kg/hr of >40-mesh size
granular carbon with <20% carbon loss and restoration of >75% of the adsorptive
capacity.
Although indirect heating of such a kiln offers many advantages, weight
and size restrictions forced the selection of direct firing for the kiln. Even
then, an initial goal of a mobile system capable of regenerating 91 kg/hr of
carbon could not be achieved and the size was reduced by 50%.
System design parameters were imposed by trailer limitations of size and
weight and expected over-the-road stesses. Lightweight and resilient fabri-
cation was emphasized throughout, including light structural containment,
ceramic fiber thermal insulation, and flexible piping.
The system is constructed in a specially modified semi-trailer van that is
2.44-m-i-wide, 13.7-m-long, and 4.1-m-high when closed for storage or transit.
When fitted for transport by tractor-truck or rail piggy-back, the van meets
all applicable requirements, regulations, and conventional load limits.
Using the system and a carbon contaminated with Toxaphene and minor amounts
of other chlorinated hydrocarbons, regeneration gave an 88% volume yield of a
completely reactivated carbon based on iodine numbers.
-------
tyJBff'^F^Pi'y^^iTO^
A pilot run was made with an activated carbon material contaminated with
toxaphene (CjoHjoClfc) and minor quantities of other chlorinated hydrocarbons
Carbon loadings were 13% contaminant and 52% water. Regeneration gave an
83% volume yield of a completely reactivated material based upon iodine
numbers. Vent gas analysis showed CO and hydrocarbons to be below detectable
limits.
-------
SECTION 3
PHASE I - PRELIMINARY SYSTEM DESIGN
The first phase of the program was directed to the development of a data
bace which would permit the' design1' of" a field scale~portable carbon
regnerati'on system. It was divided into three tasks : a technology
assessment of carbon regeneration, a review of adsorbate/carbon
characteristics, and a preliminary system design.
TASK I - TECHNOLOGY ASSESSMENT
Work in this task was directed to the identification of those techniques
which could be considered for carbon regeneration and management of the
stripped waste and assessment of their suitability to a mobile system.
Based on the desired characteristics of a mobile carbon regeneration
system and the expected behavior of the major categories of spill-chemicals
during thermal regeneration, the major components of a basic system are
given in the schematic flow diagram of Figure 1. This system is based upon
the laboratory unit previously built by MSA and includes the following
representative operations,
1. Carbon feed
2. Regeneration/activation
3. Carbon recovery and sizing
4. Condensation
5. Decantation/distillation or other method of
concentration
6. Thermal incineration
7. Scrubbing and scrubber liquid treatment.
Of these major system operations, the regenerator/activator, is most
j critical. Other component requirements are dependent on the design and
.1 characteristics of the regeneration/activation operation. Illustrative
calculations were made to define this dependency and to allow preliminary
assessment of candidate systems.
:"' ' Classification of Spill Chemical
vj
Surveys conducted on spillage of polluting materials show a great
variety of chemical substances that may become objects of water cleanup
operations. The objective of this part of the study was to evaluate this
great variety of chemicals relative to their ease or difficulty of
-------
^^iK:j^r{;^r^.^^^-*"*'.^^c?«^^r-";^r^^l^^*^^i^^ "~^ "* ^^"^^^"^ i^:>->.'^"7-^T~";j.^
(•/ •
Off-gas Line f
Noncondensable
Gas Line
_ 1
Thermal
Regenerator
^Granular
I Water Slurry of
Spent Carbon,
Silt and Other
Foreign Material
x- ' Powder
As Above
Filter
Indirect Heat t
Propane I
Combustion
On-size
jgenerat
Carbon
Spray
Sol'n '
Steam
Condenser f
Condensabjes j
Direct Heat
Propane
Combustion
urr-size / \/Vater x.uHea,y,ieir
Material , .nLaf soi'n than Water
Lighter 001 n rhom;A«i
Vent Gas
than
Water
Chemical
Scrubber
Incinerator
Air or
Propane Combustion
Gas
Figure I. Flow diagram of proposed carbon regeneration system.
-------
desorption from Che carbon and to determine what steps must be taken to
collect the desorbed chemical; and avoid further air or water pollution.
Spill chemicals were classified into five groups according to their
distinctive regenerative patterns. Table 1 presents a partial listing of
each group showing typical chemicals and range of properties.
Groups 1 and 2 are relatively volatile and have small molar volumes.
Both properties correlate with relatively low to moderate adsorptive
affinity, hence these two groups will be desorbed quite readily. At a 350°C
final regeneration temperature, recovery of over 80% of the adsorptive
capacity can be expected. Some decomposition of the heavier members of the
• groups' can be expected to occur.
Group 2 differs from group 1 in that each chemical contains chlorine,
sulfur or nitrogen. For group 1, an incineration step to convert the
desorbed- chemical to C02 and H20 is sufficient to avoid air pollution. For
group 2 a water or caustic scrubber downstream of the incinerator is
required to remove the HCl, S02 or N02 from the exhaust gas stream.
Groups 3 and 4 are considerably less volatile than groups 1 and 2.
Since they do not desorb readily, higher regeneration temperatures are
required and considerable decomposition of the adsorbed chemical can be
expected. One of the decomposition products will be free carbon. Its
accumulation over a few adsorption-regeneration cycles can drastically
decrease the carbon adsorptive capacity. To remove the free carbon, by
steam or C02 oxidation, final regeneration temperatures near 930"C are
required. Thus, for groups 3 and 4, a regenerator is required that can
withstand a considerably higher temperature than that necessary for groups 1
and 2. Similar to groups 1 and 2, group 4 differs from 3 in that the
chemicals contain chlorine, sulfur and nitrogen, requiring water or caustic
scrubbing after incineration to avoid air pollution.
Group 5 resembles groups 3 and 4 in its regenerative behavior except
that the group 5 chemicals contain nonvolatile metals and leave a metallic
oxide or hydroxide residues. Alkaline metals form hydroxides and phosphorus
forms an acid. These leach out to a large extent during the adsorption
phase. Calcium, mercury, lead, copper, nickel and chromium form oxides or
hydroxides which do not leach out. Mercuric chloride and lead acetate are
strongly adsorbed, but are difficult to remove. Mercuric chloride sublimes
at 270°C in the free state but may not when adsorbed. In this case,
regeneration may not be beneficial. Lead acetate can be expected to
decompose producing a lower lead oxide with a molar volume change from 100
to 28 cc/vol. After a second adsorption-regeneration cycle, further
regeneration cycles may be of no benefit. Regeneration attempts on carbons
containing copper sulfate, nickel chloride, chromic chloride and similar
inorganic salts can be expected to be ineffective.
-------
i. TABLE 1. CLASSIFICATION OF POTENTIAL SPILL CHEMICALS RELATIVE TO THEIR
<- EASE OF DESORPTION FROM ACTIVATED CARBONS DURING REGENERATION
.
'• . . .. .^Group Chemical
t- Acetaldehyde
'• Acetic acid
: . Acrolein
= „. - Allyl alcohol
• . Propanone
J ,' 1 Methyl acetate
> ' Propyl alcohol
• ; Benzene
'•' . 2-Butanone
; Toluene
<-.> Pentane
:".
;
^
L
•'
';.
i :
?••
il.
l't ".
?
'•V
fc.
V;
\^
•V
vt-
X
j£
rv
5>
'"t'
-T-.
;3
I
Hexane
Acetonitrile
Acrylonitrile
Dichl or ome thane
Carbon disulfide
Propanenitrile
Dimethyl sulfide
1,2, Dichloroethane
2 Chloroform
Allyl chloride
Pyridine
Propyl amine
Propanethiol
Carbon tetrachloride
Diethylamine
Ethylene glycol
Benzyl alcohol
Hydroquinone
Maleic acid
Ben zoic acid
Phenyl acetate
Methyl pentylacet ate
3 Octanediol
Carbitol
Butyl carbitol
Phthalic acid
Ethylhesyl acetate
Glucose
Sucrose
^' • "' ' .'- • ^^»-
Density
(g/cc)
0.78
1.05
0.86
0.85
0.79
0.97
0.78
0.88
0.80
0.87
0.63
0.66
0.78
0.80
1.33
1.26
0.77
0.85
1.26
1.49
0.94
0.98
0.72
0.84
1.59
0.71
1.18
1.04
1.32
1.59
1.32
1.09
0.87
1.12
0.96
1.59
0.87
1.56
1.59
Solubility
in water
(g/g at 20°C)
0.4
0.33
0.0007
0.23
0.0005
0.0004
0.0001
0.075
0.020
0.0012
0.10
insol .
0.0087
0.008
insol .
very sol.
slight
0.0008
very
Solubility
in water
(g/g at 20eC)
0.038
0.012
0.788
0.0029
slight
slight
slight
soluble
0.0070
insol.
0.49
2.04
8
Liquid
molar
Vapor
pressure
vol(cc/moleXmm Hg at 25°C)
56
57
65
68
73
76
77
88
90
106
114
130
53
54
64
66
71
73
78
80
81
81
82
90
97
103
Molecular
weight
(g/mole)
62
108
110
116
122
136
144
146
106
162
166
172
180
342
750
15
270
24
170
160
19
97
100
28
490
150
90
190
400
350
45
480
220
130
350
18
240
170
105
230
Boiling or
melting
point C°C)
198 BP
205 BP
258 BP
138 MP
249 BP
266 BP
148 BP
172 BP
245 BP
231 BP
206 ht>
199 BP
146 MP
206 MP
(continued)
• • . -. ' i
-------
TABLE 1. (CONTINUTED) CLASSIFICATION OF POTENTIAL SPILL CHEMICALS
RELATIVE TO THEIR EASE OF DESORPTION FROM ACTIVATED CARBONS DURING
REGENERATION
Solubility
Density in water
Group Chemical
Ammonium acetate
Benzonitrile
Benzamidine
Benzyl Chloride
Dichlorobenzene
Benzenesulfonic acid
Dicambia
4 2,4, D acid
Ammonium lauryl sulfate
Chlordane
Ethyl enediamine-
tetraacetic acid
Diazinon
DDT
Aldrin
Trimethyl phospate
Calcium acetate
Mercuric chloride
Soli urn stearate
Mercuric thiocyanate
Lead thiosulfate
Lead thiocyanate
5 Lead acetate
Hexadecyl sulfate,
sodium salt
Mercuric sulfate
Zinc phenyl sulfate
Potassium oleate
(g/cc)
1.17
1.01
1.10
1.30
1.57
1.25
5.44
7.56
5.18
3.82
3.25
7.56
™~™
Molecular Boiling or
weight melting
(g/g at 20°C) (g/mole) point CO
1.48
slight
soluble
insol .
<0.0005
soluble
slight
slight
insol .
slight
insol .
insol.
very sol.
0.37
0.069
sol
0.0007
0.0003
0.00005
0.44
soluble
0.0006
0.655
soluble
77
103
120
126
147
158
220
221
267
248
292
304
334
364
140
158
271
306
317
319
323
325
344
497
556
603
114 MP
191 BP
80 MP
179 BP
179 BP
525 MP
138 MP
322 BP
108 MP
104 MP
197 BP
302 BP
280 MP
95 MP
I-
ft.
-------
Review of Potential Regeneration Systems
Regeneration of activated carbon requires the removal of the sorbed
•material. Three methods are available; thermal, solvent extraction and
vacuum. The thermal method is the most widely used.
Solvent extraction uses leaching by acids, bases or solvent. It
requires either chemical reaction and product solubility in the aqueous
solution or high solubility in the solvent. Of the great variety of
potential spill-chemicals, only a small portion can be effectively desorbed
by this method. In addition, the leaching -agent -leaves the carbon in a
spent condition and hence must in turn be removed with water (for acids or
bases) or with heat or vacuum (for solvents).
Vacuum desorption is effective for some hard-to-desorb chemicals if the
carbon is relatively dry and some heat can be applied. It is assumed in
vacuum desorption that the deposited species can be removed by lowering the
ambient pressure with no decomposition in the desorption process. The high
water content in the carbon imposes a heavy load on the vacuum system. The
process is usually restricted to regeneration of carbons used to purify
gases.
Thermal regeneration offers the greatest versatility. It is able to
cope with a wide range of materials from those easy to desorb to those most
difficult to desorb, including those that may decompose in place. The
former can often be vaporized along with the water. The other materials
involve pyrolysis of the sorbed material, converting them to a volatile
portion, and a carbon residue. To control the carbon, pyrolytic
regeneration is performed at temperatures as high as 871°C. The carbon
residue is oxidized to CO and hydrogen by r^eans of steam or carbon dioxide.
A diluent gas is often necessary to prevent excessive loss of the base
carbon material when exposed to these reactivation conditions.
Thermal regenerators can be designed to cover a range of desorption
conditions with simultaneous activation. Equipment varies primarily in the
way in which heat is applied (direct or indirect firing) and the way in
which the carbon is physically moved through the heated zone (rotary tube,
multiple hearths or fluidized bed) . There are specially designed reactors
that use microwave heating; or wet air oxidation (the Zimpro process). The
latter two are developmental and not commercially proven.
The rotary tube furnace or kiln utilizes internal vanes that lift the
carbon part way up the wall, where it then falls back to the bottom. This
effects a stirring condition while the carbon moves through the tube.
Indirect firing involves heating on the outside. It has relatively poor
heat transfer and a more bulky construction than a direct-fired unit where
hot gases are passed directly through the tube and directly contact the
carbon. As the carbon passes through the tube, it is dried in the first
portion, regeneration is completed in the central portion and reactivation
(if necessary) occurs in the latter protion. Temperature regimes are varied
to meet the requirements for removal of specific sorbed materials.
10
-------
A multiple hearth furnace has a vertical stack of platforms with wiping
arms. The carbon on each platform is pushed toward an opening where it drops
to the next lower level. The platforms are alternately arranged so that
carbon moves from the center to the edge of one platform and from the edge to
the center of the next lower platform. Gases pass upward through the furnace.
Thermal distortion of the multiple hearths is a known problem. For that
reason, this type of construction would not appear to be attractive for a
mobile unit.
In a fluidized bed, streams of gas and carbon flow countercurrently.
The carbon tends to float as a layer in the gas. Fluidized bed processes
provide excellent heat transfer and a minimum of seals and moving parts
compared to the rotary tube and multiple hearth units. A fluidized bed is a
high energy consumer because of the increased pressure drop involved in
maintaining a fluid bed condition. There is also the possibilty of increased
attrition of the granular material, necessitating particulate removal equip-
ment downstream. Multiple beds may also be required to provide the appro-
priate zones and conditions for drying, regeneration and reactivation. Although
some development work has been done, there has been no commercial application
of fluidized bed techniques to carbon regeneration/reactivation.
Treatment of the offgases will probable be required regardless of the
type of regenerator used. There are differences to be recognized between the
direct and indirect firing units with respect to such treatment. An indirect
fired unit adds only heat to the interior of the system, and offgases are
evaporated moisture and desorbed pollutants only. To help control the process,
a sweep gas would most probably be required. The sweep gas could be steam,
giving a condensable material that could be removed prior to treament of
desorbed pollutants or their decomposition products. In direct firing, the
flue gas will contain a significant amount of non-condensables (CO, C02. N£,
some 02) that will add to the total gas that must pass through the treatment
system after condensation of steam.
On the basis of the preliminary review, thermal activation clearly
appeared to be the only logical choice, with the further conclusion that only
rotary type units could be built on a scale small enough to fit the size and
weight limitations of an over-the-road trailer.
Two basic designs were considered — a rotary kiln and a rotary hearth
furnace. Initially there was interest from manufacturers of rotary hearth
furnaces to investigate design modifications which would allow trailer mounting.
WJth time, however, it became clear that if modifications were possible at all,
they would take considerable time and effort. Rotary kilns appeared better
able to meet near term requirements and the program was directed to that option.
Direct and Indirect Heating
To evaluate the potential of direct and indirect firing modes on system
design, preliminary calculations were made for four modes of operation. For
11
-------
these calculations, a process throughput of 91 kg/hr (dry carbon basis) was
assumed. These modes of operation are defined as follows:
1. Combination of direct-fired regenerator arid carbon loaded (saturated)
with groups 4 and 5 chemicals.
2. Combination of indirect-fired regenerator and carbon loaded with
groups 4 and 5 chemicals.
3. Combination of direct-fired regenerator and carbon loaded with groups
1, 2 and 3 chemicals.
4. Combination of indirect-fired regenerator and carbon loaded with
groups 1, 2 and 3 chemicals.
In mode 1 high temperatures up to 927°C are required to give an active
carbon product. Three reactions occur: (1) vaporization of th» water; (2)
pyrolysis of the chemical, releasing a volatile product and depositing free
carbon and (3) steam oxidation of the free carbon (steam activation).
In mode 2, temperatures and reactions are the same as in mode 1 except
that indirect firing considerably reduces the volume of gases that are
emitted from the regenerator. As a result, the downstream components of the
system are smaller with more efficient condensation recovery.
In mode 3, lower temperatures are used, the maximum being about 316°C.
In this case only vaporization of the water and desorption of the chemical
are involved. The large volume of flue gas from direct firing limits the
efficiency of condensation recovery.
In mode 4, the temperature and reactions are the same as mode 3 except
that with indirect firing, the volume of noncondensable gases in the regen-
erator effluent is very small. When the incoming carbon slurry is clean
(i.e. it contains only the carbon, clean water and the adsorbed chemical),
the operating conditions can be very close to those used in solvent recovery.
A virtually complete recovery of the chemical is possible and the incinerator
and scrubber may not be necessary to the process.
System Functions
The various functions the system is required to perform are addressed in
the following paragraphs in the sequence in which they are performed.
Carbon Feed—
The spent carbon is delivered to the system as a 5% to 30% carbon-water
slurry. To reduce the water load on the regenerator, excess water is first
removed by mechanical procedures.
Granular carbon slurvy can be extracted by transferring the slurry into
the feed tank which has provisions for drainage. Drainage reduces water con-
tent to about 40%, or 60% by weight dry carbon. The drained carbon is then
12
-------
fed into the regenerator by an upward Inclined screw feeder that allows
further drainage during feeding. . •'.:' .
• >'•** ,"'.-'
For design purposes, the feed rate to the regenerator is:
Carbon (granular) - 91 kg/hr
Adsorbed chemical - 9.1 kg/hr
Water - 65.8 kg/hr
Other adsorbates - variable but small
Silt - variable, unknown
Powdered carbon as a feed would consist of 10% or less of carbon as an
aqueous slurry. In order .".o make carbon manageable in a screw feeder, the
water content of the mixture would have to be reduced from 90% to 60% or
lower. This can be done by centrifuging or filtering and drawing air through
the filter cake. The dried centrifuge or filter cake can then be broken up
and fed as lumps into the regenerator.
For design purposes, the feed rate would be:
Carbon (powdered) - 91 kg/hr
Adsorbed chemical - 9.1 kg/hr
• Water - 150 kg/hr
; Other adsorbates - variable but small
: Silt - variable, unknown
Details on powdered carbon feed are given here because evidence from MSA
: " work on carbon spent in tertiary sewage treatment (Juhola, 1970) showed that
' powdered carbons could be regenerated in a modified indirect-fired rotary
I tube.
>.
i Regeneration—
(
!. The regenerator should perform the following functions and must perform
t the first three:
r
| a. Vaporize water from carbon;
f
| b. Desorb the adsorbed chemical if group 1, 2, and/or 3;
c. Pyrolyze the adsorbed chemicals if group 4 and/or 5;
]
f d. Deeorb group 1, 2, or 3 with minimum loss in recovery and pyrolyze
i group 4 or 5 in a mixture of group 1, 2, or 3 with group 4 or 5
present; and
•
e. Reactivate carbon with loss of carbon not over 10% or activity not
f over 30%.
Material and heat balance calculations were made for two situations:
(1) only group 1, 2 or 3 chemicals are present and completely desorbed
(Table 2), and (2) only group 4 or 5 chemicals are present with carbon
reactivated to 100Z original activity at 5% carbon weight loss (Table 3).
13
-------
TABLE 2. MATERIAL AND HEAT BALANCE FOR GRANULAR CARBON
REGENERATION AND ADSORBATE RECOVERY
INPUT
\
Material
Act. carbon
Adsorbate
Water
Steam sweep
AH= -1060 ral/mole
Propane
AH= - 21,700 cal/mole
Air
kg/hr
91.0
9.1
65.8
22.8
(22.8)
7.8
(7.8)
126.6
323.1
°C
16
16
16
104
16
16
Kcal/hr
•
_«~
832—j
13,457-J"
93,492
—
107,781
Heat %
13-3
86.7
100.0
Act. carbon
Steam from carbon
AH= +1060 cal/mole
Desorbed vapor
AH= +2760 cal/mole
Steam sweep
AH= +1060 cal/mole
Heat loss, 10%
Propane flut: gas
N2
C02
H20
AH= +1060 cal/mole
91.0
65.8
(65.8)
9.1
(9.1)
22.8
(22.8)
OUTPUT
316
149
149
149
316
316
316
322.8
9,551
3,956
39,060
302
13,910
1,336
13,457
8,064
7,409
1,109
1,688
7,510
107,352
8.9
40.0
27.0
24.0
99.9
14
-------
For the adsorbate recovery operation (Table 2), the following operating
temperatures were assumed to be adequate to accomplish the desorptlon of the
chemicals:
Heat Source Temp. (°C)
Base (ambient) 16
Steam input 105
Act. carbon output 316
Effluent steam and chemical 149
Based on the data shown in Table 2 and the assumed feed composition,
about 40% of the heat requirement is for water desorption and 27% for chemical
desorption.
The heating requirements for regeneration by desorption only can be met
by direct heating or Indirect heating. The means of heating affects the
offgas flow, as shown in Table 3:
TABLE 3. EFFECT OF DIRECT AND INDIRECT HEATING ON GAS
FLOWS FOR THERMAL DESORPTION
Flow (kg/hr)
Gas Direct Indirect
Steam from carbon
Chemical desorbed
Steam sweep
Combustion steam
Combustion C02
Combustion No
65.7
9.1
—
12.7
23.1
98.4
65.7
9.1
22.7
—
— —
Totals 209.0 97.5
When the heating is direct, the offgas flow to the condenser is about 209
kg/hr. Of this, 121.5 kg/hr are noncondensable CO2 and N£. These gases will
continue downstream and carry with them considerable amounts of group 1 and 2
chemicals. When group 2 chemicals are present, incineration becomes a neces-
sary next step to convert the chlorides along with nitrogen and sulfur com-
pounds to a form that can be scrubbed out ot the gas stream by water or
caustic solution.
When heating is indirect, the offgas flow to the condenser Is 97.5 kg/hr.
If the gas system is devoid of foreign noncondensable gases, virtually all of
the chemical and water would be condensed with very little or no gas going to
the incinerator.
When group 4 or 5 chemicals are present, pyrolysis and reactivation are
part of the regeneration process. Materials and heat balance for this
situation are shown is Table 4. In this case, because of higher temperatures,
a 45% carbon loss is assumed. The operating, temperatures noted below were
assumed to be necessary to accomplish pyrolysis and carbon regeneration.
15
-------
v.' " ..;•".
f..
;
^
I
,1
[
:i
i
i
.'
|l
I
ti
i
•'
•.
: ._
;
•
:' • '
- ' ]
1
I
: ;|
.'
y
w
i
• • :
TABLE 4. MATERIAL AND HEAT BALANCE FOR GRANULAR CARBON
Material
Act. carbon
Adsorbate (-CH2-)
Water
Steam for acct.
AH= -1060 cal/mole
Propane
AH= - 21,700 cal/mole
Air
Act. carbon
Steam from carbon
AH= +1060 cal/mole
Desorbed gas
(-CH2-)
AH= +2760 cal/mole
CO
H2
Free carbon
AH= +4800 cal/mole
CO
H2
Act carbon loss
AH= +4800 cal/mole
CO
H2
Excess Act steam
AH= +1060 cal/mole
Heat loss, 7.5%
Propane flue gas
N2
C02
H20
AH- +1060 cal/mole
.''•'••.--'' •
REACTIVATION
INPUT
kg/hr °C Kcal/hr Heat %
91.0 16
9.1 16
65.8 16
22.8 104 832—, g-9
(22.8) — 13,457-J
12.1 16
(12.1) — 145,656 91.0
199.0 16
399.8 159,945 99.9
OUTPUT
86 899 34,272 21.5
65.8 316 8,820-, 30.0
(65.8) — 39,060-J
(5.8) 8,820-,
11.4 316 857
1.6 316 1,462 ;
(3.3) — 8,870 ;
7.8 316 580 23.3 -
0.5 316 479 _ <
4.5 — 12,096 ;
10.6 316 781
0.8 316 680
3.8 316 504 ;
3.8 2,092—1
11,844-1 ;
'*.
154.2 316 11,592 _
35.8 316 2,520 25.1 ;
19.5 316 2,621
(19.5) — 11.592-1
402.3 159,542 99.9
16 " : . . i
i
-------
^Y-ft^P'?15?^^
' ' " " •' ' "-""--"• •"-' •'- ' '--' '"•""' ' '
Heat Source Temperature (°C )
Base (Ambient) 16
Steam input 105
Act. carbon output 899
Effluent gases 316
In this case per Table 4, 30% of the heat requirement is for water
desorption and 23% for chemical pyrolysis and carbon reactivation. It also
assumes a 5% carbon loss.
For pyrolysis and activation, as was the case when desorption was the
only concern, the heating requirements can be met by direct or indirect
methods. The affect on the amount of offgas flows is shown in Table 5:
TABLE 5. EFFECT OF DIRECT AND INDIRECT HEATING ON GAS FLOWS DURING
PYROLYTIC REGENERATION
Flow (kg/hr)
Gas Direct Indirect
Steam from carbon 65.7 65.7
Chem. decorap. products
CO 19.1 19.1
H2 2.3 2.3
Act. carbon loss (5%)
CO 10.4 10.4
H2 0.9 0.9
Excess steam — 3.6
Combustion steam
Combustion C02
Combustion N2
Totals 307.9 102.0
When the heating is direct, the offgas flow to the condenser is about
308 kg/hr. Of this, 218 kg/hr are noncondensable gases (C02, CO, H2 and
N2) and will pass through the condenser. For ease of calculation, it was
assumed that the gaseous pyrolysis products reacted with the steam forming
CO and H2. in actual practice, this happens only partially. If NOx, SOX
or chlorides are present in the adsorbed chemical, these groups would be
partially emitted as inorganic compounds. An incinerator and scrubber are
needed downstream to strip them from the vent gas.
When the heating is indirect, the offgas flow to the condenser is
about 102 kg/hr, of which 32.7 kg/hr is noncondensable and combustible CO
and H2. Any NOx, SO* or chlorides would be partially or completely
decomposed to release either the inorganic acid gas or organic decomposition
product which must be further decomposed in the incinerator.
17
-------
Powdered carbon will have a water content over twice as large as that
for the grsnular carbons. The heat requirement will be correspondingly
higher, but both desorption and reactivation will require much milder
conditions. It can be expected that the regeneration condition would not be
much greater than those for the granular carbons. Powdered carbon is not
often used in wastewater treatment because-of high physical losses in air
and water streams.
The four ways -of carrying out the regeneration process will have a
bearing on the size of tha condenser and other components downstream of the
condenser as shown in Table 6.
TABLE 6. EFFLUENT GASES FROM REGENERATORS OPERATED
UNDER DIFFERENT CONDITIONS
Mode Process Condensable Noncondensable
(kg/hr) (kg/hr)
1 Direct heat, adsorbate ri
pyrolysis and carbon M
reactivation 86.2 222.3 j.;
i:
2 Indirect heat, adsorbate
pyrolysis, and carbon
reactivation 68.0 31.8
3 Direct heat and adsorbate 86.2 121.7
desorption only
4 Indirect heat and adsorbate
desorption only 95 .3 nil
Gases from nodes 1 and 2, pyrolysis and reactivation, leave the
.regenerator at 315°C and those from modes 3 and 4 at about 149°C. For
powdered carbon, the condensable input rate to the condenser would be over
150 kg/hr, primarily water.
Indirect Fired Rotary Kiln—
The most favorable conditions for spilled chemical reccvery/detoxificatlon
and control of air and water pollution appeared to lie with an indirect fired
regenerator. With indirect firing, the combustion gas;s are not contaminated
by the adsorbate or its decomposition products and thus r?r be emitted directly
to the atmosphere (assuming propane or other clean buvning fuel) without
further treatment. The contaminated- gas stream onslsts only of desorbed
materials, steam and a small volume of sweep gas (whic'n may be steam or flue
gas). This allows recovery by condensation and incineration/scrubbing of a
small volume of non-condensibles or, alternatively, incineration/scrubbing of
the total stream with minimally sized equipment.
18
-------
Indirect fired rotary kilns are employed in a variety of industrial
operations and design data are readily available. The major design criteria
are the kiln areas and volume required for efficient heat transfer and the
solids loading, as percent of total volume. In industrial practice, volume
loading ranges from about 2 or 3% to perhaps as high as 15%, depending on the
moisture content and flow properties of the solid being handled. Based on
experience with rotary tube regeneration, a loading of 10% appears feasible
for granular carbon having a drained moisture content of about 40%.
At 10% loading, the minimum size of rotary kiln that could handle both
regeneration and activation of 91 kg/hr carbon (dry basis) would be a tuba
of 0.6 m inside diameter with a 4.6 in heated length. From..discussions with
manufacturers of this type of equipment, it was estimated that a complete
kiln of this throughput would have overall dimensions of 2.1 m x 2.1 m x
10.7 m and would weigh approximately 22.7 kkg (25 tons). Estimated minimum
delivery time was about one year.
This is both too heavy and too large to fit on the semitrailer and
still leave work space and room for the other components of the system. It
was apparent that either the regeneration rate would have to be drastically
reduced from the 91 kg/hr rate or chemical recovery would have to be sacri-
ficed. Since the primary objective is to reduce environmental damage uti-
lizing carbon adsorption the correct decision appears to be to maintain the
regeneration rate level as high as possible. (Chemical recovery is of
secondary importance).
Direct Fired Rotary Kiln—
Direct fired kilns can be operated at higher throughputs than can be
attained with indirect fired units of similar dimensions because of more
efficient heat transfer. For this reason, direct fired kilns have more sizes
and designs commercially available.
Two quotations were obtained from a well known manufacturer; one for a 91
kg/hr (dry) carbon and one f c a 45.4 kg/hr carbon system. Both are packaged
systems consisting of the rotary kiln proper, dewatering feed screw, after-
burner, offgas scrubber, and quench tank. The smaller kiln with a 380 mm
inside diameter refractor>-lined tube 3.80 m long, appeared capable of being
fitted onto a semitrailer with all the other required equipment.
In view of the space limitations, the smaller (45.4 kg/hr) direct fired
rotary kiln was selected as most appropriate for further analysis.
Incinerator—
Noncondensable gases entering the incinerator ov afterburner are heated to
760° C for 0.5 sec in an oxidizing atmosphere. The size of the incinerator
can vary considerably depending on the regeneration processes used upstream.
Of the 222 kg/hr of noncondensable gases produced in mode 1, about 15% will
be combustible gases, consisting of spill chemical, CO and U2« Of the 31.8
kg/hr, mode 2, most is combustible with a high percentage of CO and H£. Of
the 121.7 kg/hr, mode 3, only about 3% will be combustible gas. In each case,
19
-------
air with or. without supplementary heating is required to raise the tempera-
ture to 760°C to oxidize the gases<
Table 7 gives estimates of total final gas volume in the incinerator at
760°C STP and anticipated interior volume when residence time is 0.5 sec.
TABLE 7. INCINERATOR SIZE ESTIMATES FOR 0.5 SEC
RESIDENCE TIME AT 760°C
Process
No.
1
2
3
Gas Volume Incinerator Volume
m^/min m^/min
at STP
13.3
5.0
2.5
at 760°C
50.4
19.0
10.7
m3
0.4
0.2
0.1
In calculating the gas volumes for modes 2 and 3, it was assumed that
about one half of the desorbed chemical passes through the condenser because
of the high volume of noncondensable gases. For heating calculations, the
desorbed chemical was assumed to be composed of -CU2~ groups. In each
process, it is also assumed that enough combustibles are present to make the
incinerator self supporting, i.e., no supplementary heat is required, even
if the incinerator has to operate at temperatures above 760°C. At the
postulated temperatures and 0.5 sec residence time, it is expected that
well over 90% of the gases will be decomposed to C02 and H^O, and HC1, N02,
and SC>2 if chlorine, nitrogen, and sulfur are present.
The probable amount of CO2 emitted from each mode, assuming 50% of the
organics in modes 1 and 3 are condensed out, will be as follows:
Mode 1-82.1 kg/hr
Mode 2 - 46.3 kg/hr
Mode 3 - 37.2 kg/hr
Scrubber—
The combustion products from the incinerator are passed through a scrubber
containing water or a 20% caustic solution to remove any UC1, NO, 502 or
other acid gas present. The gas volumes entering the incinerator will be as
given in column 2 or 3 in Table 7.
TASK II - PRELIMINARY SYSTEM DESIGN
Design studies were concentrated on a direct fired rotary kiln system.
The preliminary design calculations for carbon regeneration/activation were
based on a system throughput of 45.4 kg/hr (dry regenerated carbon), assuming
5% carbon loss in processing. The carbon feed was assumed to have an average
moisture content of 40% and an adsorbate loading of 0.1 kg per kg of carbon.
A commercial kiln designed to handle 45.4 kg/hr of carbon for sugar puri-
fication was used as the basis for design. By modifying the equipment
arrangement, a suitable spatial arrangement was devised to allow trailer
20
-------
mounting. A schematic of this arrangement Is shown In Figure 2. The
Incinerator Is mounted directly above the kiln tube for space saving.
Calculations based on the capacity of the kiln tube size (380 mm inside
diameter (ID) x 3.80 m length) would Indicate the feasibility of handling
more than 45.4 kg/hr of granular carbon. Heat transfer considerations may,
however, limit operation to near design capacity. Availability data on
direct fired heat transfer are not sufficient to allow accurate prediction.
Major factors such as initial moisture content and ease of regeneration
(final temperature requirements) will significantly affect throughput.
A rotary tube of 380 mm by 3.80 m long will provide a holdup In the
range of 10 to 15% of tube volume, and retention times in the range from 0.5
to 0.75 hr. Previous data on granular carbon regeneration/activation in-
dicated these retention times are sufficient for complete reactivation.
Retention times and holdup can be varied however by adjustment of tube slope
and rotation rates.
After the kiln size and system layout were defined, regeneration/activa-
tion process design calculations were completed for several different cases
of operation of a direct fired rotary kiln. The calculation summary for two
of these cases Is given in Table 8. The major factor in process requirements
and sizing of downstream components (incinerator, etc) is the physical form
of the water required to cool the hot furnace gases and provide steam for
activation. Lower fuel requirements, gas velocities and gas flow rates are
achieved by Injecting liquid water rather than steam. Both heat of vapor-
ization and the sensible heat of the steam can be utilized.
Case I conditions, liquid water injection, were selected for continued
design development. The incinerator required for treatment of the kiln off-
gas at 982°C and 0.5 sec residence time was calculated to be 0.23 m3 in
volume, 305 mm inside diameter by about 3.4 m long.
Manufacturer's data on lightweight, high temperature Insulation were
reviewed and kiln tube weight tradeoff calculations were made. Alumina-
silica fiber insulation with desirable heat transfer and mechanical prop-
erties Is available in both standard and custom fabricated shapes. At
a bulk density of 0.45 g/cm^, this Insulation provides considerable weight
savings over clay-based refactorles with densities of the order of 2 g/cm^.
A 76 mm layer of this type of insulation enclosed within a 457 mm diameter
steel pipe was estimated as being adequate to insulate and allow mounting
the incinerator above the kiln.
Using these design calculations along with weights of the supporting
equipment, a weight analysis was made. The values employed are given in
Table 9. The total weight of 42,185 kg is clearly in excess of the 33,113
kg trailer limit for over the road operation without special permits.
Consequently, the design was reviewed to find areas where weight could be
reduced and/or eliminated. These areas were identified as the kiln feed,
the scrubber and kiln barrel accessories.
21
-------
IO
y
8
1
2.14m
1
' - 1
3
GG
5
f
b
9^^^X ^^^ ^^^/ J"^
II 12 13 14
16
i 10 i LJUU
R7m - .1 , .,_ .. , .
Equipment
I. Generator (electric)
2. Slurry pump
3. Dewatering screw
4. Cond water pump
5. Kiln
6. Quench tank
7. Elevator
8. Dryer
9. Screener
10. Panel board
II. Steam generator
12. Still
13. Scrubber
14. Scrubber pump
15, Still feed pump
16. Transfer pump
17. Transfer pump
17
Figure 2. Mobile Regeneration Unit (45.4 kg/hr Carbon).
-------
TABLE 8. PROCESS CALCULATION SUMMARY*
Case I Case II
Item offgas at 315°C offgas at 315°C
water added steam added
Propane Required, kg/hr 10.8 40.2
Inlet Gas Velocity, in/sec 2.7 13.7
Heat Load, kcal/hr 129,600 485,000
Offgas Flow Rate, m3/hr 606 2,566
Offgae Velocity, m/sec 1.5 6.2
Steam or Water Added, kg/hr 49.0 420
*Basis: 45.4 kg/hr carbon output; 2.3 kg/hr carbon loss; 4.5 kg/hr adsorbate;
40% water content in feed; activation temperature 899°C; kiln furnace
temperature 1093°C
23
-------
TABLE 9. ESTIMATED WEIGHT FOR MOBILE CARBON
REGENERATION UNIT WITH DIRECT FIRED ROTARY KILN
Item Estimaced Weight
. (kg)
Diesel Generator, 50 KW 1,814.4
Rotary kiln, 380 mm (I.D.) x 457 mm (O.D.) x 3.80 m (Long) 11,340
Dewatertng feed screw
Afterburner/Incinerator
Offgas scrubber
Carbon quench tanic
Process Equipment 5,443.2
Elevator dryer
Screener
Auxiliary scrubber
Still
Pump (5)
Control panel
Misc. Equipment 11,340
Steam generator
Holding bags and hose
Trolley hoists and winches
Lighting and ventilation
Piping (material)
Fire and safety equipment
Special framing, walkways, etc.
Tractor
Trailer, 13.7 m long
Total
7,257.6
4,989.6
42,184.8
24
-------
The initial kiln feed system proposed for handling slurries or wet feeds
employed a dewatering screw feeder. Consideration of potential field sit-
uations, volumetric feed rates, and granular carbon drainage rates, showed
the possibility of completely eliminating the requirement of a dewatering
screw feeder. The initial draining of the carbon feed could be achieved
external to the trailer mounted system. The drained carbon could then be
transferred to a feed system on the trailer. This would consist simply of a
feed hopper and associated feed screw to transfer the drained carbon to the
kiln. Model studies indicate this modified hopper and feed screw would
extend only about 1 m beyond the feed housing of the kiln versus 3 m for the
• dewatering screw.
\
I j An investigation of the jack arrangement which provides adjustment of the
> i kiln barrel slope was made, with the decision that it could be eliminated.
The nominal slope had been set at 208 cm/m with adjustment possible both in
slope and speed to vary conveying rate. The analysis showed that sufficient
variation could be provided by the adjustment of rotational speed alone, with
some adjustment possible by shimming the channel supports.
A design review was also made on the scrubber water system to investigate
weight savings. Ion exchange treatment of total process makeup water was
considered first but exchange resin requirements were deemed to be prohibi-
tive In both cost and size, The design selected employs continuous caustic
treatment and filtration of all makeup water, with scrubber recycle through
an air-cooled heat exchanger to minimize makeup requirements. Process
calculations considered both normal and anticipated maximum water require-
ments. Maximum makeup water required for the kiln and scrubber system is
8.3 1/min.
Caustic treatment of both makeup water and scrubber recycle and waste
streams will require a maximum of 45 kg/day of NaOH or 30 I/day of 50%
caustic. This assumes a maximum acid gas formation (as HC1) from adsorbate
incineration. Normal operation may produce less acid and lower caustic
treatment requirements. Additional water makeup is required for product
carbon cooling. Wet processing of the product increases the total water
makeup requirements to 19 1/min.
Changes in the three areas noted plus a variety of small modifications
enabled a reduction in the estimated equipment weight to 18,144 kg. This
allows for a standard trailer configuration without the need for special
operating permits. The final design and spatial arrangement were as shown
schematically in Figure 3.
TASK III - BENCH SCALE STUDIES
s
! A number of the decisions to be made in the delineation of the system
' design depended on an understanding of operating requirements. To provide
j this particular data, a series of studies were conducted using the laboratory
i
•i
25
fr
-------
. 2:
:. d
• O
•- rn
• to
' X
. m
•, m
•• H
5 > !
fl!
:•* w
r 1 i
-< a
n k i
.
0 f
•s
ro
o>
A. ROTARY KILN
B. INCINERATOR
C. SCRUBBER-LIQUID COOLER
D. CARBON FEEDER
Figure 3. Trailer design
-------
scale regeneration unit built during the earlier EPA program. These studies
had two objectives: an evaluation of the regeneration/incineration design of
the proposed mobile unit, and the testing of certain proposed materials of
construction under simulated conditions.
Bench Scale System
The design and operational characteristics of the laboratory-sized regen-
erator are given in the final report, "Laboratory Investigation of the Regen-
eration of Spent Activated Carbon" (Juhola, 1970). The unit was built as an
indirect fired system. The regenerator is an electrically heated rotary
(stainless steel) tube 82-5 mm in diameter with a 1080 mm heated length. An
auger type volumetric feeder is used to meter the carbon into the rotary
tube. To approximate the direct firing characteristics of the proposed
regenerator/incinerator, a countercurrent flow of gas was added to the
laboratory unit to simulate the products of propane combustion.
The laboratory regenerator operates at 18% loading with a .590 kgm/hr
(regenerated) carbon throughput. The exhaust gas cleanup train consists of
a thermal incinerator, an air-cooled condenser, and a water spray scrubber.
The thermal incinerator is designed to operate at 900°C, and gives a 0.5 sec
residence time to the effluent gas from the regenerator. It is heated with
natural gas at startup, the flow of which is reduced as the concentration of
desorbed flammable products increases and comes to a steady-state flow rate.
Air input is metered to maximize combustion efficiency.
The condenser is a finned tube and gives about 0.5 sec residence time to
the gases from the incinerator. Some 25 wt % of the gas is water vapor and
separates from the noncondensable gases, N£ and CC>2, in the scrubber. The
condenser reduces the heat load on the scrubber and minimizes the amount of
scrubber liquid which needs to be recirculated. The cleanup train is
arranged for operation with or without the scrubber or condenser in the
train. In this way, data can be obtained on various alternate routes for
exhaust gas cleanup.
The C02, N2» air and part of the gas stream are metered in to simulate
the products of combustion of propane in air. Extra steam is added by a
steam generator to simulate moderation of the flue gas temperature. The
water from the carbon and most of the chemicals are desorbed before the
carbon reaches the hot zone of the rotary kiln. This mixture of input
materials approximates the proportions that will be seen by the mobile
unit.
Chemical Classification and Selection
Evaluation of the regeneration/incineration capabilities of the proposed
mobile system required the selection of representative chemicals from each
of the five categories previously identified. Each group has specific
properties which effect its behavior in the thermal cycle.
27
-------
For each of the five groups of spill chemicals, representative materials
were chosen for study in the bench scale rotary kiln. The following are the
selected test chemicals for each group.
Group
1
2
3
4
5
Representative Chemical
Methyl acetate
Die hlorome thane
Diethyleneglycol diethyl ether
Benzenesulfonic acid
Hexamethylene tetramine
Tris(2-chloroethyl) phosphate
Mol . Wt . (g/Mole)
74
85
162
158
• 140
277
\
'. . Regeneration/Incineration Tests
! Regeneration runs were made with each of the five materials. Three
consecutive runs were made on each of the chemicals except for hexamethylene
i tetramine, using Pittsburgh CAL carbon. The hexamethylene tetramine runs
i were not completed because of corrosion problems in the incinerator.
In preparing the spent carbons for the tests, the carbon was treated
{ with 10% by weight of chemical and 50% by weight of water. For water
! soluble materials the additions were made as aqueous solutions. For water
i insolubles, the chemical was added first and then the water. In each case,
j the mixture was allowed to stand overnight to allow equilibration of the
chemical with the carbon.
For the dichloromethane runs, the feed rate was 1.26 kg/hr,-giving a
residence time of 45 minutes and a volumetric loading in the rotary tube of
about 16%. When the necessary flows of N2, 002. 02 and steam to simulate
direct firing were fed to the rotary kiln, the gas velocity was such that
considerable quantities of carbon dust were carried downstream to the
incinerator and scrubber. In the runs for the other chemicals, the carbon
feed was decreased to 0.72 kg/hr, giving a volumetric loading of about 9%
for a 45 minute residence time. This alleviated the problems generated by
carbon dust carryover.
Satisfactory results were obtained i-n all test runs. The methyl acetate
and dichloromethane were completely decomposed on passing through the
incinerator. The incinerator was found to operate best between 816°C and
871°C with an air flow of 18 1/min. The HC1 formed from the dichloromethane
was removed from the gas stream along with the steam condensate by the air
cooled condenser. The scrubber was not used in these tests. HC1
concentrations in the condenser effluent gas stream ranged from 10 to 60
ppm.
Similar tests were carried out for the other chemicals. Typical
concentrations of gases from the regenerator and scrubber for each of the
other test chemicals are given in Table 10. Some fog appeared in the vent
gas for the benzene sulfonic acid and tris(2-chloroethyl)phosphate runs.
28
-------
TABLE 10. TYPICAL CONCENTRATION OF GASES EXITING FROM THE
LABORATORY SCALE REGENERATOR AND SCRUBBER
txic bas concentration
(ppm)
Test Chemical
Regenerator
Scrubber
Benzenesulfonic acid
Diethyleneglycol diethyl ether
Tris(2-chloroethyl)phosphate
CO - >2000
HC*- >2000
S02 ~ >400
H2CO - 30
CO - 4000
CH3COOH - >1000
H2CO - 100
CO ->4000
HC1 - 300
NO - none
CH3COOH - 27
CO - 50
HC - none
H2CO - none
S02 - 30
H2CO - 5
CO - 35
CH3COOH - none
H2CO - none
NO - 6
N02 ~ none
CO - 45
HC1 - 90
- 24
j
*HC = Hydrocarbon
29
-------
r^fv~VV™^'*vSf^,"?^^^^^y3?W^y^**^^^^*\w^^
Apparently the S02 and HC1 are difficult to remove completely, and they
result in acid mist formation. . ,
The carbon yields were about 97% and a slight increase in density
occurred. The conclusion is .that the type of mobile unit being designed,
when operated under the planned procedures, should give satisfactory
results.
Materials of Construction
The principal item of concern in the regenerator design was the thermal
insulation for the kiln. The system requires a high efficiency ceramic
insulation which can withstand the anticipated vibrations of over-the-road
travel as well as the heat and abrasion of the regeneration operations.
Extensive discussions with insulation manufacturers led to the
conclusion that only silica-alumina fiber refractories could provide the
combination of properties required. Samples of other materials and forms
were tested. Alumina-silica coated with a hard surface proved to be the
best but it was ultimately found that this refractory could not withstand
the direct loading of the wet carbon. For this reason, the design was
changed to provide a metal liner in the tube to protect the insulation.
The final step in this task evaluated material requirements for the
liners in the kiln barrel and firing breech. The basic material selected
was, 300 series stainless steel. Since some material problems had been
experienced in the hot corrosive sections of the laboratory unit, Inconel
601 was chosen for those areas in the mobile unit.
30
-------
" " -• ""' ' ' ":' "" ""' "
SECTION 4
PHASE II - COMPONENT SPECIFICATION DEVELOPMENT
With completion of Phase I, the basic system configuration had been
delineated, the performance requirements of each system component had been
defined, and weight and size limitations had been established. Taken
collectively, these activities allowed the conclusion that a mobile carbon
regeneration system could be assembled successfully, incorporating an
onboard incinerator and scrubber to handle the materials stripped in the
regeneration process.
The next step in the process was the preparation of detailed specifica-
ions for each component and subassembly. This effort was initiated with
the understanding that commercially available equipment or modifications
thereof would be specified wherever possible.
During the proposal stage and the early tasks of this program, the intent
had been to modify an existing thermal treatment device for the basic
regeneration unit. As the program progressed it became clear that this
would not be possible. The requirements and limitations established for the
rotary kiln, the incinerator and the scrubber were found to be outside the
manufacturing limits of commercial suppliers. Thus, the first task in this
phase was the development of a detailed design and manufacturing specifica-
tions for these three key items.
KILN
The kiln specifications had been set by the results of the laboratory-
scale test program and the program goals. The principal factors were a 45.4
kg/hr throughput based on dry regenerated carbon weight and a final
regeneration temperature of 982°C. With this starting point, a series of
design calculations were made with the following results. For a 45.4 kg/hr
recovery of dry regenerated carbon, the input will be 54.4 kg/hr of spent
carbon together with 4.5 kg/hr of adsorbate and 36.3 kg/hr of water. Carbon
losses will be 2.3 kg/hr due to burnoff and 6.8 kg/hr as unusable fines.
A temperature profile was constructed within the kiln barrel, starting
with 100°C in the initial drying section and increasing to 982°C in the
final regeneration stage. These temperature choices are based upon a 1093°C
temperature for the tempered burner gas. Maximum temperature in the liner
and firing breech was set at 1038°C with a 1260°C design temperature for the
liner insulation. Maximum outside shell temperature was set at 93°C which
requires & 731 Kcal/hr-m2 heat loss.
31
-------
• -"• I •''..' * .' ' "'' • " "'•','•"••-..•"•-,-' .••••"•'-'.••••,', 5 .':•*,•-- '.•••-;• •• .. '•*''•'•• "' "jja/'j
Baaed upon these requirements and experience with the laboratory unit,
materials of construction were selected. Carbon steel was chosen for the
majority of components. For most of the elevated temperature components,
low carbon 304 and 316 stainless steel were specified. Inconel 601 was
chosen to provide greater high temperature strength for the hottest region
of the kiln.
The kllr. is a three stage rotary barrel with carbon flowing in one
direction and hot gas in the opposite direction. Countercurrenl" gas flow
provides for efficient heat utilization. In the first stage, wet spent
carbon is cascaded by lifting flights into the hot gas stream. The contained
water and low boiling adsorbates are removed here. In the second stage
increasing temperatures are encountered which vaporize the remaining adsorb-
ates begin thermal decomposition of some species. In the final stage, car-
bon is gently tumbled and the temperature is raised to the maximum of 982°C
to complete regeneration.
To achieve a process flow of 45.4 kg/hr of dry carbon, a kiln barrel
size of 371 mn inside diameter and 3.9 m long was calculated to be necessary.
A slope of 0.6 cra/m and a rotational speed of 5 rpm at an 11.7Z kiln loading
gives the desired throughput. This is a 0.14 m3/hr feed with a calculated 22
mln residence time. The design operating parameters are tabulated In Tabl-a.
11*
The major concern with the 'iiln was the thermal insulation. It must be
lightweight, resilient, and have a high maximum temperature along with a
high K factor. Originally it had been Intended to use a ceramic kiln
interior, but laboratory-scale testing made it clear that no available
material could provide the necessary insulating properties and still with-
stand the abrasion and mechanical stresses which would be imposed. A metal
inner barrel alleviated some of these problems.
The anticipated road stresses eliminated all of the normal rigid case or
molded Insulating media. Alumina-silica ceramic fiber forms were Identified
early in the program as the most promising material. Evaluation of commer-
cially available materials resulted in choosing Kaowool for the kiln Incala-
tlon. This was available as vacuum formed cylinders with a 7.25 kg/m3
density. Based on its thenaal ratings and the kiln characteristics, a
minimum thickness of 114 mm would be required to maintain the 93°C design
shell temperature•
The basic kiln must be fitted with several accessory Items for operation.
A feed unit is needed to deliver spent carbon to the firet stage of the
kiln. A 4 hr operating supply was chosen as the design point, which
is equivalent to 0.57 m3 of product. A feed system waa chosen with an
appropriately sized hopper and a dual pitch screw feed. A 58 mm pitch was
chosen to meter the carbon from the hopper and a 76 mm pitch to provide
movement through the transfer tube without plugging. An adjustable drive
was added to allow delivery rates between 0.06 and 0.28 m3/hr of wet
dewatered carbon. The feed system used 300 series stainless for all
components In contact with the wet carbon.
32
-------
TABLE II. KILN DESIGN OPERATING PARAMETERS
Parameter Identification Value
Flowrates Carbon content of spent feed 54.4
(kg/hr) Adsorbates vaporized 4.5
Residual water vaporized 36.3
Carbon burnoff 2.3
Carbon fines 6.8
Carbon product recovered 45.4
Fume gases to incinerator 272.2
Fuel Propane @ 1.3-2-psig 9.1
(kg/hr) Air @ 2-12% excess 150-168
Steam 32-54
;
Pressure Water Vacuum, @ Firing Breech 3.0-3.8
(mm)
Temperatures Ambient feed 21
(°C) Initial drying section 100
Final regeneration section 982
Liners @ firing breech 1038
Tempered burner gas 1093
Liner and insulation rating 1260
Fumes to incinerator 204
Shell, @ 731 K cal/hr-m2 93
33
-------
At the other end of the kiln, a product recovery system is required to
collect the regenerated carbon. This unit must cool the carbon, remove the
unusable fines, and discharge the regenerated product.
Cooling is best accomplished by water quenching. Further, it can pro-
duce a slurry which is an effective form for both pumping and separating
into size fractions. The selected design was a quench task using recircu-
lated water. The slurry formed, normally about 7% solids, is pumped to a
two deck screening separator. An air cooler was incorporated to insure
adequate cooling before entering the separator. Carbon material larger than
40 mesh is removed in the first stage for recycle. Fines less than 325 mesh
are separated in the second stage for disposal. The effluent is filtered to
remove all participates greater than 10 u and recirculated to the quench
tank.
A slurry pump (Moyno) was selected to provide a flow of 0.6-0.9 m/sec.
This was a compromise to provide high enough flow to prevent particle sepa-
ration and plugging but below the rate which would cause exaggerated erosion
of pipe and containment components.
The kiln is also fitted with a variable speed drive which permits
adjustment of rotation between 1 and 10 rpm. This allows variations in the
residence time which might be necessary for carbons with high loadings or
difficult to remove adsorbates.
INCINERATOR/AFTERBURNER
Incineration of the offgases from the kiln had been selected as the
most appropriate mechanism for detoxification and disposal of the desorbed
contaminants• The gases would enter the incinerator from the kiln through a
fume breech.
The initial design parameters were based upon a nominal fume inlet of
272.2 kg/hr effluent gases of which 4.5 kg/hr would be adsorbates or their
decomposition products. Calculations gave 1.0 sec as an adequate residence
time for mixing with burner gas and combustion.
For this rasldence, the firebox was sized at 0.61 m x 0.61 m x 1.5 m
long. The associated burner was sized to provide up to 126,000 Rcal/hr using
11.3 kg/hr of propane and 227 kg/hr of air. These are maximum values, all
components can be varied to achieve efficient combustion, based on stack gas
composition.
The incinerator is designed so that fume gases enter below the burner to
achieve good mixing. The design fume inlet temperature Is 204°C with a flue
gas outlet temperature of 982°C. The same temperature, 982°C, is the design
temperature for the insulation hot face. The outer shell temperature
maximum is 93°C.
The hot side of the incinerator is Inconel 601 to assure adequate cor-
rosion resistance and hot strength. For insulation, alumina-silica Kaowool
was chosen. A 3.63 kg/m-* density wrap-around blanket form was deerood the
34
-------
best choice with a mineral wool backup. After consideration of over-the-
road stresses, an external lever of ICaowool board was added for mechanical
strength. The overall insulation thickness was set at 142 mm in a carbon
steel shell.
SCRUBBER
Although the incinerator is the primary means of degrading the offgases
from the kiln, the products of combustion also may contain acid gases such
as HC1 and SC>2, particulate matter, and certain condensables. To remove the
undesirable materials from the flue gases, a scrubbing system was felt to be
necessary.
The scrubber consists of five components — a spray tower, a caustic
scrubber, a cooling tower, a liquor recirculator, and a stack blower.
Flue gases entering the scrubber system are adiabatically quenched in
the spray tower, passed through the caustic scrubber to remove acid gases
and into a packed tower to cool and demist. The gases venting from the
packed tower should be composed primarily of C02, N2, and 02. Through
monitoring of these gases, changes are made in the system, incinerator,
and/or scrubber to correct any deficiencies. Vent gases are discharged
through a stack using a blower to maintain an induced draft.
The reclrculating scrubber liquor passes through a filter to remove
particulate matter, through a cooler for temperature control, and into a
makeup chamber to provide pH control. The scrubber design parameters are
given in Table 12.
Material selection for the scrubber system was made principally for
corrosion resistance except for the spray tower where elevated temperature
strength was required. The material selected for the spray tower was
Inconel 625. It was sized at 406 mm diameter and 1.8 m high to accomodate
maximum inlet gas flow. It is flanged to the incinerator, thermally
insulated and contains 6 spray nozzles spaced equally around the circumfer-
ence.
The tower outlet discharges to a sump of approximately 380 L (100 gal)
capacity. Anticipated operating conditions would have the sump two-thirds
full of spent scrubbing liquor. The upper one-third acts as a duct to direct
the quench gases into the caustic rubber.
The caustic scrubber is designed as a .61 mm diameter, 1.5-m-high cylinder
packed within .15 mm of the top with Pall rings. The liquid distributor is
located above the Pall rings. The scrubber is vented through a MSA mist
eliminator using Teflon fiber spirals. The material of construction is fully
annealed low carbon 316 stainless steel to minimize the possibility of
chloride stress corrosion.
The liquor cooler is designed to handle 151,200 Kcal/hr (600,000 BTU/hr)
using ambient air at 332 m-Vmin (11,700 cfm). The cooler is located above
the kiln at roof level to facilitate venting the cooling air.
35
-------
TABLE 12. SCRUBBER DESIGN PARAMETERS
Parameters
Identification
Value
Gas Flows Spray column inlet 315
(m^/min)
Packed column inlet 15
Scrubber effluent 8.4
Spray Water @ 50% vaporization . 7.6
(L/mln)
Scrubbing
Liquor Scrubber inlet 114
(L/min)
Bleed pH stream 7.7
29% NaOH addition 5.7
Pressures Fan Suction 106.
(mm water
vacuum) Scrubber outlet 51.
Packed column inlet 25.
Spray tower inlet 23.
Temperatures Spray tower inlet gas 1016
(°C)
Packed column inlet gas 84
Scrubber effluent gas 66
Hot scrubbing liquor 79
Cooled scrubbing liquor 57
Ambient spray/caustic 21
36
^
-------
The scrubbing system is designed to use available water with an onboard
supply of caustic. Space and weight are available to carry 757 liters (a 7
day supply) of 20% NaOH solution. Although water would be drawn from a
local source, the onboard system contains a receiver tank with a pressurized
reservoir. This is to allow operation using adjacent water sources if a
local pressurized supply is unavailable.
Water drawn from a local supply is discharged back to its source.
Activated carbon filters are employed in line to insure acceptable quality
for discharge.
REGENERATOR POWER AND CONTROLS
The total system was designed to draw power and gas from local utilities
if available. Power requirements were estimated at 50 KVA. To compensate
for peak needs, the design value was increased to 75 KVA using 240V 3-phase
power.
The gas requirements for the kiln and incinerator combined were
estimated at .025 x 106 Kcal/hr. This is equivalent to 1.3 m3/hr of
propane.
It is expected that operation of the carbon regenerator will often be at
a remote site where local utilities are unavailable. For this reason the
system must be capable of independent operation. The gas requirements can
be readily filled by liquefied propane trucked to the site. The 11.3 m3/hr
of gas is only 41.6 1/hr of liquid. A 1.25 to 2 psi pressure is adequate
for the. onboard system.
The electrical requirements can be met by an onboard diesel generator
unit., The proposed physical arrangement allowed sufficient room in the
forward part of the trailer tor a commercial unit delivering 60 KVA
continuously with a peak output to 75 KVA. The generator was located so
that the exhaust could be ducted through the front of the trailer. The
muffler was suspended on the front of the trailer above the cab of the
tractor.
A diesel fuel storage tank was provided below the trailer deck. It was
sized for 1514 liters (400 gal) to provide 3 days of continuous operation.
The diesel generator area was expected to generate sufficient noise and
heat exclusive of the. exhaust system that a soundproof enclosure would be
required. Therefore, the generator enclosure was acoustically insulated.
To maximize space utilization, the wall of the generator ' enclosure
separating it from the main trailer area was used as the support for the
control panel of the regenerator/incinerator.
The control panel contains the necessary instruments and recorders for
operation of the regenerator. All components operated at elevated
temperatures are continuously monitored by means of thermocouples. Gas
flows, gas temperatures, and effluent gas compositions are also continuously
monitored. Water temperatures, water flows, and pH are measured and
37
-------
recorded. The diesel generator output is also instrumented. Fail safe
interlocks are provided to protect against runaway reactions or other'
abnormal operating conditions. All interlocks are connected to visible and
I audible alarms. Where possible, interlocks are also connected to automatic
f| shutoff-shutdown operators.
U TRAILER DESIGN
»j
•i In the early phases of the program a decision had been reached to base
[i the trailer design on available configurations. Complete trailer design was
i felt to be beyond the scope of this program. The basis for the development
', was a standard enclosed platform van 2.44 m wide x 13.7 m long x 4.1 m high.
IJ
':; Four significant modifications to the standard design were found
;, necessary:
!i
I; (1) To reduce over-the-road stress on the regenerator/incinerator
ij unit and allow at least limited off-the-road movement, the
'i platform support was strengthened and reinforcements were
\ added to minimize bed deflections.
j.j (2) The sides and the rear of the van were redesigned so that
J they could be opened for ventilation when the trailer was
I positioned for operation. In the design chosen, the sides
• are made in two segments. The lower segment drops down- to
•i provide a walkway on both sides. The upper segment raises to
;{ provide an awning structure. Thus along with ventilation,
| the design improves access and movement within the trailer
• and worker protection during inclement weather.
(3) A separate access door is provided for the diesel generator
enclosure. This area has its own overhead ventilation port.
[ (4) A series of jacks are incorporated to allow leveling of the
trailer on sloped or uneven ground.
In addition to the above modifications, the trailer was fitted with
several systems to accomodate specific components. Included in these are
mounting supports for the generator muffler, the vent stack from the
scrubber, and piping penetrations through the floor for water, gas and
electricity. The underside of the trailer accomodates a series of tanks
for water, diesel fuel, and caustic solution and the pumps associated with
each of these services. The minimum road clearance required for these units
was calculated to be 368 mm.
38
-------
GENERAL REQUIREMENTS
The unit as designed is completely telf-contained for field operation.
It requires a parking area of 12.2 m by 24.4 m, a local source of fresh
water, propane or natural gas, and a source of caustic replenishment. Power
can be drawn from the local utilities or from its onboard generator. Also,
a mechanism is necessary to dewater the spent carbon and drums are needed to
receive the regenerated product.
39
-------
SECTION 5
PHASE III - PROCUREMENT, ASSEMBLY, AND TESTING
With completion of the design phase, the acutal task of building the
mobile regenerator was initiated. This involved the procurement of all of
the required components, their assembly, the verification of system opera-
tion and finally delivery of the regenerator to the Environmental Protec-
tion Agency.
PROCUREMENT
The specifications prepared in the previous design phase were arranged
in appropriate form for the preparation of purchasing documents. For all
possibile components solicitation was for existing commercially available
equipment. Where minor differences existed between requirements and avail-
able units, the differences were resolved in favor of the avilable equip-
ment where possible. Where modifications were absolutely necessary, these
were negotiated with suppliers. Although price was a consideration, per-
formance was the major factor. Where modifications were necessary, the
design requiring the least revision was the usual choice.
In all procurement and equipment selection the Quality Assurance Group
at MSA was part of the decision procens. They in turn verified that all
incoming components complied with specifications.
In the course of the procurement process some problems with incoming
equipment were revealed. In all caces the difficulties were minor and
successfully resolved without compromising the system design or operational
specifications.
These significant items are as follows:
(1) The carbon product separator as delivered exhibited unaccept-
able blinding (screen plugging). A vibratlonal antlbllnding
accessory was Incorporated. In addition, to assist in effec-
tive separator operation, a 9.5 mm mesh basket screen was
installed in the slurry tank.
(2) The spent carbon feed screw experienced some binding when the
carbon contained an appreciable quantity of fines. Increasing
the pitch of the feed tube essentially eliminated this diffi-
culty. It was recognized, however, that carbon feed had to be
reduced below the design level to prevent plugging when the
feed contained a large percentage of carbon powder rather than
granules. Two 6.4 mm water taps were added to facilitate wash-
out operations in the feed mechanism.
40
^•jJy . _ *.-.
-------
Gnat"*'-""-''^-."*"*• T--*!».t-*"•n>ify'^»prtt»~t«ryi^r*]^s*?"*••«»•; «*"•*•••""••• ?«w" •^^•--^^^^'j'»^^tiyji«if^i^ij!* wFvJ;r»yy^"?fsT'?^^3^^?i^f*i^MiJB»^
(3) In the final portion of the kiln additional tumbling flights
had to be. added to get a gentler turning of the carbon and
prevent fragmentation.
ASSEMBLY
With the delivery of the trailer, installation of system components was
begun. With only minor changes, components were located as identified in
the spatial arrangement. All equipment mounted within the trailer was
bolted through the floor of the trailer to the steel cross members.
A common framework was installed below the trailer bed to support the
6 pumping units. Below deck supports were also installed for the 7 tanks;
3 for fuel, 2 for caustic and 2 for water.
Lugs were welded to the main trailer beams. The support work was bolted
to the lugs. Pumps were bolted to the support work. Tanks were held in
place by steel straps connected to the support work. Service piping was
interwoven with the supports and passed through the floor to mate with the
appropriate equipment.
As the assembly work progressed, certain accessories were added where
a need was indicated. A slide control valve was added to the scrubber
outlet to control the air from the blower. Roof hatches were Incorporated
over the two circulating cooler fans. Flexible mountings and connections
were made at several points in the piping to minimize shock and twisting
stresses.
ASSEMBLY REVIEW
Upon assembly of all components on the trailer, an operational plan was
prepared and reviewed with the Technical Project officer at EPA. This
included system drawings, flow sheets, operating parameters, and maintenance
specifications. The system was approved with only a few minor changes in
the arrangement and components. The Project Officer requested a level-
measuring mechanism with each support jack and elapsed time meters for the
kiln and the dlesel generator for proper maintenance scheduling.
SYSTEM CHECKOUT
Following approval of the operating plan by EPA, a system shakedown was
conducted using pure carbon containing only water. The system operated as
designed with only minor changes required in the operating procedures.
Operating conditions are given in Table 13.
41
-------
TABLE 13. OPERATING CONDITIONS OF SYSTEM
Condition Description
Regenerator Temperature 916°C
Incinerator Temperature . 1010°C
Carbon Residence Time 45 minutes
Carbon Feed rate .150 kg/hr
(dry generated product)
Steam Input Rate .156 kg/hr
Gas Input Rate
C02 to regenerator .142 kg/hr
N2 to regenerator .690 kg/hr
Air to Incinerator 1.420 kg/hr
Water to scrubber 54.0 kg/hr
The only significant difficulty came from localized overheating in the
kiln during heatup. The temperature excursions were not severe but they
did trip automatic shutdown operators. This interlock was disconnected
while leaving the alarms in place. Whether this interlock should be in
place during system operation is a decision which will require some
operating experience.
The successful shakedown run was followed by regeneration tests using
contaminated carbon.
In preparation for the system demonstration, EPA had shipped to MSA
drums of wet spent activated carbon used In the cleanup of a spill in
Plains, VA. A preliminary run had been made with this carbon in the
laboratory-scale regeneration unit during the design phase of the mobile
regeneration unit.
The pollutant in the carbon was primarily Toxaphene (C^o^ioClg) plus
smaller quantities of other chlorinated compounds such as Aldrin
Dieldrin (C^^ClgO), Heptachlor (Cio^Cly), and Chlordane (CioH6cl8)
carbon was discolored with a brown material that appeared as a thin layer on
the outside surface of the granules. On the basis of bulk density
measurements, the pollutant content was about 13% and water content was
about 52% by weight. Both percentages are based on the dry weight of the
regenerated product.
Two runs were made In .which an 88% bulk volume yield of reactivated
carbon was attained. The large loss is believed to be due to removal of the
brown mater la', from the surfaces of the granules. Table 14 presents the
Iodine numbers and bulk densities of the spent and regenerated carbons from
these runs. The carbon was not identified, but on the basis of density and
Iodine number appears to be Pittsburgh type CAL carbon. If this is the
case, the carbon was completely regenerated. All HC1 was removed in the
scrubber.
-------
TABLE 14. IODINE NUMBERS
Iodine No. Bulk Density
'Carbon "" (mg/g) (g/cc)
Run 1 Run 2
Pittsburg CAL 960 960 0.500 dry
Spent Carbon 690 710 0.577 dry, 0.844 wet 1
Reg. Carbon, Run 1 1030 850 0.507 dry •
Reg. Carbon, Run 2 960 940 0.510 dry ]
- - - ^
The success of these tests demonstrated the operating status of the
mobile regeneration unit. On February 8, 1979 the unit was picked up by an
EPA contracted hauler at the Evans City plant of MSA for transportation to
the Hazardous Waste Engineering Research Laboratory at Edison, New Jersey.
Eight photographs are provided to give general views of the mobile j
regenerator/incinerator. Figure 4 shows the van closed for storage or
transit. Figure 5 shows a rear view of the van open for operation. Carbon
is being transferred from a drum into the carbon feed hopper. Figure 6 is a
side view of the carbon feed and the kiln with the incinerator mounted above
it. Figure 7 is a head end view of the kiln showing firing breech and gas
burner. Figure 8 shows the side of the unit opposite the kiln with the
scrubbing towers and water filtration units. Figure 9 shows the tank and
pump assemblies slung beneath the bed of the trailer. Figure 10 is the
control panel in the forward end of the van with the carbon screening unit
in the right foreground. In Figure 11 the van is shown coupled to a stub
tractor leaving the MSA plant enroute to EPA.
43
-------
SfcGIN
Figure 4. Enclosed trailer
ecctN
LAST
(ji- Tc.x
Figure 5. Trailer open for system operation
2 3-'8" w
EPA Form 23DO-4 14-801
(PRKVIOUSUV C.IN. tl'A I OHH
PA<.it MUMRFfl
TYPING GUIDE SHEET
R.'IH fOM OF
IVAi';U Af:t A;
OUI3ID6
Oir.TNSKJN
FOR TAUlt-S
I AM) ILLU5-
THATIOfJS
. ^.£.*-- i
. L
-------
TOP ')f
A. W.E
|7
; Figure 6. Profile of kiln and incinerator
" • I
Figure 7. Head end of kiln
~r
I
EPA form r.fjO-4 (t-601
II-KI vio>rjt_v CIN. KPA
TYPING GUIDE SHEET
-------
ECSN
Figure 8. Scrubber and filtration equipment
i -
Figure 9. Tanks and pumps beneath trailer bed
,. u.,. _..,j_
Li_3:i_i m
Form Z.I.'XI-'t M-80i
PAGE (;v-v,ijin
vr-•'.';. nKioN
i , r.., TA,- ?s
Tvr-»»(.-» ^*i,.--.*- „...-__
-------
BEGIN
MRST
LINE Of- 1
IcXT *"f
HEHE.
TOP or
BEGIN
LAST LINE:
or i i-xi
Figure 10. Control panel and product screener
Figure 11. Trailer In transit
~1 3 'p-'
L.ri
T.
EPA Form E3&0-4 (1 BO)
[PKt;VIOU£L.y CIN. UPA FOHM 247)
PAGE NUMliEH
i;OTTOM or
IMAGE AHKA;
D'.J'i'SIL'ir.
lMtNSION
FOR TABLES
I ANi> IM.LI3-
1 RATIONS
TYPING GUIDE SHEET
i
i _
------- |