EPA-660/2-73-030
DECEMBER 1973
Environmental Protection Technology Series
Treatment Of Sulfite Evaporator
Condensates For Recovery Of
Volatile Components
I
55
V
S3ZZ
\
UJ
o
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five broad
categories were established to facilitate further
development and application of environmental
technology. Elimination of traditional grouping
was consciously planned to foster technology
transfer and a maximum interface in related
fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental studies
This report has been assigned to the ENVIRONMENTAL
PROTECTION TECHNOLOGY series. This series
describes research performed to develop and
demonstrate instrumentation, equipment and
methodology to repair or prevent environmental
degradation from point and non-point sources of
pollution. This work provides the new or improved
technology required for the control and treatment
of pollution sources to meet environmental quality
standards.
-------
EPA-660/2-73-030
December 1973
TREATMENT OF SULFITE EVAPORATOR CONDENSATES FOR
RECOVERY OF VOLATILE COMPONENTS
Kenneth W. Baierl
Nai L. Chang
Bernard F. Lueck
Averill J. Wiley
Robert A. Holm
Grant No, S80120T
Program Element 1B203T
Project Officer
Ralph H. Scott
Chief, Paper and Forest Industries
U.S. Environmental Protection Agency
National Environmental Research Center
Corvallis, Oregon 97330
Prepared for
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D.C. 20^60
-------
EPA Review Notice
This report has been reviewed by the Environmental
Protection Agency and approved for publication.
Approval does not signify that the contents neces-
sarily^ reflect the views and policies of the
Environmental Protection Agency, nor does mention
of trade names or commercial products constitute
endorsement or recommendation for use.
-------
ABSTRACT
A pilot plant study of a process to recover the volatile constituents
of the condensate derived from the evaporation of a sulfite spent wood
pulping liquor has been made. The data from this one-year evaluation
confirm prior work demonstrating that recovery of sulfur dioxide, fur-
fural, methanol, and acetic acid (in the form of ethyl acetate) will
yield reusable and salable materials, and provide either 60 or 90% BODs
reduction on the condensate depending on whether the condensate is
contaminated by using it as wash liquor.
The work reported covers four major sections:
1. Assay of condensate samples from supporting mills,
2. Operation and data of a pilot system comprising steam
stripping, activated carbon adsorption, and fractional
distillation,
3. Mass, heat, and BODs balances made according to the
actual operating condition of the pilot plant at the
Appleton Division mill of Consolidated Papers, Inc. to
January 1973,
U. Low temperature (200°C-390°F) regeneration of carbon.
Assays of the condensate samples indicated a large variation in con-
densates from different mills which would necessitate tailoring of the
complete process to the individual mill.
Operation of the pilot system, an extension of work previously done
at the Scott Paper Company, has shown that the above-mentioned materials
can be recovered as relatively pure products. Mass and heat balances,
recoverable product values, and credits for BODs removal combine to
show the process to be a favorable avenue for the elimination of the
pollution potential of the sulfite condensate waste.
111
-------
The low temperature regeneration of carbon was an extension of work
previously performed at The Institute of Paper Chemistry. This approach
continues to be of interest and is considered to be technically feasible.
However, all attempts to use the principles and equipment for electrical
induction heating, as developed at the Lowell Technological Institute,
failed due to mechanical design problems encountered in the pilot trials
and which could not be developed and corrected within the time and fundr-
ing available for this project.
This report was submitted in fulfillment of Grant No. S-801 2Q7 under
the partial sponsorship of the Environmental Protection Agency, by The
Institute of Paper Chemistry with the Wisconson Department of Natural
Resources and a group of pulp and paper mills cooperating. Work was
completed as of May 1973.
IV
-------
CONTENTS
Sections
I Conclusions 1
II Recommendations 2
III Introduction 3
IV Assay of Condensate Samples From Supporting Mills 5
V Design, Construction and Operation of Pilot System
At The Appleton Mill of Consolidated Papers, Inc. l6
VI Mass, Heat, and BOD5 Balances 103
VII Low Temperature Regeneration of Carbon 138
VIII Invention Record and Publication 16?
-------
FIGURES
No.
1 Scott Paper Company Pilot Plant Installation
Everett Mill 18
2 IPC Pilot Plant — Consolidated Papers, Inc.
Appleton Mill 20
3 Activated Carbon Colutnn Furfural Regeneration 21
U Activated Carbon Column Acetic Acid Regeneration 2,2
5 Methanol Removal vs. Stripping Steam Used — Scott
Paper Company - US-Inch (121.9 cm) Packing 28
6 Methanol Removal vs. Stripping Steam Used — Scott
Paper Company - Variable Packing 29
7 Sulfur Dioxide Removal vs. Stripping Steam Used —
Scott Paper Company - 48-Inch (121.9 cm) Packing 30
8 Sulfur Dioxide Removal vs.. Stripping Steam Used —
Scott Paper Company — Variable Packing 31
9 Methanol Removal vs_. Stripping Steam Used — Consolidated
Papers, Inc. — US-Inch (212.9 cm) Packing 32
10 Methanol Removal ys. Stripping Steam Used — Consolidated
Papers, Inc. -12-Inch (30.US cm) Packing 33
11 Sulfur Dioxide Removal vs. Stripping Steam Used —
Consolidated Papers, Inc. —US-Inch (121.9 cm) Packing 3U
12 Sulfur Dioxide Removal vs. Stripping Steam Used -
Consolidated Papers, Inc. -12-Inch (30.US cm) Packing 35
13 Acetic Acid Concentrations — Steam Stripping Column •
Feed and Bottoms U6
lU Acetic Acid Concentrations — Furfural Activated Carbon
Column Feed arid Effluent UT
15 Acetic Acid Concentrations — Acetic Acid Activated
Carbon Column Feed and Effluent US
16 Methanol Concentration — Steam Stripping Column Feed .
and Bottoms ' U9
17 Methanol Concentration — Furfural Activated Carbon
Column Feed and Effluent 50
18 Methanol Concentration — Acetic Acid Activated Carbon
Column Feed and Effluent 51
19 Biochemical Oxygen Demand — Steam Stripping and
Activated Carbon Column Flows 52
vi
-------
Figures (Continued)
No..
20 Organic Loosely Combined Sulfur Dioxide Concen-
trations — Steam Stripping and Activated Carbon
Column Flows 53
21 Total Inorganic Sulfur Dioxide Concentrations —
Steam Stripping and Activated Carbon Column Flows 5^
22 pH — Steam Stripping and Activated Carbon Column ,
Flows , '55
23 Acetic Acid Concentrations — Steam Stripping and
Activated Carbon Column Flows Using Regenerated
Carbon 57
2k Methanol Concentrations — Steam Stripping and
Activated Carbon Column Flows Using Regenerated
Carbon 58
25 Total Inorganic Sulfur Dioxide Concentrations — Steam
Stripping and Activated Carbon Column Flow Using
Regenerated Carbon 59
26 Organic Loosely Combined Sulfur Dioxide Concentrations —
Steam Stripping and Activated Carbon Column Flow Using
Regenerated Carbon 60
27 Biochemical Oxygen Demand — Steam Stripping and
Activated Carbon Column Flow Using Regenerated Carbon 6l
28 pH — Steam Stripping and Activated Carbon Column
Flow Using Regenerated Carbon 62
29 Furfural Concentrations — Steam Stripping and
. Furfural Activated Carbon Column Flows 6U
30 Modified IPC Pilot Plant — Consolidated Papers, Inc.
Appleton Mill 75
31 Both Activated Carbon Columns Regenerated with Ethanol 83
32 Steam Stripping, Furfural Activated Carbon and Acetic
Acid Activated Carbon Columns 85
33 Steam Stripping, Furfural Activated Carbon and Acetic
Acid Activated Carbon Columns 87
3^ Process Flow 89
35 Regeneration — Step 1 . 90
36 Regeneration — Step 2 91
37 Regeneration - Step 3 92
38 Steam Stripping, Furfural Activated Carbon and Acetic
Acid Activated Carbon Columns 95
vii
-------
Figures (Continued)
No..
98
39 Suggested Regeneration '
Uo Projected Plant — Consolidated Papers, Inc.
Appleton Mill 10°
hi Suggested Commercial Application Consolidated Papers,
Inc. Appleton Mill ' 101
42 Overall Mass Balances of the Pilot Plant 10T
1*3 Separation Process — Mass Flow Sheet 1Q8
hk Activated Carbon Regeneration and Furfural Recovery.
1. Furfural Recovery 109
45 Activated Carbon Regeneration and Furfural Recovery.
2. Methanol Recovery HO
U6 Activated Carbon Regeneration and Ethyl Acetate
Recovery. 1. Draining of Column HI
UT Activated Carbon Regeneration and Ethyl Acetate
Recovery. 2. Recovery of Ethyl Acetate and
Recovery of Ethanol li2
U8 Separation Process llU
49 Activated Carbon Regeneration and Furfural Recovery 115
50 Activated Carbon Regeneration and Ethyl Acetate
Recovery 116
51 Activated Carbon Thermal Regeneration Column
52 Location of Electrode and Thermocouple Terminals
Vlll
-------
TABLES
NO.I
1 pH, Sulfur Dioxide, and Color Data 8
2 COD, BOD5, and Acetic Acid and Formic Acid Data 9
3 Methanol, Acetic Acid, and Furfural Assays By Gas
Chromatography and Calcium and Sodium Assays 10
k Pollution Potential and Materials Data lU
5 Methanol Balance Steam Stripping Runs 38
6 Furfural Balance Steam Stripping Runs Hi
7 Furfural Balance Activated Carbon Adsorption 65
8 Acetic Acid Balance Activated darbon Adsorption 67
9 Evaporator Studies Data 76
10 Analyses of Evaporator Effects of Cooperator Mill 79
11 Carbon Adsorption Work at Scott Paper Company
Oconto Falls Evaporator Condensate 79
12 Carbon Adsorption Work at Consolidated Papers, Inc.
Appleton Div. Evaporator Condensate 79
13 Typical Activated Carbon Acetic Acid Column Regeneration
Data — Potentially Patentable Process 9k
Ik Typical Pilot-Plant Operating Data IQk
15 Typical Activated Carbon Acetic Acid Column
Regeneration Data 105
16 Typical Activated Carbon Furfural Column Regeneration
Data 106
17 Recoverable Values Based on the Condensate of
Consolidated Papers, Inc. 135
18 Pollutional Balance Sheet 136
19 Lowell Unit - Direct Current
20 Lowell Unit - Direct Current
21 Lowell Unit - Direct Current . lU7
22 Lowell Unit - Direct Current 1^9
23 Lowell Unit - Direct Current 151
2k Lowell Unit — Alternating Current 15H
2"5 Lowell Unit - Alternating Current 157
26 Lowell Unit — Alternating Current 160
27 Lowell Unit — Alternating Current 162
28 Lowell Unit — Alternating Current l6k
ix
-------
ACKNOWLEDGMENTS
Sponsors of this project, in addition to Environmental Protection Agency
and The Institute of Paper Chemistry, the prime contractor, were:
Authorized
Organization Representative
Wisconsin Department of Natural Resources Everett D. Cann
American Can Company Richard F. Benning
Consolidated Papers, Inc. S. H. Dorcheus
Finch, Priyn & Company, Inc. John Shema
Flambeau Paper Company Walter A, Sherman
Great Northern Paper Company . V. F. Mattson
ITT Rayonier Inc. J. J. Leithem
Masonite Corporation H. A. Spalt
Publishers Paper Company Zenon Rozycki
Scott Paper Company Lyle J. Gordon
Wausau Paper Mills Company Richard L. Post
The Project Director was A. J. Wiley, with direct supervision and
engineering of the pilot plant operation under K. W. Baierl, assisted
by John E. Baumann, pilot plant operator. The low-temperature carbon
regeneration investigation was under the supervision of B. F. Lueck.
Those responsible for and assisting in the laboratory phases were
George A. Dubey, Arthur A. Webb, Gerald W. Hovind, Mrs. Linda Benkers,
and Miss Gay Thurner. Engineering balance calculations were made by
Nai L. Chang. With the exception of Mr. Chang, all are members of the
IPC Effluent Processes Group, which is a section of the Division of
Industrial and Environmental Systems headed by Robert A. Holm.
The support of the personnel of the Appleton Division of Consolidated
Papers, Inc., especially W. R. Durdell, Plant Manager, and Dennis G.
Wilch, who acted as liaison between the mill and IPC where the pilot
plant was located, is acknowledged with sincere thanks.
-------
SECTION I
CONCLUSIONS
Analyses of sulfite mill evaporator condensates submitted by sponsoring
mills revealed that most of their condensates contained larger quan-
tities of recoverable chemicals than were present in the relatively
weak condensates available for processing at the Appleton Division
mill of Consolidated Papers, Inc.
Data were compiled to assist interested sulfite pulp manufacturers
toward making feasibility studies for utilizing steam stripping,
fractionation, and activated carbon adsorption systems as tools toward
solving their condensate pollution problems. With minimum additional
pilot plant work tailored to individual mill requirements, these units
can be reasonably scaled to commercial sizes with more accurate
development of process economics. The newly developed activated car-
bon regeneration system appears economically feasible, but more
optimization work could be worthwhile.
Low-temperature regeneration of activated carbon utilizing direct
application of electric current through the carbon bed is not ready
for commercial evaluation. The problems were identified, and can be
used to evaluate further research potential.
-------
SECTION II
RECOMMENDATIONS
Steam stripping, fractionation, and activated carbon adsorption have
been demonstrated to be commercially feasible, and these unit processes
are considered ready for a commercial demonstration plant. It is
recommended that pilot work be continued on an individual mill basis,
with the unit operations tailored toward establishing design data
specific to each mill's ultimate objective. The data generated to date,
including previous work at The Institute of Paper Chemistry, Scott
Paper Company, and work under this project, should be used as the basis
for future individual mill studies of plant design factors.
Future research work, supported collectively or individually, should
follow the Process Options presented on page 73. These can be
accomplished on laboratory or pilot plant bases. If time permits, a
modular stepwise approach is recommended.
The established and developed analytical techniques can be used and/or
extended in future studies.
The heat and material balances presented should be used as the bases
for individual mill optimization work recommendations and future
economic evaluations.
The newly developed activated carbon regeneration system should be
optimized and fully evaluated with further pilot plant studies.
Finally, the same unit operations can be utilized to treat kraft
evaporator condensates, which are also a potential source of stream
and/or air pollution.
-------
SECTION III
INTRODUCTION
Five years of research in two separate but complementary projects di-
rected to removal of pollution-contributing materials from sulfite
evaporator condensates have been conducted by The Institute of Paper
Chemistry and by Scott Paper Company. The original studies have been
advanced in this project by installation and operation of the pilot
plant units in the Consolidated Papers, Inc.'s Appleton Division mill
located in Appleton, Wisconsin. The systems developed at Scott
Paper Company incorporated steam stripping, activated carbon adsorption
and fractional distillation as means for removing and recovering sulfur
dioxide, methanol, acetic acid, furfural, and other pollution-contrib-
uting materials, as well as regenerating the activated carbon.
The system previously studied at The Institute of Paper Chemistry
consisted primarily in the removal of BODs and COD contributing
materials from these sulfite condensates by steam stripping and
adsorption on activated carbon. Successful regeneration of the carbon
was attained with low temperature heat (200°C-392°F) and the addition
of superheated steam. Work under this project, herein described,
consisted of the evaluation of a means of heating the carbon during
regeneration by use of a system developed by Messrs. Bela M. Fabuss
and Wilson C. Dubois of Lowell Technological Institute.
Mass, heat and BODs balances are made according to the actual operat-
ing conditions of the pilot plant at the Appleton Division Mill of
Consolidated Papers, Inc. and are based on data taken prior to
January 1?, 1973.
As an adjunct to this project, the sponsoring mills were invited to
send samples of their evaporator condensate wastes to The Institute of
Paper Chemistry for evaluation and comparison of their various
constituents. Such a compilation of data would aid the personnel of
-------
the various mills to evaluate the process under study by comparison of
their condensate to that of the Appleton Division of Consolidated
Papers, Inc., where the pilot vork was undertaken.
Data and discussion are presented in the following order:
1. Assay of condensate samples from supporting mills.
2. Design and construction of Pilot System at the .Appleton
mill of Consolidated Papers, Inc.
3. Mass, heat, and BODs balances.
k. Low temperature (200°C-390°F) regeneration of carbon.
-------
SECTION IV
ASSAY OF .CONDENSATE SAMPLES PROM SUPPORTING MILLS
METHODOLOGY
Samples from ten mills of eight companies sponsoring this project, in-
cluding samples from the Appleton Division of Consolidated Papers, Inc.
taken during operation of the pilot-scale chemical recovery system,
were received. The samples have been analyzed for 12 parameters. Most
of the condensate samples showed greater potential for acetic acid and
furfural recovery than is available in the condensate at the Consoli-
dated mill. The content of a condensate varies with the pulping
process and is substantially affected, first, by the type of wood
cooked, second, by the nature of the cook and, third, by the pH of the
liquor being concentrated.
METHODS OF ANALYSIS
The procedures used in these analyses are as follows:
pH was determined on a Beckman pH meter.
Sulfur Dioxide
Two types of sulfur dioxide were determined; TI (total inorganic)
and OLC (organic loosely combined). The total inorganic sulfur di-
oxide which is sometimes designated as free sulfur dioxide was deter-
mined by direct titration with standard iodine at 0°C (32°F). The
organic loosely combined sulfur dioxide is that sulfur dioxide which
is combined but which is readily releasted by treatment with alkali
followed by acidification. It is determined by making the sample
strongly alkaline with sodium hydroxide followed by acidification
after one hour and titratiofl with standard iodine at 0°C (32°C). This
is the standard TAPPI procedure.
-------
Color
A Hellige aqua tester was used to determine color. The sample is di-
luted as necessary and the color compared against standards which are
equivalent to the cobalt platinum standards of "Standard Methods for
the Examination of Water and Waste Water" of the American Public Health
Association (APHA).
Chemical Oxygen Demand (COD)
The standard method of APHA was used, which consists of digestion of
the sample with concentrated sulfuric acid and standard dichromate
solution for two hours followed by back-titration with standard fer-
rous ammonium sulfate. Silver sulfate is used as a catalyst since
without it the acetic acid would not be oxidized.
Biochemical Oxygen Demand (BODs)
The five-day BOD was determined by the dilution technique of standard
methods of APHA. In all cases a series of dilutions were set up in
duplicate and the dissolved oxygen determined with a Weston and Stack
meter before and after the five-days incubation. Seeding was accom-
plished by using domestic sewage and BODs was determined on a standard
glucose-glut amic acid standard as a control for all procedures.
Volatile Organic Acids
The volatile organic acids generally present in spent sulfite liquors,
and consequently in the condensate derived from these liquors, are
acetic and formic. For many years the Effluent Processes Group has
used an ether extraction procedure for determining these total vola-
tile organic acids. The procedure consists of extraction with ether
under acidic conditions using a ratio of 9 parts of ether to 1 part of
sample, followed by titration with a standard alkali in the presence of
a small amount of C02 free water. The volatile organic acids are then
calculated in terms of equivalent acetic acid.
6
-------
Formic Acid
The: 'formic acid was determined from the residual remaining after the
titration for the volatile organic acids. It was determined by removal
of the ether by evaporation followed by acidification and reduction of
mercuric chloride to mercurous chloride. The mercurous chloride is
separated by filtration and determined by the addition of an excess of
standard iodine followed by a back-titration with standard thiosulfate.
Acetic Acid
The acetic acid was determined by difference between total volatile
organic acid and the formic acid*
Gas Chromatography - ./
Methanol, furfural, and. acetic acid were determined by gas chromatog-
raphy using an Aerograph 1520 instrument. Separation was on a PORA
PAG Q column. Butanol was used as an internal standard with a flame
detector.
Calcium and. Sodium
Both calcium and sodium were determined on a Beckman Flame Spectro-
phbtometer reading at the appropriate wavelengths.
ASSAYS
Data covering the assays on the various samples received from the
cooperators are given in Tables 1, 2, and 3. Origin of the samples is
given only-under a mill code letter. However, in order to compare with
data being obtained in the operation of the process at the Consolidated
mill, the composite data from that mill are identified as Sample No. 21,
or Mill G. The data under Sample No. 21 are either averages of eight
samples or assays performed on a composite of those eight samples.
These samples were taken at periods during normal operation, which
included channel switching of the Rosenblad evaporator.
-------
Tattle 1. pH, SULFUB DIOXIDE, AMD COLOR DATA
Mill
A
A
A
A
B
B
B
B
B
B
B
B
C
C
C
C
D
D
D
D
E
E
E
F
P
P
0
H
H
H
H
I
J
Sample
number
1
2
3
h
5
6
T
8
18
19
32
33
9
10
11
12
13
lit
2l»
25
15
16
IT
20
22
23
21
26*
27*
28
29°
30
31
S0». BlK/liter
Sample tdentifi cation
Dense hardwood 9/T/T2
Dense hardwood 9/8/72
Dense hardwood 9/9/72
Aspen-dense hardwood 9/10/72
3U? Hardwood, 66% softwood
Same as Ho. 5 but used for wash
93? Poplar, 7? spruce
Same as No. 7 but .used for wash
lU? Hardwood, 86? softwood
Same as Ho, 18 But used for wash.
SSL, 31*? Hardwood, 66% softwood
3l*? Hardwood, 66% softwood
100? Aspen 9/19/72
100* Aspen 9/20/72
100? Aspen 9/22/72
100? Aspen 9/26/72
95? Aspen, 5? spruce combined
condensate
' 95? Aspen, 5? spruce second effect
95? Aspen, 5? spruce-combined
condensate 1/2U/73
95? Aspen, 5? spruce -second effect
100? Hardwood
100? Softwood
Total discharge including
surface condensate
Exploded wood pulp
Hardboard mill 12/20/72
Hardboard mill l/U/73
Softwood
Softwood, high solids effect
Softwood, low solids effect
Softwood, 26 and 27 blend
Softwood, total condensate
Magnefite softwood 2/2/73
Magnefite softwood 2/5/73
pH
2.15
2.17
2.18
2.12
1.78
3.02
2.1*1*
3.21
2. lit
2.67
2.30
2.15
2.63
2.57
2.50
2.55
2.63
2.38
2.1*3
2.18
2.38
2.27
2.32
3.5l»
3.25
3.30
2.20
6.19
5.U5
5-69
2.1*9
2.79
2.37
TI
180
150
160
1*30
787
6
28
8
52
5
75
29
102
73
236
6k
13
36
102
270
12
596
5U
0
2
1
336
2.0
1.0
1.3
1.0
76
380
OLC
1*30
U90
510
1*90
919
108
270
600
1*97
181*
3398
575
261*
181
1*71*
229
110
269
161
357
212
1*27
1061
0
16
1*
722
1.0
0.5
0.2
31*3.
71*
107
Colo:
250
1*30
600
280
60
1*00
90
100
to
200
20,000
100
500
150
750
200
15
<5
<5
<5
5
5
1000
650
300
300
200
250
50
150
15
<5
10
*APHA Standard Color Units.
From prestripped neutralized SSL.
Total oondensate from prestripped unneutralized SSL.
-------
Table 2, COD, SODj, AMD ACETIC AM) FORMIC ACID DATA
Via ethen>*xtraetion
Mill
A
A
A
A
B
B
B
B
B
B
B
B
C
C
C
C
D
D
D
D
I
E
I
P
P
F
0
H
H
H
H
I
J
Sample
number
1
2
3
It
5
£
7
e
18
19
32
33
9
10
11
12
13
lit
2 It
25
15
16
IT
20
22
23
21
26*
27b
28
29"
30
31
Sample identification
Dense hardwood 9/7/72
Dense hardwood 9/8/72
Dense hardwood 9/9/72
Aspen-dense hardwood
9/10/72
66% Hardwood, 3lt* softwood
Same as So , 5 but used
for wash
93* Poplar, 1% spruce
Same as Ho. 7 but used
for wash
iW Hardwood,- 86* softwood
Same as No, 18 but used
for wash
SSL, 3W hardwood, 66*
softwood
3W Hardwood, 66* softwood
100* Aspen 9/19/72
100* Aspen 9/20/72
100* Aspen 9/22/72
100* Aspen 9/26/72
95* Aspen, 5* spruce
combined condensate
95* Aspen, 5* spruce
second effect
95* Aspen, 5* spruce
combined condensate
l/2l*/73
95* Aspen, 5* spruce
second effect 1/2U/73
100* Hardwood
1009! Softwood
Total discharge includ-
ing surface condensate
Exploded wood pulp
Hardboard mill 12/20/72
Hardboard mill l/U/73
Softwood
Softwood, high solids
effect
Softwood, low solids effect
Softwood, 26 and 27 blend
Softwood, total con dene ate
Hagnefite softwood 2/2/73
Magnefite softwood 2/5/73
COD,
mg/1
1011(0
11016
10622
8769
12098
9828
12062
10592
9"(87
77,1(3
1731(25
101)00
11268
7859
161*22.
9780
5830
5830
5162
6330
9272
6152
f
30192
3225
3W5
3393
7620
900
2251*
I69"t
3191
2981
3885
BODS,
*g/l
6090
6252
677>4
5870
8881t
61t8o
7890
7lil6
7113
6258
37"i25
6772
5586
5088
7lt56
61(38
Itli76
1*350
3870
l*761t
713l»
1*506
11100
1730 .
1332
2205
3718
118
1566
1031
233"l
15lt8
2358
TOAa,
tag/1
5998
5876
5l»15
5326
6627
5680
6578
7006
1)282
1(532
5879
1*633
1*703
1(978
5528
589lt
2761
3023
262lt
3581(
7280
1(025
5803
1136
162U
1567
257"(
61
38
52
1953
1325
2 1*1*9
Acetic
acid,
mg/1
5875
5733
5l(ll(
5308
6550
5570
6523
6961*
1(277
1(529
5550
1(572
1(650
1(960
51(71
5813
2618
2976
2591
3551
7266
1(022
5800
967
1301
ll*29
2511*
1(0
17
33
1896
121(6
2350.
Formic
acid,
mg/1
95
110
1
lit
59
Bit
1)2
32
.1*
2
252
1*6
1*1
ll*
1*1*
62
110
36
25
25
11
2
2
130
2lt8
106
1(7
17
16
lU
1(3
60
60
/Volatile organic acids (VOA) in terms of acetic acid.
From prestripped neutralized SSL.
"fetal condensate from prestripped unneutralized SSL.
-------
Table 3.
METHANOL, ACETIC ACID. AMD FURFURAL ASSAYS BY GAS CHBOMATOGRAfH*
AND CALCIUM AND SODIUM ASSAYS
Via aaa chromatoeraphy
Mill
A
A
A
A
B
B
B
B
B
B
B
B
C
C
C
C
D
D
D
D
E
E
E
F
T
F
G
R
H
H
H
I
J
Sample
number
1
2
3
u
5
6
7
8
18
19
32
33
9
10
11
12
13
Ik
Sk
25
15
16
17
20
22
23
21
26a
27*
28
29b
30
31
Sample Identification
Dense hardwood 9/7/72
Dense hardwood 9/8/72
Dense hardwood 9/9/72
Aspen-dense hardwood
9/10/72
66* Hardwood, 31** softwood
Same as No. 5 but used
for wash
93* Poplar, 7* spruce
Same as No. 7 but used
for wash
ll;* Hardwood, 86* softwood
Same as Ko. 18 but used
for wash
SSL, 3fc* hardwood, 66*
softwood
3W Hardwood, 66* softwood
100* Aspen 9/19/72
100* Aspen 9/20/72
100* Aspen 9/22/72
100* Aspen 9/26/72
95* Aspen, 5* spruce
combined condensate
95* Aspen, 5* spruce
second effect
95* Aspen , 5* spruce
combined condensate
1/2V73
95* Aspen, 5* spruce
second effect 1/2U/73
100* Hardwood
100* Softwood
Total discharge .including
surface condensate
Exploded wood pulp
Hardboard mill 12/20/72
Hardboard mill 1A/73
Softwood
Softwood, high solids
effect
Softwood, low Bolida
effect
Softwood, 26 and 27
blend
Softwood, total con-
densate
Magnefite softwood 2/2/73
Magnefite softwood 2/5/73
MeOH,
ag/l
U20
530
1»20
370
1877
390
915
U05
1035
'tis
633
850
3UO
338
1(23
1(97
3U7
210
1(10
177
130
520
1)80
127
185
200
620
None
1,20
220
21.3
233
217
Acetic
acid,
mg/1
6260
5630
5000
1(790
6057
6200
7390
7790
6100
601(5
5533
6000
5372
5U72
6088
6U73
3837
U2I7
2730
3770
7650
5075
6930
1007
11(50
1300
2760
None
<300
<300
191(3
1533
2100
Furfural,
mg/1
630
7i(0
7UO
530
2273
575
1K85
575
Ik70
535
1000
1800
283
365
362
tt7
323
235
300
133
255
325
517
77
285
295
120
None
None
Hone
190
160
107
Ca,
90.9
92.5
119.5
108.5
Trace
2U.6
Trace
5.0
Trace
55.0
k.8
Trace
298.5
123.0
1(98.5
191.0
Trace
Trace
Trace
None
Trace
Trace
89l(.0
3.8
1.5
1.8
219.5
Trace
Trace
Trace
Trace
Trace
Trace
Na,
2.0
2.1
2.2
2.1
O.U
0.6
0.5
0.7
0.5
0.7
751
Trace
1,4
0.7
3.0
1.2
0.6
O.lt
1
Trace
•O.lt
0.6
2.2
12.9
5.9
7.8
3.7
20.5
2.1
9.8
1.3
0.5
0.8
prestripped neutralized SSL.
Total condensate from prestripped unneutralized SSL.
10
-------
Mill A cooks all hardwood and the samples submitted were daily compos-
ites taken by sampling at 15-ainute intervals. It is not known if the
samples from Mill B are grab samples or composites. However, there is
a. distinction between normal condensate and condensate used for wash.
The samples from Mill C were from 100$ hardwood and represent 8-hour
qomposites C9verlng a switching cycle on the Eosenblad evaporator. The
samples from Mi^l D were frpm a mill cooking 95$ aspen and 5$ spruce
by the magnesium bisulfite process. The combined condensate contained
condensate from the second and third effects of a 3-effect evaporator,
along with condensate from a surface condenser. The samples were 22-
hour composites. Of the three samples submitted by Mill E, one is
from a 100$ hardwood cook, another from a 100$ softwood cook and the
third is a sample of the total condensate discharge. The samples from
Mill F are condensates derived from a non-chemical cook. The four
samples from Mill H were unique in that these condensates were derived
from liquors that had received pretreatment prior to evaporation.
Samples 26, 27, and 28 were derived from liquors that had been pre-
stripped and neutralized, while Sample 29 was derived from a liquor
that was prestripped only. All four samples were grab samples of 100$
softwood origin. The samples from Mills I and J were derived from the
evaporation of magnefite softwood liquors. Both were grab samples.
Table 1 contains pH, S02, and color data. The COD, BOD5, and volatile
organic acids, consisting of acetic and formic acid data, are given in
Table 2. The data for the remainder of the assays, consisting of meth-
anol, acetic acid, and furfural, as determined by gas chromatography,
and assays for calcium and sodium are given in Table 3.
DISCUSSION
The. pH of the various samples (Table l) was between 2.12 and 3.30 for
a spread of about 1.2, except in Sample 5 which had a pH of 1.78 and
Samples 26, 27, and 28 which were preneutralized. There was a large
variation in S02 content, both total inorganic and organic loosely
combined, from the various mills, and even from samples derived from
11
-------
the same mill. This is the result of the cooking characteristics of
each individual situation. From the viewpoint of the condensate chem-
ical recovery process, those condensates containing relatively high SOa
content will have to be steam stripped.
Most condensate samples upon aging will turn darker in color. So the
color levels given in Table 1 are not necessarily indicative of what
the fresh condensate would be. However, it is evident that some of this
color is derived from lignin carry-over during evaporation or in some
cases lignin entrainment when the condensate is used to wash the evap-
orator.
The BODs (Table 2) is largely derived from the acetic acid, although in
some cases the furfural and methanol are heavy contributors. As would
be expected, the BODs on condensates derived from hardwood liquors is '
considerably higher than those derived from softwoods. It is much
easier to assess the source of the COD than for BODs. As an example,
if one calculates the COD derived from the acetic acid, the formic acid,
furfural, methanol, and S02 of Sample 21, a calculated COD of 1*100 mg/1
is obtained. Comparing this with a determined COD of 7620 mg/1 shows
that a good deal of the COD must be derived from liquor carry-over or
from liquor entrained during washup of the evaporators. The high color
content of 250 color units verifies this hypothesis. If one takes
several other samples at random and determines the calculated COD on
the same basis and compares it with the determined COD, the following
results are obtained: Sample 1, calculated COD 8050 mg/1, determined
COD 10,lUo mg/1 COD. Again the difference indicates relatively high
lignin content and again this is verified by the high color content of
200 color units. Sample 18 had a calculated COD of 8,631 mg/1 and a
determined COD of 9^8? mg/1. Obviously, this sample contained much
less liquor carry-over and this is verified by the lower color content
of 70 color units. Sample 29 had a calculated COD of 2,8ll mg/1 and a
determined COD of 3,191 mg/1, vith a low color of 15 color units.
Sample 31 had a calculated COD of 3,162 mg/1 and a determined COD of
12
-------
3,885 mg/1, with a color determination of 10 color units. It is quite
obvious that these two latter samples contained a very small amount of
liquor carry-over.
Assays performed by gas chromatography are given in Table 3. The
methanol and furfural varied greatly in samples from various mills, as
well as from samples from the same mill. The data from Mill D which
submitted samples of condensate during normal operation and condensate
that was used for washing, indicates a considerable loss in the meth-
anol and furfural during the washing operation. Calcium levels varied
considerably from mill to mill and from sample to sample within the
mill. The two primary sources of calcium present in the condensate are
from liquor carry-over and from removal of calcium from the evaporator
in the washing cycle. The sodium content was so low that it appears
to be of little or no consequence in these studies.
The data measuring the volatile pollution potential and materials
found in these samples are given in Table h under categories of conden-
sate derived from hardwoods and from softwoods. As would be expected,
the acetic acid content of those samples derived from softwoods is
generally lower than those derived from hardwood. There is no real
trend in the methanol, formic acid, or SOa content of the condensates.
The SOa content, of course, is mostly affected by the type and method
of cooking the wood. The furfural content of the hardwoods is general-
ly higher than those of the softwoods but there are exceptions. Since
the furfural is derived from the pentoses present in the original liquor,
it would be expected that the hardwoods would contain the higher fur-
fural content, but the level of furfural present is greatly affected by
the type of cook or method of evaporation utilized.
It will be noted that the acetic acid content was determined by two
methods and that the gas chromatography method usually produced higher
results than the ether extraction procedure. Extremely high acetic acid
results by the gas chromatography method are attributable to inter-
ferences by unknown materials present in some of these condensates. No
13
-------
Table k. POLLUTION POTENTIAL AND MATERIALS DATA
Acetic
Sample
number
1
2
3
1*
7a
Qa,e
9
10
11
12
13d
ll*d'f
15
2l*d
25d,f
16
I8b
19b'e
21
29°
30
31
COD,
mg/1
1011*0
11016
10822
8769
12062
10592
11268
7859
161*22
9780
5830
5830
9272
5162
6330
6152
91*87
771*3
7620
3191
2981
3885
BODS,
mg/1
6090
6252
6771*
5870
7890
71*16
5586
5088
71*56
61*38
1*1*76
1*350
7131*
3870
1*761*
1*506
7113
6258
3718
2331*
151*8
2358
acid,
mg/1
5875
5733
5l*ll*
5308
6523
6961*
1*650
1*960
5U71
5813
2618
2976
7266
2591
3551
1*022
1*277
1*529
2511*
1896
121*6
2350
MeOH,
mg/1
Hardwood
1*20
530
1*20
370
915
1*05
31*0
338
1*23
1*97
31*7
210
130
1*10
177
Softwood
520
1035
1*15
620
2l*3
233
217
Formic
acid,
mg/1
95
110
1
11*
1*2
32
1*1
11*
1*1*
62
110
36
11
25
25
2
1*
2
1*7
1*3
60
60
-
Furfural ,
mg/1
630
7l*0
7l*0
530
11*85
575
283
365
362
1*1*7
323
235
255
300
133
325
ll*70 f
535
120
190
160
107
Total
S02,
3Dg/l
610
61*0
670
920
298
608
366
25!+
710
293
123
305
221+
263
627
1023
5l+9
189
1058
3l*U
150
1*87
Hardwood, 1% softwood.
86j< Softwood, ll*>5 hardwood.
,From prestripped SSL.
,,, Aspen, 5? spruce.
°Used for wash.
Second effect.
11*
-------
effort was made to alter the gas chromatography procedure to alleviate
this problem, since the ether extraction procedure had also been used.
CONCLUSIONS
The analysis of the condensate samples submitted by sponsors of this
project have indicated the potential amount of S02, acetic acid,
methanol, and furfural that would be available for recovery. For most
of the participating mills, the output of these components would exceed
the recoverable chemicals available in the condensate from the Consoli-
dated mill where the system was under pilot-scale study.
15
-------
SECTION V
DESIGN, CONSTRUCTION AND OPERATION OF PILOT SYSTEM
' AT THE APPLETON MILL OF CONSOLIDATED PAPERS, INC.
PREVIOUS WORK
Initial work in Scott Paper Company's Everett Mill was done using 4
inch D x U ft H glass pipe for all operations. Four activated carbon
columns were placed in series to study the selective adsorption of
methanol, furfural, and acetic acid in aqueous solutions by granular
activated carbon. Filtrasorb ilOO was used. This work showed that:
1. Methanol came out of Column 1 in 15 minutes, Column 2
in 30 minutes, Column 3 in h5 minutes, and Column U in
an hour.
2. Acetic acid came out of Column 1 in 1 hour, Column 2
in 2-1/2 hours, Column 3 in 3 hours 50 minutes, and
Column k in k hours 50 minutes. Furfural did not come
out of Column 1 after more than 10 hours of operation.
At the start all columns had the activated carbon flooded with water.
It takes approximately 10 minutes for the water to be displaced with
feed in each column (Uo minutes for total displacement assuming plug
flow), so only the break through, not full loading, was used for cal-
culating the following carbon loadings.
Methanol Loading — Four Columns
CH3OH = 12.6 gph (1 hour) (8,33 Ib/gal.) (
= 0.051^ lb 23.3 g
CH3OH Loading = 0.051^ lb/28A lb carbon
= 0.00181 Ib/lb carbon
= 0.00181 g/g carbon
Acetic Acid Loading — Four Columns
CH3COOH = 12.6 gph (5 hours) (8.33 Ib/gal.) (0 0053)
= 2.T8 lb 1.26 kg
16
-------
CH3COOH Loading = 2.78 lb/28.1* lb carbon
= 0.098 Ib/lb carbon
= 0.098 g/g carbon
Furfural Loading — First Column
CUH3OCHO = 12.6 gph (10.1 hours) (8.33 Ib/gal) (O.OOOU8)
= 0.508 Ib 0.23 kg
C^HaOCHO Loading = 0.508/7.08 = 0.0718 Ib/lb carbon
= 0.0718 g/g carbon
t
Therefore, it was^ necessary to remove methanol by steam stripping and/or
fractionation. Using various combinations of steam stripping and
fraqtionation column operations, the following products were recovered:
85$ + by weight S02 vapor
+ by weight methanol
+ by weight furfural
These results provided two possibilities for furfural removal and
potential recovery, fractionation and/or adsorption by activated carbon.
Using these data, a pilot plant was built at Scott's Everett Mill using
new, existing and scrap equipment to study recovery of volatile chemi-
cals from the Everett mill hot water accumulator overflow (blow gas
condensates). Essentially, it consisted of k inch (10.2 cm D) diam-
eter glass piping with steel plates and reflux splitters, and two 10
inch (25.U cm) diameter stainless steel adsorbers (also used as re-
actors during regeneration) complete with column packings, heat ex-
changers, and pumps. This equipment presented the opportunity to in-
expensively experiment with various combinations of steam stripping,
fractionating and/or adsorbing operations. Scott Paper Company used
this pilot plant equipment in their Everett, Washington mill to study
chemical recovery from clean evaporator and blow gas condensates. The
equipment had been installed and operated as shown on Fig, 1. All the
above mentioned chemicals, as well as 90$+ by weight ethyl acetate,
were recovered. Patents covering the processes, as well as their
17
-------
SOz Vapor
Methanol
*,
1%
OK
m
_OL
Steam
Stripping
Column Fee1
Drum
Evaporator
Condonsate
Feed
Water
a
Stripping
Column
Feed Pump
Steam
Strip-
ping
Column
Furfural
Fractionation
Column
en
Fractionation
Coltunn Feed
Pump
1 I
1st
a
3
vated Car
mn Furfur
tt J
-o o
*$ o
fn
1)
O
s
1 '
1
2nd
a
,i
1
S
C -H
° -S
Hi U
O -H
-P
•^3
-------
adaptability to kraft condensates, were applied for by the Scott Paper
Company in 1971-
METHODOLOGY
In July, 1972 the pilot plant at Everett was dismantled and shipped to
Appleton. Since preliminary analyses of Consolidated's evaporator con-
densates showed their methanol, acetic acid, and furfural compositions
considerably lower than those encountered at Scott, the initial pilot
plant units were arranged as shown on Fig* 2. The activated carbon
regeneration systems were built as shown on Fig. 3 and U. No more than
50 minutes storage capacity was provided for storing the evaporator
condensate fed to the pilot plant area. No special arrangements were
made with Consolidated Papers, Inc.'s operating personnel. The evapor-
ator condensate was sent to the pilot plant in the same manner that it
was drained. The main differences between this work and that done
previously were:
1. The evaporator condensate was processed within minutes after
it was produced in the evaporators. Previous work was done on
condensates that had set long period of time (weeks) before
they were processed.
2. At times, portions were received of this condensate which had
been used to wash scale, fibers and spent sulfite liquor from
the evaporator surfaces. Previously only clean condensate
was used.
One purpose of this pilot plant project was to prove the processes
developed could handle evaporator condensate after it had been used to
wash the evaporator surfaces. It. was also.desired to provide a choice
of commercially feasible alternatives for processing of evaporator con-
densates, with recovery of saleable values wherever possible to help
pay the costs of disposal processing of this waste flow from acid
sulfite mills equipped with evaporators. It was anticipated that more
j
than one possible route to recovery of these volatile components would
19
-------
SO; Vapor
n I
Steam Stri
ping Colum
Feed Drum
SSCF
130-155°F
5l».5-68.3°C
Feed Pump
Stripping
98.5-100.5°C
200-213°F
97°F
36.7°C
Con-
Fractiona-
tion Columnl
65.5-T6.6c
150-170°
ndensate
eed
Water
a
HI
1
Feed Pump
Distillation
•1213°F
100.5\
FCO — FCFRO
Methanol
FCSC — FCFRSC
Furfural
33
ACCFE
SI
135
57-1
lUO°F
I
~OTJI
°c
1st
Carbon
•fural
ivated
1WI« T?n»
3
=4
-4
•P H
O O
< 0
11*0°
Cooler
F
60°C
m
\ '
133°F
2nd
Column
1
§•3
cd o
O -H
tj
-------
Methanol from
Fracti onation Column
Methanol
Feed
Drum
® g
§•
*«*
3|
•P J3
oj -p
(U 0)
ACCFRI
ro
H
Water
Methanol
Feed Pump
1st
jh
o "M
•d a
o£
II
-P H
O O
< O
ACCFRE
'ractionation
Column Feed
Drum
FCRF
To Fraotionation
Column
Transfer
Pump
'•'ractiona.tion
Column Feed
Pump
Figure 3. Activated carbon column furfural regeneration
-------
ro
ro
Make Up
Ethanol
Ethanol
Feed
Drum
ACCACBI
Ethanol
Feed Pump
Water
2nd
8,
a -H
o o
o
O TH
0) O
•p <:
4J 0>
0>
aa o
01
a>
a
o
o
0)
O
4>
K
ACCACRE
CM
o
FCACRSC-1
Transfer
Pump
Conden-
ser
FCACRO
Ethanol — Ethyl Acetate
Fractionation Column
(Methanol — Furfural
Fractionation Column
Modified)
•Water to Drain
TCACSB
Figiire U. Activated carbon column acetic acid regeneration
-------
be developed in ways which would fit the variety of operating conditions
in the various sulfite pulp mills supporting this project.
The initial pilot plant constructed as per Fig. 2 was not a replica of
the one at Scott's Everett Mill. The flow pattern was more advanced
and tailored to the needs of the Appleton Div. of Consolidated Papers,
Inc. For simplification it will be discussed in five parts: the feed
system, steam stripping column system, fractionation column system, the
activated carbon adsorption system, and the activated carbon regenera-
tion system.
Feed
The evaporator condensate was pumped by means of a transfer pump from
the evaporator area through the pilot plant area to the river through a
2-inch diameter stainless steel pipe. The feed was tapped off this
transfer line and run through a screening system to remove particulate
material (fibers and evaporator scale) to protect the stainless steel
mesh packing in the stripping column. From the screening system it
went into a 55-gallon (208-liter) stainless steel feed drum. From this
drum it was pumped by two small Eastern pumps in series through the
feed preheater to the steam stripping column. Two pumps were required,
because one pump could not constantly deliver the required flow when
back pressures exceeded 15 psig (1.05 kg/cm3 gage). No effort was made
to determine the scale and fiber content as the amounts varied. It was
established, however, that scale and fiber were in the evaporator con-
densate and had to be removed^ Excellent screening performance was
attested to, since the steam stripping column did not require cleaning
for a period extending back over three years of operation in Appleton
and Everett.
Steam Stripping
The steam stripping system consisted of the preheater, steam stripping
column, reboiler, condenser, and receiver. The preheater was used to
bring the evaporator condensate up to its boiling point, 209 to 211°F
23
-------
(98.3 to 99.U°C). The stripping column is U inch (10.2 cm) diameter
and contains a U ft (121.9 cm) height of packing* The packing used was
Goodloe 316 S.S. mesh. The evaporator condensate was fed into the top
of the column and steam was fed into the bottom. The steam;rate was
controlled by pumping water into the reboiler and super-heating it in-
directly with steam. In that way all the Water was converted to steam.
The vapor left the top and flowed down through the condenser. The con-
densed vapor was collected in the k inch (10.2 cm) diameter glass re-
ceiver. By controlling the temperature of the vapor leaving the1re-
ceiver at 205 (96.1) to 190°F (87.8°C), the S02 concentration in the
exhaust vapor was controlled. The steam stripped evaporator condensate
was pumped from the "bottom of the column to the first activated carbon
column. The feed rate capability was 1.5 gpm maximum (5.7 liters per
minute). Plans were made to evaluate the steam stripping column system
while varying the feed rates from 0 = 5 to 1.5 gpm (1.9 to 5.7 liters per
minute) and using 2 to 12$ by weight of feed stripping steam.
Fractionation
The fractionation system consisted of a fractionation column, a re-
boiler and a condenser. The fractionation column had four packed
sections of k inch (10.2 cm) diameter x U ft (121.9 cm) high glass
piping between the feed and bottom of the column [l6 ft (U87.7 cm) of
packing total], one packed section of k inch (10.2 cm) diameter x U
inch (121.9 cm) high glass piping between the feed and the first reflux
splitter [U ft (121.9 cm) of packing] and one packed section of h inch
(10.2 cm) diameter x 2 ft (6l.O cm) high glass piping between the first
splitter and top splitter [2 ft (6l.O cm) of packing]. All packing were
Goodloe S.S. mesh. The condenser was directly above the top reflux
splitter. The reboiler was operated the same as that on the steam
stripping column to ensure complete vaporization.
Initially, condensed overheads from the steam stripping columns were
fed into this system. Sulfur dioxide as a vapor was vented through the
top of the condenser, a methanol crude was withdrawn from the top
2k
-------
splitter and a furfural crude from' the bottom splitter. Disposal of
the bottoms from the fractionation column was to be studied. In a com-
mercial operation they might be sent back to the evaporators, since the
material they contain were no longer volatile. However, this could in-
volve about a 10% increase in evaporation load. This raised the ques-
tion of how many participating mills could handle a 10% increase in
evaporation load. Plans were to operate this column in conjunction
with the feed and steam stripping column systems, and study the effect
of the steam stripping operation on this column and its product purities,
Activated Carbon Adsorption
The activated carbon adsorption system consisted of a feed pump, heat
exchanger to control the temperature of the steam stripped evaporator
condensate to the carbon columns, and 10 inch I.D. (25.^ cm) stainless
steel columns. The first one was fed up-flow and the second one, which
was Jacketed, down-flow. The liquid level in the second column was
controlled above the carbon by running the discharge tubing higher than
the top of the column. Disengaging sections were provided over and
under the columns. Temperature and pressure measurements were taken in
these sections in both columns. In the nonjacketed column there was a
connection for temperature and pressure readings in the middle. The
discharge condensate was piped to the drain. The nonjacketed column
contained 98 pounds (kU.5 kg) of Filtrasorb kOO granular carbon and the
jacketed one 110 pounds (U9-9 kg). The purpose was to study the bene-
fits of selective adsorption of acetic acid, furfural, and polymerized
material in the two carbon columns, and to evaluate the effectiveness
of the developed carbon regeneration.systems.
Activated Carbon Regeneration
The initial activated carbon regeneration systems were built as shown
on Fig. 3 and k. These systems were identical to those developed and
used in Scott Paper Company's Everett Mill's pilot plant. For recovery
of furfural, polymerized material and that adsorbed portion of spent
25
-------
sulfite liquor, methanol was pumped from' the feed drum through a heat
exchanger where it was vaporized and slightly superheated. The super-
heated vapor entered the top of the carbon column and forced the con-
densed methanol, water, furfural, polymerized material and that adsorb-
ed portion of spent sulfite liquor out the bottom. The first portion
of liquid removed, approximately 10 gallons (37.8 liters) was drained.
A i ' 'j
Analyses had shown it to contain less furfural and roughly the same
amount of methanol and acetic acid as in the feed to the column. Event-
ually, higher concentrations of methanol, furfural, and residual
material came through as a liquid. Finally, the methanol vapor came
out the bottom, and was condensed and collected in the receiver. Once
the condensed methanol vapor contained only trace quantities of furfur^
al, water was fed into the vaporizer to produce steam. The steam
forced the methanol vapor out of the carbon. The methanol was conden-
sed, collected in the receiver, and pumped into the fractionation col-
umn for eventual recovery. When only steam remained in the column, the
steam stripped evaporator condensate, cooled, was pumped into the carbon
column displacing and condensing the steam. The carbon column was thus
back on its normal adsorption cycle. The methanol and furfural were
recovered in the fractionation column in the same manner as described
for recovering them from the steam stripping column condensed overhead
vapors. The polymerized material and residual spent sulfite liquor
were held in the bottom of the fractionation column. When the concen-
tration (over 20% by weight) affected the pumps recirculation rate or
when there was an excessive amount of water, it was discharged to the
drain.
Activated carbon serves as a catalyst for the esterification of acetic
acid with ethanol in the vapor phase and the resultant ethyl acetate is
a readily marketable product. These facts provided the bases for choos-
ing ethanol as the regenerating agent for the second activated carbon
column. Figure k shows the manner in which the vaporized ethanol was
passed upward through it. This column was jacketed for steam to pre-
vent condensation in the carbon. The overhead vapors were then
26
-------
condensed and collected in a receiver. From the receiver the conden-
sate was fed into a fractionation column. Ethyl acetate was collected
overhead, and ethanol was recycled from a point about one third of the
height from the bottom of the column back through the carbon column.
Water recycled back through the activated carbon system' and was removed
from the bottom of the column. The reboiler was used to control the top
reflux and the vaporization of ethyl acetate.
EXPERIMENTAL OPERATION AND EVALUATION OF PILOT PLANT
Steam Stripping
At the September, 1972 meeting of the project sponsors, plans to evalu-
ate the steam stripping column system varying feed rates from 0.5 to
1.5 gpm (1.9 to 5.7 liters per minute) and using 2 to 12$ by weight of
feed stripping steam were reported. The height of the column packing
was to be varied under similar operating conditions. This work was
completed with the equipment set up as shown in Fig. 2. The trouble
points when operating this system were:
1. Controlling the temperature of the evaporator condensate feed
entering the column at its bubble point.
2, Maintaining a constant pressure drop across the column packing.
3. Controlling the condenser so as to maintain a constant temp-
erature for the condensed steam stripping column overhead
vapors.
h. Maintaining a constant back pressure from the S02 venting
system.
All of these were related to the concentration of the volatile chemi-
cals in the evaporator condensate feed.
To compare Consolidated Papers, Inc.'s Appleton Div. Mill evaporator
condensate steam stripping data with that of Scott Paper Company's
Everett Mill blow gas condensate steam stripping data, refer to Fig. 5
to 12. For a fair comparison keep these points in mind. At Everett
27
-------
100
Scott Paper Company
Everett Mill Blow Gas Condensate
!+8-Inch (121.9 cm) Goodloe Packing
3 1* 5 6 7 8 9
Stripping Steam, % weight of feed
Figure 5- Methanol removal ys_. stripping steam used - Scott Paper
Company — l*8-inch (121.9 cm) Packing
28
-------
100
90
80
70
60
o
9
fi
-p
Scott Paper Company
Everett Mill Blow Gas Condensate
• 6-Inch (15.2*1 cm) Goodloe Packing
+ 12-rInch (30.1*8 cm) Goodloe Packing
X 18-Inch. (1*5-72 cm) Goodloe Packing
2l»-Inch (60.P6 cm) Goodloe Packing
O 30-Inch (76.20 cm) Goodloe Packing
3 ' 1* 5 6 7 8
Stripping Steam, % weight of feed
10
11
12
Figure 6. Methanol removal vs_. stripping steam used — Scott
Paper Company — variable packing
29
-------
100
90
80
70
* 60
S 50
S
3
I
3
co
30
20
10
SoOtt Paper Company
Everett Mill Blow Gas Condensate
l»8-Inoh (121.9 en) Ooodloe Packing
3^56789
Stripping Steam, % weight of feed
Figure 7« Sulfur dioxide removal vs.. ; stripping steam
used — Scott Paper Company ~ ^8-inch (121.9 can) packing
30
-------
10Q-
8C~
7< -
6ct-
1
K
%
•rl
S
w
2C ~
1C ~
3C-
I
I
I
4
X
Scott Paper Company
Everett Mill Blow Gas Condenaate
6-Inch (15.2U can) Goodloe Packing
12-Inch (30 J+8 cm) Goodloe Packing
18-Inch (^5.72 cm) Goodloe Packing
2U-Inch (60.96 cm) Goodloe Packing
30-Inch (76.20 cm) Goodloe Packing
I
I
I
3^56789
Stripping Steam, % weight of feed
10
11
12
Figure 8. Sulfur dioxide removal vs. stripping steam
used — Scott Paper Company — variable packing
31
-------
100
Consolidated Papers, Inc.
Appleton Division Mill Evaporator Condensate
US-Inch (121.9 cm) Goodloe Packing
I
I • I
JL
±
_L
3 !» 5 6 7 8 9
Stripping Steam, $ weight of feed
10
11
12
Figure 9- Methanol removal vs_. stripping steam used -
Consolidated Papers, Inc. - k8-inch (212.9 cm) packing
32
-------
ioo r
90
80
70
60
50
30
20
10
Consolidated Papers, Inc.
Appleton Division Mill Evaporator Condensate
12-Inch (30.U8 cm) Goodloe Packing
I
I
I
_L
I
I
I
1 2 3 1* 5 6 7 89 10 11 12
Stripping Steam, % weight of feed
Figure 10. Methanol removal vs. stripping steam used —
Consolidated Papers, Inc. —12-inch (30.^8 cm) packing
33
-------
IOC
•**.
i
g
w
Consolidated Papers, Inc.
Appleton Division Mill Evaporator Condensate
US-Inch (121.9 cm) Goodloe Packing
OLO,
O
o
_L
_L
3 1* 5 6 7 89
Stripping Steam, % weight of feed
10
11
12
Figure 11. Sulfur dioxide removal vs. stripping steam used
Consolidated Papers, Inc. — ^8-inch (121.9 cm) packing
-------
ioo r
90
80
70
TI
Consolidated Papers, Inc.
Appletbri Division Mill Evaporator Condensate
12-Inch (30.U8 cm) Goodloe Packing
** 60
1
o
I
-H 50
x
o
a
OLC.
uo
30
20
10
I
I
I
I
I
I
3 1» 5 6 7 8 9
Stripping Steam, % weight of feed
10
11
12
Figure 12. Sulfur dioxide removal vs. stripping steam
used — Consolidated Papers, Inc. — 12-inch
(30.U8 cm) packing
35
-------
there was a feed tank holding more than 1500 gallons (5670 liters) of
blow gas condensate, and the feed entered the column at a temperature
of 205°F (96.1°C). At Consolidated a 50-gallon (189 liters) feed drum
was used, and the feed entered the column at a temperature of approxi-
mately 211°F (99-H°C).
Figures 5 and 6 show the percentage of methanol removal from Everett
Mill's blow gas condensate as a direct straight-line function of the
amount of stripping steam used. Note that none of the points are more
than the equivalent of 5$ methanol removal away from the line drawn,
and both lines go through zero. This indicates the trial runs were
made under ideal operating conditions. Figures 7 and 8 show 90$ or
more SOa removal, when using 2% or more weight of feed stripping ,
steam. The better than 99% SOa removal in Fig. 7 versus the poor
results on Fig. 8 was due to the following:
1. When the feed entered the column at 205°F (96.1°C), less than
2% by weight of feed stripping steam was not adequate to over-
come the column heat losses and sensible heat required to
raise the liquid in the column to its bubble point and effec-
tively strip-out the SOa.
2. The data plotted in Fig. 7 were the result of samples analyzed
immediately after they had been taken. Analyses in Fig. 8
were conducted hours or even the following day after the
samples were taken. The free S02 shown in Fig. 8 was believed
to be that regenerated from S02-OLC which was not removed.
This was verified in future runs. S02-OLC became a\i important
part of the analyses from August, 1971 to date.
The clusters of points in Fig. 6 at 3 and 6% by weight of feed strip-
ping steam showed very little benefit derived from increasing column
packing height. At 1.0% stripping steam rate some advantage was shown
by increasing column height. Compare straight-line points in Fig 5
and 6. However, it was doubtful that the increase would justify th
additional construction costs.
36
-------
Figures 9 and 10 show the percentage of methanol removal from the Con-
solidated Mill evaporator condensate. The points were very widely scat-
tered, and the lines were discretely drawn. On Fig. 9 the two zero
points were neglected as being bad data. The line was drawn keeping as
many points on or near it as possible and an equal number on either
side. Note that neither line on Fig, 9 and 10 goes through zero. This
indicated the bubble point of our feed was probably exceeded arid it
partially vaporized in the preheater. Figures 11 and 12 show the per-
centage S02 removal of both TI and OLC. The S02-TI removal was similar
to that shown for Scott in Fig. 8. The analyses were done in The
Institute of Paper Chemistry laboratory hours or days after the samples
were taken. An important fact shown in Fig. 11 and 12 was that no more
than 75% of the SQa-OLC was removed. Similar work on Scott's blow gas
condensate showed 90+$ SOa-OLC consistently being removed. The dif-
ference was attributed to the SOa-OLC in Consolidated Mill's evaporator
condensate as the result of residual spent sulfite liquor removed from
the evaporator by the condensate in the washing cycle. This SOa-OLC
was much more firmly combined.
This concluded the steam stripping experimental portion of this pro-
ject. These data can assist any pulp and paper company interested in
steam stripping sulfite mill evaporator condensate with their design,
construction, and operation of a steam stripping system. The pilot
plant also was in a position, with Consolidated Papers, Inc.'s approval,
to test individual mill's evaporator condensates under predetermined
operating conditions.
Fractionation
During the steam stripping runs the condensed overheads from the steam
stripping column were fed into the fractionation column. Methanol and
furfural crudes were recovered. Balances based on gas chromatographic
analyses of the steam stripping column's feeds and condensed overheads
are given in Tables 5 and 6. From the methanol balance, Table 5
methanol analyses showed *Kl96 lb (1.90 kg) to be in the condensed
37
-------
Table 5. METHANOL BALANCE STEAM STRIPPING RUNS
9-11-72 to 9-20-72
9-11-72
In Feed
U hr (1.52 gpm) (60 min.) (8.33 Ib/gal.) (0.000lt5) - 1.37 lb (0.622 kg)
In Overhead
It hr (0.008) (60) (8.33) (0.027) = 0.1*3 lb (0.195 kg)
% Recovery = (0.1+3/1.37) 100 = 31.1+$
9-12-72
In Feed
It (1.368) (60) (8.33) (0,000575) - 1-575 lb (0.715 kg)
In Overhead
It (0.038) (60) (8.33) (0.00385) = 0.290 lb (0.132 kg)
% Recovery - CO.290/1.575) 100 • 18.60
9-12-72
In Feed
It (1.215) (60) C8.33) (0.0001*5) = 1.095 lb (0.1*97 kg)
In Overhead
1* Co.Ol+9) (60) (.8.33) CO.0030) = 0.290 lb (0.132 kg)
% Recovery = CO.290/1.095) 100 » 26.5$
9-11+-72
In Feed
It (0.76) C60) (.8.33) CO.00055) • 0.81+ lb (0.381 kg)
In Overhead
It (0.0lt5) (60) (8.33) (0.00355) • 0.32 lb (O.Ht5 kg)
% Recovery = (0.32/0.81t) 100 = 38.0$
9-1U-72
In Feed
It (0.608) (8.33) (60) (0.000725) = 0.88 lb (0.399 kg)
In Overhead
1+ (O.OU3) (8.33) (60) (0.001+225) • 0.36 lb (0.163 kc)
% Recovery * (0.36/0.88) 100 = 1+1.2$
33
-------
Table 5 (cont'd). METHMOL BALANCE STEAM STRIPPING RUNS
9-11-72 to 9-20-72
9-15-72
In Feed
1* (0.502) (8.33) (60) tO.000575) = 0.578 Ib C0.263 kg)
In Overhead
U (0.01*09) (8.33) (60) (0.00335) - 0.328 Ib (0,ll*9 kg)
% Recovery • (0.328/0.578) 100 • 56.8#
9-18-72
In Feed
1* (1.52) (8.33) (60) (0.00091*) - 2.ffc Ib (1.2^5 kg).
In Overhead
1* (0.01*7) (8.33) (60) (0.001*3) - 0.1*05 Ib (0.18U kg)
% Recovery = (0.1*05/2.71*) 100 - ll*.8$
9-18-72
In Feed
U (1.368) (8.33) (60) (0.00056) m 1.53 Ib (0.695 kg)
In Overhead
1* (Q.0l*5) (8.33) (60) (0.00351*) " 0.318 Ib (0.11*1* kg)
% Recovery = (.0.318/1.53) 100 • 20.8$
9-19-72
In Feed
2 (1.11*0) (8.33) (60) (0.00061) - 0.695 Ib (0.316 kg)
In Overhead
2 CO.050) (8.33) 160) (0.00l*2l*) - 0.212 Ib (.0.096 kg)
% Recovery = (.0.212/0.695) 100 - 30.5$
9-19-72
In Feed
2 (1.215) (8.33) (6p) CO.00056) = 0.680 Ib (0.309 kg)
In Overhead
2 (0.057) (8.33) C60) (0.00361*) = 0.207 Ib (0.09U kg)
% Recovery » CO.207/0.680) 100 = 30.5%
9-19-72
In Feed
1* (0.76) C8.33) C60) CO.00035) - 0.532 Ib (0.2fc2 kg)
In Overhead
1* (0.01*6) C8.33) C60) CO.00270) - 0.21*8 Ib (0.113 kg)
% Recovery = (.0.21*8/0.532) 100 = 1*6,7^
39
-------
Table 5 Ccont'd). METHANOL BALANCE STEAM STRIPPING RUNS
9-11-72 to 9-20-72
9-20-72
In Feed , ,
It (0.608) (8.33) (60) (0.000725) - 0.881 Ib (0.40 kg)
In Overhead
1+ (0.050) (8.33) (60) (0.001*12) = 0.1*12 Ib (0.187 kg)
% Recovery = (0.412/0.881) 100 = 46.8$
9-20-72
In Feed
1* (0.502) (8.33) (60) (0.0007) = 0.704 Ib (0.32 kg)
In Overhead
1* (0.051) (8.33) (.60) CO.00369) = 0.376 Ib (0.171 kg)
% Recovery = (0.374/0.704)'100 = 53.5%
\
Total methanol in overhead
0.430 Ib
0.290 Ib
0.290 Ib
0.320 Ib
0.360 Ib
Ib Collected - 4.3 Ib (1-95 kg)
0.318 Ib
0.212 Ib
0.207 Ib
0.248 Ib
0.412 Ib
0.376 Ib
4.196 Ib (1.90 kg)
40
-------
Table 6. FURFURAL BALANCE STEAM STRIPPING RUHS
9-11-72 to 9-20-72
9-11-72
In Feed
k hr (1.52 gpm) (60 min.) (8.33 Ib/gal.) (O.OOOl) = 0.30 Ib (0,136 kg)
In Overhead
U (0.008) (60) (8.33) (0.0012) = 0.019 Ib (0.0086 kg)
% Recovery = (0.019/0.30) 100 = 6.3$
9-12-72
In Feed
1* (1.368) (60) (8.33) (0.00025) - 0.681* Ib (0.310 kg)
In Overhead
U (0.038) (60) (8.33) (0.00155) = 0.117 Ib (0.053 kg)
% Recovery = (0.117/0.681*) 100 = 17.1*
9-12-72
In Feed
1* (1.215) (60) (8.33) (0.00020) = 0.1*86 Ib (0.221 kg)
In Overhead
1* (0.01*9) (60) (8.33) (0.001385) = 0.133 Ib (0.0601* kg)
% Recovery = (.0.133/0.1*86) 100 = 27 .'M
9-1U-72
In Feed
It (.0.7-6) (60) (8.33) (0.000175) • 0.267 Ib (0.121 kg)
In Overhead
1* (0.01*5) (60) (8.33) (0.001) = 0.090 Ib (O.OkOQ kg)
% Recovery • Co.090/0.267) 100 - 33.7#
In Feed
1* (0.608) (60) (8.33) (0.0002) = 0.21*3 Ib (0,110 kg)
In Overhead
U (0.01*3) (60) (8.33) CO.00105) • 0.089 Ib (O.Ql*Ol* kg)
% Recovery = (0.089/0.21*3) 100 = 36.8$
-------
Table 6 (cont'd). FURFURAL BALANCE STEAM STRIPPING RUNS
9-11-72 to 9-20-72
9-15-72
In Feed
1* (0.502) (60) (8.33) (0.0002) = 0.201 It (0.0912 kg)
In Overhead
1* (0.0^9) (60) (8.33) (0.000925) = 0.0906 Ib (0.01*11 kg)
% Recovery = (0.0906/0.201) 100 = 1*5.0%
9-18-72
In Feed
1* (1.52) (60) (8.33) (0.000265) = 0.772 Ib (0.351 kg)
In Overhead
1* (O.OU7) (60) (8.33) (0.001335) = 0.126 Ib (0.0571 kg)
% Recovery = (0.126/0.772) 100 = 16.3$
9-18-72
In Feed
^ (1.368) C60) (.8.33) (0.000175) = 0.1*78 Ib (0.217 kg)
In Overhead
It (0.01*5) (60) (8.33) (0.000885) = 0.0795 Ib (0.0361 kg)
% Recovery - (0.0795/0.1*78) 100 = ±6.6%
9-19-72
In Feed
2 (i.iko) C60) (.8.33) (0.00015) = 0.171 Ib (0.0776 kg)
In Overhead
2 (0.050) C60) (8.33) CO.00091) » 0.01*55 Ib (0.0207 kg)
% Recovery = (0.01*55/0.171) 100 = 27.6$
9-19-72
In Feed
2 (1.215) (60) (8.33) (0.0002) = 0.2**3 Ib (0.110 kg)
In Overhead
2 (0.057) (60) (8.33) CO.00071) = O.Ql*ol» ib (0.0183 kc)
% Recovery = C0.0l*ol*/0.2l*3) 100 = ±6.6%
9-19-72
In Feed
1* (0.76) (60) (8.33) (0.000125) = 0.190 ib (0.0862 kg)
In Overhead
1* C0.0l*6) (60) (8.33) (0.000635) = 0.0583 Ib (0.0265
% Recovery = CO.0583/0.190) 100 =30.7$
1*2
-------
Table 6 (cont'd). FURFURAL BALANCE STEAM STRIPPING RUNS
9-11-72 to 9-20-72
9-20-72
In Feed
1» (0.608) (60) (8,33) CO.0001) = 0.122 Ib (0.055** kg)
In Overhead
U (0.050) (60) (8.33) (0.000825) = 0.0825 Ib (0.0371* kg)
% Recovery = (0.0825/0.122) 100 = 67-65?
9-20-72
In Feed
k (0,502) (60) (8.33) (0.00015) = 0.151 Ib (0.0685 kg)
In Overhead
k (0.051) (60) (8.33) (0.000825) = 0.08U Ib (0.0381 kg)
% Recovery = (0.08U/0.151} 100 = 55-6$
Total furfural in overhead
0.019 Ib
0.117 Ib
0.133 Ib
0.090 Ib
0.089 Ib
0.091 Ib
0.126 Ib Collected - 0.925 Ib (O.H19 kg)
0.080 Ib
0.0^6 Ib
, Q.OkQ Ib
0.058 Ib
0.082 Ib
0.08U Ib
1.055 Ib (O.U79
-------
overheads. The methanol recovered was U.3 lb (1-95 kg)- Although
analyses showed it to be in the 85 to 90% by weight range, the bubble
point and specific gravity indicated a 95+$ by weight methanol crude.
From the furfural balance, Table 6 furfural analyses showed 1.055 lb
(O.U8 kg) to be in the condensed overheads. Furfural recovered was
0.925 lb (O.U2 kg). Analyses showed it to be in the 8U to 92% by
weight range. This work was done before we developed our gas chromato-
graphic programs for analyzing high purity methanol, ethanol,-ethyl
acetate, methyl acetate, and furfural products. No sample of the
fractionation column bottoms contained even trace quantities of
methanol or furfural. There was zero pressure drop from the feed port
to the bottom of the fractionation column indicating this column was
grossly oversized. No attempt was made to preheat the condensate to
its bubble point. The reboiler and condenser were more than adequate.
The work at Scott Paper Company utilized steam stripping and fraction-
ation on the evaporator condensate before it was pumped through the
activated carbon columns, and the fractionation column was the piece
of equipment that limited the pilot plant capacity to processing less
than 1/2 gpm (1.9 liters per minute) of the condensate. No attempt was
made to optimize the size of the fractionation-column, since the pro-
ject priority was shifted to getting activated carbon adsorption and
regeneration data.
Activated Carbon Adsorption
After the steam stripping work was completed, steam stripped evaporator
condensate was processed through the activated carbon columns at the
rate of 1 gpm (3.8 liters per minute). Six percent by weight of feed
stripping steam was used. The overhead vapors were condensed and prp-
cessed through the fractionation column. The steam stripped evaporator
condensate was processed as shown in Fig. 2 (p. 20).
The first run was made^ October 9, 1972. Efforts were concentrated on
the steam stripping and activated carbon adsorption columns. Almost
100 samples were taken over a l6-hour period to effectively study these
-------
unit operations. Hundreds of separate analyses were made on these
samples. The data accumulated have been plotted in Fig. 13 to 22.
Figure 13 shows the acetic acid concentrations in the feed to and in
the bottom liquid from the steam stripping column. Interestingly, the
acetic acid concentration was consistently lower than that in the feed,
most of the time much lower than can be accounted for by the 6$ strip-
ping steam dilution. Since a darkening of color in the processed con-
densate was apparent, the acetic acid lost may be responsible in part
for some of the polymerized material collected. No analytical effort
has been made to verify or disprove this. Figure lU shows the acetic
acid concentrations of the feed to and effluent, from the first activat-
ed carbon column. There was no acetic acid in the effluent for over 3
hours. From previous Scott work the breakthrough point had been cal-
culated to occur after 2-3A hours. The increased time may have been
due to lower concentrations of acetic acid in the feed. Note that
after 5 hours there was more acetic acid in the effluent than in the
feed. Then, after 8 hours the carbon was adsorbing it again. During
this period the feed concentration was rising. After 13 hours there
was more in the effluent once again. Figure 15 demonstrates some
analytical problems. Up until this time one injection of the sample
was made into the gas chromatograph. Since the acetic acid concen-
trations of the samples were relatively close, no problems had arisen.
Now injections of samples with acetic acid concentrations varying from
1% or better were mixed with those containing no acetic acid. Since
it was illogical that the acetic acid would break through both activat-
ed carbon columns at the same time, two of the effluent samples fortu-
nately saved were checked. They showed no acetic acid, ith the improv-
ed injection procedure, so it was concluded the acetic acid shown
previously was coming from the injection needle and/or gas chromato-
graph column. Poropak Q columns require some column conditioning with
acetic acid. Similar experiences were encountered at Scott. All later
quantitative work on the gas chromatograph utilized 3 sample injections.
With this modification of analytical procedure it could be shown that
little or no acetic acid was adsorbed in the first U hours because it
1*5
-------
0.1)1-
Steam Stripping Column
Feed Rate = 1 gpm (3-78 liter/minute)
Stripping Steam • 6% by Weight of Feed
•w.
o
3
V
o.J
Feed e o e
Bottoms x X—— X
6 ' T 8 9 10 11 12
13 lU
Figure 13. Acetic acid concentrations — steam stripping column
feed and bottoms
1*6
-------
O.Ul-
0.3
s
§1
*
~ 0.
•o
•H
o
I
0.1
Acetio Acid Activated Carbon Column
Feed Rate • 1 gpm (3.78 liters/minute)
!••••••••••
-o_o—o— o—o— o- Feed
-K—X—X—it—X—*• Effluent
I I I I I
I
I I
0 1 2 3 ** 5 6 7 8 9 10 11 12 13
Hours Column Used
Figure lU. Acetic acid concentrations — furfural activated
carbon column feed and effluent
-------
0.3
« o.
•d
•H
O
V
•rl
o.i
Acetic Acid Activated Carbon Column
Feed Rate = 1 gpm (3-78 liters/minute)
!••!••• »««»i
-o—o—o—o—o—o- Feed
-S—X—X—X—X—K. Effluent
1 1 L I
01231*56789 10
Hours Column Used
12 13
Figure 15. Acetic acid concentrations — acetic acid activated
carbon column feed and effluent
U8
-------
0.1
o.os
0.0
0.0
0.0
s
1
V
0.0
0.01 _
o.o: .
Steam Stripping Column
Peed Hate - 1 gpm (3.78 liters/minute)
Stripping Steam = 6% by Weight of Feed
0.0£
0,0:1-
•o-
-X-
-o-
-X-
-e Feed
-X Bottoms
I
J_
I
I
56789
Hours Column Used
10
11 12 13 Ik
Figure 16. Methanol concentration — steam stripping column
feed and bottoms
-------
0.10,
0.09K
Furfural Activated Carbon Column
Feed Rate • 1 gpm (3-78 liters/minute)
0.081
0.071
0.061
0.051
H
I
o.oi*|
0.031
0.02
0.01
X^-x-"X\s / o-o
x
/—e e
v\... A
,'N
e o
o \ / b-
o
•
o o Feed
x ~x~ •* * Effluent
01231*56789 10
Hours Column Used
12 13
Figure 17- Methanol concentration — furfural activated carbon
column feed and effluent
-------
O.lOr-
Acetle Acid Activated Carbon Column
Feed Rate = 1 gpm (3.78 liters/minute)
o « o » Feed
X x K 'X Effluent
111
II
8 9 10 11 12 13 ll»
Hours Column Used
Figure 18. Methanol concentration — acetic acid activated carbon
column feed and effluent
51
-------
8000 r-
7000
6000
5000
•a
9
Uooo
£1
a
o
fr
in
3000
2000
100C
Steam Stripping, Furfural Activated Carbon and
Acetic Acid Activated Carbon Column
Feed Rate • 1 gpm (3-78 liters/minute)
Stripping Steam « 6% by Weight of Feed
_L
Steam Stripping Column e , «
Feed
Steam Stripping Column x—X X
Bottoms
Furfural Activated
Carbon Column Effluent
Acetic Acid Activated
Carbon Column Effluent
±
5678
Hours Column Used
10
11
12
13
lit
Figure 19
Biochemical oxygen demand — steam stripping and
activated carbon column flows
52
-------
0.8
0.7
0.6
fc o.
-p
v -a
•d o.
o
0.2
0.1
Steam Stripping, Furfural Activated Carbon and
Acetic Acid Activated Carbon Columns
Feed Rate = 1 gpm (3.78 liters/minute)
Stripping Rate = 6% by Weight of Feed
Steam Stripping Column Feed •
Steam Stripping Column
Bottoms
Furfural Activated Carbon
Column Effluent
Acetic Acid Activated
Carbon Column Effluent
0 ! 2 3 * 5 6 7 8 9 10 11 12
Hours Column Used
Figure 20. Organic loosely combined sulfur dioxide concentrations
steam stripping and activated carbon column flows
53
-------
Steam Stripping, Furfural Activated Carbon arid
Acetic Acid Activated Carbon Columns
Peed Rate = 1 gpm (3.78 liters/minute)
Stripping Steam = 6% by Weight of Feed
Steam Stripping Column •-
Feed
-a • »
Steam Stripping Column X-
Bottoms
-X X
Furfural Activated Car- O
bon Column Effluent
Acetic Acid Activated ®
Carbon Column Effluent
©
678
Hours Column Used
Figure 21. Total inorganic sulfur dioxide concentrations
steam stripping and activated carbon column flows
-------
10
Steam Stripping, Furfural Activated Carbon and
Acetic Acid Activated Carbon Columns
Feed Rate * 1 gpm (3.78 liters/minute)
Stripping Steam » 6$ by Weight of Feed
Steam Stripping Column
Feed
Steam .Stripping Column
Bottoms
Furfural Activated Car-
bon Column Effluent
Acetic Acid Activated
Carbon Column Effluent
_L
JL
-L
±
_L
o i
10
11
12 13
Figure 22. pH -
U 5 6 7 8 S
Hours Column Used
steam stripping and activated carbon column flows
55
-------
had been adsorbed on the previous column. For the next five hours
acetic acid was being absorbed, and in the final 5 hours it was being
pushed out of the carbon at a faster rate than it was being adsorbed.
The next three graphs, Fig. l6 to 18, show the methanol concentrations
and are similar to the first four on acetic acid. About 50$ of the
methanol in the feed was expected to be removed by steam stripping.
Figure 16 shows this was not always achieved. Since this was the first
run with the activated carbon columns, too much effort was placed on
their operation, and the steam stripping was not as efficient as it
should have been. Figure 17 shows that methanol goes through activated
carbon quickly with little adsorption. It also absorbs and readsorbs
depending on feed concentrations.^ Figure 18 is similar to Fig. 17,
but deals with the second carbon column.
Figures 19 to 22 are composite graphs covering the. BOD and sulfur
dioxide analyses of the samples, as well as their pH. Figure 19 shows
the BOD reduction over the lU-hour operating period. The 700 mg/liter
range indicates the BOD of the residual methanol. The 3000 mg/liter
range is indicated for the acetic acid and residual methanol, the 3500
to U800 range includes acetic acid, residual methanol, residual fur-
fural, and some polymerized material, and the kOOO to 5700 range con-
tains all the acetic acid, methanol, furfural, and other materials.
Any commercial acetic acid activated carbon adsorption unit that will
be, or has been recommended, would be designed on the basis of 2 1/2 to
3 hours data.
Figures 23 through 28 represent data obtained during repeats of the lU-
hour run, except the activated carbon column in which the acetic acid
was adsorbed had its carbon regenerated using ethanol. Since furfural
had not broken through the activated carbon in the first column, its
use was continued. Figure 23 shows the acetic acid breaking'through
the column at about 3 hours operating time, not appreciably different
from that of the virgin carbon indicating very good regeneration
56
-------
Steam Stripping, Furfural Activated Carbon and
Aoetto Acid Activated Carton Columns
Feed fcate = 1 gpm (3.78 liters/minute)
Stripping Steam » 6% by Weight of Feed
Steam Stripping
Column Feed
Steam Stripping
Column Bottoms
Furfural Activated
Carbon Column
Effluent
Acetic Acid
Activated Carbon
Column Effluent
5 0123t567
Hours Column Used
Figure 23. Acetic acid concentrations — steam stripping and
activated carbon column flows using regenerated carbon
57
-------
O.lOi—
0.09
0.08
0.07
0.0
-p
si
5
1
1
0.0
0.04-
0.03-
0.0
0.Oil-
Steam Stripping, Furfural Activated Carbon and
Acetic Acid Activated Carbon Columns
Feed Rate = 1 gpm (3.78 liters/minute)
Stripping Steam = 6% by Weight of Feed
Steam Stripping Column
Feed
Steam Stripping Column
Bottoms
Furfural Activated Car-
bon Column Effluent
Acetic Acid Activated
CarbonColumn Effluent
I- ».
Oct. 23, 1972
J
L
_L
5 01
Hours Column Used
Nov. 6, 1972
-X — X
Figure 2k. Methanol concentrations — steam stripping and activated
carbon column flows using regenerated carbon
-------
2.0 r-
1.8
1.6
l.U
fc
go
•H tH
Q 3 i.o
o
H C
3 H
I
0.8
0.6
O.U
0.2
Steam Stripping, Furfural Activated Carbon and
Acetic Acid Activated Carbon Columns
Feed Rate = 1 gpm (3.?8 liters/minute)
Stripping Steam = 6# by Weight of Feed
Oct. 23, 1972
0 1
Steam Stripping Column
Feed
Steam Stripping Column
Bottoms
Furfural Activated Car-
bon Column Effluent
Acetic Acid Activated
Carbon Column Effluent
Nov. 6, 1972
fc
2 3 l, 5 01
Hours Column Used
Figure 25. Total inorganic sulfur dioxide concentrations — steam
stripping and activated carbon column flow-
using regenerated carbon
59
-------
0.8r-
v -d
S 3
H V
P M
CO O
a
0.7
0.6
0.1*
o
•rl
00
fi 0.3
0.1
Steam Stripping, Furfural Activated Carbon and
Acetic Acid Activated Carbon Columns
Feed Rate * 1 gpm (3.78 liters/minute)
Stripping Steam » 6% by Weight of Feed
Steam Stripping Column
Feed
Steam Stripping Column
Bottoms
Furfural Activated Car-
bon Column Effluent
Acetic Acid Activated
Carbon Column Effluent
Oct. 23, 1972
I I I
5
Hours Column Used
1 1
D 1
Nov.
2
6, 1972
' |
3 U
_., . . 1 ,_
5
1
6
7
Figure 26. Organic loosely combined sulfur dioxide concentrations
steam stripping and activated carbon column flow
using regenerated carbon
60
-------
8000r-
7000-
Steam Stripping, Furfural Activated Carbon and
Acetic Acid Activated Carbon Columns
Feed Rate =° 1 gpm (3.78 liters/minute)
Stripping Steam = 6% by Weight of Feed
Steam Stripping Column
Feed
Steam Stripping Column
Bottoms
Furfural Activated Car-
bon Column Effluent
Acetic Acid Activated
Carbon Column Effluent
X X
-®
Nov. 6, 1972
123^5 0123^567
Hours Column Us,e,d
Figure 27. Biochemical oxygen demand — steam stripping and activated
carbon column flow using regenerated carbon
61
-------
10
Steam Stripping, Furfural Activated Carbon and
Acetic Acid Activated Carbon Columns
Feed Bate • 1 gpm (3.78 liters/minute)
Stripping Steam « 6% by Weight of Feed
Steam Stripping Column
Feed
Steam Stripping Column
Bottoms
Furfural Activated Car-
bon Column Effluent
Acetic Acid Activated
Carbon Column Effluent
•8,5
Oct. 23, 1972
Nov. 6, 1972
J I
J
It 3 01
Hours Column Used
Figure 28. pH -
steam stripping and activated carbon column flow
using regenerated carbon
62
-------
Figure 2U shows methanol being adsorbed in both activated carbon columns
over the entire 5-hour operation on October 23, 1972. The run of Novem-
/
ber 6, 1972 was more normal. Since methyl acetate was recovered during
regeneration, methanol reactions were possible as these chemicals were
allowed to set in the column. Figure 27 is interesting in that no ace-
tic acid appeared in the 2 and 2 1/2-hour samples, and yet the BOD is
almost as high as that in the steam stripping column bottom liquor.
This BOD should be characterized in further studies. If the 700 mg/
liter BOD range that has been attributed to residual methanol is a
problem area of concern, remember that the steam stripping column was
operated utilizing 6% by weight of feed stripping steam. Higher steam
utilization will reduce the methanol. The effluent was to be recycled.
Any recycle problems would have to be evaluated against the cost of
using more stripping steam. .
Two more regenerations of the carbon in the column adsorbing acetic
acid were made, but no attempt was made for chemical recovery or
material balances, since the multipoint recorder malfunctioned spoiling
the 3rd regeneration cycle and Uth adsorption cycle. More than 5 weeks
pilot plant operation time was lost, before this unit was repaired.
Since furfural had not broken through the carbon in the first activated
carbon column, that carbon column and the steam stripping column were
operated every day until a breakthrough did occur. This required 70 1/2
hours total operating time. Actually, the furfural was held in the
column from October 9, 1972 until December 21, 1972. The data compiled
in this belated recovery run was plotted in Fig. 29. Graphically inte-
grating the area under the steam stripping column bottoms line indicated
a furfural concentration average of approximately 0.01$ by weight. As
shown in the Furfural Balance, Table 7» the furfural adsorbed was cal-
culated to be 3.52 Ib (1.6 kg). That recovered was 3-7^ lb, (1.7 kg) or
106$ of that calculated to be adsorbed. Considering the time it was
held in the column, furfural is quite stable under the operating con-
ditions to which it was subjected.
63
-------
0.05
o\
O.oU
I
•g 0.03
0.02
Steam Stripping and Furfural
Activated Carbon Columns
0.01 -X—X
Steam Stripping Column
Feed
Steam Stripping Column
Bottoms
Furfural Activated Carbon
Column Effluent
12 16 20 2k
28 32 36 ho
Hours Column Used
U8 52 56 60 6k 68 72
Figure 39. Furfural concentrations — steam stripping and furfural activated
carbon column flows
-------
Table 7« FURFURAL BALANCE ACTIVATED CARBON ADSORPTION
10-9-72 to 11-15-72
Furfural adsorbed
Furfural « 1.0 gpm (60 min./hr) (8.33 Ib/gal.) (70.5 hr) (O.OOOl)
= 3.52 Ib (1.596 kg)
Furfural recovered — methanol regeneration
Sample No. 359 = 330 g
Analyses = 88.8$ furfural
91.0$ furfural
Average = 89.9$ furfural
Furfural = 330 g (0.899) = 296 g
Sample No. 360 = 1*35 + 555 + 530 - 1520 g
Analyses = 88. 5%
- 91-0$
- 90.5$
= 93.8$
= 93.6$
= 93.1$
s 92.5$
Average = 922. VlO = 92.2$
Furfural = 1520 (.0-922) = 1^00 g
Furfural
recovered = (296 + 1^00 )/U5U - 3.7^ Ib (1.697 kg)
$ Furfural
recovery " C3.7V3.52) 100 = 106$
-------
Activated Carbon Columns-Regeneration,
The activated carbon was regenerated as shown in Fig. 3 and U. That
adsorbing primarily acetic acid was regenerated vaporizing ethanol,
while that adsorbing furfural was regenerated vaporizing methanol. The
Furfural Balance, Table 7 was discussed. The Acetic Acid Balance,
Table 8, shows 6.65 Ib (3.02 kg) of acetic acid adsorbed in 3 trial
runs. This is the equivalent of 9-75 Ib (U.U2 kg) of ethyl acetate.
Since no analytical program was prepared to analyze the wide range of
ethyl acetate samples recovered in each run, they were batch distilled,
and the overhead samples analyzed for ethyl and methyl acetate concen-
tration. The results were used to calculate the total ethyl acetate
recovered. The equivalent of 9.13 Ib (U.lU kg) of ethyl acetate, or
93-7% of that retained on the carbon was recovered. Ethyl acetate
purity of 5>9% by weight was accomplished. The remaining 11$ was water
and ethanol, indicating pure ethyl acetate was possible.
It took approximately 16 and 2h hours operating time to regenerate the
carbon adsorbing acetic acid and furfural, respectively. While this
A
presented no problem for that retaining the furfural, it must be reduc-
ved for that retaining acetic acid. The limiting factor for regenerat-
ing the activated carbon retaining acetic acid is the heat exchanger
vaporizing the ethanol. This unit was too small. An optimum vapor
rate through the carbon has not been established. The l6-hour period
was used to assure maximum conversion of acetic .acid to ethyl acetate,
maximum ethyl acetate recovery, and maximum ethanol recovery. Since no
gas chromatograph was available for pilot plant use, no effort was made
to optimize regeneration time.
While efforts were expended to maximize S02, methanol, furfural and
acetic acid recovery, as much polymerized material was recovered when
regenerating the activated carbon holding furfural as furfural. Poly-
merized material was also recovered from regenerating the activated
carbon containing acetic acid. No efforts were made to determine the
quantity. Whereas most of that coming from the furfural carbon was not
66
-------
Table 8. ACETIC ACID BALANCE ACTIVATED CARBON ADSORPTION
10-9-72 10-23-72 11-6-72
Acetic acid absorbed
10-9-72
By graphic integration — large squares
5 + 8-5/8 + 7-3A + 3-1/2 + l*-l/2 + 3-1/2 = 32-7/8 - 11-7/8
= 21
One square = 0.02$ by weight per hour
Acetic acid absorbed = 21 (0.0002) (l gpm) (60 min./hr) (8.33 Ib/gal.)
= 2.1 Ib (0.951* kg)
Drainage = 60 gallon
Acetic acid in drainage = 60 (8.33) (.0.00199)
= 500 (0.00199) - 0.995 Ib (0.1*52 kg)
Total acetic acid absorbed = 2.1 - 1.0 = 1.1 Ib (0.50 kg)
10-23-72
By graphic integration — large squares
16.8 + ll*.5 + 9.0 + 2.5 = U2.8 squares
Acetic acid absorbed =1*2.8 (0.0002) (l) (60) (8.33)
= 1*.28 Ib (1.9U2 kg)
Drainage = 60 gallon
Acetic acid in drainage = 60 (8.33) (0.0021)
= 500 (0.0021)
= 1.05 Ib (0.1*76 kg)
Total acetic acid absorbed = 1*.28 - 1.05 = 3.23 Ib (l.l*65 kg)
11-6-72
By graphic integration — large squares
12.0 + 12.9 + 9-3 - 1.0 = 33.2
Acetic acid absorbed = 33.2 (0.0002) (l) (60) (8.33)
= 3.32 Ib (1.507 kg)
Drainage = 1.0 Ib (0.1*57 kg)
Total acetic acid absorbed = 3-32 - 1.0 = 2.32 Ib
Total acetic acid absorbed in 3 runs
Acetic acid = 1.10 + 3.23 + 2.32 = 6.65 Ib (3.02 kg)
Total acetic acid absorbed as ethyl acetate
Ethyl acetate = 6.65 (88/60) = 9«75 Ib (1*.1*2 kg)
-------
Table 8 (cont'd). ACETIC ACID BALANCE ACTIVATED CARBON ADSORPTION
10-9-72 10-23-72 11-6-72
Total ethyl acetate recovered
Sample No. 3^9 = ^30 g (0.55) • 236 g
No, 350 = 1*30 g (0.81) = 3W g
No. 351 * 1»30 g (.0.85) = 365 g
Sample broken = 860 g (0.85) = 730 g
No. 352 = 2170 g (0.78) - 1692 g
No. 353 • 910 g (0.55) = 501 g
No, 351* = 570 g (0.17) = 97 g
No. 355 = 385 g (0.01) = k g
3973 g
Total ethyl acetate recovered • (3973A5*0 = 8.75 Ib (3.97 kg)
Ethyl acetate in ethanol • 30 gal. (8.33) (0.0005) =0.12 Ib (0.05^5 kg)
Methyl acetate - • U30 g (0.20) = 86 g
* U30 g (0.03) = 13 g
- 0.22 Ib - 99 g
(0.10 kg)
Methyl acetate as ethyl acetate = 0.22 (88/710 = 0,26 Ib (0.116 kg)
Ethyl acetate accounted for = 8.75 + 0.12 + 0.26
= 9.13 Ib (l*.lU kg)
% Accountability « (9.13/9-75) 100 = 93.1%
68
-------
volatile, that coming from the acetic acid carbon was. However, it
polymerized or reacted during the processing in the fractionation
column and lost its volatility. The nonvolatile material from both
columns are to be recycled back to the concentrated evaporator liquor
for burning or spray drying purposes. In the pilot plant this material
was discharged to the drain with the water at the bottom of the frac-
tionator at 15% by weight solids concentration. No attempt was made to
increase this solids concentration.
Evaporator Studies
Throughout the period that the multipoint temperature recorder was
being repaired and the pilot plant was sitting idle, arrangements were ,
made to discuss in detail Consolidated's evaporation system with their
supervisors and engineers. After these discussions it was agreed that
it would be beneficial to analyze the evaporator condensates from each
effect of their Rosenblad evaporator system. With Consolidated Papers
Inc.'s permission the results of these analyses are included as part of
this report. They are expressed as % by weight as follows:
Methanol Acetic Acid Furfural
Weak spent liquor feed 0.090 O.UlO 0.0^5
Condensate from 50% solids
effect 0.005 O.U63 0.005
35$ Solids effect 0.010 O.U21 0.010
Solids effect 0.006 0.210 0.005
Solids effect 0.071 0.31^ 0.021
The condensate accumulated from one effect to another, but not
necessarily directly from the 50$ solids effect through the 35$ and 20%
effects to the 10% effect. No detail about the path of the condensate
will be given, since time will not permit an adequate description of
it. Consolidated has switching procedures that are too complicated to
describe without drawings. The important factor was that the condensate
from the 10$ solids effect includes the condensate from all other
69
-------
effects. Therefore, it is apparent that the "bulk of the methanol and
furfural were removed in the 10$ solids effect. Note that, although
condensate flow is less than weak spent liquor feed flow, the condensate
had 19% of the methanol concentration, 11% of the acetic acid concen-
tration, and ^1% of the furfural concentration that was in the feed.
This indicated an appreciable loss of those chemicals in the evapora-
tion system. If equal evaporation is assumed in each evaporator effect,
it is apparent that almost k times the methanol and furfural concen-
tration, or a 0.28$ methanol and 0.08**$ furfural, would occur in the
condensate from the 10$ solids effect if the other condensates were
separated. Examination of Fig. 9 shows that the other 75$ of the
evaporator condensate contains substantially less methanol than our
best steam stripped evaporator condensate. Essentially the same thing
is true for the furfural (see Fig. 29).
Project Status in_ February. 1973_
The merits of these unit operations have been established. A high
percentage of SOj vapor that can be reused in the mill has been pro-
duced. Crude methanol, furfural, and ethyl acetates have been re-
covered. Four important areas of study must be further developed
before this process is ready conclusively for commercialization:
1. A market for the crude products produced must be
established.
2. The life of the carbon must be determined when feeding
evaporator condensate that has been used as a backwash
in the evaporators.
3. Regeneration procedures for the activated carbon used in
adsorbing acetic acid must be optimized.
U. An overall economic evaluation of the process is required.
To date the evaporator condensate used was readily available from the
Consolidated Papers, Inc.'s mill without suggesting any changes in
their operating procedures for producing and/or utilizing their
70
-------
evaporator condensates. The condensate feed was less than ideal for
their purpose. Since they have several evaporator systems, a composite
of their condensates was processed after these flows had been used for
evaporator wash. The feed is dilute and contains lignosulfonates and
other spent liquor solids. The chemical concentrations expressed as %
by weight varied as follows:
High Low Mean
Methanol 0.170 0.025 - 0.065
Acetic Acid 0.36U 0.170 0.268
Furfural O.OUo 0.010 0.016
Sulfur Dioxide-TI 0.233 0.017 0.06U
Sulfur Pioxide-OLC ;0.076 0.015 0.03^
Biochemical Oxygen Demand 65^0 2270 ^063 milligrams
per liter
The methanol, furfural, and acetic acid balances, Tables 5, 6, 7, and
8, were made using the instantaneous values, graphically integrating
to determine recoverable quantities. The main function of these
balances was to determine operating and analytical proficiency. The
evaporator studies had already indicated that splitting the evaporators
and/or their condensates was the path to pursue for Consolidated Papers,
Inc.'s Appleton Mill. No other mill showed any enthusiasm for this
approach.
Since the establishment of a market for the crude products was essential
to the economic evaluation of the process, the following information was
presented. Assuming 300,000 gallons (1,13^,000 liters) per day evap-
orator condensate capacity for the Consolidated Mill and using the mean
concentrations list in the "Previous Work" section and 350 days oper-
ation per year, Consolidated Papers, Inc. produced the following:
Methanol 569,000 Ib/year 258,500 kg/year
Acetic acid 2,3^5,000 Ib/year 1,063,000 kg/year
Furfural 1^0,000 Ib/year 63,600 kg/year
71
-------
Sulfur dioxide (Tl) 560,000 Ib/year 25^,500 kg/year
Sulfur dioxide (OLC) 298,000 Ib/year 135,300 kg/year
Conversion of the acetic acid to ethyl acetate would give 3,^0,000
Ib/year (1,560,000 kg/year).
The product value is:
Ethyl Acetate
Market value - 3,^0,000 ($0.12) = $1*12,000
Mill value - 3,^0,000 ($0.09.) = $310,000
Furfural
Market value - 1^0,000 ($0.175) = $ 2*f,500
Mill value - lUO.OOO ($0.138) = $ 19,300
Methyl Alcohol
Market value - (569,000/7-5) ($0.12) = $ 9,100
Mill value - (569,000/7.5) ($0.10U) = $ 7,900
Sulfur Dioxide
Market value - (858,000/2) ($0.001) =$ U ,290
Mill value - (858,000/2) ($0.001) = $ U,290
Total Market Value = $Ul2,000 + $2^,500 + $9,100 + $i+,290 = $UU9,890
Total Mill Value = $310,000 + $19,300 + $7,900 + $1*,290 = $31H,U90
Cost of Ethanol = $166,000
Money left to pay
for operating costs,
capital expenditure,
maintenance, etc. = $175,^90
No credit was taken for reusable water and/or elimination of pollution
problems. It was believed that the capital costs will compare favorably
with the secondary treating process for this waste flow.
A more complete analysis of this same information based on actual
complete material and heat balances on the pilot system are presented
in a later section.
72
-------
Process Options
The following process options were presented at a meeting of the co-
operators on January IT, 1973 and remain as valid options for appli-
cation of the system.
1. Separate the evaporator condensates from the various effects, so
that only that portion high in methanol, furfural, and sulfur
dioxide is steam stripped, and those high in acetic acid are
adsorbed on the activated carbon.
2. Steam strip the weak spent liquor to remove SOa, methanol, and
furfural before the evaporators. Recycle the portion of conderisate
weak in acetic acid concentration, and send that which is more
concentrated to the activated carbon adsorbers.
3. Adsorb the furfural directly from the weak spent liquor with
activated carbon before evaporation. Separate the evaporator
condensates, so that only that portion with concentrations high
in methanol and sulfur dioxide is steam stripped, and those with
concentrations high in acetic acid are adsorbed on the actiyated
carbon.
U. Steam strip that portion of the evaporator condensate with con-
centrations high in methanol, furfural, and sulfur dioxide.
Neutralize the overhead with Ca(OH)2 to destroy the furfural
and sulfur dioxide. Recover the methanol in the fractionation
column. Adsorb those-portions of evaporator condensate high in
acetic acid concentration on activated carbon. Regenerate the
carbon with methanol. Treat the regeneration effluent with
Ca(OH)2 to destroy the furfural and methyl acetate. Recycle and
recover the methanol in the fractionator. Spray dry the Ca(OH)2
reaction products.
5. Adsorb furfural and acetic acid in one activated carbon column,
regenerate the carbon with ethanol, recover the ethyl acetate,
furfural, and ethanol.
6. Recycle the ethyl acetate-ethanol regenerating agent until the
ethyl acetate concentration is at its economical peak. Recover
73
-------
• chemicals at a central location. Over 8 1/2% by weight ethyl
acetate in ethanol was used in the pilot studies at Scott with
no adverse effects in carbon regeneration.
T. Design and do economic evaluation on the process as it is now
operated. Use the best available data.
8. Identify and evaluate the acids and other chemicals removed with
the furfural.
9- Rerun equipment as it is now installed to establish carbon life.
Additional Experimental Operation and Evaluation of Pilot System
Methodology
During the January IT, 1973 meeting of the members of this Group Pro-
ject it was agreed that Mr. Lueck continue with his low temperature,
thermal, activated carbon regeneration work, while Mr. Baierl pursue
Process Options 5 and 9 of the nine options presented. Unfortunately,
it was not possible to run these experiments simultaneously. Initial
emphasis was placed on Mr. Lueck's work. The pilot plant was modified
to provide processed evaporator condensate to his unit as shown on
Fig. 30. Recirculation pumps were added at the bottom of the steam
stripping and fractionation columns, and water lines were eliminated.
Simultaneously, the steam stripping column condensed overheads were fed
into the fractionation column in an effort to optimize its size and
operation, and studies were made on Consolidated Papers Aqua Chem No. 1
and Rosenblad evaporation units.
Evaporator Studies
The data in Table 9 were compiled during the evaporator studies, while
the Aqua Chem No. 1 units were operated in series with and ahead of the
Rosenblad units.
It was readily discernible from these data that it would be advantag-
eous for Consolidated Papers, Inc. to separate these condensates. The
condensate flow was from Effect 3 to Effect 2 to Effect 1 to Effect !+
-------
vn
m i
Evaporator
Condensate
Feed
Steam Strip
ping Column
Feed Drum __
130-155F
5l*.5-68.3°C
Feed Pump
Stripping
FCO — FCFRO
fi
Feed Pump
Activated Carbon
SO2 Vapor
Methanol
FCSC — FCFKSC Furfural
_ [3 i
Q 1 '
3a i35°F
71 57. 0^
/ \
\
3H
°F| '
Opt
• 5(|C
/
/ l!*0°F
/ 60°C
/
/ 1
/ 1
f
\
'F
5°fc
iU
210°F
^98.8°C
-
1st
c
H
c3
a
c
£
c
^
IK
"S
CJ
^o
H Cooler
^
*
f
s-
J^
J^l
C
F
i
i
13
i
133°F
2nd
H
O
CJ
a
o
a
o
*&
01
d
-p
a
<
•a
o
-------
Table 9. EVAPORATOR STUDIES DATA
1. Operation on Channel A
cr\
% by Weight
Sample
Aqua Chem Feed
Aqua Chem Prod.
Rosenblad Feed
Rosenblad #3
Rosenblad #4
Rosenblad #1
Rosenblad $2
Aqua Chem Cond.
Aqua Chem
Scrubber Cond.
Rosenblad
$3 Cond.
Rosenblad
$2 Cond.
Rosenblad
#1 Cond.
Rosenblad
#4 Cond.
MeOH
0.069
0.035
0.018
0.01
0.113
0.205
0
Trace
0
0.02
HAc
0.340
0.343
0.330
0.386
0.210
0.240
0.219
0.431
0.373
0.255
0.252
Furf.
0.028
0.023
0.023
0.020
0.03
0.06
0
0
0
0.12
PH
2.85
3.70
3.89
4.05
4.03
4.05
2.89
2.52
2.61
2.55
2.88
2.21
7. Solids
Neut.
10.34
12.87
15.16
20.58
31.82
54.73
0.29
0.27
0.57
0.50
0.30
0.48
24 Hr
9.90
12.60
14.71
20.39
32.19
55.57
0.07
0.03
0.12
0.09
0.06
0.14
S02,
TI
0.804
0.252
0.183
0.207
0.411
0.604
0.040
0.084
0.037
0.131
0.024
0.796
8/1
OLC
3.996
3.975
4.179
5.421
6.054
9.171
0.048
0447
0.253
0.227
0.103
0.267
VOA, me/1
Total
4212
3792
3901
4602
6340
6736
1969
2006
4054
3590
2201
2871
HAc
4177
3583
3717
4100
5691
5295
1948
2006
4054
3577
2193
2860
Formic
27
160
141
385
498
1105
16
0
0
10
6
8
BOD5.
mg/1
3357
4241
3798
2892
1618
2409
-------
Table 9 (cont'd). EVAPORATOR STUDIES DATA
2. Operation on Channel B
T. by Weight
Sample
Solids
Meth. HAc Furf . pH Neut. 24 Hr
Aqua Chem Feed 0.066 0.336 0.032
0.054 0.321 0.029 2.83 11.88 11.53
Aqua Chem Prod*
Rosenblad Feed 0.028 0.337 0.016 3.59 14.55 14.44
Rosenblad #3
Rosenblad #4
Rosenblad £1
Rosenblad 42
Aqua Chem Cond.
Aqua Chem
Scrubber Cond.
Rosenblad
#3 Cond.
Rosenblad
#2 Cond.
Rosenblad
#1 Cond.
Rosenblad
#4 Cond.
0.010
Trace
0.108
0.132
0
0.01
0
0.021
0.328
0.311
0.273
0.242
3.387
0.380
1
0.293
0.284
0.015 3.85
0.016 3.88
3.98
4.45
0.023 3.01
0.028 2.64
0 2.48
0 2,50
0 2.80
0 2.25
16.86
22.09
31.85
53.27
1.15
0.37
0.59
0.51
0.31
0.42
16.95
23.03
34.53
54.69
0.91
0.09
oao
0.04
0.03
0.07
SOz. g/1 VGA, pig/1 BODs,
TI OLC Total HAc Formic mg/1
0.507 4.588 4359 4330 22
0.181 4.182 3962 3900 48
0.157 4.305 3962 3886 58
0.206 5.920 4633 4552 62
0.281 7.020 6096 6031 50
0.583 11.162 7254 7150 80
0.053 0.375 2390 2377 10
0.056 0.142 2512 2496 12
0.054 0.140 4151 4138 10
0.080 0.094 3718 3705 12
0.013 0.111 2384 2371 10
0.452 0.262 2786 2773 10
-------
cumulatively. Steam was added to that in Effect 1. Consequently, con-
siderable methanol and furfural were added in Effect U to bring the
cumulative condensate to their final concentrations. To verify this,
samples of the condensates made in Effect k only were analyzed, with
the following results:
% by Weight S02. g/1
Sample MeOH HAc Furf. EtOH TI OLC
Channel A - 2:30 p.m. O.l8 0.23 0.05 0.01 1.39 0.&8
Channel A - 3:30 p.m. 0.19 0.23 0.06 0.01 1.51 0.51*
Channel B - 2:30 p.m. 0.20 0.28 0.08 0
Channel B — 3:30 p.m. 0.26 0.26 0.06 0
Prior composite condensates showed methanol and furfural to be in the
O.OU to 0.05 and 0.01 to 0.02 concentration ranges, respectively. Al-
most h times the methanol and furfural concentrations had been con-
servatively predicted.
Since all condensates from Effects 3, 2, and 1 had less methanol and
furfural in them than the majority of the steam stripping column
bottoms, they should be kept separate and out of the steam stripping
column system. Thus, the equipment size in both the steam stripping
and fractionation systems will be substantially lowered, as will the
steam requirement.
The data in Tables 10, 11, and 12 were submitted by one of the coop-
erators and are compared with data from Scott (Table 21) and from
Consolidated (Table 22).
First note the ineffective steam stripping at Scott. This was the
initial work in 1969. Sixteen ft (1+87.7 cm) of packing and 10% steam
were used. Now look at Consolidated in April, 1973 using 1+ ft (121.9
cm) of packing and 6% steam. This re-emphasizes what over-design can
do. No problems with scaling or plugging were encountered in this
column, either at Scott or Consolidated, after the screening system
was installed. The figures from the cooperator were sent to demon-
strate little differences in acetic acid concentration in their
78
-------
Table 10. ANALYSES OF EVAPORATOR EFFECTS OF COOPERATOR MILL
Solids
content
Volatile acids
on solids
wt. % as acetic acid, wt. %
Feed
From 5th
Hth
3rd
2nd
13.2
IT. 66
21.88
31. 2U
55.51
10.6
9.2
7.2
5.9
2.6
Calculated volatile
acids in condensate
as acetic acid, wt. J
1.5U
1.83
0.9^5
2.19
Table 11. CARBOJT ADSORPTION WORK AT SCOTT PAPER COMPANY
OCONTO FALLS EVAPORATOR CONDENSATE
As received Steam stripped
Methanol 0.119# by weight 0.01+9$ by weight
Acetic acid 0.600$ by weight 0.530$ by weight
Furfural 0.122$ by weight O.OW by weight
S,team stripping column — ^-inches deep x 16 feet high
using 10$ by weight of feed stripping steam.
Table 12. CARBON ADSORPTION WORK AT CONSOLIDATED PAPERS, INC.
APPLETON DIV. EVAPORATOR CONDENSATE
As received April 23. 1973 ~ 11:30 AM
mean As received Steam stripped
Methanol 0.065$ *>v weight 0.06$ by weight 0.01$ by weight
Acetic acid 0.268$ by weight 0.21$ by weight 0.22$ by weight
Furfural 0. 016$ by weight 0.01$ by weight 0.00$ by weight
Steam stripping column — h feet deep x k feet high using 6% by weight feed
stripping steam.
79
-------
evaporator effects. There is a dual purpose for showing them. First
the number of hours the activated carton adsorbed acetic acid was al-
most identical for these condensates. The carbon load at Scott was 10%,
and 3 to h% at Consolidated. That is about 20 times the concentration.
That means the carbon loading for this condensate showing 1.5U to 2.19$
by weight would be 30 to kO%. The second fact that this could be U($
versus 30$ shows the increase in concentration is significant for the
developed process. Furfural has a multiple of over 100. The loading
at Consolidated was 1 to 2% from 0.01 to 0.02$ concentration. At Scott
it was 20 to 30$ for a 0.2 to 0.3$ concentration.
F ract ipnati on
As the steam stripping system and activated carbon column-furfural were
providing feed to the Lowell Unit, the fractionation column system was
run to establish optimum equipment size. On February 1, 1973 evaporator
condensate was pumped through the steam stripping column preheater to
the fractionation column at a point about Ik feet above the bottom of
the column. The rate was 0.76 gpm (2.88 1/min.).
From the following data it was apparent that the fractionation column
rapidly lost its efficiency:
$ by Weight
Sample
Feed
Bottoms
Feed
Bottoms
Time
11:35 a.m.
11:35 a.m.
3:30 p.m.
3:30 p.m.
MeOH
0.030
0.010
0.055
0.030
HAc
0.2UO
0.260
0,215
0.210
Furfural
0.012
0
0.010
0.007
Column flooding occurred about U feet below the feed. The following
data showed a similar run in the l*-foot steam stripping column with
equivalent steam to that used in the fraction column:
80
-------
Sample
Feed
Bottoms
Time
IikQ p.m.
1:UO p.m.
MeOH
0.01*5
0.005
% by Weight
HAc
0.216
0.189
Furfural
0.013
0
This pointed out the disadvantages of overdesign; it can be worse than
underdesign. These results verified similar work at Scott Paper Com-
pany's Everett Mill.
From February 12 through February 22, 1973 the evaporator condensate
was pumped through the steam stripping column system at the rate of
0.5 gpm (1-9 1/min,). Approximately 10% by weight of feed stripping
steam was used, and the condensed overhead was processed in the frac-
ctionation column system. The stripping column bottom liquid was
either processed through the activated carbon column and/or drained.
The main objectives were to determine the capacity of the fractionation
column system, to establish carbon loading times and to provide loaded
activated carbon units for further regeneration work. Using a 1 to 1
ratio of feed to steam in the fractionation column, furfural was re-
moved as a 90$+ by weight crude from the bottom splitter, while no
methanol or furfural was detected in the fractionation column bottom
liquor. This indicated that only 9 ft (27^ cm) of Goodloe Packing was
necessary to remove all the furfural in the steam stripping column con-
densed overhead vapors. The 9 ft (27^ cm) of packing was probably
conservative, since the fractionation column feed was introduced in the
column above the furfural take-off point. It was interesting to note
that the furfural coming off the middle splitter, above the feed, had
about 1.0% by weight methanol in it, whereas that coming off the bot-
tom splitter, under the feed, showed little or no methanol in it. Ap-
proximately 1565 g, or 3.5 pounds? of 90$+ by weight methanol was held
in the top of the fractionation column.
Activated Carbon Adsorption and Regeneration
Since the activated carbon column-acetic acid was loaded and the
activated carbon column-furfural was partially loaded, it was decided
81
-------
to regenerate them in series with ethanol in accordance with Process
Option 5. The regeneration proceeded as shown in Fig. 31. Ethanol,
at the rate of 1 gpm, (3.8 1/min.) was pumped into the top of the
activated carbon column containing essentially acetic acid for 12
minutes. By plug flow essentially 10 gallons (37.8 l) of water con-
taining less methanol and acetic acid than in the evaporator conden-
sate feed was sent to the drain. In a commercial installation this
liquid would be pumped to the activated carbon column in parallel with
this one for acetic acid adsorption. The steam valve feeding steam to
•
the column jacket was cracked open and that feeding steam to the vapor-
izer was opened to control the steam pressure to the vaporizer at about
20 psig (l.UO kg/cm2 gage). Pressure in both columns built up to about
Ik psig (0.98 kg/cm2 gage). Liquid was- forced out the activated carbon
column containing the furfural through the condenser to the receiver.
From there it was pumped at the rate of 0,3 gpm (l.lU 1/min.) to the
fractionation column, which was partially filled with ethanol. Ini-
tially, everything looked great. The liquid to the receiver rapidly
discolored. The temperature at the top of the fractionation column
was..quickly dropping its 173°F (78.3°C) temperature. What appeared to
be furfural was collecting in the tap-off section on the bottom split-
ter of the fractionation column. No furfural was detected in the liquid
from the bottom of the fractionation column. By 11:00 a.m. the temp-
erature at the top of the fractionation column had reached 158°F
(70.0°C), indicating a good grade ethyl acetate. Since alcohol was
building up in the system, a recirculation stream from the middle split-
ter back to the ethanol feed drum was started. When what appeared to
be furfural was removed from the bottom splitter and tested, it was
not furfural but highly colored water which had displaced the alcohol.
Then the problems arose. To maintain 0.3 gpm (l.lU 1/min.) recircu-
lation rate of alcohol, it was necessary to recirculate from the bottom
splitter. To avoid carrying water and furfural back to the ethanol
feed drum, another receiver was hooked up in parallel with it. This
corrected one problem but created another. Furfural and polymerized
material was being carried into the activated carbon column that had
82
-------
CD
to
Make Up
Ethanol ^
/
i
Ethanol
Feed
Drum
Et
Fee
Wa-
ACCACRI
a
lanol
i Pump
ter ^
\ >
2nd
Activated Carbon Column
Acetic Acid
CD
*" G
Hi C
'' To Drain
V
1st
c
l-l
o
o
a 3
O
-------
previously contained only methanol and acetic acid. By now it was ap-"-
parent that ethyl acetate, ethanol, furfural, and water removal in one
fractionation column was going to be too complicated, so the recovery
of furfural was abandoned. It was allowed to go out the bottom of the
fractionation column with the water and residual spent liquor solids.
The ethanol feed and the steam to the activated carbon column system was
shut off. The fractionation column was held at total reflux overnight.
The next day there were two partially regenerated activated carbon
columns, about Uo gallons (151.5 l) of ethanol in the complete regen-
eration system and flow rate limits to the fractionation column of about
0.15 gpm (0.5T 1/min.). The regeneration was continued, and the flow
rates to the carbon columns were varied from 0.1 (0.38) to 0.3 (l.lU
1/min.) gpm. The ethyl acetate was allowed to accumulate in the frac-
tionation column. Although not practical, this system worked well and
the carbon columns were regenerated, see Fig. 32. Unfortunately, about
mid-morning the water pressure in the main header dropped substantially,
and no cooling water reached the condenser at the top of the fraction-
ation column. Over a 15-mihUte period all the ethyl acetate and
ethanol remaining in the top of the column vaporized and discharged
through the vent. Even so, the recovery of ethyl acetate was the high-
est of any run. Although the level in the ethanol storage tank showed
almost full recovery of ethanol, results of analyses (two days later)
showed water and ethyl acetate in the alcohol and ethanol in the
activated carbon column effluent.
Since Process Option 9 was the second choice, it was decided to run the
adsorption and regeneration cycles on the two columns in series. Evap-
orator condensate feed rate was held at 0 = 5 gpm (1.9 1/min.) and 10$ by
weight of feed stripping steam was used to assure maximum removal of
methanol and furfural. For regeneration, a line was built from the
ethanol feed pump to the top of the activated carbon column-furfural
and from the top splitter on the fractionation column to the ethanol
feed drum. Water and residual spent liquor was to be removed from the
-------
1
•H
V
f>
Vi
V
o
0.6 r_
0.5 -
0.2 -
0.1 -
X-
o-
-• Steam Stripping Column Feed
-J( Steam Stripping Column Bottoms
-O Furfural Activated Carbon Column Effluent
Acetic Acid Activated Carbon Column
Effluent
March 8, 1973
10 12
Hours Column Used
Figure'32. Steam stripping, furfural activated carbon
and acetic acid activated carbon columns
85
-------
bottom of the fractionation column and ethyl acetate would be allowed
to build up in the alcohol. Three successive adsorption runs, March 8,
13, and 26, showed the acetic acid coming through progressively earlier;
10 hours, 8 hours, and 7 hours, respectively, see Fig, 32 and 33. How-
ever, complete acetic acid loading was still extended beyond ten hours.
The real problem became apparent when traces of ethanol appeared in the
effluent from the activated carbon column containing acetic acid after
it was 7 hours on the adsorption cycle. Black material appeared in the
initial condensate effluent. It became apparent that residual spent
liquor in the activated carbon column containing acetic acid was not
being removed, and that the ethanol was not being quickly removed from
the carbon in the activated carbon column containing furfural. Also
the ratio of ethanol to water was not high enough. Too much acetic acid
was being removed as acetic acid and too much ethanol [almost kQ (151.5
l) gal.] was held in the system. Piping modifications and regeneration
system changes for recovery of acetic acid as ethyl acetate and the re-
moval of water and residual spent liquor simultaneously were.mandatory.
Four things were learned from these runs:
1. Ethyl acetate, ethanol, furfural, and water were not going to be
recovered in one fractionation column.
2. Acetic acid and furfural were effectively removed from both
carbon columns.
3. Channeling occurred when going downstream in the Act. Garb. Col.
— Fur. and the water and steam with the ethanol made the ethanol
recovery expensive.
U. Most importantly the carbon life was deteriorating, due to
processing dirty condensates, especially in the Act. Garb. —
Acetic acid column.
New Activated Carbon Regeneration System
It was decided to move the heat exchanger from directly under the
activated carbon column-acetic acid to a position on the side of the
86
-------
••" •• Steam Stripping Column Peed
"X X Steam Stripping Column Bottoms
•O O Furfural Activated Carbon Column Effluent
.0 Acetic Acid Activated Carbon
Column Effluent
0.6T
March 26, 19T3
Ithanol Trace
I I I
10 0 2 1»
Hours Column Used
Figure 33. Steam stripping, furfural activated carbon
and acetic acid activated carbon columns
10
87
-------
column. The vapors would then enter the bottom disengaging section from
the side similar to the reboiler on the fractionation column. When this
heat exchanger was removed and inspected, better than 6Q% of its tubes
(mostly those around its outer circumference) were completely plugged
with the black polymerized material. This verified that there was no
escape path for the nonvolatile material and indicated the heat exchanger
was too large for the job it was doing. The smaller heat exchanger that
was used for vaporizing the methanol when regenerating the activated
carbon column-furfural was used to replace this one. Piping modifica-
tions and changes to the activated carbon column-acetic acid regenera-
tion system were made in an effort to recirculate liquid through the
smaller heat exchanger now installed on the side of the column's bottom
disengaging section. Existing equipment, with the exception of new
brass pressure gages to replace the older corroded ones, was used. A
check of the new system operation, shown schematically in Fig. 3^ to 37>
indicated that theoretically it functioned well. However, so much
black material, plastic in appearance, was removed it plugged the rotpm-
;'
eter, pumps, and packing in the fractionation, column. Operational
problems, such as controlling the steam pressures and feed rates, were
encountered. Steam pressure regulators had to be changed,,and since
all but one pump that was available lost! their flows when discharge
pressures exceeded 15 psig (1.05 kg/cm2 'gage), maintaining constant
flows was almost impossible. With erratic operation, it was extremely
difficult to know what the concentrations of ethyl acetate, ethanol,
and water were in the various streams. Attempts were made to get a gas
chromatograph capable of determining these in extreme concentrations.
None was available.
During the latter part of March and the early part of April a concen-
trated effort was made to perfect the new regeneration system. Ethanol-
water runs were made to establish steam pressure controls and constant
pumping rates. In previous regeneration experiments no reflux was used
in the activated carbon column. The new system is demonstrated in Fie.
3^ through 37. In this system steam rates had been reduced to the
-------
Feed
2nd
Activated
Carbon
Column
Acetic
Acid
Vent
To Drain
Figure 3k. Process flow
.89
-------
Ethanol
Storage
Tank
2nd
Activated
Carbon
Column
Acetic
Acid
Pump
To Drain
Figure 35- Regeneration — step 1
90
-------
Vent
Liquid
Ethanol
Storage
Tank
r
Pump
2nd
Activated
Carbon
Column
Acetic
Acid
•H
9)
a
Pump
Conden-
ser
X
Rt.hvl
Ethanol - Ethyl
Acetate Fractionation
Column
JTo Drain
Pump
Figure 36. Regeneration — step 2
91
-------
Vent
'
Liquid
1
Ethanol
Storage
Tank
r
a
i
p
^
i
fapor >
2nd
=====
Activated
Carbon
Column
Acetic
Acid
=====
1
Pump
c
Pump
Conden
ser
X
Ethanol - Ethyl
Acetate Fraetion-
ation Column
'o Drain
Pump
Figure 37. Regeneration - step 3
92
-------
extent that more steam leaded by the original steam pressure regulator
than was required now to vaporize the ethanol. A smaller pressure
regulator was installed. The ethanol held in the system was reduced to
10 gal. (37-9 1) from the ^0 (151.5) previously used, and the ethanol
rate was 0.05 gpm (0.18 1/min.) instead of 0.30 gpm (l.ll* 1/min.). This
necessitated the installation of smaller rotameters. A successful
ethanol-water run was made establishing operating parameters during the
week of April 2-6, 1973. These operating parameters (see Table 13) were
the basis for the newly developed, possibly patentable activated carbon
regeneration system.
On April 10, 1973 a steam stripping and activated carbon adsorption run
was made. Samples of the acetic acid carbon column effluent were taken
after one hour of operation and then every half-hour until a total of
3-1/2 hours adsorption time had elapsed. The feed rate was 1 gpm (3-79
1/min.). The results shown on Fig. 38 were interesting in that no ace-
tic acid was detected in any sample. Previously acetic acid was found
in the samples after 2-1/2 hours of adsorption operation and had com-
pletely broken through after 3 hours. This indicated the new regener-
ation 'system was very effective, and the piping changes may have provid-
ed better liquid distribution over the carbon, thereby getting better
adsorption.
Unfortunately, during the regeneration run the new by-pass valve to the
vent was left open. All the alcohol vaporized, other than that pumped
into the column as a liquid, by-passed the carbon column and discharged
through the vent. Ethyl acetate was made as the alcohol in the system
was vaporized; however, it also worked its way into the recirculation
system and was vented. Since it was impossible to tell whether or not
the carbon had been regenerated, this regeneration run was repeated.
Interestingly very little ethyl acetate was made during the repeat run,
indicating either the alcohol in the carbon column was sufficient to
regenerate the carbon in the previous regeneration test, or the new
regeneration system was ineffective as far as producing ethyl acetate.
93
-------
Table 13. TYPICAL ACTIVATED CARBON ACETIC ACID COLUMN REGENERATION
DATA - POTENTIALLY PATENTABLE PROCESS
Fractionation Column Reboller, 1st Pressure Gage - 21 psig (l.kj kg'/om2)
Fractionation Column Reboiler, 2nd Pressure Gage - 20 psig (l.l*0 kg/cm2)
Fraotionation Column Bottom Pressure Gage - 10" H20 (25.^ cm H20)
Fractionation Column Feed Station Pressure Gage - 3" H20 (7-6 cm H20)
Fractionation Column Middle Splitter Pressure Gage — 0" H20
Fractionation Column Top Splitter Pressure Gage - 0" H20
Fractionation Column Vent Pressure Gage -<• 0 to V H20
(0 to 1.6 cm H20)
Acetic Acid Activated Carbon Column Top Pressure Gage — 0 psig
Acetic Acid Activated Carbon Column Jacket Pressure Gage — 0 psig
Acetic Acid Activated Carbon Column Bottom Pressure Gage — 1% psig (.0.12 kg/cm2)
Acetic Acid Activated Carbon Column Vaporizer Pressure Gage — l6.5 psig Cl.l6 kg/cm2)
Acetic Acid Activated Carbon Column Rotameter to Vaporizer — 0.05 gpm (.0.19 1/min.)
Acetic Acid Activated Carbon Column Rotameter to Fractionator — 0.05 gpm (0.19 1/min.)
Temperatures
Fractionation Column Liquid Bottoms - 213°F (100.5°C)
Fractionation Column Bottom Splitter - 212°F (100.0°C)
Fractionation Column Feed — 17U°F (78.9°C)
Fractionation Column Middle Splitter — 173°F (78.3°C)
Fractionation Column at Feed Plate - 17U°F t78.9°C)
Fractionation Column at Top Splitter — l61*°F (73.3°C)
Fractionation Column Vapor after Condenser — 92°F (33.3°C)
Fractionation Column Reboiler Discharge — 213°F (l00.5°C)
Acetic-Acid Activated Carbon Column Top — 175°F (.79.1*°C)
Acetic Acid Activated Carbon Column Bottom — 207°F t97.2°C)
Ethyl Acetate Removed Top Fractionator Splitter. Periodically
Ethyl Alcohol Removed Bottom Fractionator Splitter to 50 gal. Drain Periodically, 189 liters
-------
0.6,
"• Steam Stripping Column Feed
-X Steam Stripping Column Bottoms
Furfural Activated Carbon Column Effluent
Acetic Acid Activated Carbon Column
Effluent
0.5
o.u
i
April 10, 1973
April 23, 1973
X
J?
. 0.3
•a
•H
O
<
o
ft
43
4)
o.:
o.it
j
Ethanol in All Samples
L
5 0
Hours Column Used
Figure 38. Steam stripping, furfural activated carbon
and acetic acid activated carbon columns
95
-------
The next steam stripping and carbon adsorption run was made April 23,
1973. The evaporator condensate was fed at the rate of 1 gpm (3.79
1/min.) into the steam stripping column. Once again samples of the
acetic acid activated carbon column effluent were taken at the end of
the first hour and every half hour thereafter for U hours total. No
acetic acid was detected in the samples after 3 hours, a trace amount
in the 3-1/2-hour sample and only 30$ by weight of that in the feed in
the U-hour sample. See Fig. 38. These results indicate that the new
regeneration method does an efficient job regenerating the carbon.
There was still no evidence that the acetic acid was being effectively
removed as ethyl acetate. Since all effluent samples showed ethanol,
it was obvious the regeneration purge of ethanol was stopped premature-
ly. The previous run had shown an excellent purge of ethanol, so this
is no problem.
Another regeneration attempt to verify the production of ethyl acetate
was made on April 2U, 1973. This time the odor of ethyl acetate was
strongly detected. A check for leakage was to no avail. Finally, the
pipe from the steam trap taking care of condensed steam from the acti-
vated carbon vaporizer and jacket was disconnected from the drain. The
ethanol and ethyl acetate coming from the pipe was of high concen-
tration, indicating a leak in the reboiler or the jacket. The two
systems were separated and the leak was pin-pointed as being in the
jacket. Analyses of the water discharging-from the carbon column and
finally leaving the bottom of the fractionation column showed 0.03$ by
weight acetic acid, indicating little loss as acetic acid. The run was
continued April 25, 1973 to verify the removal of acetic acid as ethyl
acetate. Approximately 2 quarts (1.9 liters) of 50$+ by weight ethyl
acetate were collected. Once the analyses showed sufficient ethyl
acetate was made, the regeneration was stopped, as the regeneration
method was shown effective and there was no need to risk further column
damage. The activated carbon was removed from the column. Inspection
showed that the inner shell of the column was pushed in opposite the
steam inlet to the jacket. A hole was under the convexed section about
96
-------
3 feet (91.U cm) from the top of the column. Further checking revealed
that the installed pop-off valve was for 70 psig (It.9 kg/cm2 gage) in-
stead of 7 psig (O.U9 kg/cm2 gage) as was expected. It is probable
that during the runs when trouble controlling the steam pressures occur-
red, the combination of high jacket pressure and a vacuum inside the
column occurred simultaneously buckling the inner shell. Since the pop-
off valve never released, the high pressure was not revealed. This was
the second time the inner column had collapsed, so no attempt was made
to repair it.
The new regeneration system appears more complicated than it is. A
simplified form is shown in Fig. 39- Essentially, the system involves
using the activated carbon column as a distillation column during the
carbon regeneration cycle. A smaller fractionation column is added to
the top of the carbon column for product purity improvement. This may,
or may not, be required. Alcohol (ethanol is used) is pumped into the
carbon column below the fractionation column and above the activated
carbon. Water containing acetic acid, polymerized material, and portions
of spent sulfite liquor comes out the bottom of the carbon column, and
d,s collected in the receivers. From there it is recircxilated by means
of a pump through a vaporizer and disengaging section at the bottom of
the carbon column. The vapor from the partially vaporized recirculating;
liquor passes up the carbon column- By controlling the heat .to the
vaporizer the carbon column is gradually converted into a reactor and
distillation column. As the vapors move up the column, water condenses
and is displaced, while the alcohol vaporizes and reacts with the acetic
acid to form esters (ethyl acetate). As the alcohol and acetate are
rectified at the top of the carbon column and in the fractionation
column by means of refluxing the condensed overhead vapors, the water
containing furfural, polymerized material, and residual spent sulfite
liquor are collected in the receivers. Once all the acetic acid has
been converted to ethyl acetate, the alcohol is recovered by vaporizing
the recirculating liquor and concentrating the collected materials in it.
97
-------
Vent
Crude Ethyl Acetate
Ethanol
Storage
Tank
T
Pump
Practiohation
Column
Conden
ser
2nd
Activated
Carbon
Column
Acetic
Acid
1
Pump
'TV. Dra.1 f)
Figure 39- Suggested regeneration
98
-------
When only steam remains in the carbon column, it is ready for its next
adsorption cycle.
Figures Uo and kl show the application of what was learned in this
project used as the bases to project the utilization of the developed
unit operations for clarifying evaporator condensate at Consolidated
Papers, Inc.'s Appleton Mill.
Present Project Status
During Project 3100 enough data were compiled to assist interested
sulfite pulp manufacturers toward making feasibility studies for uti-
lizing steam stripping, fractionation, and activated carbon adsorption
systems as tools toward solving their pollution problems. The major
project achievements were: ;
1. The development of a steam stripping system capable of removing
by weight of the sulfur dioxide in evaporator condensates. Only
the evaporator condensate data for interested mills need be estab-
lished to effectively design this system.
2. The fractionation system developed and patented by Scott Paper
Company was optimized. This system is capable of producing high
purity sulfur dioxide for reuse back in the mill, methanol for
reuse and eventual sale, and furfural for sale.
3. The verification of selective activated carbon adsorption data
compiled at Scott Paper Company and used as the bases for their
selective adsorption patent. Carbon loadings were established.
k. An activated carbon regeneration system capable of producing a high
purity ethyl acetate crude (&9% by weight ethyl acetate) from ad-
sorbed acetic acid was demonstrated.
5. The use of alcohol and water for regenerating the activated carbon
when using relatively clean evaporator condensate was shown to be
effective. The carbon life showed signs of deteriorating when dirty
evaporator condensate was processed, and the carbon was regenerated
99
-------
SO 2 Vapor
•3 I
V -p
0> 01
^ >5
o o
CO
Steam Strip •
ping Column
Tank
O
o
Evaporator
Condensate
Low Solids
Effects
-P
-------
Crude Ethyl Acetate
Fractionation
Column
Ethanol
Storage
Tank
sr
Pump
Con-
den-
ser
2nd
Activated
Carbon
Column
Acetic
Acid
U
1
Vent
Pump
To Spray Drier
Figure Ul.
and/or Pulp Washers
Suggested commercial application Consolidated
Papers, Inc. Appleton mill
101
-------
by the system developed and patented at Scott Paper Company, These
regeneration systems were not optimized.
6. A new activated carbon regeneration system was developed using
alcohol and water. This system involves using the activated carbon
column as a distillation column. It varies from the Scott method
in that reflux of alcohol and water is utilized in the carbon col-
umn. It completely regenerated a carbon that had been showing
definite carbon life deterioration, and may be patentable. This
process has not been optimized, although substantial cost reductions
when compared to the Scott process are apparent. Data .on this pro-
cess are limited and more work is necessary to establish its full
value.
The urgency required for pollution abatement is the factor most influ-
encing the movement of this project to commercialization in its en-
tirety. The next step is a commercial demonstration unit. Preferably
it should be constructed modularly. The steam stripping and fraction-
ation systems first, and the carbon adsorption and regeneration systems
next. The highest priority pilot plant study should generate data
specifically for this demonstration unit.
102
-------
SECTION VI
MASS, HEAT, AND BOD5 BALANCES
METHODOLOGY
Mass, heat and BOD5 balances are made according to the actual operating
condition of the pilot plant at the Appleton Division Mill of Consoli-
dated Papers, Inc. and are based on data taken prior to January IT, 1973.
See Tables lU, 15, and 16. Since the individual items of the equipment
are not optimized and this general process may not be particularly ex-
act for the evaporator condensate processed, it is believed that total
energy balances of this pilot plant would be of little value and could
be misleading. Only heat balances are made. From this basic infor-
mation, an estimate of the value of the products was obtained.
A pollutional balance sheet is also included. This sheet shows the
BODs values of the flow streams shown in the overall mass balances of
the pilot plant (Fig. U2). The references for those theoretical con-
versions are also listed.
The balances presented are for the particular arrangement which was
studied during these experiments. The particulars of product value,
equipment arrangement, and operating costs for other evaporator systems
will, of course, be somewhat different depending upon the individual
pulp mill's opportunities and major objectives in the installation of
this process.
MASS BALANCES
The mass flows are shown in six separate mass-flow sheets: Figure U3
covers the separation process; Fig, kh and 1*5, the recovery of furfural
and methanol; and Fig. 146 and U?, the recovery of ethyl acetate and
ethanol. The three operations are then combined as Fig. U2 which shows
the overall mass balances of the complete process.
103
-------
Table lU. TYPICAL PILOT-PLANT OPERATING DATA
Evaporator Condensate Transfer Pump Rotameter Reading
Evaporator Condensate Transfer Pump Pressure Gage
Steam Stripper Feed Tank Level
Steam Stripper Feed Pressure Gage
Steam Stripper Feed Rotameter — 67$
Steam Stripper Steam Rotameter — 10 setting
Steam Stripper Manometer
Main Line Steam Pressure Gage
Preheater, 1st Pressure Gage
Preheater, 2nd Pressure Gage
Preheater, Dial Thermometer
Reboiler, 1st Pressure Gage
Reboiler, 2nd Pressure Gage
Steam Stripping Column Top Pressure Gage
Steam Stripping Column Bottom Pressure Gage
Water to Steam Stripping Column Condenser Rotameter
Steam Stripping Column Condensed Overhead Rotameter, 9.5
Fractionation Column Reboiler Rotameter — 8.5 setting
Fractionation Column Steam, 1st Pressure Gage
Fractionation Column Steam, 2nd Pressure Gage
Fractionation Column Bottom Pressure Gage
Fractionation Column Manometer (Feed to Bottom)
Fractionation Column Top Pressure Gage
Furfural Activated Carbon Column Feed Pressure Gage
Furfural Activated Carbon Column Feed Rotameter
Furfural Activated Carbon Column Bottom Pressure Gage
Furfural Activated Carbon Column Top Pressure Gage
Acetic Acid Activated Carbon Column Top Pressure Gage
Acetic Acid Activated Carbon Column Bottom Pressure Gage
Temperatures
Steam Stripping Column Feed before Preheater
Steam Stripping Column Feed after Preheater
Steam Stripping Column Liquid Bottoms
Steam Stripping Column Top Vapor
Steam Stripping Column Vapor after Condenser
Steam Stripping Column Steam
Fractionation Column Liquid Bottoms
Fractionation Column Steam
Fractionation Column Feed
Fractionation Column at Feed Plate
Fractionation Column 1st Splitter
Fractionation Column 2nd Splitter
Fractionation Column Vapor after Condenser
Activated Carbon Column Bottom (Furfural)
Activated Carbon Column Center (.Furfural)
Activated Carbon Column Top
Activated Carbon Column Bottom (Acetic Acid)
Activated Carbon Column Top (Acetic Acid)
20
15 psig
Full
1*1 psig
1 gpm
0.060 gpm
0" H20
130 psig
130 psig
73 psig
211°F
83 psig
52 psig
No good .
l.k" H20
7.5 Setting
0.057 gpm
O.Ql+9 gpm
135 psig
125 psig
0" H20
0.3" H20
2.5" H20
18 psig
l.Oo gpm
5 psig
k psig
^ psig
6.7 psig
(1.05 kg/ cm2)
(2.88 kg/cm2)
(3.79 1/min.)
(0.23 1/min.)
(0 cm HaO)
(9.1A kg/cm2)
(9.11* kg/cm2)
(5.13 kg/cm2)
(99.1»°C)
(5.83 kg/cm2)
(3.65 kg/cm2)
(18.8 cm H20)
(0.22 1/min. )
(.0.18 1/min. )
(9.U9 kg/cm2)
(8.79 kg/cm2)
(0 cm H20)
(0.76 cm H20)
(6.35 cm H20)
(1.26 kg/ cm2)
U.01 1/min.)
(0.35 kg/cm2)
(0.28 kg/ cm2)
(0.28 kg/cm2)
(O.Vr kg/cm2)
13l*°F
2l2°F
212°F
213°F
203°F
2l*7°F
210°F
213°F
166°F
213°F
213°F
ll*9°F
97°F
lUo°F
lUO°F
135°F
129°F
133°F
(56.7°C)
(100. 0°C)
(100. 0°C)
(100. 6°C)
(95.0°C)
(ll9.1*°C)
(98.9°C)
(100. 6°c)
(7U.lt°C)
(100. 6°C)
(100. 6°c)
(65.0°C)
(36.l°c)
(60.0°C)
(60.0°C)
(57.2°C)
(5.3.9°C)
(56.1°c)
-------
fable 15. TYPICAL ACTIVATED CARBON ACETIC ACID COLUMN REGENERATION DATA
Fraotionation Column Reboiler, 1st Pressure Gage _ gT pslg (ijt71
Fractionation Column Rebolier, 2nd Pressure Gage - gg p8ig (^
Fractionation Column Bottom Pressure Gage _ 0 psig (0.00 kg/cm?)
Fractionation Column Manometer (Peed to Bottom) - i" Hz0 (2.5^ cm Hzo)
Fractionation Column Top Pressure Gage _ 2V H20 (6.35 cm H20)
Fractionation Column Middle Pressure Gage — 3" H 0 (7 62 cm H 0)
Acetic Acid Activated Carbon Column Top Pressure Gage — o psig (0.00 kg/cm2)
Acetic Acid Activated Carbon Column Jacket Steam Pressure — It psig (0.28 kg/cmz)
Acetic Acid Activated Carbon Column Bottom Pressure Gage — 7 psig (O.U9 kg/oma)
Acetic Acid Activated Carbon Column Vaporizer 1st Steam
Pressure Gage _ 135 psig (9.1^9 kg/cjn2)
Acetic Acid Activated Carbon Column Vaporizer 2nd Steam
Pressure Gage - 115 psig (8.08 kg/cm2)
Acetic Acid Activated Carbon Column Rotameter to Acetic
Carbon Column - 0.30 gpm (l.lU l/mi,n.)
Acetic Acid Activated Carbon Column Fractionator (Alcohol) — 0.30 gpm (l.ik 1/min,)
Temperatures
Fractionation Column Liquid Bottoms - 212°F (100.0°C)
Fractionation Column Steam - 213°F (100.6°C)
Fractionation Column Bottom Splitter - 173°F t78.3°C),
Fractionation Column Feed - 17U°F (78.9°C)
Fractionation Column at Feed Plate - 172°F l77-80C)
Fractionation Column at Middle Splitter - l62°F (72.2°C)
Fractionation Column at Top Splitter - 157°F (.69.^°C)
Fractionation Column Vapor after Condenser - 100°F (37.8°C)
Acetic Acid Activated Carbon Column Top - 175°F C79.t»°C)
Acetic Acid Activated Carbon Column Bottom - 173°F (78.3°C)
105
-------
Table l6. TYPICAL ACTIVATED CARBON FURFURAL COLUMN REGENERATION DATA
Fractionation Column Reboiler, 1st Pressure Gage
Fractionation Column Reboiler, 2nd Pressure Gage
Fractionation Column Bottom Pressure Gage
Fractionation Column Manometer (Feed to Bottom)
Fractionation Column Top Pressure Gage
Fractionation Column Middle Pressure Gage
Activated Carbon Furfural Column Top Pressure Gage
Activated Carbon Furfural Column Bottom Pressure Gage
Temperatures
Fractionation Column Liquid Bottoms
Fractionation Column Bottom Splitter
Fractionation Column Feed '
Fractionation Column at Feed Plate
Fractionation Column Middle Splitter
Fractionation Column Top Splitter
Fractionation Column Vent
Activated Carbon Furfural Column Bottom
Activated Carbon Furfural Column Middle
Activated Carbon Furfural Column Top
Flow Rates to Activated Carbon Column
Flow Rates to Fractionation Column
70 psig (U.92 kg/cm2)
- 67 psig (U.71 kg/cm2)
— 0 psig (0.00 kg/cm2)
— • 1" H20 (2.51* cm H20)
- 2V H20 (6.35 cm H20)
- 3" H20 (7-62 cm H20)
3 psig (0.21 kg/cm2)
- 0 psig (0.00 kg/cm2)
- 213°F (100.6°C)
(65.0°C)
(65.0dC)
150°F 165.6°C)
96°F (35.6°C)
r 163.9°C)
? t63.9°C)
151°F (66.1°c)
0.10 gpm (0.38 1/min.)
0.10 gpm (0.38 1/min.)
106
-------
TF, Xb 1*3.8 Sulfur Dioxide K^O 3-90; Furf. O.900; MeOH 1.50; S0a 37-5
UEGEND
water
furfural
methanol
sulfur dioxide
acetic acid
EtOH ethanol
EtOAc ethyl acetate
Poly Matls. polymerized material!
TF, lb total flow, lb
Furf.
MeOH
MeCOQH
Evap. Cond. TF, lb 50,000
HaO 49,800; Furf. 8.00;
MeOH 3«!.!>; ^6_ 4y.oo;*
MeCOOH "'-
Water TF, lb 6000
HgO 5000
Water TF, lb
H,0 tOQ
Water TF, lb 16,100
«80-
Ethanol TF, lb 64.7
-«*r?
Q
ON
TF, lb 15.2 Methanol MeOH 15.0; SO 0.200
TF, lb 7.30 Furfural HgO 0.200; Furf. 7-10
W
o
&fr
TF, lb 68,900 Sewer (or recycle)
HaO 68,900; S0a 5-30; MeCOOH 38.2;
MeOH l6.0; unaccounted HAc 11.4;
Poly Matls. 6.00
Figure U2. Overall mass balances of the pilot plant
-------
^ TF, Ib 0.438 Sulfur Dioxide Furf. O.O
LEGEND
TF, Ib
Furf.
MeOH
SO,
HsO
MeCOOH
IF, Ib 500
total flov, Ib
furfural
methanol
sulfur dioxide
water
acetic acid
to be retained
unaccounted
Ib 500
I-1
O
CD
Furf. 0.0800;
MeOH 0.325;
, 0.14.90;
HaO 498;
MeCOOH 1.3k
TF, Ib 500
Evaporator
Condensate
Storage
Furf. 0.0800;
MeOH 0.325;
SOa 0.^90;
HaO t98;
MeCOOH 1.3k
A Furf. 0.0800 ;
| MeOH 0.325;
Oa 0.^90;
HsO 1*98;
MeCOOH 1.3k
TF, Ib 500
Furf. 0.0800;
MeOH 0.325;
SOa 0.11-90 ;
HaO lt-98;
MeCOOH 1.3k
TF, Ib 30
HsO 30
Vteter
Basis:
Scale:
TF, Ib 30
Furf. O.Ol(OO;
MeOH 0.165;
SOa O.lUQ;
HaO 29.2;
MeCOOH 0.0560
(MeCOOH
1
0.0390
so,
0.21)0;
MeOH
0.0120
Furf.
0.0090(
1TF, Ib 0
MeOH 0.0
80s 0
.138
0300
.135
MM
i Me
MB
OH 0.01
50; HftO 0.0390; S0a 0.575
TF, lb 0.152 Methauol
MeO
H 0.150; SO, 0.002
TF, Ib 0.033 Furfural
Furf. 0.031O; HaO O.00200 —
TF, Ib 500; MeOH
-tpy; SOa
COOH'LIT
^99J
ri
1 hour
none
i
o
O
I
Oi
S
-rj
fc
I*
I
s
&
TF, Ib 500
Furf. 0.01(00;
MeOH 0 .160; ^
Furf.
0.0090(
>;
\ S TJ?', J-6
0.300
IF, Ib •"*' lb 29'7
Furf.
O.OtoO;
MeOH
0.165;
so»
O.lHO;
^a HaO
29-2;
MeCOOH
0.0560;
(MeCOOH
Furt. "
0.0310;
MeOH
0.153;
SOa
0.170;
HsO
29
.2
MeCOOH
0.0560;
(MeCOOH
O.Tlll^
-
SOa 0.08OO;
HaO It99
MeCOOH 1.17
,
fa
\
i
\K
TF, Ib 30 1
— — •/
c
I
4
!
•3
O
§
•H
1
Fractlot
a
§1
£)
i
[TF, It
0.0900]
ri?!ti~r
C
C
i.otoo;
^m^m
fi
%
^
1
tivated C
O
%
J^
-------
Make-up
Methanol
TF, Ib 0.639
MeOH 0.639
IP, Ib 639 r
MeOH 639 *
TF. Ib 64O
£££
O
VO
" TF, Ib &£>
MeOH 650
none
Scale:
Basis:
Recovery of furfural:
Recovery of methanol:
8 hours
LEGEND
TP, Ib
Furf.
MeOH
Poly Matls.
total flov
furfural
methanol
polymerized materials
water
to be retained
originally retained
Vent
> M
••
§
-1
1
>
:
ITS
CS
eOH 61K)
(IF, hr
100)
(HaO
92.0)
(Furf.
4.00)
(Poly
Matls.
4.00)
[MeOH
0.639]
[IP, hr
0.639]
v
1st Activated Carbon Column
Tl
Condenser 1— >.
1 >
v IF, it 739
, Ib 639
639
TF, Ib 739
MeOH 639;
HaO 9S.O;
4.00;
Poly Matls.
,00
MeOH 639; H20
92.0; Furf. 4.00;
Poly Matls. 4.00
IF, Ib 4.00 Furfural
Furf. 4.00 '
Ib 73
MeOH 639;
92.0;
Furf. 4.00;
Poly Matls.
4.00
TF Ib 739
MeOH 639;
O 92.0j
Furf. 4.00;
Poly Matls. 4.00
IF, Ib 96.0
*-Sewer
HaO 92.0;
Poly Matls.
4.00
Figure
Activated carbon regeneration and furfural recovery.
1. Furfural recovery
-------
o
Scale: none
TF, Ib
Furf.
Poly Matls.
HsO
[ J
( )
TF, hr 0.639
MeOH 0.639 TF, hr too
> ' HaO too
o j* I Jj
41 3 I g
§•
ir too
too
le
2JD
total flow
furfural
Polymerized materials
water
to be retained
original 1 y retained
1 '
s
o
0
o
•P
m
c-1
i' TF,
MeO
(TF, hr
0.639)
(MeOH
[TF, hr <
92.0]
[HaO 92.0]
i-U
FH
hr 309
308;
H 0.639
*
^
\
'i
i
c
c
P
1
p
M
H
1)
>
-1
1)
J
S
F
j
IF, hr 309
HsO 308 ;
MeOH .0.639
1 1
jlj r
t
I
Vent
A
JL,
-------
H
H
LEGEND '
TF, Ib total flow.
EtOH
HaO
Ib
ethanol
water
MeCOOH acetic acid
Poly polymerized
Matls . materials
[ ] to be retainei
I
>
'
originally
retained
it *r* CS
G^ 1
p «i £
^H PCI PH
" TF, Ib 56.0
^ J ^— — ^ -ir- -•> —
EtOH 56 ,0~~
g
3
"3
o
a
*
•a!
O
1
5
j^>
^4
"2
a
OJ
In
EtOH 56.0; h
Poly Matls SJ
0 .006*10 -5}
o
TF, Ib 367 ^
HsO k)3; EtOH*~ HgO 3.66; MeCOOH "
Basis: 56.0; MeCOOH 0.978; Poly Matls.
Recovery of ethyl acetate: 12 hr 0.978; Poly O.O136
Recovery of ethanol: k- hr Matls. 0.0200
0)
•a
o
o
a
I
o
•H
-P
1
L I
Sever
i!
Figure U6. Activated carbon regeneration and ethyl acetate recovery.
1. Draining of column
-------
_TF, lb 1470
EtOH 1470
(TF, lb 2.57)
(Poly Matls. O.b400
(MeCOOH 2.53)
[TF, lb 483]
[SaO 433]
TF. lb 1470
Vent
ro
Make-up
Ethanol
LTF, lb 1.
EtOH 1.94
fEtOH 56.0]
!
I
1
a
•s
&
TF, lb 14l6
EtOH 1*H6
Water vQ displace ethanol TF. lb 483
-jjg-
Scale; none
HB0 433
Figure Vf.
2.
TF, lb 1417
EtOH 1414;
EtOAc 3.71;
H=0 0 75Q
(TF, hr 173)
(BtOH 56.0)
(HaO 117]
(Poly Matls. 0.00640)
0BMBI
n
•
C
on
y
13
^
5
j
. ITF,
Bt<
EtC
IH&C
1
8
£
I
EtOH
lb 1419
)H l4l4;
lAc 3.71;
> 0-759 ;
1470
3.71;
H30 117;
Poly
Matls.
0.0464
±
M
I
0)
•8
o
o
I
a
o
•H
I
•ri
£
\
Ethyl Acetate
TF, lb 3.71 *•
BtOAc 3.71
LEGEHD
TF, lb, to
EtOH et
HfeO wa
MeCOOH ac
EtOAc .etl
Poly Matls. po
[ 1 to
( ) or
-------
The individual titles of the figures and the brief calculations are as
follows:
I. Fig. U8 —Equipment flow sheet of the separation process.
II. Fig. U9 —Equipment flow sheet of the activated carbon regen-
eration and furfural recovery.
III. Fig. 50 —Equipment flow sheet of the activated carbon regen-
eration and ethyl acetate recovery.
IV. Fig. U2 -Overall mass balances -Basis: 50,000 Ib (22,700 kg)
of evaporator condensate. To combine the separation
process and the two recovery-regeneration processes,
the flows shown in Fig. k3 are multiplied by 100,
those in Fig. UU and kf by 1, and those in Fig. k6
and hi by 33.3.
V. Fig. U3 —Mass flow sheet of the separation process. The
quantities indicated on this flow sheet are based on
the following calculations:
1. Basis: The total flows shown are based on 500 Ib (227 kg) of
evaporator condensate feed. It should be noted that three
significant figures are used for the flow quantities.
2. Evaporator condensate: The following mean analyses taken from an
interim report to sponsors of the project were used.
Acetic acid 1-3U It (0.6l kg)
Methanol 0-325 Ib (0.15 kg)
Furfural 0.0800 Ib (O.OU kg)
Total sulfur dioxide 0.1*90 Ib (0.22 kg)
Water ^8 Ib (226.09 kg)
3. Mass balance around stripping column:
A. In:
a. Preheated feed
Acetic acid 1-3^ Ib (0.6l kg)
Methanol 0.325 Ib (0.15 kg)
Furfural 0.0800 Ib (O.OU kg)
113
-------
Q Thermometer
, *,*«,*„„ -..
fCft Recorded
T' Temperature
fea Pressure Gage
f Tj Trap
|| Manometer
y Rotameter
->C~ Valve
I"1
Sulfur Dioxide
Sulfur Dioxide
Methanol
Scale: none
Sewer
Figure U8. Separation process
-------
Methanol
from
Fractionation
Column of
Separation
ui
y
Pressure
Gage
Trap -IX-Valve
Rotaiaeter If Manometer
Scale: none
Figure ^9. Activated carbon regeneration and furfural recovery
-------
Vent
Vent
ON
LEGEND
Pressure Gage
Recorded Temperature
Trap
Valve
Rotameter
Ethyl Acetate
Steam
i
Scale: none
Sewer
Figure 50. Activated carbon regeneration and ethyl acetate recovery
-------
Total sulfur dioxide 0.^90 lb (0.22 kg)
Wate* ^98 lb (226.09
500 lb (227.00 kg)
b' Steam 30 lb (13.62 kg)
Total in = 530 lb (2^0.62 kg).
B. Out:
a. Bottoms — by analysis
Acetic acid 1.17 rb (0.53 kg)
Methanol 0,160 lb (0.07 kg)
Furfural 0.0^0 lb (0.02 kg)
Total sulfur dioxide 0.0800 lb (O.OU kg)
Water 1*99 lb (226.55 kg)
500 lb (227.00 kg)
b. Volatiles — by difference
Acetic acid 0.0560 lb (0.02kg)
Methanol 0.165 lb (0.07 kg)
Furfural 0.0^00 lb (0.02 kg)
Total sulfur dioxide 0.1*10 lb (0.19 kg)
Unaccounted acetic acid O.llU lb (0.05 kg)
Water 29.2 lb (13.26 kg)
30.0 lb (13.62 kg)
Total out = 530 lb (2^0.62 kg)
It should be noted that since the compositions of the preheated feed,
the gases leaving the receiver, and the fractionation column feed were
all based on actual analyses, the "unaccounted acetic acid" has to be
used here to complete the acetic acid analysis.
U. Around the fractionation column:
A. In:
a. Feed
Acetic acid 0.0560 lb (0.02 kg)
Methanol 0.153 lb (0.07 kg)
Furfural 0.0310 lb (0.01 kg)
Total sulfur dioxide 0,170 lb (0.08 kg)
117
-------
Unaccounted acetic acid O.llU lb (0.05 kg)
Water 29.2 lb (13.26 kg
29.7 lb (13.48 kg
b. Steam 30.0 lb (13.62 kg)
Total in - 59.7 lb (27-10 kg)
B. Out
a. 802 stream
S02 0.135 lb (0.061 kg)
Methanol 0.003 lb (.0.001 kg
0.138 lb (0.063 kg
b. Methanol stream
Methanol 0.150 lb (0.068)
Sulfur dioxide 0.002 lb JO.001)
0.152 lb(0.069)
c. Furfural stream
Furfural 0.0310 lb (O.OlU)
Water 0.0020 lb JO.001)
0.0330 lb(0.015)
d. Bottom
Acetic acid 0.0560 lb (0.025)
Sulfur dioxide 0.0330 lb (0.015)
Unaccounted HAc O.llU lb (0.052)
Water 59.2 lb (26.877)
59.^lb (26.968)
Total out = 0.138 + 0.152 + 0.0330 + 59.1+
= 59-7 lb (27.10U kg)
5. First activated carbon column:
A. In:
Acetic acid 1.17 ib (0.531 kg)
Methanol 0.160 lb (0-073 kg)
Furfural 0,0^00 lb (0.018 kg)
Sulfur dioxide 0.0800 lb (0.036 kg)
Water ^99 lb (226.5^6 kff^
500 lb (227.000 kg)
118
-------
B. Out:
Acetic acid ltl? lb (0.531 kg)
Methanol 0>160 lb (0.073 kg)
Sulfur dioxide 0.0300 lb (b.OlU kg)
Water ^99 lb (226.5^6 kg)
500lb (227.000 kg)
C. Retained:
Furfural 0.0^00 lb (0.018 kg)
Sulfur dioxide 0.0500 lb (0.023 kg)
0.0900 lb (0.01*1 kg)
In = Out + Retained
6. Second activated carbon column:
A. In:
Acetic acid 1.17 lb (0,531 kg)
Methanol 0.160 lb (0.073 kg)
Sulfur dioxide 0.0300 lb (O.Oll* kg)
Water 1*99 lb (226.51*6 kg)
500 lb (227-000 kg)
B. Out:
Methanol 0.160 lb (0.073 kg)
Sulfur dioxide 0.0100 lb (O.OQl* kg)
Water 1*99 lb (226.5^6 kg)
1*99 lb (226.5^6 kg)
C. Retained:
Acetic acid 1.17 1*> (0.531 kg)
Sulfur dioxide 0.0200 lb (0.009 kg)
l7l9lb (0-5^0 kg)
In = Out + Retained
VI. Fig. 1*1* -Activated carbon regeneration and furfural recovery:
1. Furfural recovery.
This flov sheet shows the mass flows during the cycle of furfural
recovery. The balances are based on a 16-hour cycle of furfural re-
covery and an 8-hour cycle of methanol recovery. The flow quantities
shown on this sheet are the total flows of 16 hours. The balances of
importance are:
119
-------
A. Around the activated carbon column:
a. In:
Methanol 6^0 lb (290.56 kg)
6kO lb (290.56 kg)
b. Removed from column:
Furfural ^..00 Ib (l.8l6 kg)
Polymerized materials U.OO Ib (l.8l6 kg)
Water 92.0 Ib (Ul.768 kg)
100Ib (U5.UOO kg)
c. Retained:
Methanol vapor = 0.639 Ib (0.290 kg)
calculated as follows':
volume of column = 60 gal. (227-1 l)
number of moles MeOH vapor in the column = n_
•
compressibility factor = 1
l6.2-(lfrU) (8.02) ' n,onc ,
' ' 2- = 15*i5 (607) = °-01"5 no' moles
.'. methanol vapor = 0.639 Ib (0.290 kg)
d. Out:
Furfural ^.00 Ib (l.8l6 kg)
Methanol 639 Ib (290.106 kg)
Polymerized materials H.OO Ib (l.8l6 kg)
Water 92.0 Ib (lq.768 kg)
739Ib (335-506 kg)
6UO + 100 = 0.639 + 739
B. Around the fractionation column:
1. In:
Furfural 1*.00 lb (l.8l6 kg)
Methanol 639 lb (290.106 kg)
Polymerized materials U.OO lb (l.8l6 kg)
Water 92.0 lb (lq.768 kp^
T39 lb (335-506 kg)
120
-------
2. Out:
a. Furfural stream
Furfural K.OO Ib (l.8l6 kg)
H.OO Ib (1.816 kg)
b. Bottom
Polymerized materials lj.00 Ib (l.8l6 kg)
Water 92.0 Ib (Ul.768 kg)
96.0 Ib (1*3.584 kg)
c. Methanol
Methanol 639 Ib (290.106 kg)
639Ib (290.106 kg)
739 (335.506 kg) = I*.00 + 9.6.0 + 639
VII. Fig. U5 —Activated carbon regeneration and furfural recovery:
2. Methanol recovery.
The balances are based on an 8-hour cycle of methanol recovery and
the flow quantities shown on this flow sheet are the total flows of 8
hours.
VIII. Fig. U6 —Activated carbon regeneration and ethyl acetate re-
covery :
1. Draining of column.
IX. Fig, 1*7 —Activated carbon regeneration and ethyl acetate recov-
ery:
2. Recovery of ethyl acetate and recovery of ethanol.
The balances shown are based on an ethyl acetate recovery cycle of 12
hours and ethanol recovery cycle of 4 hours. The flow quantities
shown are totals of 12 hours, except that of the water flow, which is
the total of U hours.
HEAT BALANCES FOR PILOT PLANT
The balances shown in this section are primarily based on the recorded
data shown in the interim report to sponsors of this project and the
mass flow quantities in Fig. U2. Since the basic flow quantities in
Fig. U2 are 100 times that of Fig. 43, the balances for the separation
121
-------
process made in this section are also 100 times that of the actual flow
quantity of 500 Ib evaporator condensate processed in 1 hour. The
calculations for the regeneration processes are accordingly proportion-
ed.
Since the individual equipment used in the pilot plant is not optimized,
and the flow quantities and conditions vary from case to case, it is
felt that a total energy balance of the pilot plant would not be help-
ful in the economy feasibility analyses. Only enthalpy balances are
made. The calculations are based on the actual operating conditions,
so that estimations of the heat requirements can be made.
Enthalpy balances:
I. Separation process
A. Preheater Sp. Heat,
Water U9.800 Ib (22609-2 kg) 1.05
Furfural 8.08 Ib (3-6 kg) 0.1*18
Methanol 32.5 Ib (lU.8 kg) 0.66
Acetic acid 131* Ib (60.8 kg) 0.5^
Sulfur dioxide 1*9.0 Ib (22.2 kg) 0.5 (estimated)
The above heated from 13^°F (56.7°C) to 212°F (100.0°C)
Heat required, HI = |>9800 (l.05) + 8.00 (O.UlS) + 32.5 (0.66) +
13^ (0.5*0 + H9-0 (0.5)1 (212-13U) = U,090,000 Btu
(1,030,680 kg-cal).
B. Stripping column reboiler
To heat 3000 Ib (1362 kg) of water from TO°F (21-1°C) to 2^T0F
(119-U°C) vapor, heat required H2 = 3000 [(2^7-212) (0.1*82) +
970 + (212-70)] = 3,390,000 Btu (85^,280 kg-cal).
Estimated amount of heat required by the stripping column:
1. Assume there is no heat loss from the point the condensate
leaves the preheater to the feed point. The feed is in the
liquid state and at a temperature which is just about the
bubble point. Therefore, no heat is required to bring the
feed up to the bubble point.
122
-------
2. Heat of vaporization * 2920 (970.3) + U.OO (107.5) (1.8) +
16.5 (262.8) (1,8) + 1*1.0 (9U.9) (1.8) + 5.60 (96.8) (1.8)
+ 11.U (300) (1.8) a 2,860,000 Btu (720,720 kg-cal).
3. The bottom stream leaves at 210°F (98.9°C). Heat given up
by the bottom stream = [(1*9900 + x) (l) •*• 1*.00 (O.Ul8) +
16.0 (0.5.66) + 8.00 (0.5) + 117 (0.1*87)1 (212-210) = 99,900
+ 2x_.
If 73 psig (51.3 kg/cm2) live stream is used: 2,860,000 -
(99,900 + 2x) = [(318-212) (O.W2) + 970.3 + (212-210)] x
1025x = 2,959,900
x = 2690 Ib (1221.3 kg)
Total heat required = 2,750,000 Btu (693,000 kg-cal). This
compares with 3,390,000 Btu (85l*,280 kg-cal).
Stripping column condenser
C of the components at gaseous state
Water vapor O.U82 Btu/lb/°F (O.U82 kg-cal/kg/°C)
Methanol 0.1*58 Btu/lb/°F (O.U58 kg-cal/kg/°C)
Sulfur dioxide 0.131* Btu/lb/°F (0.131* kg-cal/kg/°C)
Acetic acid 1.50 Btu/lb/°F (1.50 kg-cal/kg/°C)
Furfural 0.6 Btu/lb/°F (0.6 kg-cal/kg/°C)
(estimated).
C of components at liquid state
Water 1 Btu/lb/°F (l kg-cal/kg/°C)
Sulfur dioxide 0.5 (estimated)
Methanol 0.566
Acetic acid 0.1*87
Furfural 0.1*18
Unaccounted HAc 0.5 (estimated)
Heats of vaporization of components
Water 970.3 Btu/lb (539-0 kg-cal/kg)
Acetic acid 96.8 cal/g (ca) (17^.2 Btu/lb)
Methanol 262.8 cal/g (ca) (1*73.0 Btu/lb)
Furfural 107-5 cal/g (ca) (193.5 Btu/lb)
Sulfur dioxide 9^.9 cal/g (ca) (170.8 Btu/lb)
123
-------
1. Those gases cooled from 213°F (100.6°C) to 203°F (95.0°C)
without condensing
Water vapor 3.90 Ib (1.771 kg)
Sulfur dioxide 24.0 Ib (10.896 kg)
Methanol 1.20 Ib (0=545 kg)
Furfural 0.900 Ib (0.409 kg)
H3 = [3.90 (0.482) + 24.0 (0.13*0 + 1.20 (0.458) + 0.900
(0.6)] (10) = 61.9 Btu (15-6 kg-cal).
2. Those gases cooled from 213°F (100.6°C) to 212°F (100°C),
condensed, and cooled to 203°F (95°C)
Water 2,920 Ib (1325.7 kg)
Sulfur dioxide 17.0 Ib (7•7 kg)
Methanol 15.3 Ib (6-9 kg)
Furfural 3.10 Ib (1.4 kg)
Acetic acid 5-60 Ib (2.5 kg)
Unaccounted HAc 11.4 Ib (5-2 kg)
Hi»i = [2920 (0.482) + 17-0 (0.134) + 15-0 (O.U58) + 3.10
(0.6) + 5.60 (1.5) + 11.4 (1)] (1) = 1440 Btu (362.9 kg-
cal).
It should be mentioned that the sulfur compounds that condensed into
liquid form are taken as S02 and that the C of the unaccounted acetic
P
acid is estimated as 1.
H-2 = 2920 (970) + 17.0 (94.9) (1.8) + 15.3 (263) (1.8) +'
3.10 (108) (1.8) + 5-60 (96.8) (1.8) + 11.4 (100) (1.8) =
2,8.50,000 Btu (718,200 kg-cal).
The sulfur compounds are taken as S02 and the X of the unaccounted
acetic acid is estimated as 100 cal/g (180 Btu/l6)
H., 3 = [2920 (1) + 17.0 (0.5) + 15-3 (0.566) + 3.10 (0.418)
+ 5-60 (0.487) + 11.4 (0.5)] (212-203) = 26,500 Btu (6678
kg-cal)
IU = Hm + Ek2 + H43 = 2,880,000 Btu (725,760 kg-cal).
124
-------
D. Fractionation column
1. Reboiler - evaporates 3000 ib (1362 kg) of water from 70°F
(21.1°C) to 2l*7°F vapor. Heat required = [(212-50) + 970
+ O.H82 (35)] (3000) = 3,390,000 Btu (85!*,280 kg-cal).
Estimated amount of heat required by the fractionation
column:
a. Heat to bring the feed temperature, l66°F (7l*.l*°C) to
its bubble point, say 205°F (96.1°C) = [2920 (l) +
3.10 (0.1*18) + 15.3 (0.566) + 17.0. (0.5) + 5-60 (0.1*78)
+ 11.1* (0.5)] (205-166) = 115,000 Btu (28,980 gm-cal)
b. Heat of vaporization = (15-0 + 0.3) (262.8) (1.8) +
(0.2 + 13.5) (91*-.9) (1.8) + 0.2 (970) + 3.10 (107-5)
(1.8) = 10,300 Btu (2596 gm-cal)
c. Heat to bring the bottom stream [(x + 2920 - 0.2) Ib of
water, 3.30 S02, 5.60 HAc and 11. U unaccounted HAc]
from 205°F (96.1°C) to 210°F (98.9°C) = [(2920 + x)
(1) + 3.30 (0-5) + 5-60 (0,W7) + 11.1* (0.5)] (210-
205) = 5(2930 + x) = 1U650 + 5x (x = Ib of steam
used).
If live steam is used, and the steam comes in at 73
psig (5.13 kg/cm2) (3l8°F) (158.9°C) and leaves at
210°F (98.9°C) as liquid, then 115000 + 10300 + 1^650
+ 5x [(318-212) (0.1*82) + 970.3 x + (212-210)] x.
x = 137 Ib (62.2 kg)
Total heat required = iUI.OOO Btu (35532 kg-cal). This
compares with the heat required for the reboiler,
3,390,000 Btu (85^280 kg-cal).
2. Let lU9°F (65°C) be the reference temperature. Heat in
with steam which is at 213°F (100.6°C), HI = 3000 [(0.1*82)
(1) + 970 + (1) (63)] = 3,100,000 Btu (781200 kg-cal).
3. Heat in with the feed which contains:
Water 2920 Ib (1325-7 kg)
Furfural 3.10 Ib (l.l* kg)
125
-------
Methanol 15.3 Ib (6.9 kg)
Sulfur dioxide 17*0 Ib (7-T-kg)
Acetic acid 5-60 Ib (2.5 kg)
Unaccounted HAc 11.1* Ib (5.2 kg)
H2 [2920 (l) + 3.10 (0.1*18) + 15-3 (0.566) + 17-0 (0-5) +
5.60 (0.1*8?) + ll.U (0.5)] (166-1U9) = 50,100 Btu (12625
kg-cal).
1*. Heat out with methanol, HS = 0.
5. Heat out with furfural stream:
H^ = [3.10 (0.1*18) + 0.200 (I)] (213-11*9)'= 96.0 Btu
(2U.2 kg-cal).
6. Heat out with the bottom stream:
H5 - [5920 (1) + 3.30 (0,5) + 5.60 (0.1*87) + 11.1* (0.5)]
(210-11*9) = 36,200 Btu (9122.1+ kg-cal).
7. Heat out with sulfur dioxide stream, He =0.
8. Neglect the heat of solution and heat of dilution.
9. Heat loss of the column, H7.
3,100,000 + 50,100 = 96 + 36,200 + H7.
Heat loss = 3,110,000 Btu (783720 kg-cal).
E. Cooler before the first carbon column. The stripping column
bottom is 50,000 Ib (22700 kg) containing:
Furfural 1*.00 Ib (1.82 kg)
Methanol 16.0 Ib (7.26 kg)
Sulfur dioxide 8.00 Ib (3.63 kg)
Acetic acid 117 Ib (53.12 kg)
Water - 1*9900 Ib (2265!+.6 kg).
To cool this stream from 212°F (100°C) to ll*0°F (60°C), the
heat required • [1*.00 (0.1*18) + 16.0 (0.566) + 8.00 (0 = 5) +
117 (0.1*87) + 1*9900 (1)] (212-11*0) = 3,600,000 Btu (907200 kg-
cal) .
F. Condenser after fractionation column.
The S02 gas stream comes in at ll*9°F (65°C) and leaves at 97°F
126
-------
Heat given up • [13.5 (0.13H) + 0.300 (O.H58)] UH9-97) «
101 Btu (25.lt kg-cal).
Heat given up by the methanol stream during condensing from
vapor to liquid = 15.o (263) (1.8) + 0.2 (9H.9) (1.8) = 71^0
Btu (1799-3 kg-cal).
Total heat removed by cooling water = 721*0 Btu (182H.5 kg-cal).
II. Recovery of furfural and methanol
A. Furfural recovery
1. Vaporizer
Assume methanol available at,70°F.(2l.l°C) from feed tank.
sp. gr. = 0.792
C = 0.6 (liquid)
C = 0.1*58 (vapor)
X = 262.8 cal/g
b.p. = 6U.7°G = lU8.5°F
a. Heat required to heat the MeOH from 70°F (21.1°C) to
lH8.5°F (6H.T°C). Hi = 6UO (1U8.5-70) (0.6) = 30,100
Btu'(7585.2 kg-cal).
b. Heat required to heat the vapor from lU8.5°F (67.7°C)
to 151°P (66.1°C). H3 = 6UO (O.U58) (1.8) (2.5) •
1320 Btu (332.6 kg-cal).
2. Condenser after the activated carbon column.
The liquid coming out of the activated carbon column con-
tains 639 lb (290.1 kg) of methanol, 92.0 Ib (Hi.8 kg) of
water, H.OO lb (1.8 kg) of furfural, and H.OO lb (1.8 kg)
of polymerized materials. The temperature is at ll+8°F
(6H.H°C). After the condenser, it is cooled to an average-
temperature of 100°F (37.8°C). Heat given up • (lHS-100)
[639 (0.566) + 92.0 (1) + H.OO (0.-H18) + H.O (0.5)] -
22,000 Btu (55HH kg-cal).
3. Fractionation column
a. Reboiler
127
-------
Assume that the liquid comes in at 210°F (98-9°C) and
goes out at 2l*7°F (119.U°C) (same as the other re-
toilers in "separation process"), and that 9«60 Ib
(U.36 kg) of the liquid are reboiled. In 9.60 Ib
(1*.36 kg) of the bottom liquid there are 9-20 Ib (U.I
kg) of water and 0.1*00 Ib (0.18 kg) of polymerized
materials.
Make the following assumptions for the polymerized
materials:
C (liquid) 0.6
C (vapor) 0.3
A 300 cal/g (5^0 Btu/lb)
b.p. 212°F (100°C)
Heat required to heat liquid from 210°F (98-9°C) to
212°F (ldO°C) = 9-20 (1)
(2) + 0.1*00 (0.6) (2) = 18.9 Btu (It.76 kg-cal)
Heat of vaporization = 9.20 (970.3) + 0.1*00 (300)
(1.8) = 9ll*0 Btu (2303 kg-cal)
Heat required to heat the vapor from 212°F (lOO°C) to
2l*7°F (119.1+°C) = 35
[9-20 (0.1+82) + 0.1*00 (0.3)] = 159 Btu (1*0 kg-cal)
Total heat required = 9320 Btu (231*8.6 kg-cal)
Use of ll*9°F (65°C) as reference temperature.
Heat in = [9-20 (0.1*82) + 0.1*00 (0.3)] (213-212) +
9-20 (970.3) + 0.1*00 (300) (1.8) + [9.20 (l) + 0.1*00
(0.6)] (212-11*9) = 971*0 Btu (2l*5l*. 5 kg-cal)
Heat out = [lOl(l) + li.Uo (0.6)] (213-11*9) = 6630
Btu (1670.8 kg-cal)
Heat loss = 3110 Btu (783-7 kg-cal)
It can be seen that the heat loss is based on several
assumptions. The actual loss can be either higher or
lower.
128
-------
k. Fractionation column condenser.
No flow quantity of the vent is available. Heat given up
by the methanol = 639 (263) (1.8) = 302,000 Btu (76,10k
kg-cal).
B. Methanol recovery
Assume same operating conditions on the fractionation column.
Enthalpy balance around the fractionation column:
1. Reboiler
Assume 30.8 Ib (13.98 kg) of reboiling. Heat required to
heat the liquid (which is essentially water) from 210°F
(98.9°C) to 212°F (100°C) = 30.8 (l) (2) = 6l.6 Btu (15-5
kg-cal)
Heat of evaporation =30.8 (970.3) = 29900 Btu (753^.8 kg-
cal)
Heat required to heat vapor from 212°F (100°C) to 2^7°F
(119.U°C) = 35 (30.8) (O.U82) = 520 Btu (131 kg-cal)
Total heat required = 30,500 Btu (7686 kg-cal)
2. Condenser after fractionation column.
No vent flow quantity is available- Heat given up by
methanol = 0.639 (263) (1.8) » 302 Btu (76.1 kg-cal)
3. Heat loss — fractionation column.
Use lU9°F (65°C) as reference temperature.
Heat in = 30.8 [(0.1*82) (l) + 970.3 + (l) (63)] = 31,800
Btu.
Heat out = (308-30.8) (l) (6k) » 21,7.00 Btu (5^68.H kg-
cal).
Heat loss = 10,100 Btu (25^5-2 kg-cal).
The heat loss shown here can be either higher or lower for
similar reasons as indicated before.
III. Recovery of ethyl acetate and ethanol
A. Ethyl acetate recovery
For ethanol:
p = 0.785
b.p. - 78.5°C = 173.3°F
129
-------
This is the exit temperature of the ethahol vapor. Since the
heater is attached to the activated carbon column, this is
also the inlet ethanol vapor temperature to the column.
X = 20k cal/g
C = O.U06 cal/g/°C (O.U06 Btu/lb/°F) (vapor)
P
C = O.U56 cal/g°C (0.^56 BtU/lb/°F) (liquid)
For acetic acid
X =96.8 oal/g (ITU.2 Btu/lb)
C =1.50 cal/g/°C (1.50 Btu/lb/°F) (vapor)
C = Q.U68 cal/g/°C (O.U68 Btu/lb/°F) (liquid)
For ethyl acetate
X =102 cal/g (183.6 Btu/lb)
C = 0.371 cal/g/°C (0.371 Btu/lb./°F) (vapor)
C = 0.1*59 cal/g/°C (0.1*59 Btu/lb/°F) (liquid)
1. Heater before the activated carbon column.
Ethanol floV = ll*l6 ~±f (100) « 1*7,200 Ib (2ll*28.8 kg)
j. jj-
Assume at 70°F (21.1°C)
Heat required = U7»200 [0.>56 (1T3-TO) + 2Qh (1.8)1 =
19,500,000 Btu (^91^000 kg-cal).
2. No information for the steam consumption in the steam
jacket is available. It is believed that unless some
unusual source of high heat loss exists, the steam
requirement for this item should not be high.
3. Heat of reaction
CH3COOH(g) + C2H5OH(g) •* CH3COOC2Hs(g) + H20(g)
Assuming constant pressure and volume, the heat of re-
action is estimated as follows:
AH of CH3COOH(g) = -219.82 K cal/g-mole
C*
AHQ of C2H5OH(g) = -336.82 K cal/g-mole
AHc of CH3COOC2H5(g) = -5^7-^6 K cal/g-mole.
130
-------
These are the standard heat of combustion with reference con-
ditions of 25°C (77°F), 1 atm. pressure and gaseous substances
in ideal state.
.'. Total heat of reaction • ' • -32,100 Btu
(-816U.8 kg-cal).
Since the major final products are gaseous carbon dioxide and
liquid water, it is necessary in this case to deduct the latent
heat of vaporization of the liquid water formed.
Latent heat = -^ (l8:) (99!*) = 25', 200 , Btu -(6350-.U kg-cal)
.'.The net heat of reaction = -7200 Btu '(-l8lV.li kg-cal).
U. Fractionation column
a. Reboiler
Assume 390 Ib (177-1 kg) reboiling. Heat required to heat
the liquid which is essentially water from 210°F (98.9°C)
to 2U7°F (119.U°C) = 390 (1) (2) = 780 Btu (196.6 kg-cal).
Heat of evaporation = 390 (970.3) = 378,000 Btu (9526 kg-
cal) . .
Heat required to heat the vapor from 212°F (100°C) to 2U7°F
(119. U°C) = 390, (O.U82) (35) = 6580 Btu (1658.2 kg-cal).
Total heat required = 385,000 Btu (97020 kg-cal).
b. Heat loss around column.
Use 157°F (69.U°C) as reference temperature.
Heat in:
Heat in with feed = (17^-157) [^9000 (O.U56) + 3900 (l)
+ 12U (0.1*59)3 = WfsOOO Btu (1126UU kg-cal).
Heat in with reboiled liquid = 390 [(O.U82) (l) + 970-3 +
1 (213-157)] = UOO.OOO Btu (100800 kg-cal).
Total in = 81*7,000 Btu (2I3^kk kg-cal).
Heat out with ethanol = (168-157) U9,000) (O.U56) =
2^6,000 Btu (61992 kg-cal)'.
Heat out with bottom liquid = (3900 + 390) (213-157) (l)
= 2^0,000 Btu.(6oU80 kg-cal).
131
-------
Total heat out = 1+86,000 Btu (122^72 kg-eal).
Heat loss = 361,000 Btu (90972 kg-cal).
It should be noted again that the heat loss shown here
could be either higher or lower depending upon the operat-
ing conditions.
5. Condenser after fractionation column. No vent flow quantity
is available. Heat to be given up in condensing the ethyl
acetate = 12k (102) (1.8) = 22,800 Btu (57^5-6 kg-cal).
Summary of Heat Balances
I. Basis: To process 50,000 Ib (22700 kg) of evaporator condensate in
the pilot at the Appleton Division Mill of Consolidated
Papers, Inc.
II. Calculations are based on the actual operating conditions of the
pilot plant. It should be noted that:
A. The individual pieces of equipment are not optimized; and
B. The process may or may not be exactly suitable for the
particular evaporator condensate.
III. Heat Required
A. Separation Process
1. Preheater U,090,000 Btu (1030680 kg-cal)
2. Stripping Column 3,390,000a Btu (85^280 kg-cal)
"K
3. Fractionation Column 3,390,000 Btu (85^280 kg-cal)
B. Recovery of Furfural and Methanol
1. Furfural Recovery
a. Vaporizer 3^6,000 Btu (87192 kg-cal)
b. Fractionation Column 9,1^0 Btu (2303.3 kg-cal)
2. Methanol Recovery
a. Fractionation Column 30,500 Btu (7686 kg-cal)
^These are based on the actual steam consumptions. This indirect heat-
ing may or may not be desirable in large scale practical operation. If
live steam is used, this quantity could be reduced to 2,750,000 Btu
(693,000 kg-cal).
Similar to a. This could be reduced to 1^1,000 Btu. (35,532 ke-cal)
132
-------
C. Recovery of Ethyl Acetate and Ethanol
1% Heater 19,500,000° Btu (1I9UOOO kg-cal)
2. Fractionation Column 385,000 Btu (97020 kg-cal)
3. Heat of Reaction _Tj200 Btu (-l8lM kg-cal)
Total heat consumption = 31,100,000d Btu (7837200 kg-cal).
IV. Heat Removed
A. Separation Process
1. Stripping Column Condenser 2,880,000 Btu (725760 kg-cal)
2. First Carbon Column Cooler 3,600,000 Btu (907200 kg-cal)
3. Fractionation Column
Condenser 7,21*0 Btu (l82*K5 kg-cal)
B. Recovery of Furfural and Methanol
1. Furfural Recovery
a. Fractionation Column
Condenser 302,000 Btu (76lO^ kg-cal)
2. Methanol Recovery
a. Fractionation Column
Condenser 302 Btu (76.1 kg-cal)
C. Recovery of Ethyl Acetate and Ethanol
1. Fractionation Column
Condenser 22,800 Btu (57^5-6 kg-cal)
Total heat removed = 6,810,000 Btu (1716120 kg-cal).
COST ESTIMATE
With a "basis of calculation of 50,000 (22,700 kg) pounds of the Appleton
Division mill of Consolidated Papers, Inc. condensate, the value of the
products potentially recoverable and the economic benefit from the
°This figure is known to be too high due to inefficient pilot plant
operating conditions. According to the latest revised arrangement
and operating conditions the steam consumption for this :item should
be k,000,000 Btu. (1008000 kg-cal)
revised total heat required = 11,700,000 Btu. (29^00 kg-cal)
133
-------
removal of 200 (90.8 kg) pounds of BOD5 gave the following summary of
values to be derived from the process shown in Table 17. It should be
mentioned that no credit has been given to the value of the reusable
water or recovered sulfite liquor.
The major operating expense is steam, and with an estimated power con-
sumption and labor requirement for an automated system, the combined
expense and estimated gross net income per year are shown in the same
table.
These figures provide guidance as to which portions of the process are
most productive of byproducts and which might be simplified, modified,
or eliminated with little detriment to the overall economic performance
of the installation. Such modifications and more accurate data on costs
of operation would have to be provided by detailed full-scale design
work specific to the desires of the local mill management.
POLLUTIONAL BALANCES
To show the effect of recovering the various chemicals from the conden-
sates, upon the pollution load, the BODs values of these materials have
been calculated and assembled as a balance sheet. This is given in
Table 18. The relative BOD5 values of these materials have been taken
from tables previously published in Sewage and Industrial Wastes. Of
the 202 pounds (91-7 kg) of BOD5 admitted into the system (by analyses),
l6l pounds (73.1 kg) have been accounted for by using equivalent BODS
values of these materials.
CONCLUSIONS
Based on the analysis of the recovered materials and the operating data,
the conclusion can be drawn that the process is economically feasible.
Sizing of the equipment and more accurate economic feasibility analysis
were intended but not made because of these facts:
13U
-------
Table 17. RECOVERABLE VALUES BASED OM THE COHDENSATE
OF COHSOLIDATED PAPERS, INC.
I. Basis.: To process 50,000 (22,700 kg) pounds of the evaporator condensate of the
Appleton Division Mill, Consolidated Papers, Inc.
II. Recoverable Valuesa
1. Products
7.10 Ib (3.22 kg) of furfural, 120/lb (26.^/kg) $ 0.85
15.2 Ib (6.9 kg) of methanol, 10^/gal. C22.0tf/kg) 0.23
127.0 Ib (57.6 kg) of ethyl acetate, 9*/lb (l9.8tf/kg) 11 A3
1*3.6 Ib (19.8 kg) of S02, 0.1#/lb (0.220/kg) Q.Qk
$12.55
2. Savings on BOD 5 reduction
203 Ib (92.9 kg) BODs, M/lb (8.8^/kg) $ 8.06
Total $20.61
III. Estimated Operating Costs
1. Steam
11,700 Ib (.5311.8 kg), $1.00/1000 Ib C$2 < 20/1000 kg) ill. 70
2. Pover
Assumptions:
Capacity » 100,000 Ib 0*51*00 kg) of evaporator condensate/hour
Pumps- = 100 bp
Pover cost • Itf/kw-hr $ 0.38
t
3. Labor
Assumptions:
Number of operators * 1 operator/8-hr shift
Wage » $10,000/annum
Days operated per year » 350
Other assumptions * same as under Power cost
Labor cost $ 1.59
i
U. Ethanol
6U.7 lt> (29.* kg), 9*/» tl9.8*/kg) Uj83
Total $19-50
IV. Estimated Groas Income per Annum - t20.6l - 19.50) (2) (2*) (350) - $18,600
This net income is based on a plant capacity of 100,000 Ib (1*5^*00 kg) of condensate
per hour
credit has been given to the value of the reusable water or the recovered sulfite liquor.
135
-------
Table 18. POLLUTIONAL BALANCE SHEET
MeOH* Acetic Acid11 S02C Furfural*1 BOD.-lbdif)
Ihtkg} BODs-li(kgl Ibtkgi BODs-rlhCkgl IbCkgl BODs-lb(igl IbUigl BOD.-liCigl In Out
Into system 32.5 27.95 13^.0 100.50 1*9.0 12.25 8.0 6.16 ^'86
(1U.8) (12.69) (60.83) d5.63) (22.25) (5.56) (3.63) (2.80) (66.6?)
Recovery
Separation
S02 stream 1.50 1.29 0 0 37-5 9-38 0.9 0.69 11.36
(0.68) (0.58) (0) (0) (IT.02) (U.26) (O.Ul) (0.31) (5.16)
MeOH stream 15.0 12.9 0 0 0.2 0.05 0 0 12.95
(6.81) (5.86) (0) (0) (0.09) (0.02) (0) (0) (5.88)
Furfural stream 0.0 0.0 0 0 00 3.1 2,39 2.39
(0.0) (0.0) (0) (0) (0) (0) (l.ltl) (1.08) (1.08)
Regeneration
Furfural column 00 00 00 lt.0 3-08 3.08
(0) (0) (0) (0) f (0) (0) (1.82) (l.UO) (1.1*0)
Acetic acid column 0 0 123.7 63.25 0 0 00 63.25
(0) (0) (56.15) (28.72) (0) (0) (0) (0) (28.71)
93.03
(1)2.210
Re-usable water stream 16.0 13-76 0 0 2.0 0.; 0 0 lU.26
(7.26) (6.25) (0) (0) (0.91) (0.23) (0) (0) (6.1)7
11*
Waste material to be burned
From separation 0 0 5.60 It.20 0000
(0) (0) (2.5H) (1.91) (0) (0) (0) (0)
Unaccounted acetic 0 0 11.50 8.55 0 0 0 0
acid (0) (0) (5.22) (3.88) (0) (0) (0) (0)
From furfural regen- 00 00 0000
eration (0) (0) (0) (0) (0) (0) (0) (0)
From acetic acid 0 0 32.6 2l*.I*5 0000
regeneration (0) (0) (lU.80) (11.10) (0) (0) (0) (0)
Total 11*6,86 lW».!)9
(66.67) (65.60)
Total BODs into system (by analyses) based on mean data of Report Tvo, page 7, 202.3 lb (91.81) kg)
Total BODs into system based on calculated data, 1U6.86 lb (66.67 kg)
BODs in re-usable -water, li.26 lb (6.U7 kg)
BODs removal —based on 202.3 lb (91.8U kg) into system, 93.Ot
BODs removal - baaed on ll)6.86 lb (66.67 kg) into system, 90.3?
Difference in calculated and analyzed BODj into system largely due to carbohydrates which vould be burned with waste
material.
B0.86 lb BOD5/lb MeOH - Sewage Ind. Wastes 27(9)slOltl) (Sept., 1955);
*O.T5 lb BODs/lb acetic acid - Sewage Ind. Wastes 30(5): 677 (May, 1958).
C0.25 lb BODs/lb SOj - not a true BODs but included as BOD5.
0.77 lb BODs/lb furfural - Sewage Ind. Wastes 27(9):10U6 (Sept., 1955).
eAs ethyl acetate.
Calculated from equivalent acetic acid.
136
-------
The general process for which pilot data were available was not
specifically designed for processing the evaporator condensate at
the Appleton Division Mill of Consolidated Papers, Inc.;
The individual pieces of equipment were mostly not optimized;
The chemical constituents and physical conditions of the evaporator
condensates from different effects of the evaporating systems at
different mills would be different; and
The flow quantities would vary from case to case.
It is felt that to avoid misleading the reader, the design of full-scale
equipment, calculation of energy balance, cost estimation and the com-
plete economic analysis should be done for the individual mills. Based
upon the mass and heat balances done, the following points have caught
our attention:
1. The use of direct steam instead of a reboiler for the strip-
ping column would save some heat and capital expense.
2. More saving if direct steam is used for the fractipnation
column.
3. The elimination of the condenser after the first activated
carbon column in the furfural recovery process would be
logical from both capital arid operating cost standpoints.
U. The elimination of the condenser after the second activated
carbon column would be reasonable. A vapor feed to the
fractionation column could work out well.
137
-------
SECTION VII
LOW TEMPERATURE REGENERATION OF CARBON
METHODOLOGY
This phase of the project has to do with an attempt to scale up prior
laboratory development studies for low temperature [(200°C) (392°F)]
regeneration of carbon adsorption columns. Previous projects performed
at The Institute of Paper Chemistry in which removal of BOD5 and COD
contributing materials from the condensates derived from the evapora-
tion of spent sulfite liquors were studied utilized 2-inch (5-08 cm)
glass and stainless steel columns. Successful regeneration of this
carbon over 32 cycles by applying heat to the column by means of an
electric tape and superheated steam lead to its inclusion in this
project. It was obvious that another system for heating of the larger
column would have to be founds and without the introduction of addition-
al water which would result in diluting the concentrated materials.
Early in 1971 it had been learned that Messrs. Bela M. Fabuss and
Wilson C. Dubois of Lowell Technological Institute had applied for a
patent entitled "Apparatus and Process for Desorption of Filter Beds by
Electric Current." Basically, the process consists of utilizing the
carbon bed itself as an electrical resistance heater, permitting a
rapid temperature rise without any indirect heat transfer. Contacts
with Professor Fabuss eventually lead to an agreement with Lowell
Technological Institute to construct an activated carbon column incorp-
orating this means of heat regeneration. Operation of this unit is the
subject of this phase of the project.
The unit as constructed by Lowell consisted of a one-foot (30.5 cm)
diameter stainless steel column with a total height of 5 feet (1.52 m)"
and containing k feet (1.22 m) of activated carbon. Stainless steel
grid electrodes 1/2 inch (1.27 cm) thick were spaced at 1-foot (30.U8
cm) intervals. A stainless steel screen was fastened to the underside
of the bottom electrode to support the carbon. The column was electri-
cally insulated on the interior by a teflon coating and the electrodes
138
-------
were supported on glass laminated strips. Since the Institute had
several direct current power supply units, previously used on an electro-
dialysis project, it had been decided, with agreement from,Professor
Fabus, to utilize this source of power- Prior to shipment of the unit
to the Institute, it was tested by Lowell personnel, using a direct cur-
rent rectifier, supplied by the Institute,
A diagrammatrical sketch of the unit as it was initially set up and
connected at the Appleton Division mill of Consolidated Papers, Inc. is
shown in Fig. 51. All materials coming in contact with the condensate
were constructed of 316 stainless steel. Four thermocouples were in-
stalled midway between each electrode which were Initially inserted 6
inches (15.2 cm) or to the center of the column. The, connectors to the
electrodes, as well as the thermocouples, were insulated from the shell
of the column by means of custom-made ceramic insulators. The two con-
densers used in series consisted of a single tube.
During the absorption phase "the condensate was passed through the con-
densers and entered the column at the top and exited through the bottom.
The column was maintained in a flooded condition by means of a constant
level device. During regeneration the vapors were removed from the top
of the column through the condensers, except during several trials when
the flow was reversed.
As noted in Fig. 51, the three negative electrodes were common to both
power sources, with one positive electrode being connected to each
power unit. Initially, the column was charged with Filtrasorb 300
activated carbon which had been supplied by Lowell Technological Insti-
tute and had been used during their trial period. This was later re-
placed with Filtrasorb UOO carbon. The unit was never operated success-
fully and most of the available time was spent operating repeated dry-
ing and wetting cycles using tap water. Two runs using condensate were
made and the feed for these runs consisted of condensate that had passed
through the stripper and also through the furfural column of Mr.
139
-------
Sel-Bex
Unit 1
Sel-Rex
Unit 2
Water Out
ndenser
Vent
IT
londenser
In
Input During Adsorption
Outlet During Regeneration
_
• Open During Adsorption
Closed During Regeneration
Discharge During
Adsorption
Figure 51. Activated carbon thermal regeneration column
-------
Baierls' system, so that basically most of the S02, methanol, and fur-
fural had been removed.
DATA AND DISCUSSION
To simplify the explanation of the data and operation of the column, the
procedures will be given in chronological order. Figure 52 is a diagram
which labels all of the outlets through the side of the column. All
changes made in the electrode hookup and in the location of the thermo-
couples as the work progressed will be identified by the letters and
figures in this sketch. The letters A, B, C, D, and E were the normal
outlets for the connectors to the electrodes, while the 1, 2, 3, 1*
positions were the original locations of the thermocouples.
The data to be presented will represent only the significant operation-
al changes. Table 19 contains the operating conditions and the temp-
eratures obtained during the operation from January k through January 9,
1973. During this time the unit was operated as originally proposed by
Lowell Technological Institute and as indicated in Fig. 51; that is,
the common cathodes were A, C, and E, while the B anode was connected
to Rectifier No. 1 and the D anode to Rectifier No. 2. Prior to acti-
vating Rectifier No. 1, the carbon was wet with tap water and drained.
Initially, the voltage was set at 75, which produced 55 amps, for a
resistance of 0.68 ohms per foot (2.23 ohm/m) of carbon. Within 15
minutes the resistance had increased to 0.89 ohms per foot (2.92 ohm/m)
of carbon. After an additional 10 minutes the resistance had increased
to 2.02 ohms per foot (6.63 ohm/m), and finally after a total of 35
minutes of operation the resistance was at 28.55 ohms. However, the
temperature had risen to 95°C (203°F) after 15 minutes, which is prob-
ably at the boiling point of water, since it was later found that the
thermocouples were reading low. The second rectifier, which was to heat
the bottom 2 foot (6l.O cm) of carbon, was activated 15 minutes after
Rectifier No. 1 and approximately boiling temperatures were reached in
10 minutes.
-------
AC-
1C
BC-
2C
c c-
3C
DC
EC
3-D'
3B'
Figure 52. Location of electrode and thermocouple terminals
1U2
-------
fable 19. LOWELL UHIT - DIRECT CURRENT
Water Evaporation
Filtrasorb 300 Carbon
Carbon Bed 1 ft x U ft (30.5 cm x 1.219 m)
Date
l-U-73
1-5-73
1-8-73
1-9-73
Time
10:07
10:22
10:32
10:U3
11:00
11:20
11:1*0
12:00
1:10
2:15
Rectifier Mo. 1 Rectifier No. 2
Volts Amps Volts Amps
Temperature. °C (°F)
Power connection — cathodes — A, C, & E
anodes — B — Rectifier No. 1
-D -Rectifier No. 2
Wet column
75
98
113
111*
115
115
115
111*
111*
55
55
28
2
55
50
3
33
8
95
9:15
9:50
10:20
10:UO
11:10
11:30
11:50
12:25
1:25
2:05
3:10
3:30
50
50
75
75
100
100
125
50
50
50
100
100
8
7
5
6
9
16
9
7
6
7
16
5
0
0
20
97
0
0
25
U8
Re-wet carbon
0 0
6k 77
98 5
Re-wet carbon
0 0
0 0
Re-wet carbon
20( 68)
95(203)
30( 86)
9**(201)
95(203) 95(203)
22( 72)
95(203)
29( 81*)
95(203)
15( 59) 20( 68) 16( 6l) 1*5(113)
92(198) 90(19^) 28( 82) 58(136)
95(203) 95(203) 95(203) 92(198)
61*)
6U)
26( 79)
25( 77)
33( 91)
33( 91)
6o(ll*0)
6o(ll*0)
115 10 9U(201) 9M201) 8U(l83) 8l*(l83)
50 27 7M165) 7M165) 7»*(l65) 7Ml65)
50 15 87(189) 93(199) 93(199) 93 199
75 2 87(189) 9l*(20l) 9>*(20l) 9^(201)
75
1 s
100
At \f\f
100
116
50
50
50
0
0
39
•-t y
30
17
9
11
8
7
0
0
27( 81)
29( 81*)
9**(201)
93(199)
9^(201)
93(199)
93(199)
87(189)
88(190)
27( 81)
91(196)
9!* (201)
93(199)
9>*(201)
93(199)
91(196)
85(185)
83(181)
27( 81)
89(192)
9l* (201)
93(199)
9l* (201)
93(199)
9U(201)
91(196)
92(198)
30( 86)
50(122)
50(122)
7l*(l65)
76(169)
75(167)
75(167)
72(162)
71(160)
Attempted to dry carbon with stream of air overnight
9-10 112 6
Low amperage (high resistance), removed carbon for
inspection of the unit
-------
The carbon was rewet several times more and following each wetting good
amperage was obtained for a short period, after which there would be a
sudden increase in resistance. Following the third rewetting, however,
at 2:15 p.m. the resistance of No. 1 rectifier was at U.75 ohms and on
No. 2 rectifier at 2.88 ohms and the resistance remained high during
the trials of January 5, 1973 and January 8, 1973. On January 9, since
the resistance continued to be high, it was decided to remove the carbon
for inspection of the unit.
This inspection showed that the electrodes which had been in the posi-
tive position were completely covered with a scale or coked carbon
material, and that the teflon coating on the interior of the column was
blistered extensively. This even extended to the bottom plate. Analysis
of the corrosion, which was scraped from the anode electrodes, revealed
that on a dried basis, it contained 80.k% ash, 7-8$ iron, 9-3$ chro-
mium and 1.8$ calcium. Thus, it became apparent that the electrodes
were literally being disintegrated by electrolysis. It was realized
that electrolytic action would result in disintegration when in contact
with liquid media, but it was not certain that this would happen in a
vapor atmosphere.
The electrodes were cleaned by sand blasting and reinstalled in the
column. Initially, one foot (30.5 cm) of carbon was packed into the
column and two electrodes connected to Rectifier No. 2; the anode being
in the E position and the cathode in the D position. The carbon used
was the same that had been removed previously from the column and which
had been air dried. Table 20 contains the data obtained. Initially,
the resistance was 1.5 ohms per foot (U.92 ohm/m) of carbon but this
increased to 13.5 in 15 minutes. During this time a stream of air was
fed through the column. Surface temperatures were then determined
using a surface pyrometer. The temperatures were very consistent
around the periphery of the column. Current was applied for an addi-
tional hour, with continued low amperage, after which surface temper-
atures were again taken. Temperatures were highest in the vicinity of
-------
Table 20. LOWELL UNIT - DIRECT CURRENT
Water Evaporation
Filtrasorb 300 Carbon
Carbon Bed 1 ft x 1, ft (30.5 cm x 1>219 m)
Date
1-22-73
Rectifier No. 2
Volts Amps
Temperature,
°C
Power connection — anode — E
cathode — D
One foot (O.SOW m) of air-dried carbon in column
8:58
9:08
9:13
Location
At D
Between D & E
At E
Below E
27
27
27
18
12
2
26( 79)
35( 95)
37( 99)
Right side
26.1(79)
27.8(82)
27.8(82)
26.7(80)
Surface temperature. °C C°F)
Front
25.6(78)
27.8(82)
2l».l*(76)
Left side
25.0(77)
27.2(81)
27.2(81)
Air
on
on
on
Back
25.0(77)
26.1(79)
27.8(82)
Date
1-22-73
Time
9:33
9:1*3
9:53
10:03
10:30
Location
Between D
At D
Between E &
At E
& C
Rectifier Ho. 2
Volts Amps
50
50
50
50
50
16
13
9
5
7
Temperature,
°C (°F)
37( 99)
5K129)
57(135)
59(138)
35( 95)
Surface temperature. °C C°F)
Right side
1*2.0(107)
1*2.5(108)
36.TC 98)
30.0( 86)
Front
36.7( 98)
1*6.8(116)
35.6( 96)
28.3( 83)
Left side
35.6C 96)
1*1.1(106)
36.i( 97)
28.3( 83)
Air (%)
50
50
25
0
50
Date
1-22-73
Rectifier Ho. 1 Rectifier No. 2 Temperature, °C (°F)
Time Volts Amps Volts Amps 2 3 T"
10:1*6
10:56
0
0
0
0
50
50
3M 93)
27 ( 81)
Power connection — cathode — D
anode - C - Rectifier No. 1
- E - Rectifier No. 2
Added second foot of air-dried carbon
1:22
1:22
50
50
23
23
Air
CO
50
50
50
50
Location
At E
Between D & E
At D
Between C & D
At C
Between B & C
At B
50 18 27< 81) 1*3(109) 2l( 70)
50 32 1*8(118) 88(190) 38(100)
Surface temperature. °C
Right side
31.K 88)
33-9( 93)
1*1*.5(112)
39-0(102)
39-5(103)
33.3( 92)
25-6( 76)
Front
1*0.0(101*)
1*2.2(108)
1(2.2(108)
1*1.9(107)
36.7( 98)
35-6( 96)
2l*.l*( 76)
Left side
26.7f 80)
31.K 88)
38.9(102)
35.6( 9!*)
30.6( 87)
30.6( 87)
25.6( 78)
Back
33.9( 93)
33.9( 93)
1*7.8(118)
1*5.0(113)
1*5.0(113)
33.9( 93)
25.6( 76)
Removed carbon from column
-------
the D electrode. The current was again turned on for an additional 10
minutes, and then shut off and a second foot (30.5 cm) of air-dried
carbon added to the column. Power connections were rearranged so that
the cathode from both rectifiers was in the D position, with the anode
for Rectifier No. 1 being in the C position and for Rectifier No. 2 in
the E position. The current was activated for about 10 minutes at about
50 volts and 23 amps, after which surface temperatures were again taken.
The temperatures again were highest around the area of the D electrode
as indicated in Table 20, The carbon was then removed from the column.
The column was refilled with Filtrasorb UOO virgin carbon. It was
placed in the column in an "as received" condition without previous
wetting of the carbon or flooding of the column. Power connections
were made so that the A, C, and E positions were anodes and the B
position was the. cathode for Rectifier No. 1 and the D position was the
cathode for Rectifier No. 2. By placing an anode in the A position, it
became possible to observe this electrode by removing several inches of
carbon. The heating data obtained under these conditions is given in
Table 21. Power was on both units for a period of about 35 minutes;
voltages of both 50 and 25 were applied with the resistances ranging
from 0.18 to 0.25 ohms per foot (0.59-0,82 ohm/m) of carbon in the top
2 feet (6l.O cm) of the column and 0.31 to 0.38 ohms per foot (1.02-r
1.25 ohm/m) in the bottom 2 feet (6l.O em). The temperature at No. 1
thermocouple increased from 33 to 193°C (91 to 379°F) during this
period, while at No. U thermocouple there was no increase. The surface
temperatures were again taken and reflected the same heating pattern at
the top of the column with very little heating at the bottom. No real
hot spots, however, developed. After operating Rectifier No. 2 for 10
minutes, the power connections were changed so that only the bottom
foot (30.5 cm) of carbon was being heated. This did result in an in-
crease in temperature at the No. h position from 60°C to 210°C (lUo°F
to laO°F). Surface temperatures were fairly even except for a possible
hot spot on the left side of the column at the D electrode.
-------
Table 21. LOWELL UNIT - DIRECT CURRENT
Water Evaporation
Filtrasorb 1*00 Carbon
Carbon Bed 1 ft x l» ft (30.5 cm x 1.219 m)
Date
Rectifier No. 1
Time Volts Amps
Rectifier Mo. 2
Volts Amps
Temperature. °C (°F)
Power connection — anodes - A, C, & E
cathode — B — Rectifier No. 1
- D - Rectifier No. 2
1-22-73
2:1*0
2:U3
O • Ji Ji
3:05
3:ll*
Date
1-22-73
Time
3:21
3:31
3:35
1*:20
kt33
50
50
25
25
25
Location
At A
Between A
At B
Between B
At C
Between C
At D
Between D
At E
93
122
50
53
69
& B
& C
& D
& E
Rectifier No.
Volts
0
0
Amps
0
0
50
50
25
25
25
Right side
66.8(152)
66.8(152)
66.1(151)
65.0(11*9)
66.8(152)
61.1(11*2)
56.1(133)
1*7.2(117)
3U.U( 9k)
65 33( 91) 70(158) 81*(l83) U7(ll7)
72 95(203) 87(189) 88(190) 1*7(117)
33 88(190) 87(189) 86(187) 1*7(117)
37 123(253) 107(225) 96(205) I*0(l0l*^
1*0 193(379) 11*2(288) 116(21*1) 1*7(117)
Surface temperature, °C (°F)
Front Left side Back
63.9(11*7) 63.5(11*6) 61.8(11*3)
58.9(138) 58.5(137) 67.9(15!*)
58.5(137) 6l.7(ll*3) 67.5(153)
6l.!*(ll*2) 63.5(11*6) 61.7(11*3)
51*. 6(130) 65.0(11*9) 62.1(11*!*)
55.0(131) 58.5(137) 53.5(128)
1*7.8(118) 53-5(128) Ul*. 5(112)
1*1,8(107) 1*1.8(107) 39.5(103)
31. 1( 88) 33. 3( 92) 31. l( 88)
1 Rectifier No. 2 Temperature, °C (°F)
Volts
25
25
Power connection
o
0
^/
0
0
0
o
0
0
Location
At C
Between C
At D
Between D
At E
& D
& E
25
50
75
75
Right side
90.0(19!*)
90. 0(191* )
93.5(200)
88.9(192)
85.0(185)
Amps 1 2 3 »*
37 213(1*15) 150(302) 126(259) 33( 91)
38 2ll*(l*l7) 11*8(298) 157(315) 57(135)
— D-E only
21 — — — 60(l!*0)
1*1 213(1*15) 11*7(297) 165(329) 92(198)
1*6 197(387) 15M309) H7(2l*3) 102(216)
61 — -- — 210(1*10)
Surface temperature, °C (°F)
Front Left, side Back
92.3(198) 88.M191) 91.8(197)
89.6(193) 91.8(197) 88.!+(l9l)
89.6(193) 103.3(218) 88.9(192)
88.1*(191) 89.6(193) 88.9(192)
78.1*(173) 75-6(168) 85.6(186)
1U7
-------
Heating was continued as shown in Table 22 on the -following morning
with the power connections the same as previously used. After 10
minutes of operation, during which time there was fairly rapid heating,
power was shut off and the column was wet with warm tap water. To
facilitate drainage, air was passed through the column for one-half
hour. Heating was continued on No. 1 rectifier from 10:15 to 3:55» and
during this time the combination of voltages and amperages used repre*-
sented an electrical resistance of 0.06 to 0.27 ohms per foot (0.20-
0.88 ohm/m) of carbon. At the last reading at 3:55, the resistance
dropped to O.l8 ohms per foot (0.59 ohm/m). Power was on the No. 2
rectifier from 10:15 to 2:35, at which time the temperature at the No.
3 position suddenly increased from 9^°C to 2U5°C (201°F to U73°F).
Electrical resistances during this time were between O.lH to 0.28 ohms
per foot (O.U6-0.92 ohm/m) of carbon. Twenty-six liters (6.87 gal.)
of water had been vaporized from the column during the 5 3A hours of
operation.
The following morning on January 2U, both power supplies were activated
before wetting the carbon. No. 1 rectifier was then producing 32 volts
at 50 amps, while No. 2 rectifier produced ^7 volts at 50 amps. The
column was flooded and drained with the aid of air pressure, during
which time 17 liters (U.U9 gal.) of water were drained from the column
which is the amount of non-adsorbed water in the column when it is in a
flooded state. Current was placed on No. 1 rectifier for 20 minutes
prior to turning on the No. 2 rectifier. Initially, the resistance
was very low, there' being 13 volts on Rectifier No. 1 producing 75 amps,
while on No. 2 rectifier there were 100 amps and 37 volts. However,
shortly after the No. 2 rectifier was energized, a sudden drop in
amperage was noted and all power was disconnected. The No. 2 recti-
fier was then connected to the two bottom electrodes between the bottom
foot (30.5 cm) of carbon, the anode being in the D position and the
cathode in the E position. This configuration provided 53 amps at 52
volts. Connections were then changed to the anode in the C position
and the cathode in the E position on Rectifier No. 2 which included 2
-------
fable 22. LOWELL UHIT - DIRECT CURRENT
Water Evaporation
Filtrasorb ItOO Carbon
Carbon Bed 1 ft x I, rt (30.5 cm x 1.219 m)
Date
1-23-73
Rectifier Hq_. 1 Rectifier So. 2
Tiae Volts Amps "~Volts Amps""
Temperature, °c
Power connection - anodes -A, C, it E
cathodes - B — Rectifier Ho. 1
— D — Rectifier So. 2
8:U3
8: US
8:53
10:15
11:05
11:15
11:35
12:35
1:15
1:35
2:05
2:35
2:1*5
3:05
3:25
3:55
25
25
0
6
11
11
11
11
18
18
18
18
18
21*
50
50
50
60
0
Wet carbon
50
83
82
11
66
101
107
102
76
66
68
91
137
25
25
25
with
0
0
16
22
22
27
1*2
53
58
0
0
0
0
25
27
27
33( 91) 33( 91) 36( 97) 35( 95)
70(158) 57(135) 50(122) 5M129)
., 75(167) 6o(ll*0) 58(136) 66(151)
warm tap water
0
0
50
76
56
57
77
93
130
0
0
0
0
17( 63) 31( 88) I»l(lo6) l*l*(lll)
56(133) 66(151) !*1(106) Mi(iil)
66(151) 76(169) 1*7(117) 50(122)
83(181) 89(192) 70(158) 70(158)
79(17'*) 92(198) 91(196) 91(196)
92(198) 92(198) 91(196) 90(l9li)
91(196) 91(196) 91(196) 90(19!*)
92(198) 92(198) -9>*(201) 93(199)
90(191*) 86(187) 21*5(1*73) lHo(281*)
90(19U) 88(190) 250+(!*82) 212(Ull*)
92(198) 92(198) 250t(l*82) 235(1*55)
123(353) 103(217) 250t(UB2) 21*3(1*69)
112(231*) l61*(327) 250+(1*82) 233(1*51)
1-2U-73 8:58
9:50
10:10
10:36
26 (.6.87 gal) Liters vaporized from column
32 50 1*7 50 37( 99) 1*7(112) UU(lll) 52(126)
Wet carton with tap water
13
17
17
75
100
92
0
0
37
0
0
100
21( 70)
1*5(113)
71(160)
U3(109)
68(151*)
92(198)
U8(ll8) 51(121*)
50(122) 51(12U)
92(198) 6o(lUo)
Sudden drop In anperage on Ho. 2 rectifier
Power connection — anode — D
cathode — S — Rectifier Ho. 2
11:07
52
53 —
Power connection — anode — C
cathode — E — Rectifier So. S.
95
Immediate drop in amperage
Power connection — anode — D
cathode - E - Rectifier Bo. 2
111 111
53
53
10
,u ™ —— —
6 89(192) 90(19!*) 9H196) 91(196)
Power connection — anode — B
cathode — D - Rectifier Ho. 2
52
12
Po«er connection — anode — B
cathode — D — Rectifier Ho, 1
1111*1*
62
35
Wet carbon with tap water
Power connection •<• sane as on 1-23-73 above
2:55
3:U5
3:55
U:15
U:25
17
17
17
17
18
75 00 1*7(1121 1*7(112) 80(176) 86(187)
75 00 91(196) 91(196)' 75(167) 80(176)
75 32 50 91(196) 91(196) 760.69) 80(176)
75 32 32 91(196) 9l(l96) 91(196) 91(196)
75 00 91U96) 91(196) 90(191*) 89(192)
%-Lb pressure applied to colunn -momentary increase in amperage on Rectifier Ho. 2
It:li0 18 75 0
10 Lb/in* (0.7 kg/cm2) pressure applied to column -momentary increase in amperage on
Rectifier So. 2
-------
feet (6l.O cm) of carbon. Amperage started at 95 and immediately drop-
ped to a very low level. Again the connections were changed so that
only the bottom foot (30.5 cm) of carbon was involved, with the anode
in the D position and the cathode in the E position. With this con-
figuration at ll:lU a.m. the readings were 53 volts and 10 amps and by
11:21 the amperage had dropped to 6. The rectifier was then connected
between the middle 2 feet (6l.O cm) of carbon, with the anode in the B
position and the cathode in the D position. The power readings were
then 52 volts and 12 amps. With the same electrode connections, but to
the No. 1 rectifier instead of No. 2 rectifier, the readings were 62
volts and 35 amps.
These changes indicated that there may have been a separation of the
carbon at some point in the column, so again the carbon was wet with
tap water. All power connections were then replaced as previously,
with the anodes in the A, C, and E position and the cathode for Recti-
fier No. 1 in the B position and for Rectifier No. 2 in the D position.
While in the wet condition, resistance remained low. Prior to dis-
continuing operation for the day, pressure was exerted on the column,
first at 3A pounds per square inch (0.052 kg/cm2) and then at 10
pounds per square inch (O.?0 kg/cm2), by closing off all outlet vents.
Each time there was a momentary increase in amperage on Rectifier No. 2.
Operational data for the next day's operation, January 25, are given
in Table 23. Power connections were left the same as the previous day.
%
The carbon was again wet with tap water and power was applied for one
hour. During that time there was a rapid heating to near the boiling
point with good conductivity while the carbon was wet. The vapor flow
was then reversed to top to bottom on the assumption that this might
produce some internal pressures that would tend to pack the carbon
rather than loosen it. The amperages on No. 1 rectifier dropped con-
siderably and the application of 3 pounds per square inch (0.21 kg/cm2)
of air pressure did not restore the amperage.
150
-------
Table 23. LOWELL UNIT - DIRECT CURRENT
Water Evaporation
Filtrasorb hOO Carbon
Carlxm Bed 1 ft x U ft (30.5 cm x 1.219 m)
Rectifier No. 1 Rectifier No'. 2 Temperature. °C (°F)
Date Time Volts Amps Volts Amps I 2 —3 II
Pover connection — anodes — A, C, & E
cathodes — B — Rectifier No. 1
- D - Rectifier No. 2
Wet carbon with tap water
1-25-T3 9:02 22 jk 32 50 22( 72) 1*3(109) 50(122) 53(122)
9:20 19 75 13 25 38(100) 50(133) 69(156) 68(15*0
9:50 19 70 12 25 85(185) 91(196) 86(187) 81(178)
10:00 33 15 28 1*2 91(196) 91(196) 91(196) 88(190)
Reversed vapor flow from top to bottom
10:10 3k 12 28 18 91(196) 91(196) 91(196) 91(196)
10-Lb/in2 (0.7 kg/cm2) air pressure on column —momentary increase in amperage
Wet carbon with tap water
10:^7 32 75 00 62(ll*>*) 70(158) 67(153) 61*'(l>+7)
11:10 31* 12 kk 75 76(169) 90(l910 91(196) 91(196)
3-Lb/in2 (0.21 kg/cm2) air pressure on column —no increase in amperage
Removed carbon and cleaned electrodes
Refilled column with 1* ft- (1.219 m) of Filtrasorb 1*00 carbon
Wet carbon
1-30-73 8:17 31 50 27 50 36( 97) 38(100) 37( 99) 36( 97)
8:35 75 6 68 6 57(135) 57(135) 59(138) 78(172)
Shell of column hottest at No. C electrode
Isolated C electrode — no improvement in top foot
Ran steam through carbon to dispel air
9-20 1*5 3l+ 36 35 93(199) 93(199) 93(199) 9^(201)
l':30 l\ 27 36 32 93(199) 92(198) 92(198) 92(198)
Drop off in amperage
Tamped top foot (0.305 m) of carbon
27 76 36 20
Removed carbon from column
-------
The carbon and electrodes vere again removed from the column and again
the positive or anode electrodes were found to be corroded. The elec-
trodes were again cleaned by sand blasting, replaced in the column and
the column refilled with Filtrasorb UOO carbon. The column was flooded
with tap water and allowed to drain overnight.
The electrodes were connected as previously and current turned on at
8:17 a.m. on January 30. The wet carbon showed low resistance, there
being 31 volts and 50 amps on the Wo. 1 rectifier and 27 volts and 50
amps on the No. 2 rectifier, but within 20 minutes the amperage had
dropped to 6 on both rectifiers. It was found that the surface temper-
ature was hottest at the C electrode. This electrode was then discon-
nected from the system but there was no real improvement. Steam was then
passed through the carbon to dispel any air, but again there was a rel-
atively rapid drop off in amperage. Amperage on the No. 1 rectifier,
however, did increase upon tamping the top one foot (30.5 can) of carbon.
The power was then shut off and the carbon removed from the column. It
was decided that it was useless to continue with direct current using
stainless steel anodes. It would be necessary to use platinized elec-
trodes as anodes in order to successfully use direct current.
In preparation for using alternating current on the unit, the electrodes
were again cleaned. The work using direct current had indicated that
proper carbon packing was very important. This was almost impossible
to do with the electrode supports as originally designed, since all
five electrodes had to be inserted into the column together and tamping
through the electrode slots was next to impossible. Consequently, the
glass laminate electrode supports were cut in two places so that no more
than two electrodes had to be inserted at one time. Using air-dried
carbon, the bottom foot (30.5 cm) of the column was filled and packed
with a small wood ram rod. Following the packing of the carbon, the
leads from one of the direct current rectifiers were connected to the D
and E electrodes. This was done in order to obtain the voltage and
amperage readings for calculation of the resistance. After packing the
bottom foot (30.5 cm) > the voltage was 8 and the amperage 25, for a
152
-------
resistance of 0.32 ohm per foot (1.05 ohm/m) of carbon. Packing on the
second foot (30.5 cm) of carbon in a like manner and connecting the
electrodes in the C and D position produced the same resistance. After
packing the third foot (30.5 cm) of carbon, the resistance was 0.68 ohm,
and since this was double the resistance in the first two feet (6l.O cm),
packing was continued until the resistance was at 0.28 ohm. The top
foot (0.30U8 m) of carbon after packing had a resistance of 0.28 ohm
(0.92 ohm/m).
It had also been necessary to make other repairs and changes to the
column. Shortly after operation of the column was initiated, inspection
had shown that a considerable portion of the teflon coating had blis-
tered. These blisters were not immediately broken and the column was
used in that condition; however, when they did break, repairs were
made with high-temperature electrical insulation tape. The ceramic
electrode and thermocouple insert insulators also gradually cracked
and failed and they were replaced with a machined insulator constructed
from nylon. While nylon has a rather low melting point, these insu-
lators were easily fabricated and acted as a safety valve if localized
hot.spots developed.
A UUO-volt transformer with a step output between 110 and 220 volts was
obtained for an alternating current power source- A carbon disk rheo-
stat was also placed in the line in order to obtain voltages below 110
when desired. Having only a single power source, however, did provide
disadvantages in that operation flexibility was considerably handi-
capped. It was thus necessary to pass the current through U feet
(1.219 m) of carbon with a single unit rather than through 2 feet
(61.0 cm) each with two units, as done when using direct current.
The data for the first operation using alternating current are given
in Table 2*K
Power connections were made between A and E and the other three elec-
trodes were allowed to "float" in the system. Power was on the column
153
-------
Table 2k. LOWELL UNIT — ALTERHATIHO CUREEHT
Water Evaporation
Filtrasorb UOO Carbon
Carbon Bed 1 ft x k ft (30.5 cm x 1.219 m)
Temperature, °C
Date Time Volts Amps
2-5-73 1:27 109 56
1:1*7 109 61*
1:57 109 52
Location
At A
Between
At B
Between
At C
Between
At D
Between
At E
A &
B &
C &
D &
(°F)
12 3
Power connection — A to E
Wet carbon
33( 91) 30( 86} 32( 90)
79(171*) 67(153) 80(176)
85(185) 8U(183) 83(l8l)
Surface temperature
Right side
57.2(135)
B 51.1(121*)
!,7.8(ll8)
C 53.3(128)
86.7(188)
D 86.1(187)
69.1* (157)
E 7U.M166)
53.3(128)
Front
1,6
50
.1*7
50
' 83
83
63
67
58
.7(116)
.0(122)
.8(118)
.0(122)
.9(183)
.3(182)
.3(11*6)
.2(153)
-9(138.)
Temperature, °C
Date Time Volts Amps
2-5-73 2:17 109
2:53 121*
3:21 152
1*7
1,8
73
1
2
82(180} 78(172)
88(190) 88(190)
107(2251 88(190)
Elec . resistance
It
32( 90)
82(180)
86(187)
, °C (°F)
Left side
55
53
52
51*
88
81,
78
83
56
(°F)
3
81(178)
87(189)
96(205)
Surface temperature
Location
At A
Between
At B
Between
At C
Between
At D
Between
At E
A &
B &
C &
D &
Bight side
81.1(178)
B 79,1*(175)
88.9(192)
c 90.0(19!*)
91.1(196)
D 90.0(19!*)
91.1(196}
E 92.2(198)
66.7(152)
Front
81
83
86
90
91
89
91
90
77
.1(178)
.9(183)
.7(188)
. 0(191* )
.1(196)
.1*(193)
.1(196)
.6(195)
.8(172)
Temperature, °C
.6(132)
.3(128)
.8(127)
.1*(130)
.9(192)
.U(202)
.3(173)
.9(183)
.1(133)
ohms/ft ohm/metei1
0.1,2 1.38
0.1,2 1.38
0.52 1.71
Back
55.6(132)
62.2(11,1,)
57.8(136)
58.9(138)
85.0(185)
83.9(183)
75.0(167)
80.6(177)
58.9(138)
Elec. resistance
k
82(180)
81, (183)
88(190)
, °C (°F)
Left side
83
87
85
.3(182)
.5! 190)
.6(186) -
89.1* (193)
90
91
91
89
6k
(°F)
.6(195)
.1(196}
.1(196)
.1*(193)
.lt(l!*8)
ohms/ft ohm/meter
0.58
0.61,
0.52
Back
83.9(183)
88.9(192)
88.9(192)
88.9(192)
86.1(187)
89.9(19!*)
81.1(178)
89.1* (193)
66.7(152)
1.90
2.10
1-71
Elec . resistance
3
Power connection — A to C & E to C
ohms/ft ohm/meter
2-5-73 U:00 109 180 92(210] 87(189! 95(203) 92(198) 0.15
0.1*9
109.
Readings E to C only
1*0 127(26l]L 96(2051 99(210) 99(210) 1.36
Power connection B to E"
2-6-73
1*|23
1*133
173
173
61
61
107(225}
Wet Column
10151
111 1,0
111 50
12-.20
12; UO
12:58
1:20
113
156
156
158
157
156
156
156
UO
1,2
11
21
129
21*
32
l6C6l}
85QL851
81.CL83)
680.9,01
88(190}
Power
86(187)
85(1851
86(187)
97(207)
10l(2lUl
100(212}
169(336)
100(212)
108(226)
— Power connection A to I
I6(_6ll
88(190}
81* (1831
87(1891
87&8a)
17( 63)
81* (183)
83(l8l)
87(189)
26o+(500+}
18( 6k)
77(171)
82(180)
78(172)
77(171)
off for ten minutes
880.90}
85(185)
85(185)
163(325)
190(371*)
191(376)
75(167)
77U71)
78(172)
0.91*
0.91*
o.7l
093
3 5U
1.88
0.30
1.68
1.22
3.08
3.08
233
3 05
u 61
617
0.98
5-32
U.OO
Removed carbon from column
-------
for one-half hour at a voltage of 109 and amperage between 52 and 66,
which indicated an electrical resistance of 0.1*2 to 0.52 ohm per foot
(1.38-1.71 ohm/m) of carbon. Surface temperatures taken at the end of
one-half hour operation are given in Table 2k, and they indicate that
the column did not heat in a uniform manner, temperatures around the C
location were considerably higher than either the top or bottom, al-
though the internal thermocouples did not register this variation.
The unit was operated for an additional hour during which time the
temperatures rose only very slowly, except in the No. 1 position, which
registered 10?°C (225°F). During this time, electrical resistance was
between 0.52 and 0.6U ohm per foot (1.71-2.10 ohm/m) of carbon. Sur-
face temperatures were then read and, although there were not any hot
spots, the center of the column was considerably higher than either end,
with the bottom being the coolest.
An attempt was made to split the current by making connections between
A and C and E and C. With this 'configuration, the temperature at the
No. 1 thermocouple rose very rapidly to 127°C (26l°F). Because of this,
the power connections were changed to B to E to attempt to heat only the
bottom 3 feet (91.k cm) of carbon. With the transformer output set at
173 volts, there was a current flow of 6l amps, which was equivalent to
0.9U ohm resistance per foot (3.08 ohm/m). During the 10 minutes of
operation, the temperature rose at the No. 3 thermocouple location from
100°C to l69°C (212°F to 336°F).
On the morning of February 6, the column was again flooded with tap
water and drained. A stream of air was admitted to the bottom of the
column, but no attempt was made at the exact measurement of the air.
Air was used periodically through various trials, but at no time did i";
appear to have much effect, either favorably or adversely. The trans-
former was again connected over the entire U feet (1.2192 m) of carbon
at A and E electrodes. During the first hour of operation, electrical
resistance changed from 0.71 to 3-5^ ohms with a gradual increase in
the temperatures to around the 80°C (176°P) level. At the arid of 50
155
-------
more minutes of operation, the temperature at the No. 3 thermocouple
suddenly increased to 260+°C (500+°F) and the power turned off for ten
minutes. At 12:55 p.m. the electrical resistance was 0.30 ohm per foot
(0.98 ohm/m) but increased to 1.62 ohms within 15 minutes. After 1:20
p.m. the unit was shut down and the carbon removed so that the three
unused electrodes could be taken out of the system.
The A and E electrodes were cleaned with steelwool prior to reinstalling
in the column. The cleaning may not have been necessary as there, was no
evidence of the previous extensive corrosion which resulted when using
the direct current. The column was then filled with water and the car-
bon slowly added and allowed to settle without any physical packing,
excess water being drained from the bottom. The data for this operation
will be found in Table 25- The unit was operated for 1 hour and 10
minutes with the top flange removed, and during this time the tempera-
ture at the No. 1 thermocouple increased from 32 to lOO°C (90 to 212°F),
at the No. 2 thermocouple from 32 to &VC (90 to l83°F), at the No. 3
thermocouple from 3k to 50°C (93 to 122°F) and at the No. k thermocouple
from 29 to 9^°C (9^ to 201°F), The resistance was around 1.25 ohms per
foot (U.10 ohms/m) of carbon. From 1:10 p.m. until 3:05 p.m. operation
was with the top flange in place, and the electrical resistance varied
between 1.70 and k.3Q ohms per foot (5.58-lU.ll ohms/m) of carbon. It
was suspected that the increase in electrical resistatace was caused by
loose packing of the carbon, so the side of the column was tapped with
a mallet and immediately the amperage decreased to zero. Upon removal
of the top flange it was found that the carbon had settled below the
top electrode. Carbon was added and operation was started without the
top flange and with an electrical resistance of 0.62 ohm per foot (2.03
ohms/m) of carbon. All temperatures were at 100°C (212°F) and the
vapors were being removed from the top of the column directly to the
atmosphere.
The following day on February 8, column operation was continued and
the entire day was spent in repeatedly tapping the column and adding
carbon as necessary as the carbon packed following continued drying.
156
-------
Table 25. LOVfflL UNIT — ALTERHATIKG CUKSEHT
Water Evaporation
Filtrasorb 1*00 Carton
Carbon Bed 1 ft * U ft (0.30U8 x 1.291 m)
Date Time Volts Amps
Temperature. °C (°F)
Elec. resistance
ohms/ft ohm/meter
Pover connection A to E (clean electrodes ), other electrodes removed
Wet carbon — top flange off
2-7-73,
11:U5
12:1*5
12:55
1:50
2:12
2:25
3:05
110
110
110
138
138
138
138
25
22
22
20
2U
8
ll*
32 ( 90)
97(207)
100(212)
Top
10l(2lU)
99(210)
99(210)
100(212)
32( 90)
76(169)
81*(183)
flange in place
98(208)
100(212)
99(210)
100(212)
3>*( 93)
1*8(118)
50(122)
58(136)
99(210)
98(208)
100(212)
29( 81*)
9l* (201)
9l* (201)
101(211*)
100(212)
97(207)
100(212)
1
1
1
1
1
1*
2
.10
.25
.25
.80
.1*1*
.30
.1*6
3
It
It
5
1*
ll*
8
.61
.10
.10
.90
.72
.11
.07
Tapped side of column — amperage decreased to 0 - carbon settled below top electrode -
added carbon — left top flange off
3:1*0 12U 50 100(212) 100(212)
Power off
100(212) 100(212) 0.62 2.03
2-8-73
8:2U
8:30
8:UO
8:50
9:20
10:00
10:30
11:10
11:50
123
121)
125
12U
126
126
181*
126
126
29
17
27
1*1*
32
32
32
29
28
U6(115)
56(133)
Tapped column
6U (1U7)
85(181)
111(232)
Added carbon
115(239)
113(235)
127(261)
l62(32U)
1*8(118)
5U (129)
with mallet
60(11*0)
72(162)
10l(2lU)
repeatedly
102(216)
102(216)
108(226)
113(235)
1*9(120)
5M129)
repeatedly
60(11*0)
68(15U)
99(210)
as needed
98(208)
93(199)
100(212)
109(228)
1*7(117)
52(126)
56(133)
100(212)
100(212)
100(212)
100(212)
102(216)
1.06
1.82
1.
0.
0,
0
0
1
1
.15
.71
.98
.98
.97
.09
.12
3.1*8
5-97
3.77
2.33
3.22
3.22
3.18
3.58
3.67
1,1,0 12U 36
3:!?r
37
Power off during noon hour
200(392) 116(2UD 112(23U)
Intermittent power
25l(WU) 188(370) ll8(2HU) 107(225) 0.8U
Power on continuous!/
103(217) 0.86 2.82
2.76
3-!t9 12l* 1*2
It: 30 121* 78
1*:1»5 123 127
-
2-9-73
9:1*3
a; 55
10:1.5
lljOS
115
111
111
12B
73
22
55
58
28 ( 82)
1*0(10U)
100(212)
102(216)
35^ 95)
100(212)
101 (21<*)
28( &2)
35( 95)
81(178)
101(211*)
3l*( 93)
1*2(108)
100(212)
101(211*)
0.39
1.26
0.50
0.55
1.28
h.13
1.61*
1.80
Hydrogen sulficte odor
O 37 100(212) 100U12) 1001212) 100(212 O.J*
Small hole blown through side of column - shut down
3-08
thermocouple Verted 3 inches (7-6 cm) into column.
thermocouple inserted A ^** «•* -> ™° -1
157
-------
By 1:^0 p.m., however, the temperature at the No. 1 position was at
200°C (392°F) and the power was on intermittently for a period of 1
hour and 15 minutes in an attempt to raise the temperature of the lower
portions of the column without increasing the temperature unduly at the
No. 1 station. The No. 1 thermocouple was then partly removed from the
column so that it was inserted into the column for only 3 inches (7-62
cm) instead of the previous 6 inches (15.2U cm). It immediately regis-
tered a temperature of only 132°C (2TO°F), indicating that the temper-
ature of the carbon was high only at the center. As the temperatures
increased in the interior at the other stations, these thermocouples
were also brought closer to the periphery of the column, and in this way
the carbon was practically completely dried out without raising the
temperature at the periphery, where excessive high temperatures would
probably damage the teflon insulation.
Following this operation the column was probably packed as good as
possible but it was decided to proceed through one more drying cycle
before applying acetic acid to the carbon. All four thermocouples were
set so that they extended 2 1/2 inches (6.35 cm) into the carbon. Wet-
ting of the carbon was started by passing water into the top of the
column; however, the increased pressure upon the sides of the column
caused one of the small corks, which had been used to close up the open-
ing when the three center electrodes had been removed, to blow out.
This cork was in the B position and the dry carbon being very fluid
rapidly poured through this opening, with the result that about half
of the top foot (30.5 cm) of carbon flowed from the column. The opening
was again closed and all stoppers tightly secured and the carbon re-
placed and packed as well as possible. Flooding of the column was
continued until all of the carbon was considered to be wetted. As will
be seen later, this was not the case.
With the top flange in place, heating was started at 9:33 a.m. on
February 9» with an electrical resistance of 0.39 ohm per foot (1.28
ohm/m) of carbon. By 10:^5 a.m. the No. 1, 2, and ^ thermocouples
registered 100°C (212°F) while the No. 3 thermocouple registered 8l°C
158
-------
(178°P). by 11:05 a.m. all temperatures were at the boiling point of
water and the resistance remained fairly low at 0.55 ohm. At this point
in time, however, what appeared to be a hydrogen sulfide odor was notice-
able, but its source could not be traced. At about 11:1*5 a.m. the
resistance had increased to 0.9^ and a small jet of steam was suddenly
released from the side of the column.
The carbon was removed for inspection of the column and it became evi-
dent that localized heating had volatilized a small portion of the tef-
lon coating, which then resulted in a short circuit or arcing directly
to the shell of the column, with a subsequent breakthrough. Evidently
because of the previous disruption of the packed column, which had re-
sulted in uneven packing, the water had channeled through the column,
and one side was not wetted. Thus, this portion of the dry carbon im-
mediately became very hot, with the 'befofementioned results.
The column was repaired, charged with k feet (1.2192 m) of new Filtra-
sorb UOO carbon by adding the carbon to a flooded column and then drain-
ing. Power connections were made between the B and E position so that
the top foot (30.5 cm) of carbon would not be heated directly by the
electric current. The thermocouples were placed so that No. 1 was in
the No. 1 position 2 inches (5.08 cm) into the column, No. 2 in the C1
position 6 inches (15.2U cm) into the column, No. 3 in .the No. 3 posi-
tion 2 inches (5-08 cm) into the column, and No. U in the D1 position
2 inches (5.08 cm) into the column. A vibrator was also installed on
the outside of the shell of the column to facilitate packing. The data
obtained when operating under these conditions are shown in Table 26.
Operation during February 16 and 19 then resulted in gradual drying and
packing of the carbon. By 3:10 p.m. on the 19th the carbon was com-
pletely dry, except for about one inch (2.51* cm) around the periphery of
the column.
On February 20 the column was again wet, with power connections to the
B and E positions and the thermocouples placed the same as for February
16. By 2:15 p.m. 23-5 liters (6.21 gal.) of water had been distilled
159
-------
Table 26. LOWELL UNIT - ALTERNATING CURBEHT
Water Evaporation
Filtrasorb UOO Carbon
Carbon Bed 1 ft x 1) ft (30.$ cm x 1.219 n)
Temperature. °C (°F)
Date Tine Volts Amps
Glee, resistance
ohms/ftohm/»eter
Power connection B to E — Vibrator installed on shell of column
Thermocouple configuration:
1 — Ho. 1 position, 2 inches (5-1 en) into column
2 — C' position, 6 inches (15.2 cm) into column
3 — Ho. 3 position, 2 inches (5.1 cm) into column
I) — D' position, 2 inches (S.I cm) into column
Virgin carbon — wet
2-16-73 1:35
2105
2:15
2:55
3:25
l*:30
110
110
111
110
109
110
32
10
1(1(106)
1*1(106)
!»2(108)
82(180)
UO(IOU)
72(162)
1*1(106)
75(163)
Started Vibrator and tapped column with mallet
1*1(106)
21
6
71
1*5
86(187)
81(178)
72(162)
71(160)
Packed carbon from top
7>*(165)
100(212)
85(185)
99(210)
79(172)
99(210)
71(160)
78(172)
99(210)
1.15
3.67
1,76
6.05
0.51
0.81
2-19-73 8:15 108 27
Power off
36( 97) 36( 97) 36(, 97)
Packed carbon from top
36(. 97) 1.33
9:05
9:25
10:05.
106
106
108
10:35 106
83
5l*
33
50
38(100)
100(212)
99(210)
58(136)
108(226)
110(230)
55(131)
100(212)
100(212)
53(127)
100(212)
98(210)
0.1*3
0.65
1.09
Packed carbon from top
100(212) 110(230) 101(211*)
95(203) 0.71
3.77
12.01*
5.77
19.85
1.67
2.66
k.36
1.31
2.13
3.78
2.33
Sudden temperature rise on No. 2 thermocouple to llt3°C .(289°F) then to 181°C (358°F)
Removed top flange — packed carbon — No. 1 thermocouple out of order — flange left off
1:35
2:1(0
3:10
58
65
65
35
59
— l61*?(327) 118(21*1*) 100(212) 0.55
— 166°(331) 160(320) 105(221) 0.1*6
J- 98 (208) 106°(223) 96C(205) 0.37
Power off — carbon dry except for l.(2.5 cm) inch around periphery
Wet carbon — replaced top flange
Power connection B to E — thermocouple configuration same as for 2-16-73
2-20-73 9:00
§550
10:30
10:1*0
11:30
12;QO
1:00
2:00
2:15
88
95
105
107
107
108
107
106
80
125
100
57
1*1*
33
31
36
56
53
2lt( 75)
97(207)
98(208)
98(208)
98(208)
31( 88)
79(17"*)
95(203)
Vibrator on
28( 82)
96(205)
98(208)
97(207)
100(212)
99(210)
101(211* )
35( 95)
95(203)
103(217)
105(221)
157(315)
Vibrator off
98(2081 102(216) 102(.2l6) 151(301*)
99(210) 133(272) 105(221) 13l*d(273l
99(210) 11*7*1297) 120(21*8) l62e(32l*)
98(208) 101CC211*) 192(378) 192e(378)
0.27
0.33
0.6l
0.81
1.08
1.16
0.99
0.63
0.50
1.80
1.51
1.21
0.88
1.05
2.00
2.66
3.51*
3.80
3.25
2.07
1.61*
Power off - 23h liters (6.2 gal) of water distilled from column and 5 liters d •
drained
^Thermocouple inserted 1* inches (10.2 cm) into column.
Thermocouple inserted 3 Inches (7.6 cm) into column.
cThermocouple inserted 1*5 inches (3.8 cm) into column.
^Thermocouple inserted 1 inch (2.5 cm) into column.
eThermocouple Inserted 's inch (1.3 cm) into column.
160
-------
from the carbon and 5 liters (1.32 gal.) had been collected at the bot-
tom of the column. It was then decided that the column was ready for
adsorption of acetic acid.
v
The first run in which acetic acid was adsorbed on the carbon was made
on February 21. Condensate that had been steam stripped and passed
through the furfural carbon adsorption column was used as feed. Thus,
this condensate had most of the S02, methanol, and furfural removed.
The column was fed at 1.66 liters (O.M* gal.) per minute, which is equiv-
alent to 22.75 liters per minute per square meter (0.559 gallon per min-
ute per square foot). 1576 Grams (3.U8 pounds) of acetic acid were ad-
sorbed on the carbon. The following day, February 22, an attempt at
regenerating the column was made. The regenerating data are found in
Table 27. Power connections were in the B and E position, which re-
sulted in one foot (0.30U8 m) of carbon not being heated by the electric
current. The thermocouples were placed in the following position: No.
1 in the No. 1 position 2 inches (5.08 cm) into the carbon, No. 2 in the
C' position 6 inches (15.2k cm) into the carbon, N.O. 3 in the No. 3
position 2 inches (5.08 cm) into the carbon, and No. U in the D1 position
2 inches (5.08 cm) into the carbon. Heating was continued for 7 hours
during which time 21.62 liters'(5-71 gal.) of water were evaporated,
which contained 89.lU grams (0.20 Ib) of acetic acid. The temperature
at the No. 2 thermocouple, which was in the center of the carbon, rose
to 2l8°C (U2U°F) after 2 hours and 20 minutes of operation.. This thermo-
couple was then gradually moved toward the periphery of the column, in
order to more closely watch the temperature pattern. Since the top foot
(30.5 cm) of carbon was not heated directly, its temperature never rose
above the boiling point. The higher temperatures at the No. 2 and U
thermocouples indicated that the left side of the column was heating
much faster than the right side.
/
The second absorption run was made prior to obtaining all of the
analytical data from the regeneration of Run 1, and it was not known at
that time that such a very small amount of acetic acid had been removed
from the column. However, with a feed rate of l.TU liters (0.46 gal.)
' 161
-------
Table 27. LOWELL UHIT — ALTHIHATIHG CUBBEHT
Date
2-22-73
ON
ro
Run 1 — Desorption or Regeneration
Filtrasorb 400 Garten
Carton Bed 1 ft x 4 ft (30.5 cm x 1.219 m)
Time Volts Amps
Temperature. °C C°F)
ELee. resistance Sample
ohms/ftohm/meter No.
2-21-73 Acetic acid adsorbed on carbon -1576 grams or 3.1*8 Ibs
Pover connection B to I!
Thermocouple configuration:
1 — Bo. 1 position, 2 inches (5.1 cm] into carbon
2 — C' position, 6 inches (15.2 cm) Into carbon
3 —Ho. 3 position, 2 inches (.5.1 .cm) into carbon
4 — D1 position, 2 inches (5.1 cm) into carbon
8:20
8:50
9:00
9:20
9:30
9:50
10: 40
11:20
12:10
12:30
12:40
1:00
1:10
1:30
1:1(0
1:50
2:10
2;20
2:30
2:50
3:00
3:10
3:13
3:20
Bottom
Total
100
102
103
109
110
106
122
135
137
137
135
121
122
122
122
122
120
134
134
135
134
134
Power
—
drain
100
107
87
71
44
28
21
21
22
25
27
27
28
29
29
27
27
32
35
37
42
46
off -
—
40(104)
100(212)
100(212)
100(212)
98(208)
100(212)
100(212)
101(214)
100(212)
99(210)
99(210)
98(208)
99(210)
99(210}
99(210)
99(210)
97(207)
97(207)
95(203)
93^(199)
95^(203)
934(199)
36( 97)
96(205)
99(210)
109(228)
140(284)
167(333)
218(424)
204a (399)
I60b(320)
106°(223)
108C(226)
119C(246)
140C(284)
182°,(360)
193 (379)
206°(403)
213°(4l5)
2l4c(4i7)
218C(424)
225C(437)
226= (439)
242° (468)
spot temperature check
92d(l98)
253C(487)
40(104)
99(210)
100(212)
98(208)
100(212)
98(208)
99(210)
100(212)
98(208)
97(207)
98(208)
97(207)
98(208)
98(208)
99(210>
100(212)
99(210)
101(214)
104(219)
130(266)
169(336)
115 (239)
40(104)
98(208)
100(212)
99(210)
99(210)
99(210)
100(212)
103(217)
115(239)
118(244)
124(255)
146(295)
164(327)
128° (262)
131° (268)
135°(275)
150° (302)
158?(316)
176= (349)
200° (392)
2l8c(424)
240C(464)
0.33
0.30
0-39
0.51
0.83
1.23
1.94
2.14
2.11
1.83
1.67
1.49
1.45
1.40
1.40
1-51
1.48
1.39
1.28
1.21
1.06
0.97
on shell - 226-7°C (44o°F)
159c(3l8)
255° (491)
—
—
—
thermocouple inserted 3 inches (7.6 cm) into column.
Thermocouple inserted 2 inches (5.1 cm) into column.
"^Thermocouple inserted 1 inch (2.5 cm) into column.
Thermocouple inserted 5 inches (12.7 cm) into column.
1.08
1.98
1.28
1.67
2.72
4.04
6.36
7.02
6.92
6.00
5.48
4.89
4.76
4.59
4.59
.95
.86
.56
.20
• 97
.18
3.18
20
21
Sample Volatile organic acids
size, Concentration, Removed,
liters g/1 g
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
2.00
2.0O
2.00
2.00
2. 00
2.00
2.00
0.80
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.50
4.22
4.25
4.86
4.26
4.12
4.02
3.90
3.75
3.54
3.81
4.00
3-92
3.88
3.92
3.87
3.60
3.87
4.58
3.46
0.52
1.32
21.62
3.56
2.8ft
8.44
8.50
9.72
8.52
8.24
8.04
7.80
3.00
1.77
1.9O
2.00
1.96
1.94
1.96
1.94
1.80
1.94
2.29
1.73
1.85
3-80
89.14
-------
per minute, which was equivalent to 23.85 liters per minute per square
meter (0.586 gallon per minute per square foot), 102^ additional grams
(2.26 pounds) of volatile organic acids were added to the .carbon. This
was from a feed similar to that used in Run 1. The succeeding regener-
ation data are given in Table 28. Power connections and thermocouple
configurations were the same as at the start of Run 1. At the end of
k hours of operation the power connections were changed to the A and B
position in order to heat the top foot (30..5 cm) 'of carbon. Shortly
after, however, electrical arcing was in evidence at the top electrode.
Upon removing some of the carbon it was evident that arcing had occurred
on the left side of the top electrode. The insulation on the wall of
the column was repaired, the electrode was replaced and carbon was added
to a depth of one inch (2.5U cm) above the electrode. The column was
also insulated with 2 inches (5.08 cm) of glass wool to eliminate the
condensation around the periphery of the column. It was also necessary
at this time to wet the carbon with tap water to make necessary in-
sulator adjustments. On March 8 the power connections were made over
the entire h foot (1.2192 m) of carbon between the A and E positions,
with the thermocouple configuration as follows: No. 1 at No. 1 posi-
tion k inches (10.16 cm) into the carbon, No. 2 in the No. 2 position
2 inches (5-08 cm) into the carbon, No. 3 in the C' position 6 inches
(15.2k cm) into the carbon, and No. k in the Df position 5 inches (12.7
cm) into the carbon. After about 20 minutes of operation there was
again evidence of arcing, and so the connections were changed to the B
and E position. Heating was continued for about two hours, and the
.power connections were again changed to cover the entire U foot (1.2192
m) of carbon, to the A and E positions. After about 20 minutes of
operation, a strong unidentified odor was detected and power was shut
off. On March 9, the following day, the power was again activated
without making any changes in the configuration of the power hookup,
but after about 10 minutes a strong odor was again in evidence. The
insulation was removed and it was found that a spot had burned on the
left side, about 1 1/2 feet (H5-7 cm) down from the top of the carbon
163
-------
Table 28. LOWELL UNIT - ALTHWATIHO CURRENT
Run 2 — Deaorption or Regeneration
Flltrasorb 1(00 Carbon
Carbon Bed 1 ft * It ft (30..5 em * 1,815.ml
line Volti Amps
Tampa:
irature. "C t°r)
Elec. resistance
ohaa/ft obi/meter
Sample Volatile organic acids
Sample site. Concentration, Removed,
No. liters .g/1 g
2-2T-T3 Volatile organic acida adaorbed on carbon — 102U graaa or 2.26 Ibs
2-28-T3
3-1-73
3-8-73
Pover connection — B to H
Thermocouple configuration:
1 —Ho. 1 poaitlon, 2 inchea (5.1 en) into carbon
2 - C' poult ion, 6 inches (15.2 cm) Into carbon
3 —Ho. 3 position, 2 Inchea (.5.1 cm) into carbon
k — D1 position, 2 inchea (5.1 cm) into carbon
8:22
8:50
8:55
9:27
9:1.0
10:00
10:20
10 1 1(0
11:10
11:1.0
12:20
1:00
101
103
Pover
Pover
104
133
13k
Ik8
175
ITS
ITS
11.8
9T 39(102)
72 100(212)
off — repaired top
on
63
52
2T
25
22
23
ko
96(205!
97(207)
97(207)
97(207)
100(212)
100(212)
101 (21k)
98(208)
37( 99)
96(205)
gaaket leak
97(207)
98(208)
101 (21k)
113C235)
195(383)
Il8a(2kk)
122-C251)
171 (3kD)
35( 95)
35C 95)
_„
««.
„
„
__
_
—
36( 97)
100(212)
98(208)
100(212)
10l((219)
125(257)
I62(32k)
206(1(03)
2k7(k78)
155a(311)
0.35
0.1.8
0.55
0.85
1.31
1.83
2.33
2.65
2.5k
1.23
1.15
1,57
1.80
2.79
k.30
6.00
7.6k
8.69
8,33
k.Ok
...
1
2
3
k
5
6
7
8
""*
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.k7
"™
0.05
0.05
O.Ik
0.5k
0.65
1.3&
2.39
3.56
"*""
0.10
0.10
0.27
1.08
1.29
2.69
k.78
8.79
Changed power connection from A to B
1:30
1:50
It*
108
T5
20
99(210)
100(212)
I83a(36l)
16T"(333)
1.39
lt.56
1T.T2
3:50 98
Bottom drain
Arcing at top electrode — pover off for Borne tine
99*(210) Ii6a(2l(l) - 123*(253> 2.58 8.W
9
10
11
12
1.00
l.TO
2,80
1.30
5.22
9.T6
3.2T
1..01.
5.22
16.60
9.1T
5.26
Arcing had occurred on the left aide of the top electrode. Repaired the nil. Replaced electrode and added carbon to a depth of
1 inch above electrode, Xnaulated column vith 2 Incha (S.I cm) of glaaa vool. Vet carbon vith tap vater to make insulator
adjuatmenta
Pover connection — A to £
Thermocouple configuration:
1 —Ho. 1 poaltion, k Inchea (J.0.2 cm) into carbon column
2 — Ho. 2 poaitlon, 2 Inchea (S.I cm) into carbon column
3 — C' petition, 6 Inches (15.2 cm) into carbon column
-D' position, 5 inches
(15.2'em)
(12.T. cm)
into carbon column
1:13
1:30
l:kO
2:30
2:50
3:1.0
10k kl 37( 99) 37( 99) 37(99) 37(99)
Power off — evidence of arcing — pover connection — B to S
103 75 67(153) 57(135) 62(lkk) 72(162)
118 39 101(211.) 101(211.) 102(216) 102(216)
133 k7 101(21k) 10l(2lk) 102(216) 105(221)
13l| 35 101(21k) 101(21k) 107(225.) 119(2k6)
107 2k 10l(21kJ 106f223l llkfS37j 150(302)
0.63
O.k6
1.01
0.9k
.1.28
l.»9
2.07
1.51
3.31
3.08
k.20
k.89
13
15
3.00
3.00
3.00
0,1k
0.22
0.1.6
O.kl
0.65
1.38
Power connection — A to'E
k:05 108 18 10l(2lkl 102(216) 112(23k) 13l((273) 1,50
Power off, strong odor
U.92
16
3,00
0.75
2.82
3-9-T3 lliUO Pover on
11:50 106 21
Total
67(153) 66(151) 7k(l65) 7Kl°0) 1.26
Power off, strong odor
60.61
thermocouple inserted 1 inch (2.5 cm} into column.
-------
ted. The carton was removed and it was found that arcing had occurred
in three places.
At this point in time it was decided that, although this method of
heating the carbon column might still be successfully concluded, the
level of work required and possible additional expenditures were too
high to proceed under this particular project, and this phase of the
work was terminated.
The idea was presented that perhaps the addition of a granular metal
dispersed among the carbon particles would tend to prevent channeling
of the electric current. In order to rapidly prove or disprove this
theory, a small 2-inch diameter (5.08 cm) carbon column was set up in
the laboratory. After several runs using granular aluminum, there was
no evidence of improvement.
CONCLUSIONS
Attempted heating of the carbon column by the direct application of an
electric current through the carbon has resulted in the following
observations.
1. If direct current is used, then the anode must be at least
platinum coated to prevent disintegration.
2. The packing of carbon is very important; it may even be
necessary to selectively size the carbon particles.
3. A method needs to be devised to prevent localized heating
and/or arcing. Proper packing would be partly helpful in
preventing this.
k. Work with alternating current was hindered on this project
by having only one transformer available.
5. The column should be constructed of a more heat-resistant
electrical insulating material.
6. A change in design would be a requirement for successful
operation. This would probably require some type of pressure
165
-------
loaded electrodes to prevent loss of contact between elec-
trodes and carbon.
166
-------
SECTION VIII
INVENTION RECORD AND PUBLICATION
Invention record — A granular activated carbon regeneration
system using alcohol as the regeneration agent and the acti-
vated carbon column as a distillation unit was conceived and
reduced to practice by K. W. Baierl in the course of this
research.
Publication — Treatment of Sulfite Evaporator Condensates for
Recovery of Volatile Components, Kenneth W. Baierl, Bernard F.
Lueck, and Averill J. Wiley, Tappi 56, no. T:58-6l(July, 1973)
167
-------
GLOSSARY OF ABBREVIATIONS
SSCF Steam Stripping Column Feed
SSCF Steam Stripping Column Bottoms
SSCCO Steam Stripping Column Condensed Overhead
FCO Fractionation Column Overhead
FCB Fractionation Column Bottoms
FCSC Fractionation Column Side-cut
ACCFE Activated Carbon Column Furfural Effluent
ACCACE Activated Carbon Column Acetic Acid Effluent
ACCFRI Activated Carbon Column Furfural Regeneration Inlet
ACCFRE Activated Carbon Column Furfural Regeneration Effluent
ACCACRI Activated Carbon Column Acetic Acid Regeneration Inlet
ACCACRE Activated Carbon Column Acetic Acid Regeneration Effluent
FCRF Fractionation Column Regeneration Feed (Furfural Regenera-
tion Effluent, Acetic Acid Regeneration Effluent, or Steam
Stripping Column Overhead).
FCACRB Fractionation Column Acetic Acid Regeneration Bottoms
FCACRSC-1 Fractionation Column Acetic Acid Regeneration Side-cut No. 1
FCACRSC-2 Fractionation Column Acetic Acid Regeneration Side-cut No. 2
FCACRO Fractionation Column Acetic Acid Regeneration Overhead
FCFRSC Fractionation Column Furfural Regeneration Side-cut
FCFRO Fractionation Column Furfural Regeneration Overhead
FCFRB Fractionation Column Furfural Regeneration Bottoms
168
U.S ROUPBNMFNT PRINTING OFFICE: 1974 546-316/Z62 1-3
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
w
$, 'Report
TREATMENT OF SULFITE EVAPORATOR CONDENSATES FOR ft:.
Rnpc-it
RECOVERY OF VOLATILE COMPONENTS
Baierl, Kenneth ¥., Chang, Nai L. , Lueck, Bernard F.
TJ-i 1 gatr A-\rQ-y*T 1 ~\ T ar\r\ TTnl m PrVhov-l- A
The Institute of Paper Chemistry, Appleton,
Wisconsin, Division of Industrial and
Environmental Systems
S 801207
eriod Coveted
12- S.Bf*asdrij;. • QjrgsaA '.ti
Environmental Protection Agency report number,
EPA-660/2-73-030. December 1973.
Pilot studies were conducted on two routes for recovery of volatile components
in condensates from evaporation of spent liquors from acid sulfite pulping.
The condensates were steam stripped prior to adsorption on activated carbon.
The first route utilized fractional distillation and solvent extraction from
the carbon, while the second route used low-temperature thermal regeneration.
Relatively pure SOg, methanol, furfural, and ethyl acetate were recovered.
Estimated process economy, based upon recovery of saleable volatiles, favor-
able mass and heat balances, and credits for BOD removal, indicates the first
route may provide a favorable method for the removal of pollutants from con-
densates. Low temperature regeneration studies failed due to mechanical
design problems, but this second route continues to be considered as techni-
cally feasible. In an important auxiliary study, a substantial number of
condensate samples from cooperating mills were analyzed. Differences in
quality, quantity, and dilution of the condensates from different sources
were apparent. Individual design of processing facilities for each mill is
indicated, and especially so in those mills where backwashing of the evapo-
rators contaminate the condensate with nonvolatile materials.
i/a. Descriptors Evaporators, Condensation, Activated Carbon, Adsorption,
Solvent Extraction, Separation Techniques, Distillation, Alcohols,
Acids, Organic Acids, Sulfite Pulp Wastes, Pulp and Paper Industry,
Biochemical Oxygen Demand, Chemical Oxygen Demand, Water Pollution,
Industrial Wastes, Heat Balance, Mass Transfer.
17b. Ideati/iers
Steam Stripping, Fractional Distillation, Spent Sulfite Liquor,
Acetic Acid, Methanol, Furfural, Ethyl Acetate, Volatile Acids,
Neutral Volatiles, Nonvolatile Organics, Lignin, Lignosulfonic Acids.
c. COWRK Fi+!d & Group Q5D
U. A'-aUsfiitiiy it. v f sr'tv'Cfass.
'%).• Wo, of
Pages
.. ? Pr> . «
—^——~—*~~—
Send To :
WATER RESOURCES SCIENTIFIC INFORMATION CENTER I
US DEPARTMENT OF THE INTERIOR 1
WASHINGTON. D. C. 2O24O 1
_____ _ _^ I
------- |