United Stitw
Environmental Protection
Agency
Industrial Environmental Reaearch EPA-600/2-80-039
Laboratory January 1980
Cincinnati OH 45268
Reaeerch vxj Dcvclopnwnt
Toxicity Reduction
Through Chemical and
Biological
Modification of
Spent Pulp
Bleaching Liquors
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental 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.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-80-039
January 1980
TOXICITY REDUCTION THROUGH CHEMICAL AND
BIOLOGICAL MODIFICATION OF SPENT PULP BLEACHING LIQUORS
by
Carlton W. Dence, Chun-Juan Wang, and Patrick R. Durkin
State University of New York
College of Environmental Science and Forestry
Syracuse, New York 13210
Grant R 804779
Project Officers
Michael D. Strutz
H. Kirk Willard
Food and Wood Products Branch
Industrial Environmental Research Laboratory
Cincinnati, Ohio 45268
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
ii
-------
FOREWORD
When energy and material resources are extracted, processed, converted,
and used, the related pollutional impacts on our environment and even on our
health often require that new and increasingly more efficient pollution con-
trol methods be used. The Industrial Environmental Research Laboratory-Cin-
cinnati (lERL-Ci) assists in developing and demonstrating new and improved
methodologies that will meet these needs both efficiently and economically.
The report investigates the reduction of the toxicity of spent pulp
bleaching liquors through chemical and biological modification. Toxicity
was determined over a range of concentrations, before and after modification,
to determine their effect on the growth of fungi, alga and duckweed, and on
the survival of Daphnia Magna. Methods of chemical treatment that were in-
vestigated in this study included treatment with chlorine, chlorine dioxide,
hypochlorite, ozonation and hydrogen peroxide treatment, alum and lime addi-
tion, carbon adsorption, and by modification of the conventional chlorination
and caustic extraction bleaching stages. For further information on the sub-
ject, contact the Food and Wood Products Branch, IERL Cincinnati.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
iii
-------
ABSTRACT
Chlorophenols similar to or identical with those detected in spent
chlorination and caustic extraction liquors were synthesized and tested over
a range of concentrations to determine their effect on the growth of several
fungi, an alga (Chlorella pyrenoidosa), and duckweed (Lemna perpusilla) and
on the survival of Daphnia magna. Although differences were found in the
responses of the individual test organisms to the chlorophenols, the results
generally indicated that growth repression and toxicity increased with in-
creasing numbers of chloro substituents on the phenolic ring.
The chlorophenols referred to above were subjected to a number of chem-
ical and biological treatments to assess their susceptibility to degradation
when present as components of spent chlorination and caustic extraction li-
quors. Chlorophenol degradation was determined by gas chromatographic and/
or ultraviolet spectroscopic analysis for unreacted starting material.
Ozone and chlorine dioxide effectively degraded the chlorophenols, but com-
plete removal required the application of excessively large amounts of chem-
ical. In the case of the ozonization of tetrachloroguaiacol, decreases in
phenol breakdown were paralleled by reductions in toxicity as indicated by
Daphnia magna. Alkali treatment of the chlorocatechols at room temperature
also proved an effective means of reducing chlorophenol toxicity.
Biological treatment of the chlorophenols consisted of the application
of pure cultures of three different fungi and a mixed microbial population
for periods ranging up to 15 days. Degradation varied widely among the vari-
ous phenols and for the same phenol treated with different fungi. Under
similar conditions, aeration in the absence of any microorganisms was effec-
tive in varying degrees with catechol (1,2-dihydroxybenzene) derivatives sus-
taining the greatest amount of degradation.
The spent liquors from the chlorination and caustic extraction bleach-
ing stages of a pine kraft pulp (Kappa No. 25.5) were fractionated by means
of dialysis, gel permeation chromatography, and solvent (ether) extraction;
and the fractions bioassayed for their effect on a fungus, Aspergillus
fumigatus and Daphnia magna. Without exception, the fraction containing
jl
the lower molecular weight materials displayed the greatest repression of
fungal growth and were the most toxic to the Daphnia.
Spent chlorination and caustic extraction liquors were subjected to a
variety of chemical treatments and the resulting effects on acute toxicity
determined. Treatment with elemental chlorine, hypochlorous acid, hypo-
chlorite, ozone and hydrogen peroxide produced increases in the toxicity
of the spent liquor. A modest reduction in toxicity accompanied treatment
of spent chlorination liquor with chlorine dioxide. Toxicity reduction was
iv
-------
also achieved by the addition of alum and lime to spent caustic extraction
liquor. The concentration of toxic material expressed as total organic car-
bon (TOC) was, however, higher in each instance. As in the case of the
chlorophenols, substantial reductions of toxicity were achieved by the addi-
tion of alkali to spent chlorination liquor. Toxicity reduction in this
instance paralleled decreases in phenol and organically bound chlorine con-
tents.
Biological treatment of spent chlorination and caustic extraction li-
quors involved the application of a fungus (Candida util is), an unidentified
bacterium, and a mixed microbial population, together with supplemental car-
bon sources. The microbial mixture and the fungus effected essentially com-
plete elimination of toxicity from spent chlorination liquor and small re-
ductions in the toxicity of spent caustic extraction liquor. The bacterium
was, on the other hand, comparatively ineffective in achieving the same
objective.
Toxicity reduction through modification of conventional chlorination
and caustic extraction bleaching stages was also evaluated. The substitu-
tion of chlorine dioxide for chlorine in the chlorination stage proved to be
distinctly beneficial in reducing toxicity as was the substitution of oxygen
for chlorine in the same stage. Small reductions in toxicity attended the
introduction of hydrogen peroxide into the first caustic extraction stage
but addition of hypochlorite in the same stage was less effective than
alkali alone in this regard.
This report was submitted in fulfillment of Contract No. R804779010
by the Research Foundation of the State University of New York under the
sponsorship of the U.S. Environmental Protection Agency. This report covers
a period from September 21, 1976, to September 20, 1979 and work was com-
pleted as of September 20, 1978.
-------
CONTENTS
Foreword i i i
Abstract iv
Figures vi1
Tab!es vi i 1
Acknowledgment xi
1. Introduction 1
2. Conclusions 3
3. Recommendations 5
4. Materials and Methods 7
Chemicals 7
Biological test organisms 7
Spent bleaching liquors 8
Fractionated spent bleaching liquors 9
Chemi cal treatment of chlorophenols 9
Chemical treatment of spent chlorination and caustic extraction
liquors 11
Chlorination and caustic extraction stage modifications 12
Biological treatment of phenols and chl orophenol s 13
Biological treatment of spent chlorination and caustic
extraction liquors 13
Acute toxicity tests 14
Gas chromatographic analysis 16
Analytical procedures 17
5. Results and Discussion 19
Acute toxicity of phenols 19
Acute toxicity of whole and fractionated spent chlorination
(SC) and caustic extraction (SCE) liquors 29
Degradation of phenols by chemical treatment 48
Degradation of chlorophenols by biological treatment 60
Chemical treatment of spent chlorination and caustic extraction
liquors 64
Biological treatment of spent chlorination and caustic
extraction 1iquors 82
Reduction in SCL and SCEL toxicity through modifications
of the blcaching process 86
References 91
Appendices 97
A. Jack Meyer's modified medium for Chlorella 97
B. Hutner1s medium for duckweed 98
-------
FIGURES
Number Page
1 Effect of 2,4,6-trichlorophenol (VI) and three chlorocate-
chol s on the growth of Asperginus fumigatus 21
2 Effect of creosol (VII) and three chloroguaiacols on the
growth of Aspergillus fumigatus 22
3 Effect of age on the color of spent chlorination and caustic
extraction 1iquors stored at room temperature 30
4 Rate of loss of organically bound chlorine from spent
chlorination and caustic extraction liquors 32
5 Response of Aspergillus fumigatus to whole spent chlorination
and caustic extraction 1iquors 35
6 Gel permeation chromatograms of retentate (I) and dialysate
(II) with Sephadex G-25 38
7 Relationship between the dry weight of fungus and the total
organic carbon in the fractions obtained by single
dialysis 39
8 The effect of the ether extracts of SCL (A), SCEL (B), and
of ether-extracted SCL (C) on the growth of Aspergillus
fumigatus 42
9 A typical elution diagram for the fractionation of spent caustic
extraction liquor ether extract on silica gel 45
10 Degradation of selected phenols through reaction with chlorine
dioxide 57
11 Effect of carbon treatment on the color and TOC of lime-
treated spent caustic extraction 1iquor 72
12 Detoxication (A) of spent chlorination liquors by alkaline
treatment 81
vm
-------
TABLES
Number Page
1 Fungi Used in the Investigation 8
2 Phenolic Compounds Used in the Investigation 20
3 Minimum Concentrations of Phenols Required to Prevent Growth
of Various Fungi 23
4 Toxicity Ranges of Selected Phenols for Chlorella
pyrenoidosa 25
5 Effect of Four Phenols on the Vegetative Growth of Lemna
perpusilla 27
6 Acute Toxicity of Various Phenols to Daphnia magna 28
7 Acute Toxicity of the Whole Spent Chlorination and Caustic
Extraction Liquors to Daphnia magna 33
8 Acute Toxicity of Spen Caustic Extraction Liquor Fractions
to Daphnia magna 37
9 Acute Toxicity of Ether Extractives of Spent Chlorination and
Caustic Extraction Liquors to Daphnia magna 41
10 Toxicity of the Sub-fractions of the Spent Chlorination Ether
Extract as Indicated by Daphnia magna and Aspergillus
fumigatus 43
11 Toxicity Characteristics of Spent Chlorination and Caustic
Extraction Liquors Ether Extracts Fractionated by Chromatog-
raphy on Silica Gel 46
12 Elemental Composition of the Spent Chlorination Liquor (A)
and Spent Caustic Extraction Liquor (B) Ether Extracts
and Ether Extract Sub-fractions 47
13 Acidic Group and Total Hydroxyl Contents of the Spent Chlorin-
ation Liquor (A) and Spent Caustic Extraction Liquor (B)
Ether Extracts and Ether Extract Sub-fractions 47
1x
-------
Number
Page
14 Ozonization Rate Constants of Phenols 50
15 Detoxification of Tetrachloroguaiacol (XVI) by Ozonization
at pH 10 53
16 Ozone Oxidation of tetrachloroguaiacol-Spent Caustic Extraction
Liquor Combinations 55
17 Chlorine'Dioxide Oxidation of Tetrachloroguaiacol-Spent Caustic
Extraction Liquor Combinations 58
18 Effect of Alkaline Treatment (pH 11) on the Toxicity of Some
Chiorophenols to Daphnia magna 59
19 Effect of Aeration on Removal of Chiorophenols 61
20 Rate of Removal of Phenols durii g Aeration 61
21 Biodegradation of Phenols by Fungi in Liquid Culture Media 63
22 Effect of Sodium Hypochlorite Treatment on the Acute Toxicity
of Spent Caustic Extraction Liquor to Daphnia magna 65
23 Characteristics of Original ar.d Ozonized Spent Chlorination
and Caustic Extraction Liquors 67
24 Properties of Alum-Treated Spent Caustic Extraction Liquor 70
25 Effect of Lime Treatment on Spent Caustic Extraction Liquor ... 70
26 Properties of Lime-Treated SCEL after Reaction with Acti-
vated Carbon 71
27 Acute Toxicity of Spent Caustic Extraction Liquor Treated
Sequentially with Lime and Activated Carbon 74
28 Acute Toxicity of Carbon-Treated Spent Caustic Extraction
Liquor 74
29 Effect of Ozone Oxidation on the Properties of Lime-Treated
Spent Caustic Extraction Liquor 75
-------
Number Page
30 Effect of Chlorine Dioxide on the Properties of Lime-Treated
Spent Caustic Extraction Liquor 75
31 Effect of Hydrogen Peroxide Oxidation on the Properties of
Lime-Treated Spent Caustic Extraction Liquor 76
32 Acute Toxicity of Lime-Treated Caustic Extraction Liquor
Reacted with Ozone, Chlorine Dioxide, and Hydrogen
Peroxide 77
33 Effect of Alkali Treatment of Spent Chlorination Liquor on
Phenol and Organically Bound Chlorine Contents 78
34 Rate of Removal of Organically Bound Chlorine from Spent
Chlorination Liquor with Increasing pH 79
35 Effect of Alkali on the Toxicity of Spent Chlorination
Liquor to Daphnia magna 80
36 Effect of Prolonged (two week) Neutralization on the Toxicity
of Spent Chlorination Liquor to Daphnia magna 82
37 Effect of Biological Treatments of Spent Chlorination Liquor
and Spent Caustic Extraction Liquor on Acute Toxicity 83
38 Toxicity of Effluents from Conventional and Process-Modified
Bleaching Stages 87
39 Characterization of Selected Conventional and Process-
Modified Spent Bleaching Liquors 89
-------
ACKNOWLEDGMENT
The authors gratefully acknowledge the contributions of Dr. Kazu
Sameshima, Dr. Brian Simson, Ms. Erica Rowe, Mr. Richard Ziobro, Mr. James
Cain»and Mr. Joseph Fernandez in the experimental phase of the project.
-------
SECTION 1
INTRODUCTION
It is now a generally accepted fact that bleaching effluents are mildly
toxic to fish and other aquatic organisms. The results of a recent investi-
gation (1) have shown that the compounds contributing to toxicity are resin
acids, chlorinated resin acids, unsaturated fatty acids, fatty acid deriva-
tives (epoxy stearic and dichlorostearic acid), and chlorophenolics. The
presence of resin and fatty acids in bleaching effluents can be traced to
inefficient washing of the bleached pulp. During the chlorination and caus-
tic extraction stages, these compounds are in part converted to chloro and
epoxy derivatives.
The chlorinated phenols are derived from the residual lignin in the
pulp and appear to constitute the most toxic class of compounds comprising
bleaching effluents. During chlorination, the chlorophenols are, to a large
degree, oxidized and broken down to chlorine-substituted aliphatic compounds.
The impact of the latter compounds on the toxicity of bleaching effluents
has not been systematically evaluated.
\
Although the exact contribution of chlorophenols to the total toxicity
of bleaching effluents may be difficult to assess with any degree of confi-
dence at present, there can be little doubt but that it is at least a major
and quite possibly dominant one. The continuing identification of chloro-
phenols in spent chlorination and caustic extraction liquors serves to em-
phasize the importance of learning how such compounds respond to various
chemical and biological treatments in order that their removal from these
spent liquors may be facilitated.
In response to the aforementioned need, the investigation described be-
low was undertaken with the following objectives in mind:
1. To ascertain the relative degrees of toxicity displayed by phe-
nols and chlorophenols similar to,or identical with, those previ-
ously detected in spent chlorination and caustic extraction li-
quors when contacted with a variety of aquatic life forms.
2. To determine the feasibility of using selected chemical and bio-
logical treatments or combinations of such treatments for the
removal of those phenolic and chlorophenolic residues previously
demonstrated to have a deleterious effect on aquatic life.
-------
To pursue these objectives, we conducted experiments that involved acute
toxicity bioassays, and biological and chemical treatments of the individual
chlorophenols. These experiments were supplemented with similar tests that
used the spent liquors from the chlorination and caustic extraction of a
southern pine kraft pulp.
-------
SECTION 2
CONCLUSIONS
Several species of fungi, an alga (Chlorella pyrenoidosa), and a vascu-
lar plant (Lemna perpusi 11a) were found suitable as plant test organisms for
evaluating the toxic effects of phenols and chlorophenols. The sensitivity
of these organisms decreased in the order alga > vascular plant > fungi. An
aquatic invertebrate (Daphnia magna) also proved satisfactory for use in
acute toxicity bioassays of the phenols and chlorophenols used in the inves-
tigation. In acute toxicity bioassays involving spent chlorination and caus-
tic extraction liquors, Daphnia magna was found to be more useful than fungi
due to its greater sensitivity.
Chlorophenols similar to or identical with those present in spent chlo-
rination and caustic extraction liquors can be rapidly and extensively de-
graded with ozone and chlorine dioxide. However, their complete elimina-
tion requires inordinately large applications of chemical, particularly when
present with other oxidizable materials as would be the situation in spent
chlorination and caustic extraction liquors.
Alkaline treatment (pH 11) is effective in varying degrees in reducing
the toxicity of solutions of the test chlorophenols. Chlorocatechols can be
most successfully detoxified by such a treatment.
Chlorophenols of the type included in the investigation can be exten-
sively degraded by simple aeration (in the absence of any microorganisms in
1-week treatments). A similar effect can be achieved by biological treat-
ments involving individual fungi or a mixture of microorganisms. Fungi dis-
play widely varying degrees of ability to break down a given chlorophenol.
Some of the properties of spent chlorination and caustic extraction li-
quors were found to change as the liquors aged. Toxicity tests can be made
on reasonably aged liquors, however, without the expectation of changes
having occurred.
Based on the results of fractionation treatments, the toxicity of spent
chlorination and caustic extraction liquors gives evidence of being attri-
butable to the lower molecular weight substances contained therein.
In general, chemical treatment does not appear to represent a very
promising approach for reducing or eliminating the toxicity of spent chlori-
nation and caustic extraction liquors. However, detoxification of the
former liquor through treatment with alkali may have a future.provided the
technical and economic feasibility of such a treatment can be demonstrated
-------
by further experimentation. Lime and alum addition can be used to partially
remove toxic material from spent caustic extraction liquor. However, the
sludge that contains the toxic substances still has to be disposed of.
Modifications of conventional bleaching treatments (in which chlorine
dioxide and oxygen are substituted for chlorine in the chlorination stage,
and hydrogen perioxide is added to the caustic extraction stage) are effec-
tive in substantially reducing the toxicity of the corresponding effluents.
-------
SECTION 3
RECOMMENDATIONS
As described below in the body of this report, attempts to reduce the
toxicity of bleaching effluents through modification of conventional bleach-
ing stages produced some very encouraging results. We recommend, there-
fore, that research along these lines be continued and expanded so that the
potential of this approach may be fully assessed. In view of the general
success the paper industry has had using biological forms of waste treat-
ment, we further recommend that, in conjunction with the above bleaching
stage modifications, the effluents from such treatments also be subjected
to some form of biological treatment in order to assess the effect of the
modification on their susceptibility to biodegradation.
Improved efficiency with respect to the biological systems used in the
purification of paper mill wastes could result in an increased margin of
safety with respect to the ability of a particular treatment system to con-
sistently deliver toxicity-free effluents during mill upsets.or as a result
of other operational problems. The wide range of abilities displayed by the
various microorganisms used in the present investigation for breaking down
chlorophenols suggests the possibility of manipulating the composition of
the microbial population in the biological treatment system so as to facili-
tate this objective. On this basis,we recommend that the search for addi-
tional microorganisms having this particular capability be continued and
that the effect of environmental test conditions (i.e., pH, nutrients,
temp., etc.) be studied to ascertain how these should be controlled to
achieve the maximum rate of degradation.
In the current investigation, acute toxicity bioassays relied largely
on the response of a single species, Daphm'a magna. This approach made it
possible to screen a large number of phenols and chlorophenols as well as
spent chlorination and caustic extraction liquors before and after various
treatments. The relevance of this work could be substantially increased by
applying more sophisticated test methods and by increasing the number of
aquatic plant and animal species tested.
The feasibility of using an alga (Chlorella pyrenoidosa) and a vascular
aquatic plant, duckweed, (Lemna perpusilla) as test organisms for acute
toxicity bioassays of chlorophenols was explored in the present investiga-
tions and the findings (described in the body of this report) indicated their
potential utility in this regard, particularly from the point of view of
their increased sensitivity as compared to fungi. We propose that the tests
involving these organisms be continued and broadened to include bleaching
liquors before and after modification.
-------
Although Daphnia represent an extremely important class of aquatic in-
vertebrates and are thus a highly relevant test species, a sound program of
hazard assessment should attempt to evaluate effects on several organisms
and diverse habitats. For this reason,we recommend that bioassays of chloro-
phenols and spent chlorination and caustic extraction liquors be performed
using Gammarus, a genus of amphipods commonly found in streams and brooks.
Based on the comparative behavior of this organism and Daphnia when exposed
to other toxicants, the use of Gammarus in acute toxicity bioassay might be
expected to indicate significant differences in species susceptibility
rather than simply parallel the results obtained using Daphnia.
-------
SECTION 4
MATERIALS AND METHODS
CHEMICALS
2,4,6-Trichlorophenol, 2,4-dichlorophenol, catechol (1,2-dihydroxyben-
zene), guaiacol (2-methoxyphenol), ja-cresol and phenol were obtained commer-
cially. All other phenols and chlorophenols used in the investigation were
prepared as described by Gess and Dence (2), Cain (3), Nonni (4), Dence
et al. (5), and Willstatter and Muller (6).
Methyl dehydroabietate was prepared by methylation of the free acid
with diazomethane. This ester was not isolated, but gas chromatographic
analysis of the methylated material in the product mixture revealed the
presence of only one peak.
BIOLOGICAL TEST ORGANISMS
Animals
Daphm'a magna were purchased from Wards Natural Science Establishment,
Inc. Stock cultures were maintained at 17 +_ 1°C, pH 7.0 +_ o.l units in DM2,
a reconstituted fresh water described by D'Agostino and Provasoli (7). The
organisms were fed yeast and the photoperiod was kept on a 12-hour light/12-
hour dark cycle. Algal and protozoal contamination of the stock cultures
provided additional nutrient sources. First instar,or 1-3 day old Daphnia.
for use in the biological assays were obtained by transferring groups of 25
nongravid mature organisms to 250 ml of fresh DM2 ^7 + ^°c» PH 7-°) and
adding 5 ml of homogenized yeast (1 mg/ml) per day untfl eggs developed.
At this time, feeding was terminated and young were harvested daily.
Brine shrimp (Artemia salina) eggs, distributed by Metaframe Corpora-
tion, were purchased locally. These organisms were hatched in synthetic sea
water (8) 24 hours prior to testing. The synthetic sea water was also used
as the dilution water in all Artemia salina bioassays.
Plants
The plant test organisms used in the investigation consisted of several
fungi (listed below in Table 1), an unidentified bacterial culture, an
alga (Chlorella pyrenoidosa Chick), and a vascular plant (Lemna perpusilla
Torr.), commonly known as duckweed.
-------
TABLE 1. FUNGI USED IN THE INVESTIGATION
No. Fungus name
1 AsperglTlus fumlgatus Fres. (C-78)
2 Cladosporium herbarum Link ex Fries (C-5)
3 Paecilomyces variotj Bainer (C-64)
4 Penicinium variabile Sopp (C-78)
5 Tn'choderma koningii Pud. (C-65)
6 Aspergillus niger van Tiegh. (Wang 1152)
1_ Candida utilis (Henneberg) Lodder & Kreger-van Rij
Fungi 1-5 were organisms most commonly isolated from samples collected
at various sites of a kraft paper mill waste treatment plant. The bacterium
included in the testing was identified in the same samples.
Candida utilis is an unicellar, yeast-like organism containing 50% edible
protein.
SPENT BLEACHING LIQUORS
A pine kraft pulp (Kappa No. 25.5) was chlorinated and the resulting
pulp extracted with alkali using the conditions described below:
Chlorine or alkali applied, %
Residual chlorine, %*
Consistency, %
Time, min.
Temperature, °C
Final pH
Chlorination
5.25
0.1-0.2
3
60
22
1.7
Caustic extraction
3.6
-
10
60
70
12.3-12.4
*After filtration through a sintered glass Buchner funnel to remove
the pulp fines, the residual chlorine was completely eliminated.
The chlorinated pulp was washed with dilute (0.01 N) HC1 prior to the alka-
line extraction stage in order to maintain the same pH environment in the
pulp as existed during the Chlorination.
-------
FRACTIONATED SPENT BLEACHING LIQUORS
Dialysis
Spent chlorination liquor was dialyzed in a cellophane tube against
distilled water for 24 hours after first being concentrated to 1/10 of its
original volume. The ratio of spent liquor to distilled water during dialy-
sis was 1:40. Spent caustic extraction liquor was dialyzed directly under
the same conditions. In the case of extensive dialysis of the spent caustic
extraction liquor, the dialysis was repeated after the dialyzed liquor was
concentrated to its original volume. When sodium ion removal was required,
the retentate was further dialyzed against 0.1 N HC1 and then against dis-
tilled water.
Gel Permeation Chromatography
A concentrated neutralized sample of spent liquor (10 ml) was charged
onto the top of a 20 mm x 1 mm column of Sephadex 6-25 (40 g) and developed
by elution with distilled water. The elution of material was followed by
measuring the absorbance of the eluate at 280 nm.
Ether Extraction
Three liquid-liquid extractors (400 ml, 1-& and 10-£ capacity) were
used depending on the experimental requirements. Spent chlorination liquor
was extracted directly; spent caustic extraction liquor was extracted after
acidifying to pH 2 and without removing the precipitate which thereupon
formed.
Silica Gel Chromatography
Ether extracts of the spent liquor were further fractionated on a
column of 60-200 mesh silica gel (Grade 950, Fisher Chemical Company). A
layer of anhydrous sodium sulfate was added to the silica gel to remove
small amounts of water in the solvent extract. The ratios of sample to
silica gel and sodium sulfate were 1:150 and 1:25, respectively. The column
was eluted consecutively with petroleum ether, diethyl ether.and methanol.
CHEMICAL TREATMENT OF CHLOROPHENOLS
Chlorine Dioxide
Ten millimoles of the phenols in 30-40% aqueous ethanol were reacted
in the dark at room temperature with amounts of chlorine dioxide ranging
from 1-15 equivalents/mole of phenol until the oxidant was completely
exhausted. The product mixtures were exhaustively extracted with chloro-
form to remove any unreacted phenol and the extracts were dried over anhy-
drous MgS04, concentrated to <3 ml, and subjected to gas chromatographic
analysis. Prior to gas chromatographic analysis, the extract containing
the one catechol tested was silylated as described in the section Gas
Chromatographic Analysis.
-------
In an experiment designed to test the efficiency of phenol oxidation
in the presence of a bleaching liquor, a solution of tetrachloroguaiacol
(24 mg) was diluted with varying amounts of spent caustic extraction liquor
and reacted with 13 mg of chlorine dioxide at room temperature until the
latter was depleted. The residual phenol was recovered by extraction with
chloroform and determined by gas chroma tographic analysis as described
above.
Ozone
Ozone was generated with a Welsbach Model LOA-1 PSI corona generator
using oxygen as the gas source. Two 1-liter gas washing bottles connected
in series were attached to the generator. Both bottles were equipped with
fritted glass diffuser tubes to provide a small bubble size. The rate of
ozone generation was determined by placing a 5% KI solution in one of the
bottles and measuring the amount of iodine liberated in a given time inter-
val following introduction of the ozone stream.
The phenols (0.207 mmole) were each dissolved in 500 ml of a 20%
ethanol (v/v) solution containing 3 g of a buffer mixture comprised of 90 g
of KH2P04 and 8 g of KoHP04. This buffer concentration produced an initial
pH of approximately 6. 15 which stayed virtually constant throughout the re-
action period. The phenol solutions were ozonized at ambient temperature
(22 +_ 2°C) for 30-second intervals initially and for longer periods near
the end of the treatment. After each reaction interval, the ozone was di-
verted to the atmosphere through a t- joint in the ^ assembly and a 3-ml sample
was withdrawn from the reactor and analyzed by ionization difference spec-
troscopy
As a check on the accuracy of the above method, the residual phenol
content of a few ozonized solutions was measured by gas chromatography. In
these tests, 200 ml of a 4.14 x 10~4 M solution of each model (in 20%
ethanol) containing 1.2 g of the same buffer mixture described above was
divided into two equal portions. One 100-ml portion was set aside as a con-
trol and the 100-ml portion was reacted with ozone for one minute at a flow
rate of either 2.64 or 3.6 mg/min. At the conclusion of the reaction, both
the control and ozonized samples were extracted with 70 ml of chloroform.
The solvent phase was dried over anhydrous MgSO/i, evaporated to ^3 ml and
silylated.
In the treatment of fully etherified phenols, a new 500-ml sample was
used for each ozonization reaction period. The ozonized product mixtures
in this instance were extracted with 250 ml of chloroform. The extracts
were dried over anhydrous MgS04, concentrated to ^3 ml, and analyzed by gas
chromatography.
Tetrachloroguaiacol was ozonized in an alkaline medium. A series of
solutions was prepared, each member of which contained a 0.1 mmole amount of
this phenol dissolved in 100 ml of a buffer solution 0.08 N in Na2C03 and
0.02 N in NaHCOo (pH 10.3). These solutions were reacted at a flow rate
ranging from O.b to 0.8 mg/min. At the end of this treatment, the product
10
-------
mixtures were acidified and extracted with ether. The solvent layers were
dried over anhydrous NagSO^ concentrated to dryness, and the residues re-
dissolved in CHC13 prior to gas chroma tographic analysis.
CHEMICAL TREATMENT OF SPENT CHLORINATION AND CAUSTIC EXTRACTION LIQUORS
Chlorine-Containing Oxidants
Elemental chlorine (i.e., chlorine water), hypochlorous acid, and chlo-
rine dioxide were applied to spent chlorination liquor and allowed to react
to complete exhaustion at room temperature. In the application of hypo-
chlorous acid, the reaction mixture was buffered to pH 4.75 by the addition
of acetate. Sodium hypochlorite was applied to spent caustic extraction li-
quor at 60°C and the reaction was continued until the oxidant was completely
consumed.
Ozone and Hydrogen Peroxide
All ozone treatments of the spent liquors were performed at room tem-
perature using the ozone generator described above. A 250-ml sample of
spent chlorination liquor (pH 2) was reacted with ozone introduced at a flow
rate of 4.65 mg/min. The first 9.3 mg of ozone applied were completely con-
sumed by the liquor under the test conditions. The pH of a 250-ml sample of
spent chlorination liquor was buffered at 8.0 by the addition of 4 g of
K2HP04. The ozone flow rate was 4.75 mg/min. and the first 16.6 mg of ozone
were completely absorbed by the liquor. A 150-ml sample of spent caustic
extraction liquor was reacted with ozone (flow rate, 4.75 mg/min.) after
the pH was first adjusted to 8.0 by the addition of 2 g of KH2P04. Absorp-
tion of ozone was complete up to the addition of 143 mg (^30 min.).
Spent caustic extraction liquor was buffered (^pH 10.5), sodium sili-
cate and magnesium sulfate added, and the mixture reacted with hydrogen
peroxide at 70°C. Consumption of peroxide occurred slowly and not until
after 24 hours was the applied dosage (4 meq/150 ml of liquor) totally
consumed.
Sodium Hydroxide
The pH of spent chlorination liquor was adjusted to various levels by
the addition of sodium hydroxide and maintained at the initial level by the
addition of more alkali as required. After a predetermined reaction period
at room temperature, the phenol and organically bound chlorine contents and
the acute toxicity of 'he solutions were determined, as described in later
sections.
Aluminum Sulfate, Lime and Activated Carbon
Aluminum sulfate (1.5 g) was added to spent caustic extraction liquor
(220 ml) and allowed to react for one hour at room temperature (pH 5.5).
The precipitate was removed by centrifugation, dissolved by addition of HC1,
and subsequently dialyzed through cellophane. The supernate from the cen-
11
-------
trifugation and the retentate from the dialysis were diluted to the volume
of the original test sample, tested for phenol content, and bioassayed for
acute toxicity.
Lime treatment of spent caustic extraction liquor consisted of the addi-
tion of a saturated solution of slaked lime (Ca(OH)2) containing 2.2 g of
the latter to a 220-ml sample of the liquor (pH 11). The reaction mixture
was stirred for several hours at room temperature and the precipitate re-
moved by centrifugation. Carbon dioxide was introduced into the supernate
to precipitate the calcium as CaCOo. The combined precipitates were redis-
solved by the addition of HC1 and dialyzed through cellophane for two days
with frequent changes of water. The volumes of supernate and dialysis re-
tentate were adjusted to the original sample volume and the resulting solu-
tions analyzed for phenol content and acute toxicity.
In those instances where spent caustic extraction liquor was treated
sequentially with lime and activated carbon or a chemical oxidant such as
chlorine dioxide, ozone or hydrogen peroxide, a solution of slaked lime
(15 g/1) was added to the liquor; the mixture stirred for 20 minutes, and
then allowed to stand for one hour. The precipitate was removed by filtra-
tion through No. 41 Whatman filter paper. Subsequent treatment of the fil-
trate was proceeded by pH adjustment to 9.4 by the addition of carbon diox-
ide and to 5.0 with HC1. The CaCO^ formed after C02 addition was removed
by filtration.
Treatments of spent caustic extraction liquor with activated carbon
were performed using a sample of Nuchar S-A provided by Westvaco Corporation.
Samples of the liquor were treated in a batch process with the activated
carbon for 18 hours with continuous stirring. At the end of this period,
the carbon was removed by filtration through No. 41 Whatman paper.
CHLORINATION AND CAUSTIC EXTRACTION STAGE MODIFICATIONS
In the sequential C102/C12 treatment, 1.0% of C102 and 2.6% of Cl?
were applied to the kraft pulp at room temperature and a consistency of 3%.
These combined amounts were equivalent to 5.25% chlorine and the proportions
corresponded to a 50% replacement of chlorine with chlorine dioxide. The
bleaching was continued until the chlorine was exhausted (final pH, 2.1).
Chlorine dioxide (2%) alone was applied to the pulp under the same condi-
tions used for the sequential treatment (final pH, 2.85).
Peroxide treatment of a chlorinated kraft pulp consisted of applying
0.5% H202 (10% cons., 70°C, 3 hr) after first adjusting the pH of the bleach
to 11.2. Hypochlorite treatment as a replacement of the conventional caus-
tic extraction stage consisted of applying 2% of sodium hypochlorite to the
pulp (10% cons., 45°C, 1.5 hr, initial pH V12.5). In each case, the oxi-
dant was consumed to complete exhaustion at the conclusion of the reaction
period.
12
-------
The alkali/oxygen liquor was prepared by treatment of a sample of sou-
thern pine kraft (Kappa No. 32.9) with 2.5% of oxidized white liquor
(as NaOH) at 80 psig oxygen at 27% consistency for 30 minutes at 110°C.
The liquor used in the acute toxicity bioassay was obtained by diluting
the 27% consistency pulp to 4% and centrifuging off the liquor. The total
organic carbon (TOC) content of the liquor was 405 ppm.
BIOLOGICAL TREATMENT OF PHENOLS AND CHLOROPHENOLS
Aliquots of stock solutions of the various phenols (3 mg/ml in 40%
ethanol) were chosen, such that after dilution to the final volume (100 ml),
the concentrations of compounds having one or less chloro substituents and
those having two or more chloro groups were 10 and 50 ppm, respectively.
One ml of each fungal spore suspension (10) was used as inoculum. For each
of the phenols tested, a control sample containing no fungus was set up
and inoculated with sterile distilled water. Two replicates were prepared
at each concentration level and also for the control.
Flasks containing the solutions described above were placed on a hori-
zontal shaker and incubated at 28°C for one week. Following the incubation
period, the contents of each flask were filtered through a Seitz filter and
the filtrates of replicate tests combined to yield approximately 200 ml of
solution. Identical volumes of the filtrates were extracted with chloro-
form, and the latter was concentrated in vacuo to 3 ml and dried over anhy-
drous magnesium sulfate. The dried solvent extract was subsequently analyzed
for residual phenol by gas chromatography as described in a later section.
BIOLOGICAL TREATMENT OF SPENT CHLORINATION AND CAUSTIC EXTRACTION LIQUORS
The pH of both the spent chlorination liquor (SCL) and spent caustic
extraction liquor (SCEL) were adjusted to 7.0 with sodium hydroxide and
hydrochloric acid, respectively, over a two-hour period. Samples of the
liquors were sterilized by filtration through a Seitz filter fitted with
a #6 sterilizing pad. One hundred-mi aliquots of the filtrates were sub-
sequently transferred to sterile 250-ml Erlenmeyer flasks. One sample of
each liquor was inoculated with 5.0 ml of sludge (obtained from a bleached
kraft mill waste treatment plant), yeast or bacterial cell suspension. In
addition, one flask was left as an uninoculated control. The flasks were
placed on a horizontal shaker and incubated at 28°C for 7 days. Following
the incubation period, the flasks were filtered using a Seitz filter fitted
with a sterilizing pad. The filtrates of both the test and control samples
were then bioassayed for acute toxicity using Daphnia magna as the test
organism. The preparation of the fungal and bacterial inocula and the com-
position of the buffer and nutrient solutions are individually described
below for the treatment with sludge (mixed microbial population) yeast
(Candida utilis) and bacterium.
13
-------
Sludge (mixed microbial population
In the initial experiment,asparagine, at a concentration of 0.2 g/l>
was added as a nitrogen source and monobasic potassium phosphate at a con-
centration of 0.01 g/1, served as a phosphorous source. A second experiment
was conducted in which no supplemental nutrients were added to the liquors.
In this experiment, monobasic potassium phosphate (0.10 g/1) was added to
the spent caustic extraction liquor (SCEL) and dibasic potassium phosphate
(0.10 g/1) was added to the spent chlorination liquor (SCL).
Yeast (Candida utTMs)
Monobasic potassium phosphate (0.10 g/1) was added to the spent caus-
tic extraction liquor and dibasic potassium phosphate (0.10 g/1) was added
to the spent chlorination liquor. Preliminary experiments showed that
C_. utilis could not grow in spent bleaching liquors in which no supplemental
nutrients were added. Therefore, glucose at a concentration of 1.0 g/1 was
added to the liquors.
Inoculum was prepared according to the following procedure: Several
loopfuls of a 3-day-old culture of C_. utilis (NRRL Y-900) growing on yeast
morphology slants were dispersed in 50 ml of sterile distilled water in a
125-ml Erlenmeyer flask. Enough yeast cells were added to yield a cell
suspension having an absorbance reading of 0.32 at 540 nm measured on a
Spectrom'c-20 spectrophotometer.
A Bacterium
Mono and dibasic potassium phosphate were added to samples of the
filter-sterilized SCEL and SCL, respectively, at a concentration of 1.0 g/1.
In addition, glucose at a concentration of 1.0 g/1 was added to both liquors.
Inoculum was prepared as follows: Several loops of a 4-day-old bac-
terial culture (Ziobro, culture 2-B) growing on nutrient agar slants were
dispersed in 20 ml of sterile distilled water. Enough cells were added to
yield a cell suspension having an absorbance reading of 0.45 at 540 nm, mea-
sured on a Spectronic-20 spectrophotometer.
ACUTE TOXICITY TESTS
Acute toxicity tests were conducted on the various materials using
either first instar or 1-3 day old animals. The dilution water used in the
tests was DM2 prepared from doubly distilled water (5). This solution had
a pH of 7.0 and was aerated to saturation prior to use. Stock solutions
of the various test compounds were prepared by dissolving the appropriate
amount of the material in a suitable solvent (DMSO) and adding the desired
volume of DMSO/toxicant solution to the dilution water. Spent liquor samples
were first adjusted to pH 7.0-7.5 through the addition of NaOH or HC1 imme-
diately prior to initiating the test. The toxicant solutions were then
14
-------
serially diluted to the required concentration levels. Each test consisted
of 5-6 concentration levels with dilution factors of 0.5-0.9 between each
concentration and 10 organisms per concentration. All tests were conducted
at 17 +_ 1°C in a constant temperature bath. Lack of gross movement was used
as the criterion for death. Concentration-response patterns were either sub-
jected to probit analysis (11) using the Bio-Med computer at Syracuse Uni-
versity or through the use of tables (12,13) based on the moving average
method of Thompson (14).
Brine Shrimp
Before initiating each assay, the spent liquors were adjusted to pH
7.4-7.6 with either NaOH or HC1 and the salinity was adjusted to that of
the standard sea water formulation (22 parts/thousand) by adding NaCl. The
organisms were exposed to test volumes of 50 ml in Carolina culture dishes
having dimensions of 3.5 x 1.5 inches. Each liquor was tested at five con-
centrations with 20 organisms per concentration and a dilution factor of
0.5 between concentrations. During the tests, the pH ranged from 7.2 to 7.6
and temperature ranged from 20 to 22°C. Mortality was recorded after 24
hours and mortality-response patterns were subjected to probit analysis (11).
Fungi
The basal medium for the bioassy consisted of a solution of 3.00 g of
glucose, 1.25 g of peptone, and 1.25 g of yeast extract in 1000 ml of dis-
tilled water (pH 6.5). Each test compound was dissolved in 40% ethanol and
maintained at a stock concentration of 3.0 mg/1 (3000 ppm). The stock solu-
tion was serially diluted to concentration levels of either 200, 100, 50,
10 and 0 ppm (control) or 50, 25, 10, 7.5 and 0 ppm (control),depending upon
the number of chlorine atoms in each compound. After dilution of the test
solution with the basal medium and addition of the spore suspension, the
final volume was 20 ml and the concentration of ethanol, 2%.
Five replicate samples of eight phenols were prepared for assay of
Aspergillus fumigatus. Two replicate samples of 17 phenols were tested
against the six additional species of fungi used in the investigation. The
test solutions were placed on a horizontal shaker and incubated at 28°C for
3 days. Growth was measured by visual observation and by dry weight deter-
mination of the fungal mycelia.
Acute toxicity bioassays of spent chlorination and caustic extraction
liquors were also performed,using six common fungi as the test organisms.
Prior to testing, the pH of the two liquors was adjusted to near neutrality
by the addition of NaOH or HC1. Duplicate samples of each of the whole li-
quors and samples diluted to 1/2, 1/4 and 1/10 of their initial volumes
were added to a glucose-peptone-yeast medium and one ml of the various spore
suspensions added to each. The samples were placed on a horizontal shaker
and incubated for 7 days at 28°C. Growth of the various cultures was esti-
mated visually.
15
-------
Alga
The green alga, Chlorella pyrenoidosa, was grown in modified Jack
Myer's medium for Chlorella and Scenedesmus in acid-clean flasks on a rotary
shaker (150 rpm) at 18°C and under an illumination of 650 foot-candles of
continuous light. Inoculum was prepared as described in the EPA publication,
"Algal Assay Procedure: Bottle Test" (15), except that a larger amount of
inoculum was used (105 cells/ml). Test chemicals were dissolved in ethanol
and the final concentration of ethanol in the culture medium was 0.1%. The
absorbance of the alga in the growth medium, measured spectrophotometrically
at 650 nm, was used as a measure of the amount of growth.
The initial concentrations of the phenols tested were 100, 10, 1, and
0.1 ppm or 10, 1, 0.1, and 0.01 ppm depending on the anticipated degree of
toxicity. The range of toxicity was determined to be from the level of con-
centration at which no growth occurred to that where some growth occurred,
even if it was reduced from that of the control. Toxicity levels were de-
lineated more precisely by selecting and testing a variety of concentrations
from within the previously determined toxicity range. The absorbance of the
alga in the growth medium (measured spectrophotometrically at 650 nm) was
used as a measure of the amount of growth at the end of 12 to 14 days.
Toxicity tests of spent chlorination and caustic extraction liquors
were performed in a manner similar to those described above for the phenols
with the following modifications: The liquors were with NaOH or HC1, buf-
fered with phosphate and sterilized with a Seitz filter prior to being
tested. The intense color of the liquors prevented effective use of spec-
trophotometry to record the results. As an alternative approach, direct
counts from a hemocytometer were used. Liquor dilutions of 1:2, 1:4 and 1:8
were tested.
Duckweed
One half strength of Hutner's medium plus sucrose was used for growing
Lemna perpusilla (duckweed). The test phenols were dissolved in 4% DMSO.
2,4-Dichlorophenol; and 4,5-dichloroguaiacol were tested at concentrations
of 20, 15, 10 and 5 ppm. 4,5-Dichlorocatechol and 2,4,6-trichlorophenol
were tested at concentrations of 10, 5, 1, and 0.1 ppm. Each flask con-
taining 50 ml of the culture medium and the phenol solution was inoculated
with a 3-frond colony of U perpusilla. The flasks were incubated at 25°C
under a constant illumination of 150 foot-candles for 10 days. The number
of fronds was counted using a stereoscopic microscope.
GAS CHROMATOGRAPHIC ANALYSIS
Dehydroabietic Acid
Pine kraft pulp (Kappa No. 25.5) was washed with water, dewatered to
36.5% consistency and extracted consecutively for 24-hour periods with
8000-, 5000-, and 5000-ml amounts of sodium hydroxide solution (pH 10.5)
at room temperature. After each treatment, the pulp was filtered and washed
16
-------
with 5 x 500 ml of water. The three filtrates were concentrated separately
to ^2500 ml, acidified to pH 2 and extracted with CHC1,.
O
The chloroform extracts were reacted with CH2N2 to convert any carbox-
ylic acids contained therein to their corresponding methyl esters. The
derivatized esters were then subjected to gas chromatographic analysis using
an F and M Model 720 gas chromatograph fitted with a 40" x 1/4" O.D. stain-
less steel column filled with 3% SE-30 on 80-100 mesh acid-washed, silylated
Chromosorb G. The identification of methyl dehydroabietate was made on the
basis of spiking with the authentic ester. The yield of the ester was de-
termined by comparison of its gas chromatographic peak height with that
corresponding to a standard.
In another experiment, a sample of the kraft pulp was extracted with a
sodium hydroxide solution (pH 12.4) for 60 minutes at 70°C. The pulp was
filtered, washed with water, and re-extracted under the same conditions.
The filtrates from the two extractions were concentrated separately, acidi-
fied to pH 2, and extracted with chloroform. The chloroform extracts were
derivatized and analyzed gas chromatographically for methyl dehydroabietate
as described above.
Phenols and_Chlorophenols
Gas chromatographic analysis of the phenols and chlorophenols in the
various product mixtures was performed using either an F and M Model 720
(thermal conductivity detector) or a Varian Aerograph Model 1700 (flame
iom'zation detector) instrument. In all cases, separations were achieved
using stainless steel or aluminum columns containing 3% SE-30 on 80-100
mesh acid-washed, silylated Chromosorb G. The columns were heated, both in
the isothermal and programmed modes, and the temperatures ranged from
150-250°C.
Monohydric phenols and guaiacol (2-methoxyphenol) derivatives were
chromatographed without derivatization. Catechol (1,2-dihydroxybenzene)
derivatives were, on the other hand, converted to di-trimethylsilyl ethers
prior to injection by reaction with an excess of bis(trimethylsilyl)
acetamide ("BSA"). Quantitative estimation of the amounts of phenols was
made by comparison of peak height or area with those of standards.
ANALYTICAL PROCEDURES
BODc analysis was performed according to the procedure outlined in
"Standard Methods for the Examination of Water and Wastewater" (16). Total
organic carbon (TOC) measurements were performed on a Beckman Model 915
Total Organic Carbon Analyzer. Color measurements were based on the cobalt-
platinum scale described by Brown (17).
Acidic groups were measured by dissolving the sample in a pyridine-
water-butanol mixture and titrating with alcoholic sodium hydroxide using
a pH meter fitted with a glass-calomel electrode system. Total hydroxyl
content was estimated using the procedure described by Siggia (18) modified
17
-------
by performing the acetylation at room temperature for 17 hours.
Molecular weights were determined using a Hewlett-Packard Model 302
vapor pressure osmorneter. A non-aqueous probe thermostatted at 65°C was
used and the samples were dissolved in methyl cellosolve. Benzil was used
as a calibration standard.
Chloride was measured by potentiometric titration with standard AgNOo
solution as described by Shiner and Smith (19) and modified by Braddon and
Dence (20). Organically bound chlorine was determined by applying an appro-
priate size sample (adjusted to pH 8 if not initially alkaline) on a 5.5 cm
No. 1 Whatman filter paper while simultaneously evaporating the water with
the aid of a hair dryer. The dried filter paper was burned in a Schoniger
combustion flask containing 50 ml of distilled water. Potentiometric titra-
tion of the resulting solution with standard AgN03 solution provided the
total chlorine (inorganic and organic) in the sample. The organically
bound chlorine was calculated by subtraction of the inorganic chloride con-
tent (see above) of the original sample from this value.
Chlorine, chlorine dioxide, ozone and hydrogen peroxide were determined
iodometrically. In the case of hydrogen peroxide, the reduction was cata-
lyzed by the addition of ammonium molybdate.
lonization difference spectra were recorded on a Unicam Model SP800A
recording spectrophotometer. Semimicro matched quartz cells having either
a 1- or 10-mm light path were used.
18
-------
SECTION 5
RESULTS AND DISCUSSION
ACUTE TOXICITY OF PHENOLS
The phenols and phenol ethers used in the investigation are listed in
Table 2. Their selection was based on a number of criteria, including
mainly structural identity or similarity to compounds previously identified
in spent chlorination and caustic extraction liquors and a desire to
achieve a broader understanding of the structure-toxicity relationships
pertaining specifically to these types of phenols and chlorophenols. The
phenols listed in Table 2 thus consist principally of the various chlorine-
substituted derivatives of mono- and dihydric (i.e., catechol) ring types.
Fully methylated structures were utilized only to a limited degree and that
solely for the purpose of gauging the effect of a free phenolic hydroxyl
group in a particular measurement or treatment.
In the initial phase of the project, the phenolic compounds studied
contained a methyl substituent at the position para to a phenolic hydroxyl
group (i.e., at the side chain position). Later, these compounds were
abandoned in favor of structures containing a chloro substituent at the
para position since the latter types were concurrently being identified
in spent chlorination and caustic extraction liquors (21-23). The general
behavior of the two aforementioned structural types with respect to reac-
tivity and acute toxicity probably do not differ greatly, especially when a
comparison is made between compounds having identical numbers of chloro
substituents on the same type of phenolic ring.
Effect on Fungi
A detailed study was made of the effect of eight phenols on Aspergillus
fumigatus. The plots shown in Figures 1 and 2 indicate the effect of in-
creasing concentrations of the phenols on the amount of fungal growth.
These plots reveal that inhibition of fungal growth increased in proportion
to the number of chloro substituents attached to the phenolic ring. 2,4,6-
Trichlorophenol (VI, Table 2) had the greatest effect of the phenols tested
in reducing the growth of the fungus.
Expansion of the testing to include five additional fungi and a total
of 17 phenols produced the results shown in Table 3. Each of the fungi
used in the testing had been previously found to be present in water sampled
at various sites of a bleached kraft mill waste treatment plant. Although
19
-------
TABLE 2. PHENOLIC COMPOUNDS USED IN THE INVESTIGATION
OCH
II R1=R3=H; R2=CH3
II R1=R3=H; R2=C1
IV Rj=R2=Cl; R3=H
V R1=R3=C1; R2=CH3
VI R1=R3=R3=C1
VII R!=R2=R3=H
VIII R^Cl; R2=R3=H
IX Rj=R3=H; R2=C1
X R^R^Cl; R3=H
XI R1=R3=C1; R2=H
XII Rj=H; R2=R3=C1
XIII R1=R2=R3=C1
XIV R^R^H; R2=R3=C1
XV R1=R2=R3=C1; R^H
XVI R1=R2=R3=R,t=Cl
XVII R1=R2=R3=H
XVIII Rj=R3=H; R2=C1
XIX Rj=R2=Cl; R3=H
XX R,=K=K=LI
XXI R:= R^-H; R2=R3=C1
XXII R1=R2=R3=C1; R^H
XXIII R1=R2=R3=R1(=C1
VYT\/ D D D U
AAJ.V t\j-r\2-i\3-n
XXV R^CI; R2=R3=H
XXVI RX=H; R2=R3=C1
XXVII R1=R2=R3=C1
20
-------
XVII
2(5o"
10 50
CONCENTRATION OF PHENOL, ppm
Figure 1. Effect of 2,4,6-trichlorophenol (VI) and three chlorocatechols
on the growth of Aspergillus fumigatus. (Values for dry weight
are based on five replicates.Vertical bars represent the
standard error of the replicates.)
21
-------
10 50 100 150 200
CONCENTRATION OF PHENOL, ppm
Figure 2. Effect of creosol (VII) and three chloroguaiacols on the growth
of Aspergillus fumigatus. (Values for dry weight are based on
five replicates. Vertical bars represent the standard error
of the replicates.)
22
-------
TABLE 3. MINIMUM CONCENTRATIONS fTN PPM\ OF PHFNOI S REOUTRFD TO.PREVENT GROWTH OF VARIOUS FUNGI
IN3
CO
Phenol
Fungus
Trichoderma
koningii
Aspergillus
niger
Aspergillus
fumigatus
Paecilomyces
varioti
Cladosporium
her ba rum
Penici Ilium
variabile
oH
I
>b200
> 200
> 200
> 200
> 200
> 200
OH
II
>200
>200
>200
>200
>200
>200
oH
V
25
50
25
50
25
50
ci
OH
VI
25
10
10
10
7.5
25
oH
VII
>200
>200
>200
>200
>200
>200
CH3
oH
VIII
100
100
100
100
50
200
CM3 CH3
oM ^ 0H
IX X
100 25
200 25
200 25
200 50
50 25
100 25
In ppm
^Indicates growth was evident at the highest concentration tested.
(continued)
-------
TABLE 3 (continued)
ro
Phenol -.,
CHj
{of
Fungus f ^y*oa
o**
XI
Trichoderma bKn
koningii bf)
Aspergillus
niger > 50
Aspergillus . cn
fumigatus ou
Paecilomyces ,-n
varioti "u
Cladosporium 75
herbarum
Penicillium ,-n
vari a bile DU
;t.
OH
XII
25
25
25
25
25
50
CH3
oM
XIII
>50
25
25
25
10
>50
CH3
OH
XVII
100
>200
>200
>200
100
>200
CHi ^3
oH otf
XVIII XIX
100 >50
200 >50
200 >50
200 >50
100 25
100 >50
.^^
OH OCMj ^
XX XXIV
50 >200
50 >200
25 >200
25 >200
7.5 100
25 >200
/* fcl
5
oc«3 >
XXV
200
>200
>200
>200
100
>200
b >Indicates growth was evident at the highest concentration tested.
-------
some variations occurred in the response of the isolates, the findings gen-
erally duplicated those noted above for Aspergillus fumigatus with respect
to the effect of an increasing number of chloro substituents on the repres-
sion of fungal growth. An exception to this general trend was noted in the
case of the veratrole derivatives (XXIV - XXVII, Table 2) where increasing
substitution of chlorine appeared to exert no noticeable effect on fungal
growth.
The data in Table 3 once again underscore the strongly inhibitory
effect of 2,4,6-trichlorophenol. The potency of the latter compound as a
toxicant is well-documented and it is found on the EPA's list of hazardous
chemicals. At lower levels of chlorine substitution, the effect of the
phenols on fungal growth was similar, provided the comparison was made be-
tween compounds having the same number of chloro substituents.
Among the six fungi tested, Cladosporium appears to be the least
tolerant to the applied phenols. In general, however, this fungus and
others are more resistant to phenols of the type tested than are fish, in-
vertebrates, and other aquatic organisms.
Effect on the Alga Chlorella pyrenoidosa
The toxicity range of selected phenols toward an alga, Chlorella
pyrenoidosa, was determined and the results obtained are presented in
Table 4.
TABLE 4. TOXICITY RANGES OF SELECTED PHENOLS FOR CHLORELLA PYRENOIDOSA
Phenol
Concentration, ppm
Growth Present3 Growth Absent
Catechol (1,2-dihydroxybenzene)
2,4-Dichlorophenol (IV)
4,5-Dichlorocatechol (XXI)
4,5-Dichloroguaiacol (XIV)
Trichlorocatechol (XXII)
20
8
2
1
0.1
40
10
4
10
1
Determined with a Spectronic-20 spectrophotometer and checked by cell
count using a hemocytometer.
As in the previous tests with fungi, the enhancement of toxicity re-
sulting from the introduction of chloro substituents on the phenolic ring
is clearly evident. On the basis of the limited data in Table 4, the phe-
nol concentration levels at which the Chlorella ceased to grow are consider-
ably lower than those for similar compounds tested against fungi. For ex-
ample, a concentration of 1 ppm trichlorocatechol completely inhibited the
growth of Chlorella whereas in the case of the six fungi tested (cf. Table
25
-------
3), the corresponding range was 7.5-50 ppm. Because of its greater sensi-
tivity, Chlorella is therefore more suitable for use as a bioassay micro-
organism in the testing of low concentrations of phenols.
Effect on the Vascular Plant Lemna perpusilla (Duckweed)
Duckweed is a small aquatic plant that is widely distributed through-
out the United States. Its utility as a test organism in acute toxicity
bioassays was examined by determining the effect of various concentrations
of four phenols on its growth. The results of these tests are shown in
Table 5.
2,4,6-Trichlophenol (VI) was again found to cause the greatest sup-
pression of growth in the limited series of phenols tested and, at a concen-
ifafiippmQfaBdppm, it totally inhibited growth. The suppression of growth
thereafter decreased in the order of 4,5-dichlorocatechol (XXI), 4,5-dichloro-
guaiacol (XIV), and 2,4-dichlorophenol (IV).
Based on a comparison of a limited number of phenols, therefore, Lemna
appears to be more sensitive to such compounds than fungi, but less sensi-
tive than Chlorella. However, duckweed is much easier to work with in the
laboratory and as a biological test organism, it is less expensive to use.
Effect on Daphnia magna
The response of Daphnia magna to chlorophenols similar to or identical
with those identified in spent chlorination and caustic extraction liquors
is shown by the data compiled in Table 6. In a recent investigation,
Durkin (24) has found that several of the phenols listed in this table
interact with themselves, with certain unidentified components of spent
chlorination liquor, and with sodium chloride in low concentrations, thereby
altering somewhat the magnitude of the values reported in the table. This
finding has important implications regarding the validity of attempts to
account for the total toxicity of a particular liquor through the sum of
effects of individual toxic components; nevertheless, the finding does not
invalidate the more pronounced trends indicated by the data in Table 6.
One of the more consistent trends shown by the data in Table 6 is that
the acute toxicity of the various chlorophenols increased in proportion to
the number of chloro substituents on the phenolic ring. This same trend was
also noted above in the discussion of the response of plant test organisms
to many of the same phenols. As a class, and with respect to acute toxicity,
the chlorocatechols (XVIII, XX-XXIII, Table 6) seem to have the greatest
effect on the Daphnia. Somewhat surprisingly, 2,4,6-trichlorophenol (VI),
which ranked as having a very toxic effect on the plant test organisms,
proved to be somewhat less toxic than 3,4-dihydroxy-2,5,6-trichlorotoluene
(XX) and 3,4,5-trichlorocatechols (XXII) in the tests with Daphnia. Chloro-
catechols did, however, have a pronounced effect on retarding the growth of
the plants exposed to them (see Tables 3-5).
26
-------
TABLE 5. EFFECT OF FOUR PHENOLS ON THE VEGETATIVE GROWTH OF LEMNA PERPUSILLAg
Concentra-
tion, ppm
Number of Fronds, % of Control
ro
Phenol
Cl
0.0 (control) 0.1 1
XIV
XXI
VI
100
100
100
95 95 20
94 31
10 15 20
IV 100 -b 70 29 10 7 0C
77 55 20 10
?Ten days of incubation
°Not tested
No growth
-------
TABLE 6. ACUTE TOXICITY OF VARIOUS PHENOLS TO
Compound
24 hr LC5Q or
(95% confidence interval),
y moles/1
2,4-Dichlorophenol IV
2,4,6-Trichlorophenol VI
4-Methylguaiacol VII
6-Chloro-4-methylguaiacol VIII
5-Chloro-4-methylguaiacol IX
5,6-Dich1oro-4-methy1guaiacol X
2,5-Dichloro-4-methylguaiacol XI
2,6-Dichloro-4-methylguaiacol XII
2,5,6-Trichloro-4-methylguaiacol XIII
4,5-Dichlorogualacol XIV
4,5,6-Trichloroguaiacol XV
Tetrachloroguaiacol XVI
3,4-Dihydroxytoluene XVII
3,4-Dihydroxy-6-chlorotoluene XVIII
3,4-Dihydroxy-2,5,6-trichlorotoluene XX
4,5-Dichlorocatechol XXI
3,4,5-Trichlorocatechol XXII
Tetrachlorocatechol XXIII
4-Methylveratrole XXIV
5-Chloro-4-methylveratrole XXV
2,6-Dichloro-4-methylveratrole XXVI
2,5,6-Trichloro-4-methylveratrole XVII
51.9 (40.2-90.1)°
13.7 (11.4-16.0)b
150 (76.1-1470)
49.9 (16.4-xc)
177 (19.5-xC)
13.0 (6.77-75.4)
25.5 (13.4-212)
14.6 (8.79-46.0)
7.45 (2.74-xC)
98.3 (79.0-135)°
22.0 (15.6-26.5)°
4.96 (4.19-6.21)°
27.6 (21.0-36.5)
13.8 ( 7.6-25.1)
4.90 (4.46-5.30)
6.64 (4.76-7.81)°
3.39 (2.92-3.86)°
2.23 (1.25-2.80)°
1200 ( 557-2640)
46.3 (20-xc)
20.7 (6.89-xc)
2.36 (1.08-3.53)
0.424
0.206
0.403
0.905
2.63
0.477
0.438
0.290
1.16
0.309
0.261
0.230
_d
_d
_d
0.240
0.222
0.319
0.472
0.520
0.873
0.262
aAs discussed by Finney (11), the "g" value can be used as an index of the
quality of the bioassay, i.e., how well the observed mortality pattern cor-
"g"
values
responds to the best estimate of the probit line. In general,
of 0.4 or less indicate good agreement.
LC5Q values; all others, £50. The LCsp values are based on cardiac arrest
whereas the £50 values relate to immobility as response criteria.
*
'Meaningful upper limit not available due to low mortality at highest con-
centration tested.
j
Not determined.
28
-------
Jhe isomeric dichloro-4-methylguaiacols (X-XII) show statistically in-
significant differences in their Upvalues, a finding generally duplicated
when fungi were used as test organisms (see Table 3). On the other hand,
the effect of increasing numbers of chloro substituents that was apparent
in increasing the toxicity of the veratroles (XXIV-XXVII) to Daphnia, was
not observed in the treatment of the fungi with representatives of this
class. Veratrole derivatives are atypical of the phenols detected in
spent chlorination and caustic extraction liquors; the observed difference in
the response of the fungi and Daphnia toward veratrole derivatives probably
has little significance for our purposes in the present investigation.
ACUTE TOXICITY OF WHOLE AND FRACTIONATED SPENT CHLORINATION LIQUOR (SCL)
AND CAUSTIC EXTRACTION LIQUOR (SCEL)
In this segment of the investigation, the toxicity of the whole spent
chlorination and caustic extraction liquors was examined using a fungus
(Aspergillus fumigatus) and an aquatic invertebrate (Daphnia magna) as test
organisms. The validity of the use of fungi in acute toxicity bioassay has
been critically discussed by Windaus and Petermann (25). Daphnia magna
was selected on the basis of its comparative simplicity in bioassay appli-
cations and because it has been shown to give results that have a good cor-
relation with bioassay data that is based on the utilization of fish (26).
In an effort to identify and define more thoroughly the chemical and
physical characteristics responsible for toxicity, the whole liquors were
subsequently fractionated by various means and the individual fractions sub-
jected to acute toxicity bioassay and, to a lesser degree, chemical analy-
sis. In this phase of the investigation, the contributions of the various
fractions to the toxicity of the whole liquors were calculated using the
"toxic unit" method described by Sprague (27). As defined by Sprague and
applied throughout the course of this work, the toxic unit content was cal-
culated as follows:
Tnv-iv iin-j+c (r ii ^ - Total Organic Carbon (TOC) of Test Sample (ppm)
loxic units u-u.j LC50 expressed as TOC (ppm)
Stability of Spent Chi ori nation Liquor (SCL) and Caustic Extraction Liquor
Before subjecting the spent chlorination and caustic extraction liquors
to various fractionations, the stability of those liquors with respect to
color, BOD,-, organically bound chlorine, and acute toxicity on storage was
determined in order to evaluate the changes that would occur solely as a
result of a particular form of treatment.
tests were performed on samples of the spent chlorination and caus-
tic extraction-liquors that ranged in age from 1 to 26 days. No significant
trends were observed in either case, and the average values for spent chlor-
ination and caustic extraction liquors were 125 and 437 mg 02/1, respectively.
As shown by the plots in Figure 3, the color of both caustic extraction
liquors decreased at a greater rate initially than it did after prolonged
standing. The difference in the color of the two extraction liquors
29
-------
32--
GO
o
K>
I
O
X
CL
o>
E
CO
o:
3
8
28
24
20(
16
12
8-
4-
SCEL (CHLORINATED PULP PRE-WASHED
WITH 0.01 N HCI)
SCEL-(CHLORINATED PULP PRE-
WASHED WITH TAP WATER )
SCL
8
12
DAYS
16
20
24
Figure 3. Effect of age on the color of spent chlorination and caustic extraction
liquors stored at room temperature.
-------
is attributed to the greater retention of solubilized material in the chlo-
rinated pulp washed with dilute acid as compared to the one washed with tap
water. In contrast, the color of the spent chlorination liquor remained
virtually constant throughout the test period.
f
The stability of the organically bound chlorine in the two liquors is
shown by the plots in Figure 4. The initial loss of chlorine in the case of
the spent caustic extraction liquor is seen to be substantial at both 4° and
20°C. The spent chlorination liquor also lost organically bound chlorine on
storage at room temperature but at approximately only 1/3 the rate observed
for the caustic extraction liquor. The loss of organically bound chlorine
was reduced as the pH of the liquor was progressively decreased.
In the acute toxicity determinations, samples of spent chlorination
liquor (pH 2) were stored in the dark at 5°C and periodically tested against
Daphnia magna. For periods of storage up to 39 days, the variations in
24-hr 24 LC,^ values were not statistically significant. When allowed to
stand at room temperature for 37 days, the same liquor likewise showed little
significant change in acute toxicity. After 80 days' storage, however, a
slight decrease in acute toxicity was observed, but further testing after
more prolonged standing is required to confirm this trend. It is interest-
ing to note, however, that should the trend toward decreased toxicity on
standing be confirmed, it would parallel the decrease in organically bound
chlorine content for the liquor noted above and might indicate a cause and
effect relationship.
Spent caustic extraction liquor was also stored in the dark at room
temperature and intermittently monitored for acute toxicity. After 80 days,
no significant trend with respect to toxicity change was observed. In this
situation, loss of organically bound chlorine appears to have had no effect
on acute toxicity.
The apparent stability of the spent chlorination liquor with respect
to acute toxicity is probably due, to a large extent, to its high acidity.
As will be shown later, alkali treatment of this liquor had the effect of
greatly decreasing its toxicity in a relatively short period of time.
Smaller increases in the pH of spent chlorination liquor of the order of
magnitude produced by diluting it with chlorinated pulp washings would un-
doubtedly have produced a similar effect, albeit over a greater period of
time.
Collectively, the above findings indicate that although some of the
properties of spent chlorination and caustic extraction liquors change on
standing, acute toxicity testing can be performed on reasonably aged samples
of such liquors without changes occurring in the course of the test.
31
-------
GO
ro
_J
<
SCEL, 20° C
SCEL, 4° C
SCEL, 20°C
SCEL,4°C
58
DAYS
Figure 4. The rate of loss of organically bound chlorine for spent chlorination and
caustic extraction liquors.
-------
Responses of the Test Organisms to the Whole Spent Chlorlnation Liquor (SCL)
and Spent Caustic Extraction Liquor (SCEL)
At their original concentrations, both the spent chlorination liquor
and spent caustic extraction liquors were distinctly toxic to Daphnia maqna.
As seen in Table 7, the 24-hr LCcn value of the SCL is about six times
smaller than that of the SCEL, which means that the SCEL is considerably
less toxic than the SCL when compared on the same (TOC) basis. A similar
result has been recently reported by Pfister and Sjostrom (28). The toxic
units (TOC/LC50) were calculated as 3.21 and 4.12 for SCL and SCEL, respec-
tively. The total toxicity loads were calculated by multiplying the toxic
units by the total liquor volumes which are regulated by the pulp consis-
tencies of 3% and 10% for chlorination and caustic extraction, respectively.
The toxicity loads are about 104 (SCL) and 37 (SCEL) toxic units per kg of
pulp.
Table 7. ACUTE TOXICITY OF THE WHOLE SCL AND SCEL TO DAPHNIA MAGNA
SCL5SCEL5"
Test
No.
1
2
3
4
5
6
7
8
-------
the SC and SCE liquors, the kraft pulp was thoroughly washed with water and
then extracted successively with several portions of aqueous sodium hydrox-
ide with intermediate washes under both ambient and elevated temperature
conditions. Without making any attempt to identify and quantify all of the
resin and fatty acids in the alkaline extracts, dehydroabietic acid was iden-
tified as its methyl ester in about 0.01% yield in the primary extract. The
yields of total extracted material and methyl dehydroabietate decreased pro-
gressively with each extraction treatment. The complete removal of resin
and fatty acids (extractives) from brown stock would thus appear to be un-
attainable under ordinary technical washing conditions.
To estimate the effect of extractives carried over from the brown stock
into the spent chlorination and caustic extraction liquors on toxicity, the
pulp was first extracted with ethanol-benzene solution (1:2) and then
bleached under the same conditions as described in the Materials and Methods
section. The resulting spent chlorination liquor showed a toxicity (24-hr
LC5Q, 60 ppm of TOC) which was comparable to the value found for SCL cor-
responding to the unextracted kraft pulp (see Table 7). The SCEL correspond-
ing to the pre-extracted pulp showed, on the other hand, a slightly reduced
toxicity (24-hr LCso, 600 ppm of TOC) as compared to that found for the li-
quor corresponding to the unextracted pulp. Although the toxicity reduction
was about 30% when calculated as toxicity units, the contribution to the to-
tal toxicity load (SCL + SCEL) was calculated to be less than 10%, thereby
supporting the conclusion that the wood extractives and their chlorinated
derivatives are minor contributors to the combined toxicity of SCL and SCEL.
The response of A_. fumigatus to the original spent liquors is shown by
the plots in Figure 5. The dry weight of fungal mycelium increased in the
case of spent chlorination liquor. Because of the low concentration of
solids in the original SCL, a sample of the same liquor concentrated to
one-fifth of its original volume was also tested and a clear indication of
growth stimulation was observed (Figure 5). On the other hand, the SCEL
showed only a slight growth inhibition. Superficially, the results ob-
tained with A_. fumigatus seem to contradict those garnered by the use of
Daphnia magna; i.e., neither of the two spent liquors appears harmful to
the fungus, which seems to support the view which claims the invalidity of
fungi as test organisms (25). However, as will be demonstrated below, this
difference is more apparent than real and is a reflection of sensitivity
differences in the test organisms.
Fractionation of the Spent Bleaching Liquors According to Molecular Size
Of the various separation methods available, those based mainly on mo-
lecular size differences, i.e., dialysis and gel permeation chromatography,
were examined in order to determine the molecular size distribution of the
toxicants in the spent bleaching liquors. Although it is essentially impos-
sible to demonstrate a relationship between biological activity and indi-
vidual chemical components without isolation and purification, attempts were
made in the present investigation to effect a gross molecular size distri-
bution of the active substances in spent bleaching liquor through a frac-
tionation approach.
34
-------
ORIGINAL SCL
-0-0-0-SCL CONCENTRATED 5-FOLD
SCL
200 400 600
TOC, ppm
800
Figure 5. The response of AspergiTlus fumigatus to the whole
SCL and SCEL.
35
-------
Dialysis of SCL gave two fractions. The maximum amount of material in
the SCL passing through the cellophane tube (dialyzate) in a 24-hr period
corresponded to 95% of the whole liquor. The corresponding value for the
SCEL was 30%. This large difference may be taken as an indication that the
SCEL contained a greater percentage of higher molecular weight material than
the SCL. The large difference in the distribution of molecular sizes between
the SCL and the SCEL was also evident from the patterns of their gel permea-
tion chromatograms.
Failure of the cellophane membrane to retain more than 5% of the TOC in
SCL negated the use of dialysis as a useful technique for the fractionation
of solids in this particular liquor. SCEL was, however, fractionated using
both dialysis and gel permeation chromatography with the results shown in
Table 8.
The retentate designated as "A" indicates a retentate obtained after a
single 24-hr dialysis while the retentate designated as "B" refers to a re-
tentate obtained after extensive and repetitive dialysis. The number average
molecular weights of these retentates as measured by vapor pressure osmom-
etry were about 3100 and 7000, respectively. These numbers are close to the
value (4000) reported for the acid precipitate of a SCEL (30) but larger than
the number (500) reported for the precipitate formed by lime treatment of
SCEL (31). In the latter case, small amounts of ash or low molecular weight
organics may have contributed to the low results.
Both retentate A and dialyzate A were further fractionated by means of
gel permeation chromatography into five sub-fractions as shown in Figure 6.
The order of acute toxicity among the fractions listed in Table 8 indicates
that the smaller the average molecular size of the acute material in the
whole liquor or its fractions, the greater the acute toxicity.
The combined toxic unit contents of the dialyzates and retentates cor-
responding to the A and B treatment series revealed that 40 and 59%, respec-
tively, of the number originally present in the whole liquor were unaccounted
for. These losses may have resulted from the increased alkalinity in the
dialysates as a consequence of the residual liquor in the cellophane tubes
being concentrated.
Retentate A and dialyzate A were also subjected to bioassay using the
fungus Aspergillus fumigatus. As shown by Figure 7, the oven dry weight of
this fungus decreased when incubated with the concentrated dialysate. The
large amount of scatter in the data may be taken as an indication that sepa-
ration of materials into the dialyzate and retentate fractions was incom-
plete. The gel permeation chromatograms in Figure 6 appear to support this
contention. Regardless, the sense of the data in Figure 7 is that certain
growth inhibitors are present in the dialyzate of the SCEL; the effect of
which becomes more pronounced as the concentration of materials in the test
solution is progressively increased.
36
-------
TABLE 8. ACUTE TOXICITY OF SCEL FRACTIONS TO DAPHNIA MAGNA
CO
-si
Separation
Method Liquor Fraction
Whole SCEL
Retentate Aa
Dialyzate Aa
Retentate Ba
Dialyzate Ba
"l"
Retentate A '
R2
Gel Permeation
Chroma tography
Dlb
Dialyzate A D2b
D3b
ppm
1637
1065
572
525
1112
862
203
108
241
223
T 0 C
% of
Original
100
65
35
32
68
53
12
7
15
13
LC5Q Toxicity Units
as
TOC, ppm
397
1395
330
5000
890
>1593
> 496
NDC
781
235
T.U.
4.12
0.76
1.73
0.11
1.25
<0.54
<0.41
0.31
0.95
% of
Original
100
18
42
11
30
<13
<10
--
8
23
*A and B refer to single and multiple dialysis treatments, respectively.
See Figure 6.
cNot determined
-------
o
00
CO
LU
O
O
v>
00
6
O
oo
cvi
UJ
o
m
oc.
m
.- 30- -
20-
10--
(I) RETENTATE A
100 200 300
ELUATE VOLUME, ml.
(H) D1ALYZATE A
100 200
ELUATE VOLUME, ml.
300
Figure 6. Gel permeation chromatogramsof retentate (I)
and dialysate (II) with Sephadex G-25.
38
-------
40-
6>
330-
ID
u.
20-
h-
r
S2
UJ
10-
0
() RETENTATE A
10) DIALYZATE A
0 1000 2000 3000
TOC OF LIQUOR SAMPLE, ppm
Figure 7. Relationship between the dry weight of fungus
and the TOC of the fractions obtained by
single dialysis.
39
-------
Toxlcity Characteristics of SCL and SCEL Ether Extracts
Solvent extraction and adsorption on activated carbon have been among
the most frequently adopted techniques for the extraction of organic matter
from water. Since desorption of materials adsorbed on activated carbon is
sometimes difficult to achieve, the solvent extraction approach seemed pre-
ferable as a means of recovering the toxic materials from the spent liquors.
Moreover, the results presented in the above section suggested that the toxic
substances present in both spent chlorination and caustic extract!on liquors
were mainly low molecular weight compounds. In such a case, conventional
solvent extraction'appeared to be a logical approach for removing the toxic
compounds from the spent liquors. Although it has been shown that ethyl ace-
tate is superior to diethyl ether as an extractant for SCL and SCEL (32,33),
the latter solvent was selected for use since the dissolved solids can be
more safely recovered from it.
Ether extraction of the spent liquors was carried out for comparatively
long periods in liquid-liquid extractors. The yields of extractives increased
progressively with increasing extraction time. Thus, a 5-liter sample of the
SCL was extracted for up to 196 hours and the yield of extractives at the end
of this period was found to be 37% based on the TOC of the original liquor.
Of the TOC extracted in 196 hours, about 80% was removed in the first 96
hours. Fractions collected at the end of consecutive 96- and 100-hr extrac-
tion periods were designated A, and A2» respectively.
A 750-ml sample of the SCEL was acidified to pH 2 and similarly extracted
for 184 hours with ether. The percentage of the original TOC recovered at
the end of this extraction was about 20%. For the purpose of fractionation,
the ether extracts were collected after 62 and 184 hours of consecutive ex-
traction and designated as Bi and Bo, respectively. As in the case of the
SCL, the majority (about 70%) of TOC was extracted from the whole liquor
within the first 62 hours.
As shown by the data in Table 9, most of the acute toxicity as shown by
Daphnia magna was found in the initially collected fraction of each liquor
(i.e., in A-j and B-|). The extracted SCL was also subjected to acute toxicity
bioassay after first removing trace residues of ether by means of a flash
evaporator. The residual organics in extracted liquor represented 49% of the
original TOC and showed an LCsg of 465 ppm of TOC. These results indicate
that most of the substances toxic to Daphnia magna are extractable with
ether. The LC^Q value of 71 ppm of TOC for fraction BI (Table 9) indicates
that ether extraction is more effective than the methods adopted previously
(i.e., dialysis and gel permeation chromatography) for concentrating the
toxic materials of the SCEL.
The response of Aspergillus fumigatus to the ether extracts of the SCL
(A) and SCEL (B) and to the ether-extracted SCL (C) is shown by the plots in
Figure 8. The yields of ether extracts for SCL and SCEL were, in this in-
stance, 31% and 25%, respectively. As is clearly demonstrated by plots A and
B, both ether extracts showed growth stimulation characteristics at lower
concentrations of the extracts but distinct growth inhibition at higher con-
centrations.
40
-------
TABLE
ACUTE TOXIC EFFECTS ON DAPHNI.A- MAGNA OF
ETHER EXTRACTIVES OF SCL AND SCEL
T 0 C
Toxic units
Fraction
SCLa
A b
Al
A b
««
.ppm
231
68
15
% of
original
100
30
7
ppm of
TOC
72
23
28
T.U.
3.21
2.96
0.54
% of
original
100
92
17
SCEL3
R a
Bl
R C
B2
1637
218
92
100
14
6
397
71
204
4.12
3.07
0.45
100
75
11
aWhole liquor (see Table 7)
AI and A2 correspond to material removed from SCL after consecutive 96- and
100-hr extraction periods, respectively.
°B-j and 83 correspond to material removed from acidified SCEL after consecu-
tive 62- and 122-hr extraction periods, respectively.
Thus, there is an apparent parallel between the growth-inhibiting effect
of the ether extracts on Aspergillus fumigatus and the acute toxicity dis-
played by the same extracts toward Daphm'a magna, which is supportive of the
conclusion that the toxic components of SCL and SCEL are ether extractable.
The mutagenic compounds in SCL have likewise been found to be ether extract-
able (34). The finding that the ether extracts of SCL and SCEL display
acute toxicity toward Daphm'a magna at a concentration level where growth
of Aspergillus fumigatus is stimulated probably is ascribable to a difference
in the sensitivity of the two organisms. The fact that the ether-extracted
SCL supported fungal growth may have been the direct result of the selective
removal of a large percentage of the toxicants by ether extraction as already
demonstrated.
Toxicity Characteristics of SCL Ether Extract Sub-fractionated on the Basis
of Acidity ~~~
An ether solution of the extract from a freshly prepared sample of SCL
(yield, 31%) was further separated into 1% NaHC03 soluble, 1% NaOH soluble,
and neutral fractions (designated as Aa, At,, and Ac, respectively). The re-
sponse of Daphnia magna to these fractions are recorded in Table 10.
41
-------
0 1000 2000
CONCENTRATION AS TOC t ppm
Figure 8. The effect of the ether extracts of SCL(A), SCEL(B),
and of ether-extracted SCL(C) on the growth of Asper-
gi'Tlus fumigatus.
42
-------
45,
co
TABLE 10. TOXICITY OF THE SUB-FRACTIONS OF THE SCL ETHER EXTRACT
AS INDICATED BY DAPHNIA MAGNA AND ASPERGILLUS FUMIGATUS
Symbol
A
Aa
Ab
Ac
Fraction
whole ether extract
weak acids
very weak acids
neutrals
ppm
72
60
7
3
TOCa
% of
original
31
26
3
1
Toxic units3
24-hr LC50
ppm of
TOC
29
1500
23
18
T.U.
2.48
0.04
0.30
0.17
% of
original
77
1
9
5
Cl
w/w, %
25.0
16.0
6.8
6.6
Dry weight
of
A_. fumigatus
in ma
1.5
43.5
39.7
50.7
aBased on data for the whole SCL listed in Table 9
Compared at a TOC concentration of 2800 ppm
-------
As is apparent from the data in this table, most of the material in the
whole ether extract was collected in the Aa (weak acid) sub-fraction. The
compounds comprising this group are, in all likelihood, largely carboxylic
acids. The acute toxicity of this sub-fraction is considerably lower than
those of the whole ether extract and the very weakly acidic and neutral frac-
tions. Although small in quantity, these latter fractions are highly toxic.
The compounds in sub-fraction AK are probably phenolic in nature and
several monohydric chlorophenols, chTorocatechols, and chloroguaiacols have,
in fact, been identified by Lindstrom and Nordin (23) in SCL. Several of
these compounds have been found to be acutely toxic to fish (21). Lindstrom
and Nordin (35) have also identified a substantial number of neutral com-
pounds (several sulfur- and chloro-containing), the toxicities of which have
not been established in all cases. Presumably, sub-fraction Ac consists of
compounds of the type identified by the above Swedish researchers.
The toxic unit contents of the three sub-fractions are substantially
lower compared to that of the whole ether extract (Table 10). The loss in
toxic unit content can be ascribed to reactions (other than neutralization)
of certain components of the whole ether extract with the alkaline solutions
(NaOH and NaHCC^) used in the sub-fractionation treatments. Similar reduc-
tions in acute toxicity have been observed when whole SCL has been made al-
kaline and subjected to bioassay using fish (36) and Daphnia magna (see later
sections) as the test organisms. The data in Table 10 which indicate the
three sub-fractions as being better able to support the growth of Aspergillus
fumigatus^ than the whole extract support these findings. The discovery that
the mutagenic properties of SCL are greatly reduced by alkaline treatment
(34) is also important in this connection.
Although toxicity is often directly related to the chlorine content of
a material, the data in Table 10 indicate that the sub-fraction with the
highest chlorine content (Aa) also shows the lowest acute toxicity. In this
instance, the type of structure(s) to which the chlorine is bound may have a
greater influence on toxicity than chlorine content.
Toxicity and Chemical Characteristics of SCL and SCEL Ether Extracts Sub-
fractionated by Adsorption Chromatography
The ether extracts designated AT, A2, B-J , and 82 described in Table 10
were sub-fractionated by means of adsorption chromatography on silica gel.
A typical elution diagram is shown in Figure 9. The response of Daphnia
magna to the various fractions obtained by this type of separation is recorded
in Table 11. The results in this table duplicate the finding illustrated
previously by the data reported in Table VI, namely, that the toxic sub-
stances in SCL and SCEL are preferentially extracted by ether in the initial
stages of the treatment. As is also apparent from the data in Table 11, most
of the toxicity in A-j and B-j was recovered when these fractions were chroma-
tographed on a silica gel column using ether as the elutant. Because these
sub-fractions (A]E and BI£) represent a major fraction of the toxicity found
in the whole SCL and SCEL, respectively, they appear to be excellent starting
materials for further toxicity characterization.
44
-------
JOO
PET.
ETHER
* 3*-
O
or
^50+
u.
O
^25+
i
o
LJ
Dl ETHYL
ETHER
m »
METHANOL
B.M
BE
200 400
ELUTION VOLUME, ml.
Figure 9. A typical elution diagram for the fractionation
of SCEL ether extract on silica gel.
45
-------
TABLE 11. TOXICITY CHARACTERISTICS OF SCL AND SCEL ETHER EXTRACTS
FRACTIONATED BY CHROMATOGRAPHY ON SILICA GEL
Fraction3
SCLb
SCELb
AIE
A,M
A2E
A2M
BIE
B^
B2E
B2M
ppm
231
1637
50
18
10
4
107
in
61
31
T 0 C
% of
original
100
100
22
8
4
2
7
7
4
2
24-hr LC50
ppm of
TOC
72
397
15
77
31
38
39
138
164
47
Toxic
T.U.
3.21
4.12
3.33
0.23
0.32
0.11
2.74
0.80
0.37
0.66
units
% of
original
100
100
104
7
10
3
67
20
9
16
A.J and A2 correspond to ether extracts of SCL obtained in two consecutive
extraction treatments (see Table 9).
B-| and B2 correspond to ether extracts of acidified SCEL obtained in two
consecutive extraction treatments (see Table 9).
E and M refer to elution of material from silica gel column with diethyl
ether and methanol, respectively.
Data correspond to the mean values shown in Table 7.
In comparison with the sub-fractions of SCL ether extractives eluted
from a silica gel column with methanol (A-jM), the corresponding sub-fraction
from SCEL (B-|M) displayed a high toxic unit content. As suggested by re-
sults discussed in a previous section, the difference may be the result of
a greater contribution to total toxicity(in the case of SCEU by wood extrac-
tives.
Elemental and functional group analyses were performed on the ether
extracts of SCL and SCEL (^260 hours of extraction) and on sub-fractions ob-
tained by chromatography on silica gel with the results shown in Tables 12
and 13. The only significant differences among the fractions are found in
46
-------
TABLE 12. ELEMENTAL COMPOSITION OF THE SCL(A) AND SCEL(B)
ETHER EXTRACTS AND ETHER EXTRACT SUB-FRACTIONS
Fraction3 Yield,b
Weight, %
Number/Mn
A
AE
AM
B
BE
BM
%
58.
27.
31.
21.
6.
15.
6
5
1
8
7
1
Mn
292
307
793
225
256
746
C
38.7
39.1
36.6
46.4
50.1
44.0
H
3.4
3.6
3.4
4.4
5.0
3.6
Cl
23.0
23.2
15.3
7.4
6.1
7.6
0
34.9
34.1
44.7
41.8
38.8
44.8
C
9.4
10.0
24.2
8.7
10.7
27.3
H
9.7
11.0
26.9
9.8
12.7
26.8
Cl
1.9
2.0
3.4
0.5
0.4
1.6
0
6.4
6.6
22.1
5.9
6.2
20.9
A and B = ether extracts of SCL and SCEL, resp.; E and M = ether and metha-
nol, resp., solvents applied consecutively to eluted material from silica
gel column.
Based on TOC of the whole liquors.
TABLE 13. ACIDIC GROUP AND TOTAL HYDROXYL CONTENTS OF THE SCL(A)
AND SCEL(B) ETHER EXTRACTS AND ETHER EXTRACT SUB-FRACTIONS
Fraction3
A
AE
AM
B
BE
BM
Weak
meq./g
3.79
4.07
3.91
6.70
7.58
2.77
acids
No./Mn
1.11
1.25
3.10
1.51
1.94
2.07
Very weak
meq./g
2.60
2.04
2.50
1.99
2.00
1.94
acids
No./Mn
0.76
0.63
1.98
0.45
0.51
1.45
Total hydroxyl
meq./g No./Mn
4.82 1.41
5.00 1.54
5.18 4.11
5.67 1.28
6.27 1.61
5.31 3.96
3Symbo1s identified in footnote a, Table 12.
the chlorine content and number average molecular weight values. The higher
chlorine content of the SCL ether extract as compared to that of the SCEL
ether extract may explain the greater toxicity of the former. Comparison of
the number average molecular weight values listed in Table 12 for the E and
M sub-fractions with the appropriate acute toxicity data in Table 11 shows
that greater toxicity is associated with the sub-fractions composed of lower
47
-------
average molecular weight material as was speculatively assumed previously in
this paper on the basis of less firm data.
The results presented in this section have provided evidence indicating
that all or most of those compounds in SC and SCE liquors contributing to
toxicity are of low molecular weight and as such may be conveniently sepa-
rated from the non-toxic substances in these liquors using separation tech-
niques based on molecular size. Of the fractionation approaches attempted,
ether extraction proved to be one of the more satisfactory for effecting the
non-destructive recovery of low molecular weight (i.e., toxic) substances in
SCL and SCEL. This achievement is particularly important in the case of SCL
which is especially prone to undergo changes which affect its toxicity.
Ether extracts of SCL and SCEL therefore represent excellent starting mater-
ials for further work relating to toxicity.
DEGRADATION OF PHENOLS BY CHEMICAL TREATMENT
Treatment with Ozone
With respect to its reactions with organic materials, ozone is best
known for its ability to react with and ultimately cleave aliphatic carbon-
carbon double bonds (37,38). When applied to spent bleaching liquors,
ozone could therefore be expected to react with fatty acids and lignin frag-
ments containing such groups. Although less readily attacked than aliphatic
carbon-carbon double bonds, phenols as a class of compounds are reactive
toward ozone (39). Published findings (39-46) consistently indicate that
ozone functions by breaking down the aromatic rings of lignin and simple
phenols forming*ultimately,aliphatic carboxylic acid fragments.
An inspection of relevant literature reveals that of the phenols reac-
ted with ozone, relatively few possessed aromatically bound chloro substitu-
ents and even these were atypical of chlorophenol types identified thus far
in spent chlorination and caustic extraction liquors (21-23). In order to
obtain information more relevant to the situation actually encountered in
spent chlorination and caustic extraction liquors, chlorophenols representa-
tive of the phenolic types identified in such liquors were ozonized in a
slightly acidic aqueous medium and their rates of disappearance individually
determined.
Sufficient amounts of the aforementioned phenols were dissolved in
ethanol/water solutions to provide 0.414 mM solutions. After the flow rate
had been separately determined, the ozone was introduced into the solution
of phenol through a fritted glass diffuser tube. After varying amounts of
ozone had been applied, the residual phenol was determined by means of gas
chromatography and ionization difference spectroscopy. A detailed descrip-
tion of the apparatus and procedures is provided in the EXPERIMENTAL section.
48
-------
Calculation of Ozonization Rate Constants
The procedure used to obtain rate constants for the reaction of ozone
VJc\ thfuvarious Phenols was the same as that developed by Gould and Webber
(45). The system is assumed to be a simple one:
Phenol + 03 ka Products + 02
The rate expression the authors derived for this process is as follows:
In (PhQ/Ph) = ka Dt
where PhQ = initial concentration of phenol (moles/1)
Ph = concentration of remaining phenol (moles/1)
ka = rate constant (moles of phenol removed/mole of ozone)
D » dose rate (moles of ozone fed per minute/mole of phenol per
liter initially present)
t = time (minutes)
The plot of the natural log of Ph0/Ph versus Dt yields a straight line with
a slope equal to ka. The rate constants were calculated by taking all the
values of Phg/Ph and Dt for each sample withdrawn from the reaction vessel,
up to the point where 60% of the starting compound had reacted. This value
was chosen as the upper limit to avoid interference in the ultraviolet spec-
troscopic measurements of residual phenol by accumulating degradation pro-
ducts. The slope (ka) was then determined using an HP-55 calculator with
automatic linear regression, in order to eliminate the errors inherent in
a graphical method. Ideally, it would have been desirable to use only
those points corresponding to the first 20% of phenol removal, but in many
instances the initial reactions were so rapid that insufficient data were
available to allow the slope to be determined with any degree of accuracy.
Effect of Structure on Ozonization Rate
The rate constants for the reaction of ozone with various model phenols
are compiled in Table 14. Initially, the residual phenol concentrations
were measured gas chromatographically. This approach proved to be time con-
suming, however, because of the large number of samples involved and the
time needed to prepare each sample for analysis. Ultraviolet spectroscopy
based on measurement of ionization difference spectra appeared to represent
a much simpler and more convenient alternative approach. Analyses subse-
quently performed on the Ozonization product mixture of trichlorocreosol
(XIII) indicated that the yields of residual phenol, as determined by the
spectroscopic and chromatographic methods, agreed to within 2%. Accordingly,
the UV spectroscopic method was thereafter adopted for use in determining
the residual phenolic content of phenol/ozone product mixtures. The gas
chromatographic method continued to be used for the determination of vera-
trole derivatives (XXIV-XXVII) since they have no ionizable groups.
49
-------
TABLE 14. OZONIZATION RATE CONSTANTS OF PHENOLS
Compound3
I
II
V
VII
VIII
IX
X
XI
XII
XIII
XVI
XVII
XVIII
XIX
XX
XXIV
XXV
XXVI
XXVII
Initial pH
6.15
6.15
6.15
6.15
6.15
6.15
6.15
6.15
6.15
6.15
10.3
6.15
6.15
6.10
6.15
6.15
6.15
6.14
6.15
Final pH
6.05
5.85
5.83
5.85
5.45
5.90
5.42
5.68
5.30
5.20
> 10
5.65
5.29
6.05
6.10
5.65
6.15
6.14
6.15
Moles of ka phenol
fremoved/mole 0^
0.90
0.83
2.44
2.43
1.62
1.11
2.02
2.04
2.56
2.61
v/13.6
3.72
2.53
5.25
4.42
1.77
0.20
0.02
_c
aSee Table 2 for structures.
Compound XVI: initial cone., 1.01 mM; 03 flow rate, 0.72 mg/min.
All others: initial cone. 0.414 mM; 03 flow rate, 2.5 mg/min.
cNo reaction after application of 27 mg of 03.
As was stated previously, only data corresponding to the removal of up
to 60% of the original phenol were used to calculate the rate of constants
in order to reduce the possibility of phenol oxidation products interfering
with the measurements. In a number of instances, differential absorbance
was noted at wavelengths slightly higher than the absorption maxima found
for the phenols (297-313 nm). This differential absorbance increased with
increasing ozone input, thereby indicating the resistance of the contribut-
ing structure(s) to breakdown by the oxidant.
50
-------
. In general, the data In Table 14 indicate that the reactivity of the
phenols toward ozone decreases in the order: catechols (XVII-XX), guaiacols
(VII-XIII), cresols (II and V), and veratroles (XXIV-XXVII). This se-
quence is consistent with the order generally observed for the reactivity
of oxidants, including a sparse amount of information relating specifically
to ozone (47), toward mono- and dihydric phenols and their ethers.
The rates of reaction of phenols of the guaiacol (VII-XIII) and cate-
chol (XVII-XX) types with ozone decrease with the introduction of one chloro
substituent onto the ring. This finding is predictable since ozone func-
tions as an electrophile (see ref. 48) and the chloro substituent reduces
the electron density of the ring. However, as the number of chloro substi-
tuents increased, the rate constants (cf. Table 14) actually showed a pro-
gressive, albeit a somewhat erratic, increase rather than the expected de-
crease. An explanation for this apparent anomaly is not readily discerned
from the data at hand. However, one working hypothesis can be evolved
based on the statements of other investigators (39,48,49) indicating that
the intermediate oxidation products of phenols with ozone (e.g., o-benzo-
quinones and muconic acids) are themselves readily attacked and degraded by
ozone. Thus, there exists a competition for ozone between residual phenol
and its breakdown products. In the case at hand, it is conceivable that
the effect of chloro substituents on reducing the rate of ozone attack is
greater when they are present on phenol breakdown products than when they
are situated on phenolic rings. In applying this interpretation to the
present situation, it has to be assumed that the number and possibly the
position of the chloro substituents on both the phenol and its breakdown
products are critical variables.
Comparison of the rate constants (Table 14) for the reaction of phenol
itself (I) and pj-cresol (II) reveals that the effect of H and CH3 substitu-
ents at the position para to the phenolic hydroxyl is not pronouncedly dif-
ferent. Contrary to the behavior displayed by phenols VII-XIII and XVII-XX,
introduction of one chloro substituent onto the cresol ring did not result
in a reduction of the reaction rate with ozone, as was observed with phenols
of the guaiacol and catechol types. The reaction rate did, however, increase
systematically as the number of chloro substituents increased from 0 to 2.
The reaction sequence proposed (42,45,48,50) for the oxidative breakdown of
monohydric phenols with ozone involves a prior conversion to a catechol
through hydroxylation at a position ortho to the phenolic hydroxyl group.
Whether this additional step is sufficient to explain the slightly differ-
ent effect of chloro substituents on the observed rates of mono- and dihy-
dric phenol ozonization cannot be determined with the evidence at hand.
The problem is further complicated by the fact that conversion of monohydric
phenols to catechols may be prevented by the presence of chloro substituents
at the positions ortho to the phenolic hydroxyl group.
The fully etherified phenols (XXIV-XXVII) are atypical of the phenols
identified thus far in spent chlorination and caustic extraction liquors
(21-23) but were included in the study for purposes of comparison with
other phenolic types and with the hope of obtaining information regarding
the mechanism of oxidative breakdown with ozone. The rate data in Table 14
51
-------
clearly demonstrate the relative unreactivity of these compounds toward
ozone.
The rate constants in this series of compounds decrease progressively
with increasing substitution of chlorine as would be predicted assuming an
electrophilic attack of ozone and no other complicating side effects.
The observed stability of the trichloroveratrole derivative (XXVII)
toward ozone contrasts sharply with the behavior of the corresponding guaia-
col and catechol compounds XIII and XX, respectively, both of which were ex-
tensively degraded by ozone under the same conditions. Other evidence indi-
cating that a different mode of aromatic ring degradation is operative in
the ozonization of veratrole derivatives lies in the observation that the
ozonization product mixtures of such compounds were colorless whereas those
of the guaiacols and catechols were initially highly colored. This finding
suggests that aromatic rings in veratrole-type structures are not degraded
through an o-benzoquinone intermediate.
In other studies (41,46) of the reactions of similar (veratry1-type)
compounds with ozone, the product analysis showed that, similar to the be-
havior of the guaiacol and catechol types, ring scission occurred mainly
between the oxygen-bearing carbons. It is important to note, however, that
this is not exclusively the case since Kratzl et al. (46) have detected
dimethyl oxalate among the ozone oxidation products of veratrole and 4-
methylveratrole (XXIV). This finding indicates a rupture of the bonds be-
tween carbon atoms 2 and 3 and 4 and 5 and represents a complicating factor
in attempting to rationalize the effect of structure on ozonization rate.
In contrast to the other compounds listed in Table 14, tetrachloro-
guaiacol (XVI) was ozonized in an alkaline (pH 10.3) medium and under
slightly different conditions with respect to initial concentration and
oz6ne flow rate than those employed with the former set. Although the
ozone treatment of tetrachloroguaiacol was not performed under the rigor-
ously controlled conditions required for the acquisition of reliable kinetic
data, a rate constant for the reaction was nevertheless calculated (Table
14). After making ample allowance for the potential inaccuracies of the
measurement, the value reported for the rate constant is still sufficiently
higher than the rate constants reported for the other phenols to justify
the conclusion that the reaction of tetrachloroguaiacol with ozone is en-
hanced in an alkaline medium. Support for this interpretation is well
documented in the literature where several investigators (40,45,49,50) con-
cluded that phenols are oxidized in preference to other substances in al-
kaline media. The effect of the alkali may be ascribed to an enhancement
of the electron density on the ring as a consequence of forming a phenolate
anion, thereby promoting the electrophilic attack of ozone. An alternative
or complementary explanation is that, in aqueous, alkaline media, ozone is
decomposed into hydroxyl radicals (51) which are more reactive toward phe-
nols than ozone itself.
The relationship between the extent of removal of tetrachloroguaiacol
and the acute toxicity of the product solutions toward Daphnia magna is
shown by the data in Table 15.
52
-------
TABLE 15. DETOXIFICATION OF TETRACHLOROGUAIACOL (XVI)
BY OZONIZATION AT pH 10
Ozone consumed/
initial phenol one.
(mmoles/nmole)
Phenol loss,
ppm based on
original Phenol ppm of carbon
0
1.57
5.13
0
50
100
0.22
4.1
>10
0.06
1.31
>3.21
alnitial cone, of phenol, 259 mg/1; ozone flow rate, 0.60 mg/min.
It is apparent from the values listed that tetrachloroguaiacol can be to-
tally removed from solution by treatment with ozone in an alkaline medium.
The decrease in phenol content was roughly paralleled by a decrease in
acute toxicity as indicated by the response of Daphnia magna to the ozonized
solutions. It is also important to note that when phenol removal was com-
plete, the product mixtures were essentially non-toxic.
has been reported by Niegowski (40).
Extent of Phenol Removal
A similar finding
Continued ozonization of the phenols and phenol ethers resulted in ex-
tensive breakdown of these compounds. Although ozonization was terminated
short of 100% removal of the phenol, evidence was obtained suggesting that
this goal was attainable in all but a few instances. The fully etherified
models (XXIV-XXVII) proved considerably more resistant to oxidative break-
down than the other classes of phenols tested, and their stability increased
with increasing substitution of chlorine on the ring. In the extreme case,
4-methyl-trichloroveratrole (XXVII) was recovered essentially intact after
the application of a considerable excess of ozone.
The efficiency of phenol removal appeared to decrease with increasing
application of ozone. This situation was seemingly the result of the for-
mation of increasing quantities of phenol oxidation products which competed
successfully with residual phenol for the ozone. Because of such complexi-
ties, few if any consistent trends were apparent in the residual phenol/
ozone consumption data, and attempts to draw any reliable conclusions re-
garding the relationship between the ozone required to effect complete re-
moval of the phenols and their structure were unsuccessful.
The primary intermediate breakdown products of phenol oxidation by
ozone have not been extensively investigated. The sequence shown below is
a composite which combines experimental findings with the speculations of a
number of investigators.
53
-------
OH
Monohydric phenols (represented by A) are believed to first undergo hydroxy-
lation to a catechol (B;Ri=R2=H). The aromatic ring of B structures subse-
quently may undergo further attack by ozone directly forming a muconic acid
derivative (C) through rupture between the oxygen-substituted carbon atoms.
In the case of 4-methylveratrole (B;R1=R2=CH3), Kratzl et al. (46) have
actually identified dimethyl cis,cis-g-methylmuconate among the reaction
products. Alternatively, cis,cis-8-methy1muconic acid (E) may be formed
via an o-benzoquinone intermediate (D). Since compounds of the latter type
are colored, the formation of colored ozonization reaction mixtures could
be taken as evidence for the reality of this portion of the above sequence.
In the present investigation, ozonization of all compounds except V,
XXIV, XXVI and XXVII resulted in the formation of colored product mixtures,
thus supporting the concept that at least a portion of the overall degrada-
tion of the majority of phenols tested occurred via an o-benzoquinone inter-
mediate. The color initially generated by the introduction of ozone into
the phenol solutions was discharged when 0.8 to 2.0 moles of 03/mole of
phenol was consumed. One group of investigators (49) has suggested that
oxygen as a decomposition product of ozone plays a role in the overall pro-
cess of phenol degradation.
Another pathway in the degradation of phenols with ozone consists of
the breaking of carbon-carbon bonds other than those between the oxygen-
bearing carbons. Thus, Kratzl et al. (46) identified dimethyl oxalate
(F;R1=R2=CH,) in the product mixture from the ozonization of 4-methylvera-
trole (XXIV) in aqueous acetic acid. Bernatek et al. (47) have also postu-
lated ozone attack on a phenol ring at sites other than the oxygen-substi-
tuted carbons.
54
-------
The reactivity of o-benzoquinones, muconic acids, and other possible
phenol breakdown products toward ozone in comparison with that of phenols
toward the same oxidant has not been extensively investigated. Generally,
application of the theoretically required amount of ozone to eliminate com-
pletely a given quantity of phenol has been found to be insufficient for
this purpose (42,44,45,47). This finding has been interpreted as indicating
a greater reactivity of ozone toward the primary ozonization products than
toward the starting material (47).
A similar competition exists in the spent bleaching liquors where the
simple phenols are mixed with phenolic breakdown products, larger lignin
fragments, and carbohydrates. The specificity of ozone for a monomeric
phenol was tested by diluting an aqueous solution of tetrachloroguaiacol
(XVI, Table 2) containing 0.1 mmole of the latter with increasing amounts
of spent caustic extraction liquor and reacting the mixtures with a fixed
amount of ozone. The results of this experiment are compiled in Table 16.
TABLE 16. OZONE OXIDATION9 OF TETRACHLOROGUAIACOL-SPENT CAUSTIC
EXTRACTION LIQUOR COMBINATIONS
Carbon contributed byTetrachloroguaiacol
tetrachloroguaiacol, % of TOC removed, %
100
30.7
12.8
4.2
81
69
44
14
aOzone applied, 5.28 mg; ozone flow rate, 0.66 mg/min.; pH 10.3
These results reveal that the organic compounds present in the spent
caustic extraction liquor do indeed compete effectively with the added tetra-
chloroguaiacol for ozone. The bottom values in each of the two columns pro-
bably come closest to approximating the situation existing in spent caustic
extraction liquor with respect to the fraction of total organic carbon (TOC)
contributed by the phenolic component. In this situation, the removal of
tetrachloroguaiacol was far from quantitative. However, comparison of the
corresponding values in the two columns reveals that the removal of tetra-
chloroguaiacol (14%) was nevertheless greater than would have been expected
on the basis of its contribution to the total carbon of the system (4.2%)
and some specificity of the ozone for the phenol is indicated. However, this
preference is moderated and controlled by the relative concentration of the
phenol and ozone,and the reactivity of the other liquor components toward
ozone.
55
-------
Another parameter reflecting the character and extent of the ozonization
treatment is the formation of chloride through oxidative breakdown of phenols
containing organically bound chlorine. In the present study, the concentra-
tions of chloride ion in the ozonized solutions of the chlorophenols were
determined as a function of the amount of applied ozone. The results of
these analyses revealed that chloride ion was produced rapidly and extensively
after the addition of only a small amount of ozone to phenols of the guaiacol,
catechol, and cresol types. Low yields of chloride from the veratrole deriva-
tives again indicated their poor reactivity toward ozone. In the extreme
case, no chloride was detected when 3,4-dimethoxytrichlorotoluene (XXVII)
was ozonized.
The percent recovery of chloride based on theoretical yield was greatest
(over 60%) for the monochloro compounds VIII, IX, XVIII, and XXV. The di-
chloro and trichloro derivatives all produced chloride ion yields in the
range 30 to 50% of theoretical; individual differences were too small to
relate to structural features with any degree of confidence. In no case did
the yield of chloride ion approach the theoretical value for 100% removal,
and it must be presumed that chlorine-containing, ozone-resistant breakdown
products were formed. This conclusion is supported by the results of a study
by Gilbert (44),who ozonized several monohydric chlorophenols and recorded
chloride yields ranging from 60-97% when approximately 4 moles of ozone/mole
of phenol were applied.
Treatment with Chlorine Dioxide
The ability of chlorine dioxide to react with and degrade phenolic
lignin model compounds is well documented in the literature (4, 52-55).
This fact, coupled with the current industry trend toward greater utilization
of chlorine dioxide in the first bleaching stage provided the impetus for
performing a few experiments involving the reactions of chlorine dioxide
on phenols similar to or identical with those detected in spent chlorination
and caustic extraction liquors.
In these experiments, selected phenols (cone., 10 mM) were reacted with
varying amounts of chlorine dioxide in the dark at ambient temperature until
the oxidant was completely consumed. The residual phenol was recovered by
extraction of the product mixture with chloroform and the amount determined
by gas chromatography. The results are shown by the plots in Figure 10. The
plots indicate the likelihood of completely removing the phenols tested, pro-
vided sufficient amounts of chlorine dioxide are applied. With the possible
exception of 2,4-dichlorophenol (IV), the responses of the phenols show no
pronounced differences. Since either 2 or 4 oxidizing equivalents are the-
oretically required to convert the various phenolic rings in question to
non-aromatic moieties, it is apparent that the removal of the phenols tested
is a very inefficient process.
The efficiency of phenolic ring breakdown by chlorine dioxide was
further tested by applying a fixed amount of chlorine dioxide to a solution
of tetrachloroguaiacol (XVI) diluted with increasing amounts of spent caus-
tic extraction liquor. The results (see Table 17) reveal a decrease in the
detoxification factors ranging from 0.701 to 0.826.
56
-------
---- TO
OCH3
0 2 4 6 8 10 12
EQUIVALENTS OF CI02/MOLE OF PHENOL
Figure 10. Degradation of selected phenols through reaction
with chlorine dioxide.
14
16
57
-------
TABLE 17. CHLORINE DIOXIDE OXIDATION9 OF TETRACHLOROGUAIACOL-SPENT
CAUSTIC EXTRACTION LIQUOR COMBINATIONS
Carbon contributed by Tetrachloroguaiacol
tetrachlorogualacol. % of TOC removed. %
100
30
10
5
95
81
58 ^
19
aApplied 17 mg of C102 to 24 mg of tetrachloroguaiacol at pH 6.
efficiency of phenol removal with increasing dilution by the spent liquor. A
similar finding was previously noted in the case of the tetrachloroguaiacol/
spent caustic extraction liquor combinations with ozone (Table 16). Compari-
son of the data in the two tables leads to the conclusion that, in the pres-
ence of spent caustic extraction liquor, the selectivity of chlorine dioxide
for tetrachloroguaiacol is greater than that of ozone.
Treatment with Alkali
Data presented below and elsewhere (24,36) have demonstrated the detoxi-
fication effect of alkali on spent chlorination liquor. In an attempt to re-
late the detoxification of the liquor to specific phenol or at least phenolic
types, aqueous solutions of several phenols previously identified in spent
chlorination liquor (23) were prepared and adjusted to pH 11 by addition of
alkali and subsequently stored in the dark at 5°C for 24 hours. The appro-
priate concentrations of the phenols tested were first established by range-
finding toxicity studies. A control sample was also prepared and allowed to
stand at neutral pH under otherwise identical conditions. During the stand-
ing period, the pH of the alkali-treated guaiacols (XXIV and XXV, Table 2),
2,4-dichlorophenol (IV), and 2,4,6-trichlorophenol (VI) remained constant
whereas the pH of the alkali-treated catechols (XXI, XXII, and XXIII, Table 2)
dropped from 11 to 9.9. At the end of 24 hours, alkali-treated samples were
readjusted to pH 7 with 1 N HC1. Standard dilution water was added to each
untreated toxicant solution in a volume equal to the total volume of NaOH and
HC1 added to the corresponding treated sample. All phenol solutions were
bioassayed at 14°C using cardiac arrest as the response criterion.
The results of these bioassays are summarized in Table 18. All of the
chlorocatechols evidenced pronounced detoxification on alkaline treatment.
At concentrations greater than 20 times the LC§n of untreated chlorophenol,
alkali-treated chlorocatechols caused no significant mortality in Daphm'a.
For the chloroguaiacols, 2,4-dichlorophenol and 2,4,6-trichlorophenol, no
significant differences were apparent in the toxicities of the treated and
untreated samples. However, with each of these compounds, the toxicity of
the treated sample was consistently lower than the untreated sample, with
detoxification factors ranging from 0.701 to 0.826.
58
-------
untreated 6.54 (4.96-7.81) 0.240 <0.037
1-
TABLE 18. EFFECT OF ALKALINE TREATMENT (pH 11) ON THE TOXICITY OF SOME
CHLOROPHENOLS TO DAPHNIA MAGNA
=
24 hr LC5n
u , (q5% confidence interval)
- Chlorophenol _ In y moles/liter _ ga
4,5-Dichlorocatechol (XXI)
untreated
treated1- >178
3,4,5-Trichlorocatechol (XXII)
untreated 3.39 (2.92-3.86) 0.222 <0.055
treated >62.0 [o%]a
3,4,5,6-Tetrachlorocatechol (XXIII)
untreated 2.23 (1.25-2.80) 0.319 <0.039
Seated >57.5 |;10X]a
4,5-Dichloroguaiacol (XIV)
untreated 98.3 (79.0-135) 0.309 n QOC
treated 119 (96.6-189) 0.299 °'826
4,5,6-Trichloroguaiacol (XV)
untreated 22.0 (15.6-26.5) 0.261 n ...
treated 31.4 (26.3-40.8) 0.242 u>/m
2,4-Dichlorophenol (IV)
untreated 51.9(40.2-90.1) 0.424 .
treated 68.2 (51.2-175) 0.408 u'
2,4,6-Trichlorophenol (VI)
untreated 13.2(11.4-16.0) 0.206 n
treated 16.9 (14.8-21.5) 0.276 lt
dAs discussed by Finney (11), the "g" value can be used as an index of the
quality of the bioassay, i.e., how well the observed mortality pattern cor-
responds to the best estimate of the probit line. In general, "g" values of
0.4 or less indicate good agreement
bLCso of untreated compound LCso of treated compound
cEach treatment consisted of addition of NaOH to bring pH to 11.0
Percent mortality at specified concentration
The detoxification of the chlorocatechols (XXI-XXIII) possibly may be
associated with loss of organically bound chlorine since structures of this
kind have previously been shown to undergo alkali -catalyzed hydrolysis of
59
-------
chloro substituents (19). Conversely, 4,5-dichloroguaiacol (XIV) that sus-
tained a comparatively small loss of organically bound chlorine in the same
study, showed a statistically insignificant decrease in acute toxicity
(Table 18). This observed relationship does not, however, preclude the
involvement of other, as yet unidentified, factors.
DEGRADATION OF CHLOROPHENOLS BY BIOLOGICAL TREATMENT
The biodegradation tests involved addition of individual microorganisms
(fungi) and a mixed microbial population to solutions of various phenols; the
phenols were identical to those detected in spent chlorination and caustic
extraction liquors. Since, in the course of performing the biological treat-
ments, the aqueous phenol solutions were exposed to air throughout the test
period, control experiments were conducted in the absence of microorganisms
in order to evaluate the effect of aeration alone on the removal of the
phenols. In these tests, solutions of the various chlorophenols (cone., 20
ppm) were filtered through a Seitz filter to remove any microorganisms, then
shaken on a horizontal shaker for one week at 28°C. The residual phenol was
removed by solvent extraction and determined by gas chromatography.
The results of the aeration treatments are shown in Table 19. The ex-
tent of phenol removal appears to be related to the structure of the phenolic
ring. Thus, both guaiacol derivatives (XIV and XVI) were recovered in the
highest yield (70-75 percent), 50 percent of the di- and trichlorophenols
(IV and VI, resp.) remained after one week, while less than 5 percent of the
catechols survived after the same period. No attempt was made to character-
ize the phenolic breakdown products, but it was speculated that oxidation of
the phenolic rings to quinonoid and acyclic moieties probably occurred,
particularly in the case of the catechols (XXI and XXIII). The color devel-
oped in the aerated solutions is supportive of the proposed formation of
quinonoid units. In addition to oxidation, some hydrolysis of chloro sub-
stituents is a likely possibility.
In a somewhat similar investigation, Cain (3) aerated aqueous solutions
of several phenols for varying periods of up to 10 days at ambient tempera-
ture. Samples were withdrawn at various intervals and the amount of residual
phenol measured by ionization difference spectroscopy. The results of this
analysis are shown in Table 20. In agreement with the findings reported in
Table 19, the catechol derivatives (XVII, XVIII, XX) were degraded more
extensively than the guaiacols (IX, XIII). Among the catechols, degradation
decreased slightly with increasing numbers of chloro substituents on the
phenolic ring. In all instances, the rates of phenol removal were low.
The extents of phenol removal found in the present investigation were
slightly greater in the case of the guaiacols and very much greater for the
catechols than those indicated by Cain's results after comparable aeration
periods. The explanation for this difference may be related to the roughly
4-fold higher phenol concentrations and a slightly lower reaction temperature
that Cain used. The analytical methods employed in the two investigations,
gas chromatography and ionization difference spectroscopy, may also have
contributed to some of the difference.
60
-------
Residual Chlorophenol, %b
Aeration Aeration
Chlorophenols alone + fungus4
4,5-Dichloroguaiacol (XIV) 7n _..
Tetrachloroguaiacol (XVI) /u"/b
2,4-Dichlorophenol (IV) cn
2,4,6-Trichlorophenol (VI) bu
4,5-Dichlorocatechol (XXI) .K
Tetrachlorocatechol (XXII) b
55
16
15
40
-d
3Conc.: 20 ppm; aeration (shaking) time: 1 week; temp., 28°C
Corrected for incomplete recovery by solvent extraction
^Candida utilis (yeast)
Not determined
TABLE 20. RATE OF REMOVAL OF PHENOLS DURING AERATION
Phenol cH3
ci
Time, Days
Residual phenol, %
CH- Cri. Cti
XVII
XVIII
1
2
3
4
5
10
99
98
98
97
96
-
-
-
-
95
91
98
96
88
81
75
~
100
94
88
85
83
98
93
92
92
88
61
-------
The expectation is that the various phenolic types listed in Tables 19
and 20 will behave in a qualitative manner when subjected to aeration as com-
ponents of spent chlorination and caustic extraction liquors. However, the
rate and degree of their removal in the technical aeration treatment cannot
be accurately judged from the data in Tables 19 and 20.
In the few instances where aeration was performed in the presence of a
fungus (Candida utilis), the amount of phenol removed was increased over that
resulting from aeration alone. Under the test conditions, however, chloro-
phenol removal was incomplete in every instance. In earlier work, Leach,
Mueller, and Walden (56) found that tetrachloroguaiacol could be biodegraded
by an inoculum of acclimated activated sludge after an induction period of
approximately 5 days. The guaiacol was completely consumed after approxi-
mately 10 days' fermentation treatment. In other biodegradation tests,
three fungi, Paecilomyces varioti, Pern'cillium variabile, and Trichoderma
koningii, were examined for their ability to remove phenolic compounds from
a glucose-yeast extract-peptone liquid medium in a one-week incubation per-
iod. The amount of residual phenolic compound in the medium was determined
by gas chromatography with the results shown in Table 21. The data in this
table represent the combined effects of aeration and microorganism attack on
phenol breakdown.
The most extensively degraded phenol was 4-methylcatechol (see Table 21)
and all three fungi totally removed this compound from the media after one
week of incubation. At the other end of the reactivity scale, the veratrole
derivatives (XXIV and XXV) proved to be exceptionally resistant to fungal
attack. In between these extremes, the fungi evidenced considerable differ-
ences in their ability to degrade the same compound.
No consistent correlations were found between toxicity (see Table 3) and
susceptibility to biodegradation. Thus, the toxic effect of 2,4,6-trichloro-
phenol (VI) on Trichoderma koningii and PeniciIlium variabile were similar
whereas in the biodegradation tests, the former fungi totally removed this
phenol from the medium but the latter only to the extent of 50%.
Gas chromatographic analysis of the ethanol extract of the myc'elia
Paecilomyces varioti following incubation indicated that the fungal mycelia
had not removed the phenols from the media by physical adsorption. The re-
moval of phenolic compounds from the media can therefore be attributed to
fungal attack, coupled with aeration effects.
Cain (3) examined the biodegradation of many of the phenols listed in
Table 2 by the mixed microbial population present in the sludge of a paper
mill waste treatment plant. In these tests, solutions of the phenols were
inoculated with diluted samples of the sludge and incubated at ambient temp-
eratures for periods up to 15 days. Residual phenol content was determined
at selected reaction intervals using gas chromatography.
62
-------
TABLE 21. BIODEGRADATION OF PHENOLS BY FUNG! IN LIQUID CULTURE MEDIA
GO
Phenol
Creosol
6-Chlorocreosol
5-Chlorocreosol
5,6-Dichlorocreosol
3,6-Dichlorocreosol
3 , 5-Di chl orocreosol
Tn'chlorocreosol
4-Methylcatechol
4-Methyl -5-chlorocatechol
4-Methyl-5,6-dichlorocatechol
4-Methyl -3, 5-6- tri chl orocatechol
4-Methyl veratrol e
4-Methyl -6-chloroveratrol e
p-Cresol
4-Methyl -2, 6-dichlorophenol
2,4,6-Trichlorophenol
VII
VIII
IX
X
XI
XII
XIII
XVII
XVIII
XIX
XX
XXIV
XXV
I
V
VI
Initial
cone., ppm
50
50
50
10
10
10
10
50
50
10
10
50
50
50
10
10
Residual
Paecilomyces
varioti
70
82
75
94
89
70
77
0
70
87
40
81
83
90
53
67
phenol, %
Trichoderma
koningii
62
70
68
75
6,8
34
0
0
44
87
50
89
95
60
6.0
0
PenlclTHum
van'abile
1.4
79
83
62
6.6
16
0
0
61
11
44
94
100
66
3.0
52
-------
All of the phenols examined by Cain were at least partially degraded
under the test conditions. The rate and extent of phenol removal was found
to vary widely from compound to compound and consistent trends correlating
structure with these two parameters were not apparent. As in the previously
described tests employing individual fungi, the biodegradation with micro-
organisms in the sludge was performed in concert with aeration. The mea-
sured decrease in phenol content was therefore, in all probability, a compo-
site of the effects of biological and chemical processes. Failure to perceive
clear-cut trends in Cain's data may therefore be due to their being masked as
a result of interference on the part of one or the other type of process.
In view of the widespread use of biological processes in the treatment
of pulping and bleaching wastes, additional research aimed at optimization of
the operating parameters affecting phenol and chlorophenol breakdown would
appear to be highly desirable.
CHEMICAL TREATMENT OF SPENT CHLORINATION AND CAUSTIC EXTRACTION LIQUORS
In these treatments, the chemical was generally applied to either of the
two spent liquors but rarely to both. The choice of liquor to be used in any
given treatment was dictated by pH considerations; thus, those chemicals
known to be effective under alkaline conditions were applied to spent caustic
extraction liquor whereas those functioning in the desired manner in acidic
media were added to the spent chlorination liquor.
In the application of those chemicals (oxidants) used as pulp bleaching
agents, the treatment temperatures selected coincided with those used in the
bleaching process. The treatments were continued until the chemical was com-
pletely consumed in order to avoid its contributing to acute toxicity in any
subsequent bioassays.
Chiorination Treatment
Spent chlorination liquor was reacted with solutions of chlorine water
at room temperature without pH adjustment. In one treatment, 0.2 mg Cl2/ml
spent liquor was consumed with a corresponding 35% reduction of phenol con-
tent as indicated by ionization difference spectra measurement. In a second
treatment, 0.8 mg Cl2/ml spent liquor was consumed with an accompanying loss
of 70% of the phenol content. The reacted solutions were subjected to bio-
assay using Daphnia magna as test organism. In both cases, a distinct in-
crease in toxicity was observed over that found for the untreated liquor.
The toxicity of the two treated liquors was not significantly different.
Results reported for the phenol contents of spent chlorination and caus-
tic extraction liquors here and elsewhere in this report must be regarded as
only approximate. This stems partly from the fact that the absorptivity val-
ues of the component phenols of the liquor vary substantially among them-
selves. Thus, any attempt to calculate phenol concentration must involve the
use of an "average" absorptivity value which cannot be determined accurately.
Hence, the values for phenol content listed throughout this report indicate
"phenol" content changes resulting from a specified treatment. Even this
usage is suspect since the alkali used to effect ionization may induce other
64
-------
chemical changes in the molecule that affect its spectral characteristics.
Hypochlorous Acid Treatment
The pH of samples of chlorination liquor was adjusted to 4.75 and buf-
fered with acetate to obtain a high concentration of hypochlorous acid after
the addition of chlorine. In two treatments, 0.2 and 0.37 mg of HOCl/ml of
spent liquor were consumed, which resulted in the loss of 35 and 82% of the
phenols, respectively, as indicated by ionization ultraviolet spectroscopy.
As in the above instance, the bioassay test results indicated an increase in
toxicity as a result of the oxidation with hypochlorous acid.
Sodium Hypochlorite Treatment
A sample of SCEL was reacted with 0.7 mg of NaOCl/ml of liquor at 60°C.
After three hours, the oxidant was completely consumed and the phenol content
was reduced to 49%. With a larger application of hypochlorite (2.0 mg/ml of
liquor), consumption at the same temperature was complete in nine hours and
phenol removal was 95% complete.
After these tests had been completed, it was discovered that the commer-
cial sample of sodium hypochlorite used had a high chloride concentration that
which could have affected the LC$o values of the bioassay. Accordingly,
another experiment was performed with laboratory-prepared hypochlorite (7.8
mg NaOCl/ml liquor, 60°C, reacted to exhaustion (^24 hrs)) having an accept-
ably low chloride content. The acute toxicity of the hypochlorite-treated
liquors was determined using Daphnia magna and the results are reported in
Table 22. As in the previous instances where chlorine and hypochlorous acid
were applied, the trend shown by the data is that hypochlorite treatment ac-
tually resulted in an increase in toxicity.
TABLE 22. EFFECT TO DAPflfflEA MAGNA OF SODIUM HYPOCHLORITE TREATMENT ON
THE ACUTE TOXICITY OF SPENT CAUSTIC EXTRACTION LIQUOR
^^^^Mm^MMM^^BB^MVH^^^^HMMOB^i^BI'MV*
Liquor
Original
.MM. 1 ! HI 1 - "
Exposure time,
hrs.
24
96
LC5n in ppm TOC (95%
440 (412-469)
367 (286-490)
C.I.)
NaOCl -Treated
0.7 mg/ml liquor
2.0 mg/ml liquor
7.8 mg/ml liquor
aNot determined
24
96
24
96
24
48
65
442 (400-489)
271 (110-668)
250 (200-314)
155 (122-196)
294 -a
281 -a
-------
The explanation for the increase in toxicity of the spent liquors as a
consequence of the three aforementioned treatments is not readily apparent.
One possibility is that the treatments resulted in an increase in the amount
of organically bound chlorine, particularly that substituted on the aromatic
ring. The formation of non-phenolic oxidation products more toxic than the
phenols themselves is also feasible, especially in the case of reaction with
hypochlorite. Generally speaking, however, continued oxidation of chloro-
organic compounds to smaller fragments ultimately would be expected to have
a beneficial effect with respect to toxicity reduction since organically
bound chlorine would be converted to chloride in the process.
Treatment of Spent Chlorination Liquor with Chlorine Dioxide
Spent chlorination liquor was reacted with chlorine dioxide (0.27 meq/ml
of liquor) at room temperature until the oxidant was exhausted (^5 hours).
The pH of the reaction mixture remained at 2 or lower during the treatment.
The 24- and 48-hr 1059 values for the resulting liquor were 153 and 75 ppm
TOC, respectively, using Daphnia magna as the test organism. When compared
to the value for the original Jiquor (see Table 7), the results indicate
that a modest amount of detoxification had resulted.
Treatment of Spent Caustic Extraction Liquor with Hydrogen Peroxide
Spent caustic extraction liquor was buffered (^pH 10.5) and stabilized
with sodium silicate and magnesium sulfate, respectively, and reacted with
hydrogen peroxide (4 meq of HgOg/lSO ml of liquor) at 70°C. Consumption of
peroxide under these relatively severe conditions was very slow and was only
50% complete after three hours' reaction. Complete consumption and/or decom-
position of peroxide was achieved only after ^24 hours of heating.
The 24- and 48-hr 1X59 values for the peroxide- treated liquor were 202
and 133 (ppm of TOC), respectively, when tested with Daphnia magna. Compari-
son of these values with those of the untreated SCEL (Table 7) reveals that
the peroxide treatment actually produced an increase in acute toxicity. This
finding agrees with results reported by Betts and Wilson (57). Although
additional tests should be performed to confirm this trend, the comparative
unreactivity of the peroxide toward the spent caustic extraction liquor,
even at elevated temperature, prompts the conclusion that such a process
would not be technically feasible in any event.
Treatment of Spent Chlorination and Caustic Extraction Liquors with Ozone
Spent chlorination liquor was ozonized, both at its ambient pH (^2) and
at pH 8, while spent caustic extraction liquor was reacted with ozone only at
pH 8. Acute toxicity tests were performed on the ozonized liquors with
Daphnia magna and brine shrimp (Artemia salina). The results are summarized
in Table 23.
The ozonization consumption patterns (cf. Table 23) for the three liquor
samples are similar and consist of an essentially quantitative uptake of
ozone in the initial stage of the treatment, followed by an ever-diminishing
66
-------
TABLE 23. CHARACTERISTICS OF ORIGINAL AND OZONIZED SPENT CHLORINATION AND CAUSTIC EXTRACTION
LIQUORS
Ozone, mg/ml
Appl i ed
of liquor
Consumed
TOC,
ppm
Color
(color units)
"Phenol
removal
Spent Chi ori nation
0
0,052
0.208
0.521
0
0.034
0.121
0.145
268
254
214
204
1,383
917
575
429
0
(-)4
7
55
Spent Chi ori nation
0
0.053
0.213
0.532
0
0.087
0.232
0
0.053
0.193
0.296
0
0.086
0.175
268
261
237
195
1640
1480
1360
1,383
933
370
200
Spent Caustic
18,290
9,458
2,790
0
17
18
34
11 Org. Pound
, chlorine,
ppm
Liquor (pH 2)
91
69
67
16
Liquor (pH 8)
91
61
31
34
Extraction Liquor (pH
0
36
36
209
133
47
Acute toxicity
24-Hr LCRQ in ppm TOC (95% C.I.)
Brine Shrimp Daphnia magna
342 (126-a) 107 (81.2-212 )
-
-
284 ( 76-a) 53.2 (40.3-72.3)
-
-
26 (19-33)
Jl
1323 (466-a)
-
304 (238-434)
No meaningful estimate due to low mortality
-------
consumption with continuing introduction of ozone at a constant flow rate.
Under the same conditions, the total organic carbon (TOC) decreased in each
instance, but the total decrease observed within the test limits was only in
the range 17-27%. Thus, a sizable portion of the spent liquor constituents
resisted degradation to volatile carbon-containing compounds even in the pres-
ence of excess ozone.
A decrease in the color of the spent liquors accompanied the ozone
treatment as has been reported and commented on previously (58,59). The
overall decrease in color was greater when the ozonization was performed in
an alkaline medium.
The initial and residual phenol contents of the spent liquors were mea-
sured by means of ionization difference spectroscopy. Because of the errors
inherent in this method when applied to complex mixtures, the data for re-
sidual phenol content in Table 23 have to be viewed with some degree of
skepticism. Of particular interest with regard to these data is the apparent
increase in "phenol" content observed in the case of SCL (pH 2) in the ini-
tial stages of ozonization. The obvious implication of this finding, which
has been substantiated in replicate tests, is that either additional ionizable
groups (including phenolic hydroxyls) were created as a consequence of the
ozonization treatment or that pre-existing structures were chemically modi-
fied in such a way that their A absorbance values were increased. Although
the exact extent of phenol removal is open to question, the overall sense of
the data in Table 23 is that it decreases substantially in each instance.
An increase in chloride ion content or a decrease in organically bound
chlorine content of ozonized chloro-organic material can be considered in-
dicative of the extent of oxidative breakdown. In the case of the spent
liquors, the organically bound chlorine content (Table 23) is seen to de-
crease with increasing consumption of ozone in all but one instance. In all
cases, however, some chlorine-containing residues remain at the conclusion of
the treatment. The possibility that extended ozone treatment of the liquors
would result in complete removal of organically bound chlorine seems unlikely
in view of the observed diminished reactivity of the residual material toward
ozone. Thus, spent liquors initially either contain chlorine-substituted
structures resistant to ozone or are transformed to such during breakdown by
ozone.
The results of the acute toxicity test for the spent chlorination liquor
(pH 2) using Daphnia magna and brine shrimp as the test organisms are somewhat
surprising in that the ozonized liquor proved to be equally or slightly more
toxic than the untreated sample. An earlier study of Betts and Wilson (57)
showed that ozonization of spent chlorination liquor was ineffective for re-
ducing the acute toxicity toward salmon. Other data, collected in the cur-
rent investigation but not included on Table 23, point to a trend in which a
small decrease in acute toxicity appears with small applications of ozone but
thereafter increases with increasing ozone dosages. These data are not sta-
tistically reliable, however, and hence cannot be interpreted with any degree
of confidence.
68
-------
Because of buffer salts used to control pH in the alkaline ozonization
treatments, Daphnia magna was replaced by brine shrimp as the test organism
in these cases. The LC50 values listed in Table 23 show that ozonization of
spent chlorination and caustic extraction liquors in an alkaline (pH 8) con-
dition resulted in a decided increase in toxicity under the applied treatment
conditions. (Although no control data are available for a sample of spent
chlorination liquor adjusted to pH 8, results cited below show that alkaline
treatment of such liquors actually causes a substantial reduction in acute
toxicity.)
Viewed collectively, the results in Table 23 relating to acute toxicity
lead to the conclusion that in no instance does ozonization cause a substan-
tial reduction in the toxicity of spent chlorination and caustic extraction
liquors; and under alkaline ozonization conditions, a decided increase is pro-
duced. The reason for the failure of ozone to produce the expected benefi-
cial effect is not obvious, particularly in view of the results reported pre-
viously; showing phenols of the types found in spent chlorination and caustic
extraction liquors to be extensively degraded by this oxidant. However, the
complexity of the spent bleaching liquors with respect to component composi-
tion is a factor to be reckoned with in attempting to rationalize the effect
of ozonization on their toxicity. It is conceivable, therefore, that the
gain in toxicity reduction resulting from the breakdown of phenolic sub-
stances by ozone could be offset by the formation of toxic materials or as a
result of interactions between the liquor components as described by Durkin
(24).
Treatment of Spent Caustic Extraction Liquor (SCEL) with Aluminum Sulfate
Aluminum sulfate (1.5 g) was added to SCEL (220 ml) and allowed to react
for one hour at room temperature (pH 5.5). The precipitate was removed by
centrifugation, dissolved by addition of HC1, and subsequently purified by
dialysis through cellophane. The supernate from the centrifugation and the
retentate from the dialysis were adjusted to the volume of the original test
sample, tested for phenol content, and bioassayed for acute toxicity with
the results shown in Table 24.
The results show that,although the treatment with alum removed a con-
siderable amount of the material in the liquor, a major part of the toxicity
initially present was still retained in the supernate. This finding accords
with the concept that the aluminum coagulates the higher molecular weight
components of the liquor that appear to be less toxic than the lower molecu-
lar weight species.
Treatment of Spent Caustic Extraction Liquor (SCEL) with Lime
Sufficient Ca(OH)2 was added to spent caustic extraction liquor to pro-
vide a final concentration of 15 g/1. After standing one hour, the precipi-
tate was removed by filtration and the pH of the filtrate was adjusted to 9.4
by the introduction of C02. The lime-treated liquor, after removal of CaCC3,
had the characteristics recorded in Table 25.
69
-------
TABLE 24. PROPERTIES OF ALUM-TREATED SCEL
TOC, ppm
Phenol content, % of original
LC5n (24-hr test), ppm
TOC/LC5Q (toxic units)
Original
liquor
1362
100
425
3.2
Supernate
360
12
145
2.5
Dialysis
retentate
819
68
653
1.3
TABLE 25. EFFECT OF LIME TREATMENT ON SPENT CAUSTIC EXTRACTION
LIQUOR (SCEL)
TOC, ppm
PH
Absorbance at 280
Absorbance at 400
Ether extract
TOCb, ppm
nm
nmb
Organically bound chlorine0, ppm
LCKn, ppm of TOC)
LC(-n, % (v/v) <
O'J /
Toxic units
Toxic units, % of
24-hr test
original
Original
SCEL
1630
12.4
44.28
9.98
192
26.5
350
21.5
4.7
100
Lime-treated
SCELa
493b
9.4
13.6
2.01
141
15.8
199
40.4
2.5
53
aAfter carbonation to pH 9.4
Pleasured at pH 7.0 in phosphate buffer
°Based on original volume
As shown by the data in this table, 80% of the color of the SCEL was re-
moved by lime treatment based on the absorbance decrease at 400 nm. The ab-
sorbance at 280 nm was reduced to 30% of its original value.which is approxi-
mately the same as the decrease in total organic carbon (TOC). The yield of
ether extract expressed as TOC indicates that approximately a 30% reduction
resulted from the/Time treatment. The organically bound chlorine content
was reduced by about 40% by the same treatment.
70
-------
The LCso value for the lime-treated liquor is lower than that of the
original liquor. This means that the residual organic material in the lime-
treated liquor was more toxic than that in the original liquor. In other
words, the non-toxic fraction was removed in preference to the toxic fraction
in the lime treatment. This observation is consistent with the belief that
the lime preferentially precipitates the higher molecular weight material
which, as was demonstrated in an earlier section, is less toxic than lower
molecular weight material. The total toxicity of the SCEL, however, was de-
creased about 47% when calculated on a toxic unit basis.
Treatment of Lime-Treated SCEL with Activated Carbon and Various Oxidants
The goal was to establish methods that would use a
relatively inexpensive chemical (lime) to remove a large fraction of the dis-
solved material in SCEL, thereby necessitating the use of much smaller quan-
tities of more expensive chemicals to achieve the desired effluent quality.
The tests described below were performed.
Adsorption of Lime-Treated SCEL on Activated Carbon
The adsorption treatments were performed at room temperature for an
18-hr period; samples of spent caustic extraction liquor, with pH values of
5.0 and 9.4, were used. The results are shown in Table 26 and Figure 11.
From these results, it is obvious that pH 5.0 is superior to pH 9.4 as an
initial test level, both with respect to color and TOC removal at a given
application of carbon. The amount of ether-extractable material showed a
further decrease (cf. Table 25) and the extract of the pH 5.0 liquor had very
little or no organically bound chlorine.
TABLE 26. PROPERTIES OF LIME-TREATED SCEL AFTER REACTION
WITH ACTIVATED CARBON
Initial pH
Carbon cone., g/1
Final pH
TOC, ppm
Absorbance at 280 nm
Absorbance at 400 nm
Ether extract
TOC, ppma
Organically bound
chlorine, ppma
^^^^^^^MMiMIIIVIIIII*
5
6.4
213
1.91
0.08
34
1.1
.^^fc
c n
10
6.3
213
1.03
0.03
32
0.0
^^^^^^HMBBBHOHHmUHH^BH
15
6.3
200
0.90
0.02
30
0.0
MH^ «-«^ ^
^^l
5
9.1
340
7.30
0.88
64
5.7
_
9./I --
10
9.1
302
6.29
0.50
53
4.0
,»*^ -«
15
8.8
262
4.79
0.75
51
3.8
^m^K
^^^^^IHMIIM
aBased on original volume
71
-------
Q
UJ
o
s
LL)
a:
o
p
g
o
tr
3
O
O
COLOR
TOC
INITIAL pH
5.0
o 9.4
0
a BASED ON ABSORBANCE AT 400 nm
b BASED ON VALUES OF LIME TREATED
, LIQUOR ,
5 10 15
CARBON ADDED , q/JL
20
Figure 11. Effect of carbon treatment on the color and TOC of
lime-'treated spent caustic extraction liquor.
72
-------
As shown by tests 2 and 3 in Table 27, the liquor samples treated at pH
5.0 and 9.4 showed the same residual toxicity of 30%, thereby indicating that
pH had little or no effect on the reduction of this property. Since the LC
values of these same liquors were 150 and 245 ppm of TOC, respectively, it
is evident that a greater preference for the removal of non-toxic material
existed at pH 5.0 than at pH 9.4.
The carbon dosage of 15 g/1 at pH 5.0 (Table 27, test 4) produced essen-
tially the same reduction of toxicity as the 5.0 g/1 application of carbon.
It seems, therefore, that about 30% of the toxicity is difficult to remove
with activated carbon, even when excessive amounts of the latter are applied.
If this does indeed prove to be the case, then it may be concluded that,
similar to the results of other treatments such as lime precipitation and
alum coagulation, activated carbon is limited with respect to its ability to
effect 100% removal of toxicity.
A sample of SCEL (diluted one third) was treated directly with activated
carbon in order to compare its detoxification effect with that of lime. Com-
parison of the results (test 1, Table 27) and (test 7, Table 28) clearly re-
veals the greater effectiveness of lime for toxicity removal. The implica-
tion of this finding is that lime removed a greater amount of low molecular
substances from the spent liquor than did activated carbon. Failure of acti-
vated carbon to effect a somewhat greater reduction in toxicity is somewhat
surprising in view of the fact that Betts and Wilson (57) have reported the
successful detoxification of spent chlorination liquor on treatment with
charcoal.
The small carbon particles that passed through the filter paper when
the carbon was filtered from the treated SCEL were removed by filtration
through a Millipore (5.0 y) membrane. As shown by the results of tests 8
and 9 (Table 28), there is little difference in toxicity removal with and
without the membrane treatment, thereby excluding the possibility that re-
sidual carbon particles contributed significantly to .toxicity. The above
finding also suggests that the residual toxicity of liquors corresponding
to tests 2, 3 and 4 in Table 27 was not caused by the presence of minute
quantities of activated carbon particles.
Oxidation of Lime-Treated Liquor with Ozone, Chlorine Dioxide, and Hydrogen
Peroxide--
For the purpose of reducing the number of samples subjected to bioassay,
the properties of the lime-treated SCEL were first determined as a basis for
the screening process. The results of these analyses are reported in Tables
29, 30 and 31. The results for carbon-treated liquor (Table 28) provide a
reference for comparison.
As an initial pH value, 5.0 gave better results with respect to TOC re-
moval in the cases of ozone and chlorine dioxide whereas the opposite was
true for hydrogen peroxide. TOC removals were in all instances under 25%
based on that of the lime-treated liquor. As shown by the plot in Figure 11,
73
-------
TABLE 27. ACUTE TOXICITY OF SCEL TREATED SEQUENTIALLY WITH LIME AND ACTIVATED CARBON
Test
No.
1
2
3
4
Test
No.
5
6
7
8
9d
Cone, of act.
carbon, g/1
0
5
5
15
TABLE 28.
Gone, of act.
carbon* g/1
0
0
5
15
15
Initial
PH
9.4
5.0
9.4
5.0
ACUTE TOXICITY
Initial
PH
9.4
9.4
5.0
5.0
5.0
TOC,
ppm
493
213
340
200
24-hr LC5Q
ppm of TOC % (v/v)
199 40.4
150 72.5
245 72.5
132 65.8
Toxic
units
2.5
1.4
1.4
1.5
Residual
toxic units, %
53
30
30
32
a
OF CARBON-TREATED SCEL
TOC,
ppm
543
543
_c
_
24- hr LC50
ppm of TOC % (v/v) % (v/v)b
351 64.7 21.6
348 64.0 21.3
87.5 29.2
84.3 28.1
85.5 28.5
Toxic
units
4.6
4.7
3.4
3.6
3.5
Residual
toxic units, %
(100)
74
78
76
Diluted to 1/3 its original volume
Based on original (i.e., undiluted) volume of SCEL
cNot determined
filtered through Mi Hi pore filter
-------
TABLE 29. EFFECT OF OZONE OXIDATION ON THE PROPERTIES OF LIME-TREATED
Initial pH
Ozone added, meq/1
Ozone residual, meq/1
Final pH
TOC, ppm
Absorbance at 280 nm
Absorbance at 400 nm
Ether extract
TOC, ppm
Organically bound
chlorine, ppmb
aNot determined
56
0
5.3
370
4.12
0.85
54
5.1
en
92
0
5.5
357
2.52
0.45
74
3.2
108
0
5.6
359
2.30
0.36
41
2.3
50
0
9.1
508
_a
1.02
68
8.5
H^K^B^^^^,^^^^^^
94
.4
98
0
9.1
480
-
0.52
45
5.8
^^^^^_^M
150
0
9.0
468
_
0.33
56
3.6
Based on original volume
TABLE 30. EFFECT OF
SCEL
Initial pH
C102 added, meq/1
C102 residual, meq/1
Final pH
TOC, ppm
Absorbance at 280 nm
Absorbance at 400 nm
CHLORINE
47
0
3.2
440
2.11
0.20
DIOXIDE
5 A ,__
86
33
2.5
384
2.42
0.21
ON THE
137
87
2.5
359
2.51
0.13
PROPERTIES OF
30
10
8.7
495
_a
-
LIME-TREATED
,. q A -
52
18
7.8
457
-
126
42
7.0
359
-
Ether extract
TOC, ppmb
Organically bound
chlorine, ppmb
^
aNot determined
bBased on original volume
57
5.2
39
7.7
35
6.4
141 138 139
24.0 18.0 15.9
75
-------
TABLE 31. EFFECT OF HYDROGEN PEROXIDE OXIDATION ON THE PROPERTIES
OF LIME-TREATED SPENT CAUSTIC EXTRACTION LIQUOR
Initial pH
H202 added, meq/1
H202 residual, meq/1
Final pH
TOC, ppm
Absorbance at 280 nma
Absorbance at 400 nma
118
119
5.3
479
9.18
1.15
5n
235
238
5.6
489
8.40
1.03
353
197
5.1
451
8.14
0.99
24
0
9.0
453
9.95
0.99
.., 04--
59
4
9.1
413
9.95
0.90
118
8
8.9
373
7.95
0.69
Ether Extract
TOC, ppm
Organically bound
chlorine, ppm
129
14.6
92
12.8
102
14.7
89
11.9
97
9.5
84
9.0
aBased on original volume
carbon treatment resulted in the removal of more than 50% of TOC. This large
difference may be due to the fact that the oxidation products are mainly or-
ganic acids that still contribute to TOC, whereas in the carbon treatment,
some of these acids are actually removed from solution by adsorption onto the
carbon surface. The values for the other parameters were reduced more than
that of the TOC but less than that observed after treatment with activated
carbon at pH 5.0.
Of the three oxidants applied to the lime-treated spent caustic extrac-
tion liquor, only the ozone was totally consumed (oxidation and/or decomposi-
tion) in a reasonable length of time at room temperature. The sluggishness
of the reactions of chlorine dioxide and hydrogen peroxide with the lime-
treated liquor necessitated these reaction mixtures being heated for an ex-
tensive period of time in order to exhaust the oxidant. Complete consumption
of oxidant was imperative since residual amounts of peroxide and chlorine di-
oxide (particularly the latter) were found to have a considerable positive
effect in increasing acute toxicity.
The values shown in Table 32 reveal that of the three oxidants tested,
toxicity reduction was effected only with ozone and hydrogen peroxide. The
benefits provided by treatment with these reagents were fairly modest, how-
ever, in comparison to the toxicity reduction resulting from the initial
treatment of the spent caustic extraction liquor with lime. Contrary to the
benefits accruing from the use of peroxide and ozone, chlorine dioxide
treatment actually increased the acute toxicity of the lime-treated liquor.
As noted previously, chlorine dioxide application to whole spent chlorination
liquor produced a small reduction in acute toxicity.
76
-------
TABLE 32. ACUTE TOXICITY OF LIME-TREATED CAUSTIC EXTRACTION LIQUOR REACTED WITH
OZONE, CHLORINE DIOXIDE. AND HYDROGEN PEROXIDE
Test
No.
-
-
10
11
12
Cone, of
oxidant,
Oxidant meq/1
-
-
03 50
C102a 50
H202b 50
24-hr LC50
Initial pH
(SCEL)
(Lime-Treated SCEL)
5.0
5.0
9.4
TOC,
ppm
1630
493
370
409
434
TOC,
ppm
350
199
204
61
281
% (v/v)
21.5
40.4
55.1
14.9
64.8
Toxic
Units
4.7
2.5
1.8
6.7
1.5
Residual
toxic units, %
100
53
39
143
32
aHeated at 78°C for 12 hrs after treatment
bHeated at 78°C for 4.5 hrs after treatment
-------
The amounts of chlorine dioxide, ozone and peroxide applied to the lime-
treated liquor were equivalent to the addition of 6.1, 10.8, and 7.7 g/kg of
pulp, respectively, in a hypothetical bleaching stage in which pulp consis-
tency was 10%. Thus, even in those instances where modest reductions in
acute toxicity were achieved, the cost in terms of expenditure of chemical
would have to be considered economically prohibitive and it is concluded
that none of the above-described combination processes is likely to have
much of a future as a technical approach to waste treatment.
Treatment of Spent Chlorination Liquor with Sodium Hydroxide
The pH of SCL was adjusted to various pH levels and maintained at those
levels throughout the test by the addition of aqueous sodium hydroxide.
After a predetermined reaction period at room temperature, the phenol and
organically bound chlorine contents of the solutions were monitored. The
results are shown in Tables 33 and 34.
TABLE 33. EFFECT OF ALKALI TREATMENT OF SPENT CHLORINATION LIQUOR ON
PHENOL AND ORGANICALLY BOUND CHLORINE CONTENTS
Loss of organically bound chlorine, %
pH Phenol loss, % 16 hours 12 days
2
5
6
7
8
9
10
11
12
13
0
0
7
33
42
51
69
80
73
M57
0
11
-
14
-
19
38
65
-
-
18
36
-
44
-
55
74
75
-
-
The phenol loss was difficult to quantify accurately due to a shift in
the absorption maximum from 290 to 310-320 nm in the ionization difference
spectrum. Nevertheless, the results clearly show a large decrease in phenol
content, particularly in the pH 7-10 range. The loss in phenol content was
paralleled by a decrease in organically bound chlorine, which was time de-
pendent. The maximum values for the loss of phenol and organically bound
chlorine were identical (^75%) and might be related although this was not
established.
78
-------
TABLE 34. RATE OF REMOVAL OF ORGANICALLY BOUND CHLORINE
FROM SPENT CHLORINATiON LIQUOR WITJ-1 INCREASING pH
Loss of organically bound chlorine, %
ay 9 Day 12
2
5
7
9
10
11
<1
11
14
19
38
65
15
-
-
53
72
75
18
36
44
55
74
75
The effect of pH on the rate of removal of organically bound chlorine
from spent chlorination liquor is revealed by the data in Table 34. The
trend for enhancement of the rate with increase in pH is clearly evident,
especially in the pH 9-10 interval. The maximum loss of chlorine appears
to be reached in shorter periods as the pH is raised.
In tests designed to determine the effect of an alkaline treatment on
acute toxicity, the pH of samples of spent chlorination liquor (SCL) was
adjusted to 7, 9, 10 and 11 by the addition of sodium hydroxide solution
and allqwed to stand in the dark for 24 hours at 5°C. The above samples
and an untreated pH 2 SCL control were bioassayed after the 24-hour period
using Daphm'a magna as the test organism.
The effect of these alkaline treatments, as indicated by the data in
Table 35, was to cause a reduction in the acute toxicity of SCL. The amount
of toxicity reduction increased with increasing treatment pH. The benefi-
cial effect of alkali with respect to toxicity reduction is in agreement
with the results of an earlier investigation reported by Leach et al. (36)
who detoxified spent chlorination liquor with combined aeration and alkaline
treatment over a six-day period.
Several of the bioassays reported in Table 35, especially corresponding
to the higher pH treatment levels, gave unsatisfactorily high g values. In
all cases, this is attributable to low response rates even at the highest
concentration tested (100%, 218 ppm TOC) and is not the result of hetero-
geneous concentration-response patterns.
Detoxification values, A, were calculated as the ratio of the EC50
(immobility as the response criterion) at pH 2 for a given exposure period
divided by the EC5n of the alkaline treated sample for the same exposure
period These values suggest a consistent decrease in potency with in-
creasing pH as is illustrated in Figure 12. Although good multiple estimates
of potency over a wide pH range are available only for 48-hour exposure per-
iods, these data suggest that the relationship between detoxification and
treatment pH is linear.
79
-------
TABLE 35. EFFECT OF ALKALI ON THE TOXICITY OF SPENT CHLORINATION LIQUOR
TO DAPHNIA MAGNA
MMBKBB^KKMHIIBMI^KB^VIIIMMIH
Treatment
PH
pH 2
pH 7b
pH 9
pH 10
pH 11
pH 11
' -
Exposure
period
24
48
96
24
48
120
24
48
96
24
48
96
24
48
120
24
48
96
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
EC50 (95%
interval )
95.3
37.5
<41.4
117
51.0
41.4
338
85.4
64.5
>218
152
146
>218
>218
534
>218
233
233
(53.
(14.
(805?
(91.
(24.
(26.
(72.
(60.
,(42.
{10%
(56.
(37.
{0%}
{0%}
(92.
{10%
(90.
(90.
confidence
in ppm TOC
6 -
5 -
;} C
8 -
6 -
8 -
2 -
0 -
7 -
T **
2 -
9 -
c
c
7 -
"I
3 -
3 -
169)
97.7)
150)
106)
63.9)
1571)
122)
97.7)
412)
567)
3075)
604)
604)
Aa
1.
1.
1.
0.
0.
0.
0.
0.
<0.
<0.
0.
<0.
<0.
<0.
0.
<0.
0.
<0.
00
00
00
80
84
91
28
50
64
43
28
28
43
22
07
43
16
18
1.
0.
0.
0.
0.
1.
0.
0.
2.
2.
1.
0.
0.
g
270
518
-
219
550
438
01
386
255
_
37
97
_
-
41
_
565
565
of pH 2 liquor for given exposure period * LC$Q of liquor treated at
^specified pH for the corresponding exposure period.
pH initially adjusted to specified value and not thereafter controlled; in
all other instances, alkali was periodically added to restore pH to initial
value.
c% mortality at specified concentration.
In the course of the alkaline treatment of spent chlorination liquor,
the pH fell below the prescribed level, seemingly as the result of the hy-
drolysis of organically bound chlorine. In one bioassay, the pH was periodi-
cally readjusted back to the initial value (pH 11) by addition of alkali
whereas, in another test, the pH was not readjusted. The results in Table
35 (last two tests) indicate that this difference in test procedure did not
have a statistically significant effect on toxicity.
The effect of prolonged (two week) treatment of spent chlorination li-
quor at pH 7 is summarized in Table 36. For ECgn's at 48 and 96 hours, this
treatment resulted in a statistically significant decrease in toxicity. Com-
paring the potency values in Table 36 with the pattern in Figure 12, treat-
ment at pH 7 for two weeks appears to have about the same detoxification
effect as treatment :-at pH 10 for 24 hours.
80
-------
co
I.OT
0.8-
0.6--
0.4-
0.2- -
+
8 9 10
TREATMENT pH
II
Figure 12. Detoxification (A) of spent chlorination liquors by alkaline treatment.
-------
TABLE 36. EFFECT OF PROLONGED (TWO WEEK) NEUTRALIZATION ON THE TOXICITY
OF SPENT CHLORINATION LIQUOR TO DAPHNIA MAGNA
Sampl e
SCL,
pH 2
SCL,
pH 7
Exposure
period
24
48
96
24
48
96
hrs.
hrs.
hrs.
hrs.
hrs.
hrs.
EC50 (95%
interval)
172
70.4
55.3
>218
229
138
(92.
(56.
(48.
confidence
in ppm TOC
8
7
3
{20%}
(131
( 91
- 321 )
- 87.4)
- 63.3)
b
- 399)
- 211)
A3
1.
1.
1.
<0.
0.
0.
00
00
00
71
31
40
0.
0.
0.
0.
0.
g
620
149
185
»
721
378
LCrn of pH 2 liquor * corresponding LCcn of pH 7 treated liquor.
OU J w
Percent mortality at specified concentration.
BIOLOGICAL TREATMENT OF SPENT CHLORINATION AND CAUSTIC EXTRACTION LIQUORS
i
The microorganisms used in the treatments consisted of a mixed micro-
bial population (hereafter referred to as "seed") contained in a sludge from
a commercial pulp and paper waste treatment facility and acclimated to
bleaching wastes and of pure cultures of Candida utilis (yeast) and an un-
identified bacterium. The latter microorganism was isolated from the sludge
referred to above. The biological treatment conditions are described in the
Materials and Methods section. Following one-week treatment periods, the
liquors were subjected to acute toxicity bioassay and the results obtained
are reported in Table 37.
Treatment of spent chlorination liquor with seed in the presence of
phosphate and asparagine (tests 4 and 6, resp.) resulted in complete de-
toxification. Surprisingly, the controls corresponding to the above (tests
3 and 5, resp.) which differed from the test samples only by the absence
of seed, proved to be considerably more toxic to Daphnia magna than the
original liquor. The explanation for the increased toxicity of the control
is not immediately apparent, but in an independent experiment, it was estab-
lished that a phosphate solution of the same concentration was, by itself,
non-toxic to Daphnia magna.
In the case of spent caustic extraction liquor, the addition of phos-
phate in the absence of seed produced a liquor having a toxicity little
different from that of the original whereas a substantial toxicity reduc-
tion occurred in the presence of seed (tests 7 and 8, resp., Table 37).
When asparagine was added as a nutrient, the toxicity of the control sample
(test 9) once again surpassed that of the original liquor. In the presence
of seed, however, the toxicity of the liquor was reduced slightly as compared
to the value for the original liquor (cf. tests 2 and 10).
82
-------
TABLE 37. EFFECT OF BIOLOGICAL TREATMENTS OF SCL AND SCEL ON ACUTE TOXICITY
00
CO
Test
No.
1
2
3
4
5
6
7
8
9
10
Sample Description
Original SCL
Original SCEL
v>
c
id
*l
o
CO
S
O/
^V
OJ
'
o
1/1
S_
o
«l
to\
SCL - Control; 5.7 x 10"4 M
in K2HP04
SCL - Test; 5.7 x 10"4 M
On K2HP04 + seed
rSCL - Control; 7.4 x 10"5 M
|in Asparagine
I
/SCL - Test; 7.4 x 10~5 M in
rKH2P04; 15 x TO'3 M in
^asparagine + seed
'SCEL - Control; 7.4 x 10"4 M
Un KH2P04
* ..
)SCEL - Test; 7.4 x 10"4 M in
fKH2P04 + seed
^ -R
fSCEL - Control; 5.7 x 10 D M
\in KH2P04; 1.5 x 10~3 M in
f asparagine
/SCEL - Test; 5.7 x 10"5 M in
KH2P04; 1.5 x TO'3 M in
^asparagine + seed
Test
Period,
hrs.
24
48
24
48
24
48
24
48
24
48
24
48
24
48
24
48
24
48
24
48
L
% (v/v)
43.4 (55.3-33.6)
33.2 (45.5-21.7)
37.3 (48.3-29.0)
27.6
<13 ( )a
<13 ( - )a
>100 ( - )£
>100 ( - )D
16.9 (22.4 - 12.8)
15.6 (20.7 - 11.8)
>100 ( - )b
>100 ( - )b
45.2
45.2
74.5 (96.3-57.7)
46.9 (57.2-38.4)
9,2 ( - )a'd
9.2 ( - )a
52.7 (62.0-44.8)
48.6 (59.3-39.8)
CRn (95% C.I.)
ppm of TOC
102 (130-79)
78 (107-51)
501 (648-389)
371 '
<23 ( - )!
<23 ( - )a
>H4 ( - )k
>114 ( - )
_c
_c
542
542
708 (915-548)
446 (543-365)
_c
c
_c
(continued) -c
-------
TABLE 37 (continued)
CO
-pa
Test
o.
11
12
13
14
15
16
17
18
I/I
M
3
03
-a
re
CM
CO
1-00
< 20
55.1
36.7
- Y
80.8 (101.6-64.2)
33.8 (34.8-28.7)
4.4
- )S
>13 (too erratic to determine)
( - )a
38.6 (49.6-30.1)
25.3 (31.0-20.6)
47.3
30.0
Mortality too high to calculate.
Mortality too low to calculate.
^Incalculable because carbon source (asparagine) was added as nutrient
Erratic death pattern.
-------
Since neither asparagine, which serves as a nutrient, nor phosphate
which functions both as a buffer and a nutrient, are required for adequate
growth of the microorganisms comprising the seed, their elimination indi-
vidually and in combination in a repetition of the aforegoing tests could
help in identifying the cause of the high toxicity often found in the con-
trol tests.
When properly performed, biological treatment generally has been effec-
tive in reducing or eliminating the toxicity of bleached kraft mill wastes
(59-63). In this connection, the available toxicity data, as can best be
determined, applies exclusively to the combined chlorination and caustic
extraction effluent rather than to individual bleach streams, and thus
represents the composite property. The results in Table 37 indicate that
biological treatment of the segregated streams (SCL and SCEL in this case)
does not necessarily effect full or equivalent toxicity reduction since, in
this particular treatment, SCL was detoxified to a greater extent than SCEL.
Unlike the previous situation,involving a mixed microbial population
("seed") from a sludge, an external nutrient source (glucose was used) was
required to produce a respectable amount of microorganism growth in the
cases of the experiments with Candida utilis (yeast) and the bacterium.
With respect to the former microorganism, SCL was successfully detoxified
in the presence of phosphate and glucose based on the result from the 24-
hour test. After 48-hour exposure to the liquor, however, the Daphnia magna^
displayed an exceptionally high incidence of death (test 12). As in other
previous instances, the control sample was considerably more toxic than the
untreated liquor (cf. tests 1 and 11).
The results from the treatment of SCEL with yeast in regard to de-
gree of detoxification generally paralleled those obtained when a mixed pop-
ulation of microorganisms (seed) was applied (compare tests 13 and 14 with
7 and 8, Table 37).
In addition to its potential utility as a microorganism capable of
effecting the biodegradation of SCL and SCEL, Candida utilis was also se-
lected by reason of its being an edible protein source. Therefore, to the
extent that it would be able to use SCL and SCEL as carbon sources for pro-
tein production, this fungus offered some interesting possibilities. Unfor-
tunately, the carbon concentration provided by the organic solids in the two
liquors was insufficient to provide the amount of growth required to make
such a process fully practicable. Thus, although the future of Candida
utilis does not appear to be particularly bright with respect to the appli-
cltToF described above, additional work should be performed before this
approach is abandoned.
Generally speaking, the bacterium was ineffectual in reducing the
toxicity of either SCL or SCEL (tests 15-18, Table 37), and the death pat-
tern of the Daphnia magna was very erratic in the tests employing the
former liquor.
85
-------
REDUCTION IN SCL AND SCEL TOXICITY THROUGH MODIFICATIONS OF THE BLEACHING
PROCESS
An alternative to the approach of eliminating bleaching liquor toxicity
by chemical or biological treatment of the bleaching effluent is to modify
conventional bleaching stages in such a way that the offending chemicals are
either not produced or are formed in smaller amounts. The latter approach
has been adopted by several investigators for the purpose of achieving not
only toxicity reduction but color, COD,and BOD improvements as well. Notable
in this connection are modifications involving the partial replacement of
chlorine by chlorine dioxide in the first stage (57, 64, 65), replacement of
chlorine with hypochlorite (66), substitution of the first caustic extrac-
tion stage by a hypochlorination treatment (64), and the addition of hydro-
gen peroxide to caustic extraction stages (67). In another investigation
(65), novel bleaching sequences involving combinations of gaseous chlorine
and chlorine dioxide, oxygen and hydrogen peroxide have been explored as
possible means of reducing effluent toxicity.
In the present investigation, several combinations of the modification
treatments cited above were evaluated for their effect on toxicity. Except
as otherwise noted, the pulp used in the bleaching tests was a commercial
sample of southern pine kraft taken from the same batch used for the prepa-
ration of the spent liquors. The "conventional" and process-modified li-
quors were subjected to acute toxicity bioassays using Daphnia magna as the
test organism. The test conditions and LCsg values at various periods of
exposure appear in Table 38. In the case of the 24-hour tests, the results
are also expressed as toxic units.
When the caustic extraction stage was replaced by hypochlorite and alka-
line hydrogen peroxide treatments (Table 38, tests 3 and 4, resp.), the
acute toxicity was not substantially different from that found when a con-
ventional extraction stage was employed. The finding, with respect to
hypochlorite treatment, is contrary to that reported by Gall and Thompson
(64); who determined that introduction of hypochlorite into the second stage
extraction had the effect of reducing the acute toxicity of the resulting
effluent.
Sequential C102/C12 treatment (Table 38, test 5) caused a pronounced
reduction in effluent toxicity as compared to the situation where an equiva-
lent amount of chlorine alone was applied (test 1). Subsequent caustic ex-
traction of the C102/Cl2-treated pulp produced an effluent with a somewhat
lower toxicity than that resulting from caustic extraction of the pulp re-
acted with chlorine alone (cf. tests 2 and 6). The introduction of peroxide
into the extraction stage of the pulp from a C102/C12 sequential treatment
also proved advantageous. On the other hand, the use of an equivalent
amount of hypochlorite in the caustic extraction of the same pulp at the
very best produced no advantage and may even have had a detrimental effect
with respect to acute toxicity (cf. tests 2 and 8).
86
-------
TABLE 38. TOXICITY OF EFFLUENTS FROM CONVENTIONAL AND PROCESS-MODIFIED BLEACHING STAGES
00
Test no.
1
2
3
4
5
6
7
8
9
10
11
12
Treatment or
treatment sequence
C
C,E
C,H
C,P
D * C
D * C,E
D + C,P
D + C,H
D
D.E
D,H
A/0
LC5Q, ppm TOCb
24-hr
94
480
430
524
186
660
789
350
380
600
430
560
48-hr
51
410
280
467
55
421
476
190
150
423
265
400
72-hr
_
380
220
-
-
-
-
-
100
340
180
350
96- hr
_
-
160
-
-
-
«*
-
70
310
140
280
24-hr toxic units
(TOC/24-LC TOO
2.4
2.9
3.0
2.9
1.4
2.3
1.8
3.6
0.6
2.0
2.6
-
aC = chlorine (5.25%); E = caustic extraction (3.6% NaOH); P = hydrogen peroxide (0.5%);
H = sodium hypochlorite (2.0%); D = chlorine dioxide (2.0%); D + C = sequential chlorine
dioxide (1.0% C102); A/0 = alkaline-oxygen (80 psig oxygen).
Values corresponding to second-stage effluents in two-stage treatments.
-------
When the first bleaching stage consisted of the sole application of
chlorine dioxide (test 9), toxicity reduction was further enhanced as com-
pared to the cases where an equivalent amount of chlorine was used alone
or sequentially in combination with chlorine dioxide (tests 1 and 5, resp.).
Subsequent extraction of the dioxide-treated pulp with alkali (test 10) gen-
erated a liquor having an acute toxicity less than that shown by the caustic
extraction liquor derived from the chlorinated pulp (cf, tests 10 and 2) but
in the same range as the liquor obtained by extracting the C102/Cl2-treated
pulp. Extraction of the dioxide-treated pulp with hypochlorite (test 11)
appeared, on the other hand, to enhance slightly the toxicity of the result-
ing liquor as compared to the case where a conventional alkali extraction
was employed (cf. tests 10 and 11).
The final test recorded in Table 38 (test 12) was performed in an in-
dustrial laboratory on a sample of southern pine kraft (Kappa No. 39.2) and
a portion of the spent liquor was provided for acute toxicity bioassay.
The effectiveness of the alkali/oxygen treatment is probably best judged by
comparison with the toxicity of other alkaline spent liquors listed in Table
38. Such a comparison leads to the conclusion that the acute toxicity level.
in this case,is in the same range as those for the other alkaline liquors
listed in the table with the exception of the situations where hypochlorite
was employed. However, the advantage of the A/0 treatment lies in the fact
that it requires no prior bleaching stage as the other modification sequences
in Table 38 do, thus eliminating a source of material having a relatively
high toxicity.
Since the foregoing bioassay results indicated the value of replacing
all or part of the chlorine by chlorine dioxide in the first stage as a
means of reducing the acute toxicity of first- and second-stage effluents,
these spent liquors were subjected to additional characterization tests in
an effort to understand the reason for this effect. The results of these
analyses are shown in Table 39.
The partial or total replacement of chlorine by chlorine dioxide is seen
not to have had a particularly significant effect on either the color or TOC
levels of the first-stage effluents. On the other hand, the color of the
effluents from the subsequent caustic extraction treatments decreased signi-
ficantly with increasing substitution of chlorine dioxide for chlorine in the
first stage. The TOC contents of the extraction liquors decreased signifi-
cantly only as a result of a 100% replacement of chlorine by chlorine dioxide
in the first stage.
As expected, the chloride ion concentrations of all the liquors included
in Table 39 decreased as the proportion of chlorine dioxide used in the first
stage increased. Because of the practice followed in this investigation of
washing chlorine- and/or chlorine dioxide-treated pulps with dilute hydro-
chloric acid rather than with tap water, the chloride ion contents of the
caustic extraction liquors corresponding to pulp washed in this manner are
somewhat higher than normal. The magnitude of this difference can be judged
by comparing the final two values under the heading "chloride" in Table 39.
88
-------
TABLE 39. CHARACTERISTICS OF SELECTED CONVENTIONAL AND PROCESS-MODIFIED SPENT BLEACHING LIQUORS
00
vo
Treatment or
treatment
sequence3
C
D -> C
D
C,E
D -» C,E
D.E
Dd,E
Color ,
(pt-co units)
1290
1360
1120
17,930
13,100
7190
-
TOC
(ppm)b
228
260
226
1410
1515
1220
-
Chloride
(ppm)5
973
600
179
840
380
105
45
Org. -bound
chlorine,
ppmb
90
97
107
210
124
5
10
Org. chlorine
TOC
0.394
0.373
0.474
0.148
0.082
0.004
-
Aab*c
0.53
0.61
0.39
0.28
0.23
0.10
-
Aa ln-3
ToTx 10
2.3
2.3
1.7
0.20
0.15
0.08
»
C = chlorine; D = chlorine dioxide; E = caustic extraction; D -» C = sequential chlorine dioxide-
chlorine treatment.
Values correspond to second-stage effluents in two-stage treatments.
Measurement at 290 nm for acidic liquors and 315 nm for alkaline liquors.
dDioxide-treated pulp washed with 0.01 N HS0<
-------
As expected, the organically bound chlorine contents (Table 39) of the
first-stage liquors were considerably higher than those of the corresponding
caustic extraction liquors. The organically bound chlorine content of the
latter decreased sharply as chlorine was replaced by chlorine dioxide, and at
100% replacement, the organically bound chlorine content of the liquor was
essentially nil. This finding is in general agreement with results reported
by Rapson and Anderson (68). In the case of the first-stage liquors, the
trend is not as clear cut, but the organically bound chlorine content of the
liquor from the 100% dioxide treatment is substantially higher than for the
liquors from the C and D -> C stages. Inasmuch as the toxicity of the D stage
effluent was the lowest of the first-stage liquors (see Table 38), the ex-
pected direct relationship between organically bound chlorine and toxicity
was not demonstrated. However, the organically bound chlorine analysis in-
cludes both aliphatic- and aromatic-bound chlorine, and it is the latter
which is generally thought to be the more important contributor to toxicity.
The A-absorbance (Aa) values in Table 39 are assumed to represent prin-
cipally changes in the phenolic content of the liquors. The problems associ-
ated with the use of ionization difference spectroscopy for such a purpose
have been discussed previously, and it suffices to emphasize here that the
measured changes in Aa values at best merely reflect possible changes in
phenol content. With this in mind, the Aa data in Table 39 reveal that a
significant decrease in "phenol" content was effected only in the case where
the chlorine was completely replaced by chlorine dioxide. The corresponding
caustic extraction liquor showed a systematic decrease in "phenol" content
with increasing replacement of chlorine by dioxide in the first stage, and
this change paralleled the decrease in organically bound chlorine content
of the same liquors. In none of the tests performed in this investigation
was the "phenol" content reduced to zero, and it may be speculated that such
a condition could not be achieved without a prohibitive expenditure of
chemical.
90
-------
REFERENCES
1. Leach, J.M., and A.N. Thakore. Identification of the Toxic Constituents
in Kraft Mill Bleach Plant Effluents. CPAR Rep. No. 245-2, Can.
Forestry Service, Ottawa, Ont., 1975.
2. Gess, J.M., and C.W. Dence. The Formation of o-Benzoquinones in the
Reaction of Creosol with Aqueous Chlorine. Tapoi, 54(7):1114, 1971.
3. Cain, J.R. Ozonation and Biodegradation of Phenols and Chlorophenols.
M.S. Thesis, State University of New York, College of Environmental
Science and Forestry, Syracuse, N.Y., 1977. 89 pp.
4. Nonni, A.J. The Reactions of Creosol and Propioguaiacone with Combi-
nations of Chlorine and Chlorine Dioxide. M.S. Thesis, State Univer-
sity of New York College of Environmental Science and Forestry, Syra-
cuse, N.Y., 1979. 145 pp.
5. Dence, C.W., J.A. Meyer, K. Unger, and J. Sadowski. Steric Effects in
the Dealkylation of Phenol Ethers with Aqueous Chlorine. Taopi,
48(3):148-157, 1965.
6, Willstatter, R., and F. Muller. Zur Kenntnis der o-Chinone. Chem. Ber.,
44:2171-2191, 1965.
7. D'Agostino, A.S.,and L. Provasoli. Dioxinic Culture of Daphnia magna.
Straus. Biol. Bull., 139:485, 1970.
8. , Methods for Acute Toxicity Tests with Fish, Inverte-
brates, Microinvertebrates, and Amphibians. EPA-660/3-75-009, U.S.
Environmental Protection Agency, Washington, D.C., 1975. 67 pp.
9. Goldschmid, 0. The Determination of the Phenolic Hydroxyl Content of
Lignin Preparations by Ultraviolet Spectrophotometry. Anal. Chem.,
26(9):1421-1423, 1954.
10. Ziobro, R.J. The Microfungi of Kraft Mill Wastes and Their Biodegra-
dation of Chlorinated Phenols. M.S. Thesis, State University of New
York, College of Environmental Science and Forestry, Syracuse, N.Y.,
1978. 77 pp.
11. Finney, D.J. Probit Analysis. Cambridge University Press, London,
1971. 333 pp.
91
-------
12. Weil, C.S. Tables for Convenient Calculation of Median-Effective Dose
(LDco and ED50) and Instruction in Their Use. Biometrics, 8:249-263,
1952.
13. Thompson, W.R, and C.S. Weil. On the Construction of Tables for Moving
Average Interpolation. Biometrics, 8:51-54, 1952.
14. Thompson, W.R. The Use of Moving Averages and Interpolation to Esti-
mate Median-Effective Dose. I. Fundamental Formulae, Estimation and
Error and Relation to Other Methods. Bact. Review, 11:115-145, 1947.
15. . Algal Assay Procedure: Bottle Test. EPA- ,
National Eutr.ophication Research Proaram, U.S. Environmental Protection
Agency, , 1971. 82 pp.
16. . Standard Methods for the Examination of Water and
Wastewater. 12th Edition. American Public Health Association, New
York, 1967. p. 415.
17. Brown, H.B. Significance, Measurement and Reduction of Color in Pulp
Mill Effluents. Technical Bull. No. 85. National Council of the Paper
Industry for Air and Stream Improvement, New York, N.Y., 1957. 21 pp.
18. Siggia, S. Quantitative Organic Analysis via Functional Groups. John
Wiley and Sons, Inc., New York, 1963. 697 pp.
19. Shiner, V.J., Jr. and M.L. Smith. Rapid Argentimetric Determination
of Halides by Direct Potentiometric Titration. Anal. Chem., 28(6):
1043-1045, 1956.
20. Braddon, S.A.,and C.W. Dence. Structure and Reactivity of Chlorolignin.
I. Alkaline Hydrolysis of Chlorine-Substituted Lignin Model Compounds.
Tappi, 51(6):249-256, 1968.
21. Leach, J.M.,and A.N. Thakore. Isolation and Identification of Consti-
tuents Toxic to Juvenile Rainbow Trout (Salmo gairdneri) in Caustic
Extraction Effluents from Kraft Pulpmill Bleach Plants. J. Fish. Res.
Board Can., 32(8): 1249-1257, 1975.
22. Rogers, I.H. Isolation and Chemical Identification of Toxic Components
of Kraft Mill Wastes. Pulp Paper Mag. Can., 74:T303-T308, 1973.
23. Lindstrom, K., and J. Nordin. Gas Chromatography-Mass Spectrometry of
Chlorophenols in Spent Bleach Liquors. J. Chrom., 128:13-26, 1976.
24. Durkin, P.R. Spent Chlorination Liquor and Chlorophenolics: A Study
of Detoxication and Joint Action Using Daphnia magna. Ph.D. Thesis,
State University of New York, College of Environmental Science and
Forestry, Syracuse, N.Y., 1979. 144 pp.
92
-------
25. Windaus, 6., and E. Petermann. Determining Toxicity of Paper Mill
Wastes by Microbiological Methods. Das Papier, 29(5):200-206, 1975.
26. Tunstall, E.W., and W. Salinas. Daphnia pulex Pulls Its Weight in Pulp
Mill Toxicity Tests. Pulp Paper Mag. Can., 78(4):T93-T98, 1977.
27. Sprague, J.B. Measurement of Pollutant Toxicity to Fish. Water Re-
search, 3:793-821, 1969.
28. Pfister, K., and E. Sjostrbm. Characterization of Spent Bleaching Li-
quors. Part 1. Ultrafiltration of Effluents from Conventional and
Oxygen Bleaching Liquors. Svensk Papperstidn., 81(6):195-2n5, 1978.
29. Leach, J.M., and A.N. Thakore. Identification of the Constituents of
Kraft Pulping Effluent that are Toxic to Juvenile Coho Salmon (Oncpr-
hynchus kisutch). J. Fish. Res. Board Can., 30:379-484, 1973.
30. Sameshima, K., and T. Kondo. Study on the Color of Waste Liquor of Pulp
Industry. I. The Relationship between the Color of Waste Liquor in the
Kraft Multistage Bleaching and the Isolated Cl2-0xylignin. Mokuzai
Gakkaishi, 16:347-352, 1970.
31. Bennett, D.J., C.W. Dence, F.-L. Kung, P. Luner, and M. Ota. The
Mechanism of Color Removal in the Treatment of Spent Bleaching Liquor
with Lime. Tappi, 54(12):2019-2026, 1971.
32. Ota, M., and C.W. Dence. The Characterization of Residues in Spent
Chlorination Liquor. In: Proceedings of the 25th Industrial Waste
Conference, Part One, Purdue University, West Lafayette, Indiana, 1970.
pp. 283-293.
33. Shimada, K. Organic Compounds in Red Pine Kraft Bleaching Liquor.
J. Japan Tappi, 31(2):97-103, 1977.
34. Ander, P., K.-E. Eriksson, M.-C. Kolar, K. Kringstad, U. Rannug, and
C. Ramel. Studies on the Mutagenic Properties of Bleaching Effluents.
Svensk Papperstidn., 80(14):454-459, 1977.
35. Lindstrbm, K., and J. Nordin. Identification of Some Neutral Chlori-
nated Organic Compounds in Spent Bleach Liquors. Svensk Papperstidn.,
81(2):55, 1978.
36. Leach, J.M., J.C. Mueller, and C.C. Walden. Degradability of Toxicants
in Chlorination-Stage Effluents. Trans. Tech. Sec. CPPA, 4(3).TR-90-
TR-94, 1978.
37. Bailey, P.S. The Reactions of Ozone with Organic Compounds. Chem.
Rev., 58:925-1010, 1958.
38. Belew, J.S. Ozonization. In: Oxidation, Vol. I, R.L. Augustine, ed.
Marcel Dekker, New York, N.Y., 1969. pp. 262-272.
93
-------
39. Soteland, N. Oxidized Lignin After Ozone Treatment of Western Hemlock
Groundwood, Part II. Norsk Skogind., 25(5): 135-139, 1971.
40. Niegowski, S.J. Destruction of Phenols by Oxidation with Ozone. Ind.
Eng. Chem., 45(3):632-634, 1953.
41. Hatakeyama, H., T. Tonooka, J. Nakano, and N. Migita. Ozonization of
Lignin Model Compounds. Kogyo Kogatu Zasshi, 70(12):2348-2356, 1967.
42. Eisenhauer, H.P. The Ozonization of Phenolic Wastes. J. Water Poll.
Control Fed., 40(11):1887-1899, 1968.
43. Wingard, L.B., Jr.,and R.K. Finn. Oxidation of Catechol to cis.cis-
Muconic Acid. I & EC Product Research and Development, 8(1):65-70,
1969.
44. Gilbert, E. Ozonolysis of Chlorophenols and Maleic Acid. In: Proceed-
ings of the 2nd Int'l. Symposium on Ozone Technology, Ozone Press
Internationale, Jamesville, N.Y., 1975. pp. 253-261.
45. Gould, J.P.,and W.J. Webber, Jr. Oxidation of Phenols by Ozone. J.
Water Poll. Control Fed., 48(1):47-60, 1976.
46. Kratzl, K., P. Claus, and G. Reichel. Reactions of Lignin and Lignin
Model Compounds with Ozone. Tappi. 59(11):86-87, 1976.
47. Bernatek, E., J. Maskelind, and K. Valen. Ozonolysis of Phenols. II.
Catechol, Resorcinol, and Quinol. Acta Chem. Scand., 15:1454-1460,
1961.
48. Bailey, P.S. Reactivity of Ozone with Various Organic Functional
Groups Important to Water Purification. In: Proceedings of the 1st
Int'l. Symposium on Ozone for Water and Wastewater Treatment, The
International Ozone Institute, Inc., Waterbury, Connecticut, 1975.
pp. 101-119.
49. Nebel, C., R.D. Gottschling, J.L. Holmes, and P.C. Unangst. Ozone
Oxidation of Phenolic Effluents. In: Proceedings of the 2nd Int'l.
Symposium on Ozone Technology, Ozone Press Internationale, Jamesville,
N.Y., 1975. pp. 374-392.
50. Eisenhauer, H.R. Increased Rate and Efficiency of Phenolic Waste
Ozonization. J. Water Poll. Control Fed., 43(2):200-208, 1971.
51. Haigne, J. and H. Bader. The Role of Hydroxyl Radical Reactions in
Ozonation Processes in Aqueous Solutions. J. Polymer Research, 10:177-
186, 1976.
52. Dence, C.W., M.K. Gupta, and K.V. Sarkanen. Studies on Oxidative Delig-
nification Mechanisms. II. Reactions of Vanillyl Alcohol with Chlorine
Dioxide and Chlorite. Tapoi, 45(l):29-38, 1962.
94
-------
53. Lindgren, B.O. Chlorine Dioxide and Chlorite Oxidations of Phenols
Related to Ligmn. Svensk Papperstidn., 74(3):57-63, 1971.
54. Strumila, G.B.,and W.H. Rapson. Chlorine Dioxide Oxidation of Lignin
Model Compounds. Trans. Tech. Sec. CPPA, 3(4):TR119-TR125, 1977.
55. Strumila, G.B.,and W.H. Rapson. The Dynamics of Reaction Products of
Chlorine Dioxide Oxidation of Acetovanillon. Trans. Tech. Sec. CPPA,
4(2):TR34-TR42, 1978.
56. Leach, J.M., J.C. Mueller, and C.C. Walden. Biodegradability of Toxic
Compounds in Pulp Mill Effluents. Trans. Tech. Sec. CPPA, 3{12):TR126-
TR130, 1977.
57. Betts, J.L. and G.G. Wilson. New Methods for Reducing the Toxicity of
Kraft Mill Bleachery Wastes to Young Salmon. Pulp Paper Mag. Can.,
68(2):T53-T56, T58, 1967.
58. Whittemore, R.C.,and J.J. McKeown. Preliminary Laboratory Studies of
the Decolorization and Bactericidal Properties of Ozone in Pulp and
Paper Mill Effluents. Technical Bull. No. 269. National Council for
Air and Stream Improvement, New York, 1974. 40 pp.
59. Nebel, C., R.D. Gottschling, and H.J. O'Neill. Ozone: A New Method to
Remove Color in Secondary Effluents. Pulp & Paper, 40(9):142-145, 1974.
60. Mueller, J.C.,and C.C. Walden. Detoxification of Bleached Kraft Mill
Effluents. J. Water Poll. Control Fed., 48(3):502-510, 1976.
61. Mueller, O.C., J.M. Leach, and C.C. Walden. Detoxification of Bleached
Kraft Mill Effluents - A Manageable Problem. Tappi, 60(9):135-137,
1977.
62. Walden, C.C., and T.E. Howard. Effluent Toxicity Removal on the West
Coast. Pulp & Paper Mag. Can., 75(11):T370-T374, 1974.
63. Jank, B.E., D.W. Bissett, V.W. Cairns, and S. Metitosh. Toxicity Re-
moval from Kraft Bleachery Effluent with Activated Sludge. Pulp Paper
Mag. Can., 76(4):T117-T122, 1975.
64. Gall, R.J., and F.H. Thompson. The Anti-Pollution Sequence-A New
Route to Reduced Pollutants in Bleach Plant Effluent. Tappi, 56(11):
72-76, 1973.
65. Wong, A., M. LeBourhis, R. Wostradowski, and S. Prahacs. Toxicity, BOD,
and Color of Effluents from Novel Bleaching Processes. Pulp Paper Mag.
Can., 79(7):T235-T240, 1978.
66. Moy, W.A., K. Sharpe, and G. Betz. New Bleaching SBK Cuts Effluent
Color and Toxicity. Pulp Paper Mag. Can., 76(5):T166-T169, iy/b.
95
-------
67. DeLattre, M.G, and 6. Papageorges. Introduction of Hydrogen Peroxide
in the Alkaline Extraction Stages During the Bleaching of Kraft Pulps.
(Translation of a lecture given at the 6th Annual Convention of the
Technical Brasilian Association of Cellulose and Paper, Sao Paulo,
Brasil, 19-23 November 1973) Interox Coordination. Solvay & CIE,
Brussels, Belgium, 1974. 21 pp.
68. Rapson, W.H., and Anderson, C.B. Mixtures of Chlorine Dioxide and Chlo-
rine in the Chlorination Stage of Pulp Bleaching. Pulp Paper Mag. Can.,
67(1):T47-T55, 1966.
96
-------
APPENDIX'A
Jack Meyer's Modified Medium for Chlorella
Stock Solutions ml/I
1 M KN03 12
1 M MgS04 . 7H20 10
1 M KH2P04 9
Fe Versenol9 1
Micronutrients 1
1 M Ca(N03)2 0.1
NaHC03 (15 g/1) 1
Double distilled water 966
1000 ml
aFe Versenol: Dissolve 26.1 g Versenol in 276 ml 1 M KOH. Add
double distilled water to 500 ml. Dissolve 24.9 g
FeS04 . 7H20 in 500 ml double distilled water. Mix
the two solutions stirring rapidly and aerate over-
night. Store in a foil covered container under
refrigeration because the rotation is light and
temperature sensitive.
Micronutrients:
Ingredients Cone, in Stock Solution g/1
CoCl2 . 6H20 0.040
HB03 2.86
MnCl2 . 4 H20 1-81
ZnS04 . 7 H20 0.222
CuS04 . 5 H20 0.079
Mo03 (97.5%) 0.015
97
-------
APPENDIX B
Hutner's Medium for Duckweed
Solution 1: 17.7 g of Ca(N03)2 . 4H20 and 10.0 g of NH4N03/liter
Solution 2: 20 g of
Solution 3A: 3.295 g ZnS04 . 7H20
0.710 g H3B03
1.260 g Na2 Mo04 . 2H20 V dissolved in
0.190 g CuS04 . 5H20 f 200 ml of water
Add HC1 to disperse cloudiness
0.010 g Co(N03)2
0.897 MnCl2 . 4H20
Solution 3B: 25. Og of EDTA, add sufficient KOH to dissolve and
~400 ml water
Solution 3C: 1.25 g FeS04 . 7H20 dissolved in ~ 200 ml water
Solution C was prepared by adding 3B to 3C and
then adding 3A. The solution was diluted to one
liter and the pH adjusted to 6.
Solution 4: 25.0 g of MgS04 . 7HO/1
Hutner's medium (1/2 strength) consists of 5 ml each of solutions
1-4 and 580 ml of distilled HLO. Ten g of glucose is added, the
pH adjusted to 6.1-6.4 and the solution is au toe laved.
98
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
IT. REPORT NO.
EPA-600/2-80-039
3. RECIPIENT'S ACCESSION NO.
[i; TITLE AND SUBTITLEToxiclty Reduction Through Chemical
And Biological Modification of Spent Pulp Bleaching
Liquors
5. REPORT DATE
January 1980 issuing date
6. PERFORMING ORGANIZATION CODE
|7. AUTHOR(S)
Carl ton Dence, Chun-Juan Wang, and
Patrick Durkin
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
State University of New York
College of Environmental Science And Forestry
Syracuse, N.Y. 13210
10. PROGRAM ELEMENT NO.
1BB610
11. CONTRACT/GRANT NO.
R-804779
|12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Lab.
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
- Cinti., OH
13. TYPE OF REPORT AND PERIOD COVERED
Final: 9/71/76-9/20/79
14. SPONSORING AGENCY CODE
EPA/600/12
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Chlorophenols similar to or identical with those detected in spent
chlorination and caustic extraction liquors were synthesized and tested over a range
of concentrations to determine their effect on the growth of several fungi, an alga
(Chlorella pyrenoidosa) and duckweed (Lemna perpusilla) and on the survival of
Daphniamagna.
Biological treatment of the chlorophenols consisted of the application of pure
cultures of three different fungi and a mixed microbial population for periods rang-
ing up to 15 days. Degradation varied widely among the various phenols and for the
same phenol treated with different fungi.
Spent chlorination and caustic extraction liquors were subjected to a variety
of chemical treatments and the resulting effects on acute toxicity determined.
Treatment with elemental chlorine, hypochlorous acid, hypochlorite, ozone and hydro-
gen peroxide produced increases in the toxicity of the spent liquor.
Biological treatment of spent chlorination and caustic extraction liquors in-
volved the application of a fungus (Candida utilis), an unidentified bacterium, and
a mixed microbial population, together with supplemental carbon sources.
Toxicity reduction through modification of conventional chlorination and
caustic extraction bleaching stages was also evaluated.
ll>
DESCRIPTORS
L .
Pulping, Industrial Wastes, Phenols, Bio-
assay, chemical treatment, toxicity, bio-
logical treatment
^"""^""^
KEY WORDS AND DOCUMENT ANALYSIS
b IDENTIFIERS/OPEN ENDED TERMS
___^_ ^
Spent Chlorination
liquors, chlorophenols,
Total organic carbon
|18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
MMMMMMHMMMMM^MM^OTMM
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
20 SECURITY CLASS (This page)
UNCLASSIFIED
COSATI Field/Group
13/B
21. NO. OF PAGES
TM
PRICE
99
U.S GOV£«'
CE '980-657-146/5579
------- |