United States
Environmental Protection
Agency
Industrial Environmental
Research Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-84-055 Apr. 1984
f/EPA Project Summary
Technology Evaluation for Priority
Pollutant Removal from Dyestuff
Manufacture Wastewaters
Thomas M. Keinath
Laboratory-scale studies were con-
ducted to establish the technical fea-
sibility of employing certain processes,
including ozonation, GAC adsorption
and biophysical (PACT) treatment, for
the treatment of dyestuff manufacture
wastewaters with special focus on the
removal of organic priority pollutants.
The GAC adsorption studies showed
that GAC provided for excellent re-
movals of both color and priority pol-
lutants but cannot be expected to
concomitantly provide for high levels of
SOC removal due to comparatively high
levels of non-adsorption organics that
occasionally are present in dyestuff
manufacture wastewaters. Excellent re-
movals of organic priority pollutants
were achieved by the PACT process. In
addition, SOC and color removals were
enhanced by the addition of PAC to an
activated sludge system, generally in
direct relation to the steady-state con-
centration of PAC in the reactor. Al-
though ozonation provided for the
removal of many organic priority pol-
lutants to levels below detectability,
some proved to be comparatively resis-
tant to oxidation by ozone. As for GAC
and the PACT process, ozonation pro-
vided excellent color removals. Only
moderate organic carbon removals were
achieved by ozonolysis, however.
A priority pollutants survey showed
the presence of a total of 23 organic
priority pollutants in the raw waste-
waters. With the exception of meth-
ylene chloride, removal levels of volatile
organic priority pollutants generally ex-
ceeded 95 percent.
This Project Summary was developed
by EPA's Industrial Environmental Re-
search Laboratory, Cincinnati, OH, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back}.
Introduction
Many different types of manufacturers of
synthetic organic chemicals recently have
focused their attention on the treatment and
control of toxic chemicals that are contained
in their wastewater discharges. One such in-
dustry is the dyestuff manufacturing in-
dustry. Previous studies conducted on
wastewaters obtained from this industry
established that ozonation, adsorption on
granular activated carbon, and biophysical
(powdered activated carbon/biological,
PACT) treatment were well suited for remov-
ing conventional pollutants from dyestuff
manufacture wastewaters and had signifi-
cant potential for removing priority pollutants
from these wastewaters as well. Accord-
ingly, the principal objectives of the present
research program were (1) to exhaustively
identify and quantify those priority pollutants
contained in dyestuff manufacture waste-
waters and to establish the performance
characteristics of existing BPT (best practical
treatment) with respect to the removal of
priority pollutants, and (2) to affirmatively
establish the effect of the three selected can-
didate BAT (best available treatment) treat-
ment processes on the removal, conversion,
and/or production of priority pollutants.
To ensure that the results of this research
program were representative of the entire
dyestuff manufacturing industry, studies had
to be conducted on a representative number
of dyestuff manufacture wastewaters. To
select those that were to be included in the
-------
study, the Dyes, Environmental and Tox-
icology Organization, Inc. was asked to
solicit volunteer member companies who
would participate. This solicitation resulted
in the identification of ten dyestuff manufac-
turers that indicated a willingness to par-
ticipate. Because this number was greater
than could be accommodated within the
budgetary constraints of the cooperative
agreement, six dyestuff manufacturers,
coded A through F, were selected to par-
ticipate in the survey. This selection was
made on the basis of the following criteria:
(1) raw wastewater conventional pollutant
characteristics, (2) treated wastewater con-
ventional pollutant characteristics, (3)
wastewater treatment plant design char-
acteristics, (4) wastewater treatment plant
operational characteristics, (5) class(es) of
dyestuff produced, and (6) types of non-
dyestuff products manufactured. The six
selected were judged to be the most rep-
resentative of the entire dyestuff manufac-
turing industry.
Dyestuff Manufacturing Industry
Survey
The primary objective of this study was to
establish the technical feasibility of employ-
ing granular activated carbon adsorption,
ozonation, and the PACT process for the
removal of priority pollutants from dyestuff
manufacture wastewater. Also a study de-
signed to establish the performance char-
acteristics of existing wastewater treatment
plants at participating dyestuff manufac-
turers with respect to the removal, conver-
sion and/or production of priority pollutants
was conducted.
Survey Results
To provide the proper perspective for in-
terpretation of these treatability results, it is
important to point out that each of the
dyestuff manufacturers that were included
in the survey had extensive wastewater treat-
ment facilities. Each had provisions for
neutralization followed by a conventional ac-
tivated sludge system that was coupled with
an associated sludge handling train. The ac-
tivated sludge systems were operated at
nominal solids residence times (SRTs) that
ranged from 7 to 25 days and hydraulic
residence times (HRTs) that ranged from 18
to 36 hours. This configuration and asso-
ciated operational conditions are relatively
typical of the dyestuff manufacturing in-
dustry.
Several of the manufacturers also had ad-
junct processes such as granular activated
carbon adsorption for the liquid processing
train. These adjunct processes, however, are
not significant to this analytical survey for
priority pollutants because all samples that
herein are reported as effluent samples were
collected at the discharge of the solids-liquid
separator following the aeration basin and
not necessarily at the discharge from the en-
tire wastewater treatment plant. This was
done to ensure comparability of the data col-
lected. All influent samples were collected
either immediately before or after neutraliza-
tion, depending on which was most feasi-
ble at a particular site.
Organic Priority Pollutants
The results of all analyses conducted for
organic priority pollutants are recorded in
Table 1. Of the volatile organic priority
pollutants found in the four influent waste-
water samples, only chlorobenzene and
toluene were found in all four samples.
Ethylbenzene was noted to be present in
three of the four samples. All other volatile
organics found were present only as single
occurrences. Only toluene and ethylbenzene
were detected at concentrations greater than
the milligram per liter level.
As expected, removal levels of the volatile
organics were very high in the activated
sludge systems due to the stripping that oc-
curs as a result of aeration. With the excep-
tion of methylene chloride, removal levels
generally exceeded 95 percent. Many of the
volatile organic materials were removed to
levels below detectability.
Of the acid and base/neutral extractable
compounds that were found in the influent
samples, only phenol, 2,4-Dinitrophenol and
Bis (2-chloroethyl) ether were found to be
present at concentration levels that exceed-
ed 1 mg/l. Although most acid and
base/neutral extractable compounds were
removed significantly from the aqueous
stream by the activated sludge system, many
compounds were found to be present at
significant levels even after treatment. This
was particularly true in the case of
2,4-Dinitrophenol which was present at a
concentration of approximately 4 mg/l after
treatment in Wastewater Sample B. Only
several compounds were present at less-
than-detectable concentration levels after
treatment.
Heavy Metal Priority Pollutants
All results for heavy metal determinations
conducted on Wastewater Samples A, C,
and D are given in Table 2; no analyses were
conducted on Sample B. Analysis of these
data clearly shows that only minor removals
of heavy metal priority pollutants were
achieved by the respective activated sludge
treatment systems. Only in the case of cad-
mium, copper, antimony and nickel were
significant levels of removal achieved, but
then only in the case of Wastewater Sam-
pie C. On the basis of these results, one may
conclude that the removals achieved were
not significant.
GAC Adsorption Treatment
Technology
Adsorption Equilibria
Experimental Methodology
Adsorption isotherms were developed
using the conventional static bottle tech-
nique. This involved placing various selected
weights of Calgon Filtrasorb 400 (20 x 30
mesh) into 120 ml French Square bottles and
adding 100 ml of the appropriate waste-
water. After filling, the isotherm bottles were
placed on a gyratory shaker which was
operated at 200 cpm. All bottles were re-
moved from the shaker after a ten-day equi-
libration period since it had been
demonstrated that adsorption equilibrium
was attained in approximately seven days.
The contents of each bottle was then filtered
through a 0.45-micrometer millipore filter.
Aliquots for soluble organic carbon (SOC)
analysis were acidified, using hydrochloric
acid, to a pH less than 2 and stored at 4°C
in 30 ml screw cap vials until analyzed.
SOC Adsorption Results
Two series of isotherm studies were per-
formed. One was conducted directly on the
wastewater itself, and the second was con-
ducted on wastewater samples that had
been pretreated by ozonation in an effort to
enhance the adsorption characteristics of the
organic materials in the wastewater. These
latter samples were pretreated at ozone
dosage levels that corresponded to those
prescribed below for the ozonation studies
that were designed for priority pollutant
removal.
Analysis of the isotherm data is facilitated
by comparison of several adsorption indexes.
These include (1) the non-adsorbable frac-
tion, obtained by the intersection of the
isotherm trace and the abscissa; (2) the GAC
solid-phase concentration or "loading" at the
operational-feed concentration; and (3) the
associated calculated GAC exhaustion rate.
Values for these parameters are summarized
in Table 3.
Non-adsorbable organics, usually low
molecular weight materials, were present in
each of the six raw wastewater samples
studied. Significant differences were ob-
served in the level of non-adsorbable
organics between the various samples.
Samples A through E had non-adsorbable
concentrations that ranged from 11 to 75 g
SOC/m3, as noted in Table 3. These were
in stark contrast to Sample F which had an
-------
Table 1. Organic Priority Pollutant Concentn
Activated Sludge Treatment
Organic Priority
Sample Pollutant Observed
A Acid Extraction
2-Nitrophenol
2, 4-Dinitrophenol
Phenol
Base/ Neutral Extraction
1, 3-Dichlorobenzene
1, 4-Dichlorobenzene
Naphthalene
Acenaphthene
N-Nitrosodiphenylamine
Volatiles
Toluene
Chlorobenzene
Ethylbenzene
B Acid Extraction
2, 4-Dichlorophenol
2,4-Dinitrophenol
Base/ Neutral Extraction
1, 4-Dichlorobenzene
Nitrobenzene
Di-n-butylphthalate
Volatiles
Methylene Chloride
Chloroform
Toluene
Chlorobenzene
C Acid Extraction
2-Nitrophenol
Phenol
Base/ Neutral Extraction
Bis 12-chloroethyl) ether
N-Nitrosodiphenylamine
l/O/flfiVpo
V 1/ICHHG&
Cis- 1,3-Dichloropropene
Toluene
Chlorobenzene
Ethylbenzene
D Acid Extraction
2-Nitrophenol
Phenol
2,4-Dichlorophenol
Base/ Neutral Extraction
1, 3-Dichlorobenzene
1, 4-Dichlorobenzene
Nitrobenzene
1,2,4- Trichlorobenzene
Naphthalene
Phenanthrene
Volatiles
Carbon Tetrachloride
1, 1, 1,-Trichloroethane
Trans- 1,3-Dichloropropene
Toluene
Chlorobenzene
Ethylbenzene
*ND means not detected (no chromatograml.
ttions Before and After Onsite
Influent (ng/l) Effluent fag/l)
Dup 111 Dup 92 Dup tt1
460
2700
3200
420
800
4.0
21
150
4000
180
270
18
5300
3200
350
23
5.5
31
800
550
7.7
4900
4700
310
4.7
160
120
4000
26
150
"<20
18
68
660
230
120
4.5
55
220
130
7800
36
67
'- means not measured.
"< means a chromatogram was observed, but response
t_
_
_
-
-
-
3600
190
290
-
-
-
-
-
5.5
33
780
550
-
-
-
-
5.6
160
100
4000
_
_
-
-
-
-
-
-
-
62
200
100
7400
39
55
was insufficient to
33
420
130
3.4
27
3.9
4.6
4.0
14
*ND
ND
*ND
3900
11
91
ND
0.48
ND
7.2
12
6.6
6.7
"ND
ND
ND
27
ND
ND
*ND
<20
<20
1.2
4.9
2.6
36
<1.0
ND
ND
ND
2.9
300
ND
ND
quantify.
Dup #2
55
440
190
3.7
27
4.1
4.9
3.6
11
ND
ND
ND
4700
13
100
ND
0.38
ND
7.9
13
11.0
6.5
ND
ND
ND
19
ND
ND
ND
<20
<20
1.2
4.9
4.3
37
<1.0
ND
ND
ND
2.5
270
*/V
ND
ND
extremely high non-adsorbable concentra-
tion of 580 g SOC/m3. Pretreatment of these
wastewater samples served to alter the non-
adsorbable fraction of organic materials only
slightly. Generally, ozonation led to an in-
crease in the non-adsorbable fraction in all
samples with the exception of Sample E.
Calculated GAC exhaustion rates must be
indexed to the operational aqueous-phase
feed concentration of organics that one
would expect to be present in the feed
stream to the adsorber. For the present
studies, the feed stream concentrations were
assumed to be equal to the concentration of
organics that existed in the samples either
before or after ozonation. These are tabu-
lated in the third column of Table 3, while
the associated calculated GAC exhaustion
rates are tabulated in the last column.
Generally, it is apparent that the GAC ex-
haustion rates vary rather dramatically from
one dyestuff manufacturing wastewater
sample to another. In all but two cases.
ozonation pretreatment led to an increased
GAC exhaustion rate. In the case of Sam-
ple B, the slightly lower GAC exhaustion rate
was due to the greater non-adsorbable frac-
tion that was present after ozonation. Con-
versely, the lower GAC exhaustion rate for
Sample E was due primarily to the fact that
ozonation served to significantly increase the
adsorption capacity. This was the only sam-
ple for which this observation was made. For
all other samples the adsorption capacity
either decreased or did not change ap-
preciably.
Continuous-Flow Adsorption
Studies
Experimental Methodology
Since this phase of the research focused
on the ability of GAC to remove priority
pollutants from dyestuff manufacture
wastewaters, a continuous-upflow GAC ad-
sorption system was constructed using five
columns-in-series. The columns and their
connecting lines were constructed of glass
to minimize the adsorption of organics onto
column surfaces. Each column was 1 .2 m in
length and 2.54 cm in internal diameter. The
inlet section of each column was filled with
3 mm glass beads to a depth of 20 cm to pro-
vide for even distribution of flow across the
surface of the GAC bed. Each column was
charged with 203 gm of the 20 x 30 mesh
Calgon Filtrasorb 400 GAC. Prior to each
run, the GAC was immersed in distilled water
for 24 hours to ensure complete wetting of
the GAC prior to operation.
These experiments were conducted by
passing each of the dyestuff manufacture
wastewaters through the columnar GAC ad-
sorbers that were charged with virgin GAC.
3
-------
Table 2. Heavy Metal Priority Pollutant Concentrations Before and After Onsite
Activated Sludge Treatment
Heavy Metal
Priority Pollutant
Sample Observed
Influent (\tgll)
Effluent
A Cadmium
Chromium
Copper
Antimony
Lead
Nickel
Zinc
Thallium
Silver
Arsenic
C Cadmium
Chromium
Copper
Antimony
Lead
Nickel
Zinc
Thallium
Silver
Mercury
Arsenic
Selenium
D Cadmium
Chromium
Copper
Antimony
Lead
Nickel
Zinc
Thallium
Silver
22
142
23
33
111
140
56
45
30
114
124
12
3250
176
*ND
160
14
ND
ND
3
ND
20
8.7
180
5650
200
130
130
70
*ND
ND
17
134
24
33
66
102
56
1<5
15
83
28
12
82
59
ND
80
288
ND
ND
4
ND
12
12.5
<500
4750
200
130
130
70
125
ND
ND means Not Detected.
f< means response was observed, but was insufficient to quantify.
Table 3. GAC Adsorber Design Factors for SOC Removal
Wastewater
Sample
No Pretreatment
A
B
C
D
E
F
Ozone Pretreatment
A
B
C
D
E
F
Non-Ads.
Fraction
(g SOC/rrf)
11
40
75
70
45
590
14
50
75
91
22
600
Oper. Feed
Cone.
(g SOC/nf)
160
181
320
290
445
960
148
189
300
300
390
995
Ads. Cap. at
Feed Cone.
Ig SOC/kg GAC)
165
165
105
175
128
140
100
165
87
160
190
135
GAC Rate of
Exhaustion
(kg GAC/1000 nfl
903
850
2330
1260
3120
2640
1340
840
2590
1310
1940
2930
The wastewater was placed in a 50-liter car-
boy that was mixed with air to maintain
homogeneity, was allowed to equilibrate at
room temperature, was sampled for priority
pollutant analysis, and was passed through
the GAC columns using a peristaltic pump.
The flow rate through the columns was set
at 0.082 m/min (2 gpm/sq ft) which provided
an empty bed residence time of 45 minutes
for all five columns. The wastewater was
pumped through the system to waste for a
period of four hydraulic retention times to
ensure displacement of the distilled water.
Thereafter, the system effluent was collected
in a thirteen-liter glass carboy for a period
of four hours. At the end of each run con-
ducted on a dyestuff manufacture waste-
water, the total effluent was mixed and
sampled for subsequent priority pollutant
analysis.
Priority Pollutant Adsorption Results
The results of these GAC adsorption
studies are given in Table 4. No analyses for
volatile organic priority pollutants were con-
ducted in this study. The analytical program
was limited to the acid and base/neutral ex-
tractable priority pollutant fractions.
Results for all samples showed that in all
cases GAC adsorption effectively removed
all of the priority pollutants initially present
in the wastewater to below detectable levels.
Since these results were achieved for five
dyestuff manufacture wastewaters that have
widely divergent compositions and matrixes
and had widely varying concentrations of
priority pollutants, it appears safe to con-
clude that GAC adsorption appears to be a
reasonably universal treatment technology
for the removal of organic priority pollutants,
neglecting economic considerations. It is im-
portant to note, however, that adsorber
design criteria were not elaborated by these
studies. Such criteria can only be developed
by conducting extensive pilot-scale studies.
Biophysical (PACT) Treatment
Technology
A typical flowsheet of the biophysical or
PACT treatment system involves the addi-
tion of powdered activated carbon (PAC) to
the aeration basin of a conventional activated
sludge treatment process. The addition of
PAC to the aeration basin provides for in-
creased removals of BOD, COD, TOC, color,
and toxic compounds; provides for added
stability of the biological process; and im-
proves the settling characteristics of the
sludge. PAC addition also allows for adsorp-
tion of organic compounds, provides sur-
faces for biological attachment, and creates
the possibility for the elaboration of beneficial
biomass-PAC interactions.
Experimental Methodology
The experimental program was limited to
evaluating this process on three dyestuff
manufacture wastewaters. Samples A, B,
and F. These were carefully selected to en-
sure that the wastewaters generated from
the production of three distinctly different
product line mixes were evaluated to obtain
the broadest-based technology evaluation in-
formation possible.
-------
Table 4. Organic Priority Pollutant Concentrations Before and After GAC Treatment
Influent fug/li Effluent
Organic Priority
Sample Pollutant Observed
DupHI Duptt2 Dupttl Dup#2
A
B
C
D
F
Acid Extraction
Phenol
2, 4-Dinitrophenol
Base/ Neutral Extraction
2-Nitrophenol
1, 4-Dichlorobenzene
Naphthalene
Acid Extraction
2, 4-Dichlorophenol
2,4-Dinitrophenol
Base/ Neutral Extraction
1, 4-Dichlorobenzene
Nitrobenzene
Di-n-butylphthalate
Acid Extraction
Phenol
Base/ Neutral Extraction
Bis (2-chloroethyl) ether
N-Nitrocodiphenylamine
Acid Extraction
Phenol
Base/ Neutral Extraction
1,3-Dichlorobenzene
1, 4-Dichlorobenzene
Nitrobenzene
1,2,3- Trichlorobenzene
Naphthalene
Acid Extraction
Phenol
Base/ Neutral Extraction
Di-n-octylphthalate
110
350
40
3.4
1.6
1.3
3200
5.0
51
22
1900
2300
29
240
1.4
9.4
830
8.5
26
57
1800
120
340
35
3.1
1.4
1.3
3300
6.2
47
24
-
-
-
-
-
-
-
-
-
59
1800
*ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND means Not Detected.
Approximately 1.51 m3 of raw wastewater
was collected from the equalization basin of
each of the three wastewater treatment
systems sampled. The waste was stored at
4°C to minimize degradation and was
neutralized to pH 7.0. The requisite nutrients,
essential for biological growth, were added
prior to use as reactor feed.
Nuchar S-A 15," manufactured by
Westvaco Corporation, was the powdered
activated carbon (PAC) used in this study.
Due to the fact that biological plate counts
showed the presence of significant concen-
trations of microorganisms in the virgin PAC,
the PAC was sterilized in an autoclave prior
to use and then stored at 103°C. Daily PAC
make-up requirements were supplied to the
PACT reactors using an aliquot of a 10 kg/m3
aqueous PAC slurry. Changes in reactor PAC
concentration, however, were made by add-
ing a specific weight of dry PAC.
'Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
The laboratory-scale, continuous-flow
reactor system consisted of a glass per-
colator which was fitted with a diffuser
through which compressed air, which has
been filtered and humidified, entered the bot-
tom of the reactor vessel. This provided for
completely mixed aerobic conditions in the
reactor. Exhaust gases were drawn off
through a vacuum connection. Feed solution
was supplied to the reactors by a peristaltic
pump. The wastewater passed from a com-
pletely-mixed, 20-I plastic feed tank to the
reactor vessel through tygon tubing. Cam
cycle timers were used to aid in control of
the flowrate such that low flow rates (2.08
ml/min) could be achieved.
Treated wastewater was removed directly
from the reactor by a vacuum line connected
to a cylindrical vyon cup. The vyon cups
allowed the soluble portion of the treated
wastewater to pass into the effluent collec-
tion bottles while eliminating most biological
solids. This provided for positive control of
the biological and PAC solids contained
within the reactor.
To determine the effectiveness of the
PACT process for treating the dyestuff
manufacture wastewaters, two reactors
were run in parallel for each of the three
wastewaters studied. One system was op-
erated as a conventional activated sludge
process and the other, containing PAC, was
operated as a PACT system. Both systems
were operated at a nominal solids residence
time of 20 days and a nominal hydraulic
residence time of 24 hours.
Reactor start-up consisted of charging
both the biological and biophysical reactors
with 3 liters of mixed liquor obtained from
the aeration basin of one of the dyestuff
manufacturer's activated sludge wastewater
treatment facilities. An initial aliquot of
nutrients was also added to both of the reac-
tors to promote biological growth and to
shorten the acclimation period, while an ap-
propriate amount of prewashed, dry, steril-
ized PAC was added to the biophysical
reactor. An initial acclimation period of 40
days was necessary to develop an adequate
population of acclimated microorganisms.
During this period, the flow rate to the reac-
tors was set to maintain a one-day hydraulic
residence time (HRT). Biological and bio-
physical solids were wasted daily by remov-
ing 150 ml of the mixed liquor from each
reactor to maintain a solids residence time
(SRT) of 20 days. To account for the loss
of PAC from the biophysical system due to
wasting, an appropriate quantity of virgin
PAC was subsequently added to maintain a
steady-state PAC concentration in the
reactor.
During the acclimation period daily
measurements of pH, suspended solids, and
soluble organic carbon (SOC) were made to
follow the operating conditions of the sys-
tem. Adjustments in pH were made to main-
tain the pH of the system in the range of 6.5
to 7.5. The SOC levels in the influent to and
the effluent from each reactor were moni-
tored daily to establish the onset of steady-
state conditions. Once steady state had been
achieved (determined on the basis of SOC
analyses), effluent samples were collected
from each reactor and analyzed for color,
organic priority pollutants, and heavy metals.
SOC Removal Results
Based on PAC isotherm studies, four PAC
reactor concentration levels, 1, 3, 5, and 7
kg/m3, were selected for experimental evalu-
ation. A parallel or "control" biological reac-
tor that received no PAC feed was operated
under identical conditions throughout the
duration of the study.
Summarized data for these studies are
shown in Table 5. Qualitative analysis of the
data shows that PAC addition to an activated
-------
Table S. SOC Removals Obtained in Parallel PACT and Biological Reactor Systems
Steady-State
PAC Cone.
Ig/nfl
Wastewater Sample A
7000
3000
5900
7000
Wastewater Sample B
1000
3000
5000
7000
Wastewater Sample F
1000
3000
5000
7000
Influent
SOC Cone.
Ig/nf)
250
217
236
240
164
142
171
157
664
814
817
857
Biol. Eff.
SOC Cone.
(g/m3)
103
99
96
101
106
107
105
105
380
489
352
481
PACTEff.
SOC Cone.
(g/nf)
102
65
51
49
97
68
58
52
357
433
373
389
Biol. Rem.
Efficiency
1%)
58.8
54.5
59.3
57.9
35.4
24.6
38.6
33.1
42.8
39.9
56.9
43.9
PACT Rem.
Efficiency
<%)
59.2
70.0
78.4
79.6
41.0
52.0
66.0
66.7
46.2
46.7
54.3
54.6
Sig. Dif.
Between Systems
@ 95% Conf. Level
no
yes
yes
yes
yes
yes
yes
yes
no
yes
no
yes
sludge system appears to enhance system
performance and that system performance
improved with increasing steady-state con-
centrations of PAC in the reactor up to 5
kg/m3. Performance did not increase sub-
stantially for PAC concentrations greater
than 5 kg/m3.
To provide a more rigorous interpretation
of the data, the significance of the dif-
ferences between the observed removal ef-
ficiencies for the PACT and activated sludge
systems was determined using the "t"
statistical test. Reference to the last column
of Table 5 shows that, at steady state, the
PACT system was statistically more efficient
in removing SOC in nine of the 12 experi-
ments conducted. Exceptions were noted for
wastewater sample A (PAC level of 1 kg/m3)
and wastewater sample F (PAC levels of 1
and 5 kg/m3). In the case of sample F, this
occurrence is likely due to the fact that a
significant quantity of the soluble organic
carbon present in the wastewater (ca. 40 per-
cent) is non-adsorbable. Since one of the
predominant functional mechanisms of the
presence of PAC in a PACT system is sim-
ple adsorption, it is obvious that addition of
PAC to an activated sludge system would
have lesser beneficial effects for wastewaters
having greater proportions of non-ad-
sorbable materials.
A statistical analysis was also conducted
to establish whether or not significant dif-
ferences existed in system performance be-
tween the various steady-state PAC
concentration levels. This analysis shows
that statistically significant differences were
observed between PAC concentrations of
1/3 and 3/5 but not for 5/7 kg/m3 for
samples A and B. Because sample F had a
comparatively high level of non-adsorbable
organics, a significant difference was ob-
served only for the 3/5 kg/m3 PAC concen-
tration pair. On the basis of these results, one
may conclude that system performance was
not enhanced for PAC concentration levels
greater than 5 kg/m3 for the wastewater
samples studied.
Color Removal Results
Color analyses were performed on effluent
samples from both the PACT and biological
systems as well as the raw feed at each
steady-state PAC concentration level. The
results of these analyses are given in Table
6. Statistically significant biological removal
of color was achieved only for sample A. The
removal levels achieved for samples B and
F were not found to be statistically different
than zero. Addition of PAC to the biological
reactors proved to significantly enhance col-
or removal levels only for sample A. Only
minor increases in color removal were
observed for samples B and F as a result of
PAC addition.
It is interesting to note, furthermore,
that color removal levels generally de-
cresaed with increasing steady-state
PAC concentration levels for samples A
and F. Since physical adsorption appears
to be the principal mechanism of color
removal, this result cannot be rationally
explained. The relative removals achieved
in the case of sample B are consistent
with the anticipated performance.
Metallic Priority Pollutants
Removal Results
Because one major focus of this investiga-
tion was removal of priority pollutants and
because the PACT process has been
reported to remove heavy metals, heavy
metal analyses were performed on filtered
raw feed and effluent samples obtained from
the PACT system that was operated with a
Table 6. Color Removals Obtained in Parallel PACT and Biological Reactor Systems
Steady-State Influent
PAC Cone. Color Cone.
fg/nf) (ADMIU)
Wastewater Sample
1000
3000
5000
7000
Wastewater Sample
1000
3000
5000
7000
Wastewater Sample
1000
3000
5000
7000
A
3800
4350
4120
4040
B
16700
17000
16900
16600
F
22200
22500
22300
21700
Biol. Eff.
Color Cone.
(ADMIU)
3830
3810
3790
3680
16700
16630
16900
16500
21800
22600
20300
22100
PACTEff.
Color Cone.
IADMIU)
690
2530
1880
1780
16400
15600
15200
14100
18900
20100
20300
20200
Biol. Rem.
Efficiency
(%)
0.0
12.5
7.9
9.0
0.1
2.4
0.0
0.8
2.0
0.0
8.8
0.0
PACT Rem.
Efficiency
(%)
81.8
41.8
54.2
56.3
2.1
6.2
10.2
14.3
15.2
10.4
9.1
6.8
-------
steady-state PAC concentration of 5 kg/m3
and its corresponding parallel biological
system. The data from these analyses are
shown in Table 7.
Analysis of the data shows that it is not
particularly conducive to reaching significant
conclusions. This is likely due to the fact that
the three wastewater matrixes were dra-
matically different both in the solution- and
particulate-phase fractions. Nonetheless,
several general observations are appropriate.
In the case of sample A, two heavy metals.
nickel and antimony, were removed by the
PACT at levels that were statistically sig-
nificantly greater than the removals achieved
in the corresponding biological control reac-
tor. For sample B, the PACT system per-
formed better for chromium, copper, nickel,
antimony and zinc, while the PACT system
performed better only in the case of chro-
mium for sample F. The situation for sam-
ple F is somewhat confused by the fact that
the biological control reactor provided bet-
ter removals of antimony and zinc than did
the parallel PACT reactor. On the basis of
these results one may conclude that PAC ad-
dition to a biological reactor may result in
enhanced removals for certain heavy metals.
but that the situation is very specific and
related to the composition of the aqueous
matrix.
Organic Priority Pollutants
Removal Results
Table 7. Metallic Priority Pollutant Removals
Biological Reactor Systems
Influent Biol. Eff.
Cone. Cone.
Metal (g/rrf) (g/m3)
Wastewater
Ag
Cd
Cr
Cu
Ni
Pb
Sb
Tl
Zn
Sample A
0.08
0.005
0.05
0.10
0.97
0.58
2.65
0.36
0.38
Wastewater Sample B
Ag 0.04
Cd nnnr;
Cr
Cu
Ni
Pb
Sb
Tl
Zn
Wastewater
Ag
Cd
Cr
Cu
Ni
Pb
Sb
Tl
Zn
1.50
5.71
1.09
0.15
1.21
0.12
2.60
Sample F
0.12
0.45
.3.18
55.2
1.65
0.37
2.04
0.24
0.80
0.06
0.005
0.05
0.48
0.66
0.15
2.04
0.36
0.66
0.04
0.005
0.74
4.38
0.72
0.15
1.21
0.12
1.71
0.12
0.21
2.84
55.4
1.78
0.30
1.94
0.24
0.54
Obtained in Parallel PACT and
PACT Eff. Biol. Rem.
Cone. Efficiency
(g/rrf) {%)
0.06
0.005
0.05
0.35
0.59
0.15
1.21
0.47
0.59
0.04
0.005
0.05
1.02
0.22
0.15
1.01
0.12
0.92
0.12
0.21
2.68
53.7
2.03
0.30
2.04
0.24
0.79
25.0
0.0
0.0
*
32.0
74.1
23.0
0.0
*
0.0
0.0
50.7
23.3
33.9
0.0
0.0
0.0
34.2
0.0
53.3
10.7
#
*
18.9
4.9
0.0
32.5
PACT Rem.
Efficiency
(%)
25.0
0.0
0.0
*
39.2
74.1
54.3
*
*
0.0
0.0
96.7
82.1
79.8
0.0
16.5
0.0
64.6
0.0
53.3
15.7
2.7
*
18.9
0.0
0.0
1.2
using PACT systems for providing for the
removal of organic priority pollutants from
dyestuff manufacture wastewaters, analyses
for organic priority pollutants were perform-
ed on composite samples obtained from the
feed and the effluent of the PACT system
that was operated with a steady-state con-
centration of PAC of 5 kg/m3. The data from
these analyses are shown in Table 8.
The results for all samples showed that the
PACT system effectively removed all of the
organic priority pollutants initially present in
the wastewaters to below detectable levels,
with a single exception. This was in the case
of phenol that was present in sample B at
a level of 5300 UJQ/\. An average residual level
of 8.1 ^g/l was observed to be present after
PACT treatment. This represents a treatment
efficiency in excess of 99.8 percent. Had a
similar analyses been performed on a com-
posite sample obtained from the PACT reac-
tor that was operated with a steady-state
PAC concentration of 7 kg/m3 is it probable
that the phenol in this sample would also
have been decreased to a concentration level
that was less than detectable.
Since these results were obtained for three
dyestuff manufacturing wastewaters that
'Negative removal.
Table 8. Organic Priority Pollutant Concentrations Before and After PACT Treatment
urgantc rnoriiy
Pollutant Observed
Wastewater Sample A
2-Nitrophenol
Phenol
2, 4-Dichlorophenol
1, 4-Dichlorophenol
1. 3-Dichlorobenzene
Acenapthene
N-Nitrosodiphenylamine
Wastewater Sample B
2,4-Dinitrophenol
Phenol
1,4-Dichlorobenzene
Nitrobenzene
Fluoranthene
Wastewater Sample F
Bis 12-ethylhexyl)
phthalate
Di-n-octylphthalate
Influent l^g/l)
570
4400
2500
65
12
40
32
2500
5300
77
120
13
560
1200
Dup#1
*/VD
ND
ND
ND
ND
ND
ND
ND
8.5
ND
ND
ND
ND
ND
Dup»2
ND
ND
ND
ND
ND
ND
ND
ND
7.7
ND
ND
ND
ND
ND
*ND means Not Detected.
-------
have widely divergent compositions and
matrixes and widely varying concentrations
of priority pollutants, it is reasonable to con-
clude that the PACT process is effective for
the removal of organic priority pollutants.
Ozonation Treatment
Technology
Experimental Methodology
The ozone used was generated by a Grace
Model LG-2-L2 corona ozone generator
using bottled dry oxygen as the gas source.
Oxygen was first dried and filtered through
a TekLab FD-0235 transparent filter/drier
then directed to the ozone generator via 0.95
cm polyethylene tubing. The ozonator was
operated at a pressure of 86.2 kPa, an oxy-
gen flowrate of 10 scfh, and a variable power
setting of 0 to 300 watts.
Ozone was directed to the reactor through
polyethylene tubing, where it entered the
bottom of the column. A porous stone dif-
fuser was used for initial ozone dispersion.
To aid in gas transfer, 6.3 mm ceramic
Raschig rings were placed in the column to
a depth of 1.52 m. The reactor consisted of
two 1.52-m sections of 5.08 cm diameter
pyrex glass tubing with an expansion globe
located at the top which served as an aid in
breaking foam generated during the ex-
periments.
The test wastewater samples were re-
moved from the bottom of the column
through a glass recirculation line and re-
turned to a spray nozzle located in the ex-
pansion globe. Glass tubing was used
wherever possible to minimize the potential
for desorption of phthalates during the
ozonation studies. Tygon tubing was used
only for making connections between the
various sections of glass tubing. The
wastewater was recirculated at a rate of four
liters per minute by a peristalic pump.
The off-gas from the reactor was directed
via 0.95 cm Eastman polyethylene tubing to
two 500-ml gas washing bottles arranged in
series for ozone analysis. Three sets of gas
washing bottles were arranged in parallel.
Two sets contained a solution of buffered
potassium iodide while the third contained
distilled water to detect gas flow. Any
unreacted ozone that passed through the gas
washing bottles was vented to a 20-I tank
containing a saturated solution of sodium
thiosulfate. The exhaust from the 20-I tank
was discharged to the atmosphere.
Preliminary Studies
Preliminary ozonation characterization
studies were directed toward evaluating the
degradation of color, organics, pH changes
and ozone consumption levels to define the
experimental conditions required for the con-
duct of the subsequent priority pollutant ox-
idation study. Experimental samples were
collected from the influent to the wastewater
treatment facilities of the six participating
dyestuff manufacturers and were ozonated
batchwise to an ozone application level of
three grams ozone per gram of COD initially
present. All preliminary studies were con-
ducted at a temperature of 20 °C and all
wastewaters were neutralized to a pH of 7.0
prior to ozonation with the exception of
wastewater sample A which had been col-
lected after neutralization. The pH of this
neutralized wastewater was 7.7.
For the present studies the maximum ab-
sorbance occurred near the low end of the
visible span for all but the B and F samples.
Upon ozonation, all spectra shifted into the
ultraviolet region. Complete elimination of
the visible spectra was not achieved, even
with an ozone application level of three
grams per gram of COD initially present.
Degradation of the color value for each of
the samples was rapid as a result of ozona-
tion. Essentially complete color removal was
observed for the B and F samples for ozone
application levels of 1.76 and 2.25 g 03/g ini-
tial COD. The remaining dyestuff manufac-
ture wastewater samples (A, C, D, & E)
proved somewhat more resistant to com-
plete color value removal. Nonetheless, color
value removals ranged from 88 to in excess
of 95 percent for these samples at an ozone
application level of 3 g 03/g initial COD. A
baseline color value of 0 to 300 ADMI units
was attained in all cases.
During these preliminary studies TOC
removals ranged from 15 to 42 percent at an
ozone application level of 3 03/g COD ini-
tially present at which point each experiment
was terminated. The associated initial ozone
consumption rate expressed as percentage
of the ozone application rate ranged between
90 and 100 percent for all samples. At the
termination of each study, consumption
rates were approximately 2 to 10 percent of
the ozone application rates with the excep-
tion of sample F for which the final ozone
consumption rate was equal to approxi-
mately 40 percent of the application rate.
The ratio of ozone utilized per unit of TOC
removed did not change appreciably
throughout these preliminary experiments
although a slight increase in the ratio was
observed for several of the studies as they
progressed. The highest 03/TOC ratios, 20
to 30 g 03/g TOC, were observed for the A,
C, and D samples which proved to be most
refractory in character. The B and F samples
were observed to have comparatively low
ozone utilization ratios, in the range of 7 to
10 g 03/g TOC oxidized, while the E sample
was observed to have an intermediate ozone ^
utilization ratio. m
Organic Priority Pollutant
Removal Studies
To establish the technical feasibility of
using ozonation for removing organic priority
pollutants from dyestuff manufacture
wastewaters, a treatability study was con-
ducted on each of the six wastewaters.
These were conducted batchwise at ozone
application levels determined on the basis of
the preliminary studies.
Initially, three criteria were defined for
determining the appropriate ozone applica-
tion levels. These were: (1) mass of ozone
utilized per unit mass of total organic carbon
oxidized, (2) TOC degradation, and (3) color
value degradation. Because the time depen-
dent traces of TOC degradation and the ratio
of 03 utilized/TOC oxidized showed no
significant discontinuities that could serve as
the basis for establishing ozone application
levels, the time trace of color value degrada-
tion was selected for establishing ozone ap-
plication levels for all organic priority
pollutant ozonation studies.
Degradation of the color value could be
described by a first-order kinetic relationship.
For this reaction order, the associated time
constant of the system is equal to the time
required to attain 63 percent color value fl
removal in a batch system. To attain re- ^
movals of approximately 95 percent in a
continuous-flow, plug-flow reactor operating
at steady state, the design hydraulic
residence time for the system would have to
equal a minimum of three time constants.
Accordingly, these ozonation studies were
conducted batchwise on 4 liter samples for
a period of time that equalled three
characteristic time constants for the system
at ozone application rates that were identical
to those used in the preliminary studies.
The results of these ozonation studies are
given in Table 9 for the dyestuff manufac-
ture wastewater samples coded A through
D and F, respectively. Results for the A and
B samples showed that all of the priority
pollutants initially present in these waste-
water samples were removed to levels below
detectability. It is important to note, further-
more, that no organic priority pollutants were
produced as a result of ozonation. For sam-
ple C, both bis (2-chloroethyl)ether and N-
Nitrosodiphenylamine were removed to non-
detectable levels. Nonetheless, phenol which
was initially present at a concentration level
of 1900 pg/l was reduced to an average con-
centration of 12 pig/I representing a removal
level in excess of 99 percent.
The situation for samples D and F was
somewhat different, however. In the case of ^B
8
-------
Table 9. Acid and Base/Neutral Fraction of Priority Pollutants Prior to and
After Ozonation
Concentration
Sample Compound
Initial
Final
Dup #/ Dup #2 Dup »1 Dup #2
A
B
C
D
F
2-Nitrophenol
Phenol
2,4-Dinitrophenol
1,4-Dichlorobenzene
Naphthalene
2, 4-Dichlorophenol
2, 4-Dinitrophenol
1, 4-Dichlorobenzene
Nitrobenzene
Di-n-butylphthalate
Phenol
Bis 12-Chloroethyl) ether
N-Nitrosodiphenylamine
1, 3-Dichlorobenzene
Phenol
1,4-Dichlorobenzene
Naphthalene
1,2,3- Trichlorobenzene
Nitrobenzene
Phenol
Di-n-octylphthalate
40
110
350
3.4
1.6
1.3
3200
50
51
22
1900
2300
29
1.0
240
9.4
16
8.5
830
59
1800
35
120
340
3.1
1.4
1.3
3300
6.2
47
24
#.
-
-
-
-
-
-
-
-
57
1800
*ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
13
ND
ND
ND
15
ND
ND
ND
360
26
360
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
11
ND
ND
ND
16
ND
ND
ND
400
32
350
ND means not detected.
*- Duplicate samples were not run.
sample D, four of the organic priority pol-
lutants initially present were removed to less-
than-detectable levels. Phenol was present
in the ozonated sample at an average con-
centration of 15.5 fig/I representing a
removal efficiency of approximately 93 per-
cent. In contrast, the concentration of nitro-
benzene which was initially 830 U*Q/\ was
decreased to only 380 fig/I (54 percent
removal). The two organic priority pollutants
present in sample F proved to be even more
resistant to degradation by ozonation. For
this sample, residual levels of phenol of 29
ug/\ and of Di-n-octylphthalate of 355 ^ig/l
were observed after ozonation. This rep-
resents treatment levels of 50 and 80 percent,
respectively.
The lower treatment level observed for the
F sample is likely attributable to the fact that
the concentration of organic materials ex-
pressed as TOC was approximately three
times the magnitude of the TOC in the other
four samples. The comparatively higher con-
centrations of competing compounds would
tend to inhibit the rate of degradation of
specific organic compounds by ozone,
thereby resulting in lower removal efficien-
cies at specific ozone application levels.
Conclusions
Results of an analytical survey conducted
on the wastewaters of four dyestuff manu-
facturers showed the presence of a total of
23 organic priority pollutants in the raw
wastewaters. Of these, six were present at
concentration levels greater than 1 mg/l.
Removal levels of these organic priority
pollutants by the activated sludge systems
that existed at the sites sampled generally
exceeded 95 percent in the case of the
volatile organics with the exception of
methylene chloride. Although most acid and
base/neutral extractable compounds were
removed significantly from the aqueous
stream by the activated sludge system, many
compounds were found to be present even
after treatment. Only several of these com-
pounds were present at less-than-detectable
concentration levels after treatment.
Analyses for heavy metal pollutants con-
ducted during this survey showed that only
minor removals were achieved by the respec-
tive activated sludge treatment systems.
With the exception of copper, most heavy
metals were found to be present in the raw
wastewater in the 20 to 200 ug/\ concentra-
tion range. In two of the three wastewaters
sampled, copper was found to be present in
the 3 to 6 mg/l concentration range.
Granular activated carbon (GAC) adsorp-
tion studies conducted on six dyestuff
manufacture wastewaters showed that while
excellent removals of both organic priority
pollutants and color can be obtained, ad-
sorption of soluble organic carbon (SOC) is
not particularly efficient due to the fact that
comparatively high concentrations of
non-adsorbable organics occasionally are
present in these wastewaters. Moreover, it
was shown that ozonation pretreatment of
these wastewaters generally served to
degrade adsorption characteristics.
Parallel PACT (biophysical) and biological
(activated sludge) studies conducted on
three dyestuff manufacture wastewaters
showed that powdered activated carbon
(PAC) addition provided for enhanced SOC
and color removals, generally in direct rela-
tion to the steady-state concentration of
PAC in the reactor. Excellent removals of
organic priority pollutants were achieved by
the PACT system. Heavy metal removal
levels were small.
Ozonation studies conducted on six dye-
stuff manufacture wastewaters showed that
ozone provides for excellent color removals
but only moderate organic carbon removals
as indexed by total organic carbon (TOC).
Although many organic priority pollutants
were removed to levels below detectability
some proved to be comparatively resistant
to oxidation by ozone. It was demonstrated,
furthermore, that no priority pollutants were
produced as a result of ozonation.
-------
Thomas M, Keinath is with Clemson University, Clemson, SC 29631.
M. J. Stutsman is the EPA Project Officer (see below).
The complete report, entitled "Technology Evaluation for Priority Pollutant
Removal from Dyestuff Manufacture Wastewaters, "(Order No. PB 84-157064;
Cost: $14.50, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
10
-------
JO
5 5
0) O
=• (D
|8
O)
C
0>
-. -r
0 C 7 r
c
•i.
C"
FT-
C";
£?§
» < =
Ssg
"la
y
aiM
o
3
-.
s.S 3
i 1*
J Q> _.
Si -• 3
:± o "
00
O
3
3
------- |