-------
order of magnitude. This indicates that the waste is composed of a small
amount of strongly adsorbed material and a larger amount of weakly adsorbed
material. As a result, breakthrough will probably occur very rapidly and
treatment will require a carbon dosage far in excess of 0.28 kg/m3 (2300
Ibs/MM gal).
Breakthrough curves were measured for the REC ultrafiltrate and the
results are given in Figure 27. The TOC concentration of the feed decreased
during the test. This can be attributed to biological growth in the feed
tank. After 64 liters of the feed had been processed, the test was
terminated because of considerable slime and fungus in the feed tank.
Effluent concentrations from each of the four columns are shown in Figure
27 as a function of the volume of waste processed. As anticipated from the
isotherm, breakthrough occurred rapidly for all columns. The curves for
both columns 3 and 4 extrapolate to 38 mg/Jl at 7.5 liters processed. There-
fore, the required carbon dosage is 5.2 kg/m3 (43,300 Ibs/MM gal). Applica-
tion of a dosage as large as this would be very expensive.
Ozonation Tests
End-of-Pipe Emulsion Crumb Ultrafiltrate--
Ozonation tests on the REC ultrafiltrate were conducted at the
conditions outlined in Table 29. The Type A tests were conducted under
reaction-rate-limited conditions, while the Type B test was conducted under
mass-transfer-limited conditions. The concentration profiles (concentra-
tion vs. reaction time) for TOC, dissolved ozone, and pH are shown in
Figures A3-A6, Appendix A, for runs A1-A4, respectively. The TOC concentra-
tion profiles are compared in Figure 28.
TABLE 29. TEST CONDITIONS DURING OZONATION OF REC ULTRAFILTRATE
Test Type Run
A 1
2
3
4
B 1
Temp
(°C)
30
30
30
30
30
PH
5
9
5
9
9
UV Light
off
off
on
on
off
At all conditions, ozone was effective in reducing the TOC of the
waste. Initial TOCs ranged from 200 to 220 mg/fc, and final TOCs, after two
hours of reaction time, ranged from 10 to 30 mg/?,. At pH 5, there was
no significant effect of UV light on the rate of reaction. A good straight-
line semi-logarithmic correlation is obtained indicating that the reaction
is pseudo-first-order with respect to TOC. That is, the rate equation
-------
to
)->
c
O)
o
O
o
o
o
220
200
180
160
140
120
100
80
60
40
20
FEED
10
20
30
40
Volume Processed, liters
50
60
Figure 27. TOC breakthrough curves for carbon column treatment
of end-of-pipe emulsion crumb ultrafiltrate.
89
-------
c
0)
u
c
o
100
90
80
70
O PH 5; Mo UV
& pH 5; With UV
LJ pH 9; No UV
V PH 9; With UV
\ xpH 9
x with UV
60
.2 50
40
30
20
10
pH 9
no UV
0 10 20 30 40 50 60 70 80 90 TOO 110 120
Reaction Time, min.
Figure 28. TOC vs. reaction time at various conditions for ozonation
of end of pipe emulsion crumb ultrafiltrate.
90
-------
is given by:
where
-dC _
dt
= kC
(1)
Integration gives:
where
C = TOC concentration
t = Reaction time
k = Rate constant
log £ kt
C0 2.303
Co = Initial TOC concentration
(2)
Thus, the slope of the plot of log C vs. t is proportional to the first-
order rate constant, k.
At pH 9 without UV light, the TOC dropped rapidly over the first 15 to
20 minutes. During this time period the dissolved ozone concentration
remained close to zero indicating that mass transfer was rate limiting
during this period and that the rate of reaction was very fast. Thus, the
use of alkaline conditions dramatically increases the initial rate of
reaction. After about 20 minutes, reaction occurred at a much slower rate,
but at a somewhat higher rate than at pH 5.
It can also be observed from Figure 28 that after 30 minutes of reaction
there is no significant difference in reaction rate with and without UV
light.
The results of a Type B test at 30°C, pH 9, and without UV light are
shown in Figure 29. Because of the high initial TOC of this waste, the test
procedure was modified somewhat. During the first half hour of ozonation,
Type A test conditions were used (1.42 m3/hr @ STP, 2 wt % 03 in feed).
Over the 3.5 hours of reaction under Type B conditions, the TOC
decreased from 108 to 59 mg/£ (circles of Figure 29). From the known flow
rate of ozone to the contactor and the measured decrease in ozone concentra-
tion across the contactor, it is possible to calculate the amount of ozone
consumed per liter of waste. Based on the assumptions that there is only one
active oxygen atom per molecule of ozone, that ozone provides the only source
of oxygen, and that the oxygen demand of the contaminants is associated
entirely with the organic carbon, the theoretical ratio of ozone consumed
91
-------
To 240 mg/1
o
o
120
110
100
90
80
70
60
50
40
30
20
10
0
Expected TOC for 100%
Utilization of Ozone
Actual TOCNP
A
A.
2 3
Time (hours)
Figure 29. Ozonation of end-of-pipe emulsion crumb ultra-
filtrate at 30°Cand pH 9 without UV light (type B)
92
-------
to TOC oxidized is:
32 mg 0 48 mg 03 _ fi mg Oa (3\
12 mg TOC x 16 mg 0 mg TOC v '
With this ratio and the measured ozone consumption, one can calculate a
theoretical or expected TOC decline (triangles of Figure 29) for 100%
utilization of ozone for organics oxidation.
The ozone utilization efficiency for reaction with organics is
defined as:
(ATOC)
d x 100% (4)
where
(ATOC) = actual TOC decrease for a given
reaction time
(ATOC) = calculated TOC decrease for the
same reaction time
As shown in Figure 29, the TOC decreased more rapidly than expected
based on 100% utilization of the ozone. There are several possible
explanations. First, it is possible that more than a single atom of
oxygen per ozone molecule participates in the oxidation reaction. This
oxygen could come from the ozone molecule, from dissolved molecular oxygen,
or from organic oxygen in the contaminants (e.g., acids, ketones, etc.).
It is also possible that reaction under Type A conditions during the first
half-hour produced many partial oxidation products. These products would
require less ozone to complete the oxidation than expected on the basis
of the above assumptions and calculations. Unfortunately, there was in-
sufficient waste remaining to check this latter hypothesis.
The ozone utilization efficiency from Figure 29 is 158%. This indicates
an ozone consumption of 5 mg 03 per mg TOC rather than the assumed value of
8. Thus, ozone appears to be very efficient for the ozonation of raw
emulsion crumb ultrafiltrate. Unfortunately, the TOC level of the ultra-
filtrate is relatively high, which adversely affects the economics for
ozonation of this waste.
End-of-Pipe Emulsion Crumb Reverse Osmosis Permeate
The objective of these tests was to determine the effect of temperature,
pH, and UV light on the rate of organics removal should polishing of the
reverse osmosis permeate become necessary. Eight tests were conducted with
REC reverse osmosis permeate. The test conditions and results are given in
Table 30. Because of the low levels of TOC in the samples, there is a
rather large uncertainty in the analyses. For many of the runs, the TOC
did not decrease continuously with reaction time as would be expected.
93
-------
J_
OJ
e
o
o
o c
in !-
E
U- CL
O
GO
o GO
I-H O
I s:
<; oo
z O
o
M UJ
o oo
cc
o: L
o :
u. L
oc
I co
OO D£
UJ <_J>
o:
c/o
oo
I1 CL
Q I-H
Z Q-
O I
O Ll_
o
OO Q
o
CO
CO
X)
OJ--
> 1
w
o
o c
+^ i
o
10
h-
c
3
o:
o
en
o
vo
o
cvj
o
CM
-a
0)
4J
O
o
cor COCMCNJ
o o o o o
v v v
o un o o o o
i ro vo en CM
CD in o o CD o
r- CO kO CTt CM
o in o o o o
i n (.o en CM
COVOCTIOJ
-------
T3
CU
E
o
o
LU
_j
CQ
<=c
o
CO
>
c
o
o
o c
in «-
a)
r-
^E
O
CO
o
o
OJ-- *.
> r
O O1
« E
(D
O C
o.
i-
3
4->
s-o"
o
CO
o
10
vo ^a- ro n i i
O
CM
- O> CO O O
- ro »d- to oo
in 10
j oo r-
I CM CM
O O O O O
v y v
vo ro in in in in
CM ro co «± in i~.
in in o
in C3 o i in 10
o i o o
o in o o o o
n vo 01 CM
o in o o o o
i n t£> en CM
o in o o o o
i m i£> cr> CM
o in o o o o
r CO IO CTl CM
c
o
c
o
<*-
<»-
o
in
a-
oo
95
-------
Judging from the magnitude of the deviations from a continuous TOC decrease,
the analytical error may be as high as ± 2 mg/H for some samples. [The
specified accuracy of the instrument used (Dohrmann DC-50) is ± 1 mg/Jl in
this range.] Since the relative error for these analyses is substantial,
conclusions can be drawn only with some uncertainty. Nevertheless, the
final column of Table 30 gives the estimated time required for 50% conver-
sion of TOC to C02.
Based on these times, the following conclusions are tentatively drawn
for ozonation of raw emulsion crumb reverse osmosis permeate.
- Mithout UV light at 30°C, there is probably
no significant difference in the reaction
rates at pH 5 and pH 9.
Increasing the temperature from 30°C to 45°C
(without UV light) has no significant
effect on the reaction rate at pH 5, but
significantly increases the reaction rate
at pH 9.
The use of UV light at 30°C appears to
inhibit the reaction both at pH 5 and pH 9,
but at 45°C UV light does not inhibit the
reaction and may promote the reaction rate
slightly, particularly at pH 5.
The most rapid rate of reaction occurs at 45°C
and pH 9. At these conditions, UV light has
no significant effect.
End-of-Pipe Emulsion Crumb Carbon Column Effluent--
Both Type A and Type B ozonation tests were conducted with the REC
waste after ultrafiltration and activated carbon treatment. The TOC of the
composite effluent from the carbon columns was ~100 mg/i .
Table 31 presents the experimental conditions for the Type A tests.
The TOC vs. Time curves for tests A-l through A-4 are compared in Figure 30.
The data on which Figure 30 is based are given in Appendix A. The results
shown in Figure 30 are quite similar to those presented for ozonation of
raw emulsion crumb ultrafiltrate (see Figure 28). The rate of reaction at
low pH is very slow. Irradiation with UV light appears to increase the rate
of reaction somewhat, but still the rate of reaction is probably too slow
for economical application of ozone at these conditions. On the other hand,
the rate of reaction at high pH is reasonably rapid. Again, UV irradiation
increases the rate slightly, but not enough to justify the use of UV in a
practical system.
96
-------
10 20 30 40 50 60 70 80
Reaction Time (Min)
90 100 110 120
Figure 30. Comparison of TOC vs. time curves for ozonation of
end-of-pipe emulsion crumb carbon effluent at
various conditions.
97
-------
TABLE 31. TEST CONDITIONS DURING OZONATION OF
REC CARBON COLUMN EFFLUENT
Test Temperature pH UV Light
A-l
A- 2
A- 3
A-4
30°C
30°C
30°C
30°C
5
9
5
9
off
off
on
on
The results of the Type B tests on the REC carbon effluent are shown
in Figures All and A12, Appendix A, for pH 5 and 9, respectively. Both
of these runs were conducted at 30°C without UV light. For both runs, the
actual TOC decreased more slowly than expected for 100% utilization of
ozone for organics oxidation. Over the 6 hour reaction period, the
utilization efficiency was 80% at pH 5 and 62% at pH 9. The fact that
the utilization efficiency is less than 100% can be largely attributed to
non-productive decomposition of the ozone. The lower utilization
efficiency at high pH indicates a greater degree of non-productive ozone
decomposition; i.e., decomposition which does not lead to oxidation of
organics.
Because of the higher ozone utilization efficiency at pH 5, it would
be preferred to conduct the reaction at low pH. However, as pointed out
above, the reaction rate at pH 5 is very slow. Therefore, a compromise
must be reached between low pH for high utilization efficiency and high pH
for reasonably rapid reaction.
Table 32 presents the analytical results for ozonation of the REC
carbon effluent. The ozonation was conducted at 30°C, pH 9, and without
UV light (see Figure A8, Appendix A). Ozonation at these conditions
for 2 hours produced high removal efficiencies for TOC, BOD5, and COD.
Some removal was achieved for surfactants and color, both of which were
already at low levels in the feed. A comparison of the ozonated product
water quality and the BATEA standards is given in Table 33.
98
-------
TABLE 32. CONTAMINANT ANALYSES FOR OZONATION OF
END-OF-PIPE EMULSION CRUMB CARBON EFFLUENT
Concentration Before Concentration After
Ozonation* Ozonation*
Assay (mq/1) (mg/1)
TDS
SS
Oil and Grease
Surfactant
Fe
Pb
Color
TOC
BOD5
COD
29,600
40f
< 4.0
0.51
5.2
< 1.0
50 units
100
20
438
28,600
60
5.4
0.39
5.1
< 1.0
40
10
3
44^
Removal
Efficiency
(%)
3
< 0
< 0
23
0
-
20
90
85
90
* Ozonated for 2 hours at 30°C, pH 9, without UV light.
Concentration profiles shown in Figure A8 .
t The high reading is due to a pinhole leak in the spiral-wound
ultrafiltration module.
tt Original assay was in error and insufficient sample remained to repeat
assay, therefore, value given is based on the carbon effluent COD/TOC
ratio of 4.38.
99
-------
TABLE 33. COMPARISON OF OZONATED REC CARBON EFFLUENT
PRODUCT WATER WITH EFFLUENT GUIDELINES FOR EMULSION
CRUMB MANUFACTURING WASTES
REC carbon
Effluent Ozonated
Assay Product Water BATEA
COD (mg/£)
BOD 5 (mg/A)
Suspended Solids (mg/&)
Oil and Grease (mg/&)
44*
3
60
5.4
130
5
10
5
Based on carbon effluent COD/TOC ratio of 4.38.
Only the suspended solids content in the ozonated effluent appreciably
exceeds the BATEA standard. These solids are principally the result of a
pinhole leak in the spiral wound module used during the ultrafiltration
tests. Under actual field conditions, a suspended solids content of <4
mg/£ is to be expected.
EXPERIMENTAL RESULTS FOR SECONDARY TREATED EMULSION CRUMB WASTEWATER
Proposed BATEA Treatment
Dual-Media Depth Filtration
Filter PerformanceFiltrate turbidity and column headless are plotted
as functions of operating time in Figure 31 for the processing of the
secondary emulsion crumb (SEC) effluent through the 0.05 m diameter dual-
media depth filter. The feed turbidity was 17 NTU, as received, and 10 NTU
following pH adjustment to pH = 7.1 with IN NaOH solution. As observed in
Figure 31, the filtrate turbidity throughout the first 20 hours of column
operation remained below 3 NTU, and except for the start-up period, was
below 2.5 NTU. This represents a turbidity reduction of greater than 75%.
During the 21st hour, a substantial turbidity increase to 5.7 NTU occurred,
after which filtrate quality improved slightly. When filtrate turbidity
exceeded 6 NTU, the run was terminated. This occurred after 0.32 m3 (84 gal)
had been processed through the column and 34 operating hours had elapsed.
At no point during the run did the column headless exceed 0.21 bar
(3 psig), indicating that substantial surface straining did not occur. In
terms of both turbidity reduction and filter run length, the performance
of the anthracite/silica sand filter on SEC effluent was acceptable.
The filtrate turbidity versus time plot for the processing of the SEC
wastewater with the 0.23 m diameter column is presented in Figure A13, Appendix
A. Following an initial unsteady period, the filtrate turbidity remained
100
-------
(2_OL* ueq) '
o
CM
CO
m
CO
d
10
HO
ro
O
<£>
ro
CVJ
ro
CO
CSJ
O
CM
3
O
10 -C
CO
O
c
o
M
-P C
Q- E
O) 3
-O r
o
re o
-5 $-
cu o>
E 4J
I 0)
^ E
re re
3 -r-
o -o
s- E
^- LO
cu°.
E O
l^ !->
> re
CO
o (/>
^- re
o S
re
Ol J3
c.
E
0
o o
o -
CL>
r- O)
T3 -M
i- re
J3 Ol
S- S-
p re
re T3
i- C.
+J O
i O
oo
=3
CD
(PIN)
101
-------
stable at 5 NTU. This turbidity is higher than that for the effluent
from the 0.05 m diameter column and, with a feed turbidity of 7.2 NTU,
represents only a 30% reduction in turbidity across the column. It is
possible that some plugging and channeling occurred within the column as
indicated by the somewhat higher headless (0.2-0.4 bar) as compared with
the 0.05 m diameter column (0.17 bar).
Backflush RequirementsThe minor surface straining which occurred
during the three runs did not interfere with the backflushing operations.
In each instance, the standard backwash flow rate of 646-881 m3/m2-day
(11-15 gpm/ft2) and duration of 6 to 8 minutes were sufficient to remove
accumulated solids.
Contaminant Removal--Table 34 presents the feed, filtrate, and
backwash-water analyses for dual-media depth filtration of the SEC effluent.
Following passage through the column, the reduction of BOD5 and TOC levels
averaged 66% and 10%, respectively. The COD analyses of the filtrates are
suspected to be in error.
The secondary treated emulsion crumb effluent met BPCTCA guidelines for
BOD5, COD, suspended solids and oil and grease prior to depth filtration.
Processing by the dual-media column did not produce a filtrate capable of
meeting the BATEA effluent guidelines.
Carbon Adsorption--
A carbon adsorption isotherm determined for the SEC dual-media depth
filter effluent is shown in Figure 32. As for previous isotherms, the data
follow a Freundlich relationship. It is of interest to compare the
isotherms for raw emulsion crumb ultrafiltrate (Figure 26) and secondary
emulsion crumb depth filter effluent (Figure 32). The TOC concentration of
the secondary waste is 58 mg/Jl compared to 220 mg/2, for the raw waste. In
addition, the slope of the isotherm for the secondary waste (1/n in the
equation X/m = (C)Vn) is 1.32 compared to 3.57 for the raw waste. These
differences indicate that biological treatment removes a substantial portion
of dissolved organics and preferentially removes the poorly adsorbed organics
which caused the steep isotherm slope for the raw waste.
Three attempts were made to obtain carbon breakthrough curves for the
SEC dual-media filtrate. In the first two runs, the pressure drop across
the columns increased beyond the capabilities of the pump and the flow rate
through the column decreased to zero. This occurred after only two points on
each breakthrough curve had been obtained; thus, the breakthrough curves
could not be constructed. For the second attempt, the feed was processed
through a one-micron cartridge filter before passing it through the carbon
column, but this did not eliminate the build-up in pressure drop.
The third SEC carbon column run was performed in the upflow mode of
operation to prevent particulate plugging. Unfortunately, at this point,
only a small volume of waste remained for processing and breakthrough did
not occur. Although the feasibility of dual-media depth filtration/activated
102
-------
o
1 1
t
s
h- o:
_J LU
* * i
Tg
:n LU
I i
Q. OO
LU - s:
O ID
LU CtL
-J Z
^t O
Q OO
o; ID
o s:
Lu LU
I/O Q
LU LU
00 1
>-«=c
1 LU
^f ryf
-
' 2r
ii
^ o
Y CM CM O
** I~~
r i
CO
Or~- CMCMr OI~-CMO
r--co r^^^^v* Oi o
co ro'iiicM o
CO CTl
*~*
l/> 1^^
"O T3 O> "^^*
Q) T- a) e ra
>^ - t ^> *-~ 1
i t o ec 01 z: >,
O ^ (/> *^^ in - * ^^ +J ^-> 4J
men s- ^-^ ^ - , 4->i ,_.,- -r-^-^
, >E
Q "O "O O^D^E-^EE * ' *' *-* "*^
in c * * c * E E ^ ' O ^^t " ' ~o o m
r -O Q) i i inoir-CT>UOO S~ O (Or- S- CE
00 3E-I-EOOO 3 i- -i_JO( O. U» '
in
CU
in
(O
c
c
-p~
i~
o
CD
t3
Ol
0
CO
0.
tn
w
S
03
U
T3
C
r
^""*
*^-1'
-------
1000
900
800
700
600
500
400
300
200
to
O
-o
10
o
TOO
90
80
70
60
50
40
30
20
10
x/m at CQ=385 mg TOC/g carbon
I I I 'I
5 6 7 8 9 10 20
TOC Concentration, mg/X,
30 40 50 60 70
Figure 32. Equilibrium adsorption isotherm at 20 C for secondary
treated emulsion crumb depth filter effluent.
104
-------
carbon treatment of SEC wastewaters was demonstrated, further testing is
necessary to develop an accurate economic profile of the process.
Reverse Osmosis Tests
Selection of Preferred Membrane--
The duPont B-10 polyamide module was chosen for the initial batch
processing of the depth filter effluent because of the high dissolved
solids loading of the SEC wastewater. The B-10 module has the highest
operating pressure of the reverse osmosis modules studied during this
program and is, therefore, best suited to handle the high osmotic pressure
exerted by the dissolved salts in the feed. During a second batch concen-
tration (see below) and the total recycle tests, the ROGA CA module was
added in series ahead of the B-10. The narrow-channel, spiral-wound CA
module is less susceptible to plugging than the hollow-fine-fiber B-10
perinea tor and may be preferred if future modules are developed with higher
pressure ratings.
Module Productivity
The B-10 permeate flow rate as a function of conductivity rejection
during a batch pumpdown of the SEC dual-media depth filter effluent is
shown in Figure 33. The productivity of the B-10 module decreased from
6.6 m3/day (1.2 gpm) initially to 0.72 m3/day (0.13 gpm) at a 4.2X volumetric
feed concentration. Concentration beyond this point was judged to be
uneconomical for full-scale operation.
The non-linearity of the flux/rejection relationship implies that
osmotic pressure was not the only factor limiting module productivity.
Module fouling during the latter stages of the batch concentration appears
to have reduced permeator output.
A second reverse osmosis batch concentration was performed with the
spiral-wound CA module operated in series with the B-10 module. The test
system was operated at the maximum pressure (41.4 bars [600 psig])
recommended for the CA module. The permeator productivities for this
experiment are presented in Figure 34 as a function of the conductivity
rejection. The spiral-wound module can be seen to be less susceptible to
fouling than the hollow-fiber module since its flux/rejection curve
follows the expected linear relationship. The permeate flow rates for the
B-10 module during this test are lower since the module was operating at
an inlet pressure of 36.6-37.9 bar (530-550 psig). The characteristics of
the curves developed for the B-10 module (shown in Figures 33 and 34) are,
however, quite similar.
The CA and B-10 modules were exposed to the SEC dual-media depth
filter effluent for extended time periods during total recycle experiments
at IX, 2X, and 4X volumetric feed concentrations. The permeate flow rates
during these experiments are presented in Figure 35 as a function of opera-
ting time. At all three concentrations, the productivity of the spiral-
105
-------
CM XI O
o
in co 01
i O
+-> O-
C O
O
i
I
CQ
s- a. r~
LO S- OJ U3 1C
l/> O S- O-
O) 3
1. OJ 4-> OJ
Q- s- 03 :r <->
3 S- O- fO
4-> t/1 QJ OJ
-i CL. I U- Q-
c:
O
C£
o; i
o
10
o
in
o
-3-
O
PI
en
O 4->
i- to
-!-> S-
O 4->
C 4-
O
O (O
r-
4- -a
O 1)
O i
r" fO
P 3
o -a
re s-
o
rO C
O
> 3
-l-> O)
o
=1 T3
-O O)
O 4->
S- fO
Q. Q)
>,
T3 S-
O ro
E -o -a
E O
O O -
i O S_
i ai a)
CQ 01 Q.
n
CO
CD
3
CD
U_
/Cep/ ID
106
-------
/'ep/ ui
C 01
O C
I !
4-J CO
O 10
t-j o
cu o
s- i.
Q.
> O
r- O
P O)
o to
3
-o «
C
O to
O
C
o
O T3
C , E
t-> 3
i- S-
> O
c:
u o
3 !-
o to
o
s- 3
a. E
a;
01
r T3
3 QJ
o +J
O (O
CO +->
I
to >,
o s-
(/) "O
o c
o
a> o
to a> T3
s- to o
tu -
> s- s-
OJ O O)
a: M- a.
oo
QJ
3
05
107
-------
5.5
5.0
4.5
4.0
I I
1.5
1.0
.5
c
CA, 4X
IX 2X 4X
B-10 O D V
CA £ 0 V
Inlet pressure to CA: 41.4 bar
Inlet pressure to B-10: 36.6 bar
Temperature: 19-27°C
4 8 12 16 20 24 28 32 36 40 44 48 52 56 60
Time (hours)
Figure 35. Reverse osmosis module productivity vs. time for total
recycle life tests with secondary treated emulsion
crumb dual-media filtrate.
108
-------
wound module remained relatively stable while the productivity of the
hollow-fiber B-10 permeator declined with time. This is further indica-
tion that some fouling of the hollow-fiber membranes was occurring.
The performance history of the reverse osmosis modules during
processing of the SEC depth filter effluent is indicated by the standard
Nad test results given in Table A3, Appendix A. A 23% flux decline was
observed for the CA module during 124 hours of exposure to this waste
stream, while the B-10 module exhibited a 65% flux decline for the same
exposure. The overall NaCl rejections for both modules decreased slightly,
even though periodic increases in rejection were noted.
It is suspected that the suspended solids of the reverse osmosis feed
were not sufficiently reduced by the 5 y and 1 y string-wound cartridge
filters used for pretreatment. This probably resulted in fouling of the
B-10 module and, to a lesser degree, the CA module. Standard cleaning
procedures for removal of colloidal matter from the permeators were
employed and were successful in restoring the productivity and NaCl
rejection of both modules to pre-exposure levels. No irreversible membrane
fouling or degradation due to environmental attack was noted.
Module Rejection--
Feed, composite permeate, and final concentrate analyses for the
batch concentration of SEC dual-media depth filter effluent to a volumetric
concentration of 4.2X are presented in Table 35. Contaminant rejections
are also shown. The B-10 module exhibited rejections of 64%, 99%, and 88%
for BOD5, COD, and TOC, respectively. The analyses which are common to the
development document guidelines are compared in Table 36. The contaminant
levels in the reverse osmosis permeate are lower than required to meet the
BATEA effluent limitation guidelines for all four pollutant parameters, and
the extent of COD removal is exceptionally impressive.
TABLE 36. COMPARISON OF SEC REVERSE OSMOSIS
PERMEATE WITH EFFLUENT GUIDELINES FOR
EMULSION CRUMB MANUFACTURING WASTES
Reverse Osmosis
Assay Composite Permeate BATEA
COD (mg/A) 6 130
BOD5 (mg/£) ' 4 5
Suspended Solids (mg/Jl) <4 10
Oil and Grease (mg/&) <4 5
109
-------
CD
i i
CO
co
UJ
CJ
o
oc \
Q- z
UJ
?L?
1 U.
OQ UJ
h- C£
Z LU
O 1
Q 1
Z3 I-.
Q U.
3;
co i
> i Q.
co uj
O Q
S
CO OQ
os:
ro
uj a:
co o
a:
ij-i 2:
> O
UJ I i
C£ CO
a:=D
pS:
U. i ij
CO Q
1 1 i 1 1 i
coP
>- «=c
J UJ
«=C IX.
^ 1
-
I a:
=r
oo
UJ
[
QQ
=f
1
*
a«
«t
S-i
o
0)
U
CK
t-
UJ
"10
e1"
a:
(U
4J (U
i- +-»
co
4->
ta
i 4->
1C C
C 0)
r- O
LU. C
O
O
-a
(U
OJ
U-
+j
c
ra
c
03
4->
C
O
o
«d-
CO
en
1C
CM
CM
o
pv
m
*
a>
LO
o
*i-
CM
«
st-
i
CO
o
r
r
0
IO
-o
at
>
"3
CO
CO
!_
o
^^
r^ r~~
«3 -^
4-> en
O E
1-'
CMOcnooiovoo incM co
inoi~-ooco«*CM i r-.ro it-~
co LO co ai 10 co cncn en
A A A
CM in n in
^^COVD*±OI oinoioo
v v y v £>
us . 10 m r 10
m«d-oooioiOr Or^ioo
i .. «* CM VO 0 0
i v CM co r^ o
* M
oo *
10
ro r-. in o o
r-.coior- r-i cMOOinr-.o
CM VD c O O
in CM in
n
^3-
*E"
^~ o
r -^.
-^. CO
o> ^^ o
E x:
co ** -^ E
a en ^~ g-
i-d) E ra ^
I CO h
o ,
CO dl --^ CO . . . +J > +J
T3 CJ3 i i --. C -^ C >> >
Q) ^--^o>:3+-» <-
T3 -OcncnE-'-'EE'---''!- +J
C'-- CEE^-'U'~>^-' -0 0
(U , to ~~* -a
«»0>i OOOi-OlOr-fc. C
3E-i-ooo u i- ai o 33: o
IO- Ol CJCO(/5i-<_lt_Jt O-tJ
o
o
r
X
a)
4->
rtj C
M°
C -!->
(U
(U
U-
l»-
o
c
o
4->
03
!_
4-1
c
~,
(J
c:
QJ
f
o
<*!
<4-
LU
J,^
0]
>
c
0)
o;
X
no
-------
Ozonation
Ozonation experiments were performed only with the depth filter carbon-
treated effluent since the COD (6 mg/&) of the reverse osmosis permeate was
below BATEA standards. Two Type A and two Type B tests were conducted with
the composite effluent from the first of the partial carbon column runs
described above. The test conditions are given in Table 37.
The results for ozonation of SEC carbon effluent are shown in Figures
36 through 39 for runs A-l, A-2, B-l, and B-2, respectively. For both
Type A runs, the dissolved ozone concentration increased rapidly to a
plateau value indicating that the removal of TOC was limited by the rate
of reaction between dissolved ozone and dissolved organics rather than by
mass transfer.
TABLE 37. TEST CONDITIONS DURING OZONATION OF
SEC CARBON COLUMN EFFLUENT
Temperature
Test Type Run (°C) pH UV Light
A
B
1
2
1
2
30
30
30
30
9
5
9
5
off
off
off
off
A significant difference exists in the initial TOC concentrations for
these two runs (21 and 39 mg/&). This may have been the result of bio-
logical activity in the liquid storage container which was not refrigerated.
At both pH's the initial decrease in TOC is rapid, but at pH 5 the TOC
levels off at about 13 mg/£, whereas the TOC continues to decrease at pH 9.
There is some scatter in the measured TOC values for the Type B run at
pH 5. Based on the rather uncertain dashed line shown in Figure 39, the
ozone utilization efficiency for organics oxidation is only 14%. This can
be attributed to the very slow rate of reaction at low pH and the pre-
dominance of non-productive ozone decomposition.
The results for pH 9 are shown in Figure 38. The TOC decreased much
more rapidly at pH 9 than at pH 5. The change in slope for the theoretical
curve (triangles) results from a change in ozone flow rate from 0.028 m3/hr
(1 CFH) to 0.008 m3/hr (0.3 CFH), The actual TOC data points were correlated
with a straight line changing slope at the same point. Based on an extra-
polation of the theoretical curve, the ozone utilization efficiency for
reduction of the TOC to 5 mg/fc is 45%. For this waste, ozonation at high pH
111
-------
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure 36. Ozonation of secondary treated emulsion crumb carbon
effluent at 30°C and pH 9 without UV light.
112
-------
ffi
a
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
rH
"X
G
_e
(
0
-14
-13
-12
-11
-10
- 9
- 8
- 7
- 6
3
- 4
- 3
- 2
- 1
rH
tr>
£
0
^ O
^
-70
-65
-60
-55
-50
-45
~40(
-35
-30
?5f
L. JL
-20
- 15
- 10
- 5
/
1 1 1 1 1 1 1 1 1 1 1 1
-A
^ A °3
t /^ A A -
1 A
A /
"Ann pH n n n
tr \ LJ LJ LJ |_J LJ
y
" /\
-/ °"^0 TOC Q
/ 0
1
+
1 1 1 1 1 1 1 1 1 1 1 1
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure 37. Ozonation of secondary treated emulsion crumb carbon
effluent at 30°C and pH 5 without UV light.
113
-------
18
17
16
15
14
13
12
11
10
P 9
8
7
6
5
4
3
2
1
0
CD
£
O
o
Expected TOC for 100%
Utilization of Ozone
2 3
Time (hours)
Figure 38. Ozonation of secondary treated emulsion crumb
carbon effluent at 30°C and pH 9 without UV light
(type B).
114
-------
£U
19
18 (
17
16
15
14
13
12
11
10
^ 9
O o
o o
1
7
6
5
4
3
2
1
0
1 1 1 1 1
1 T
^v " ~ Actual TOC T
X, T
\ 1
- Expected TOC for\/\ -
100% Utilization \
_ of Ozone ^& _
^X
- -
- -
- -
-
- -
_ _
- -
- -
- -
l 1 1 1 1
3 4
Time (hours)
Figure 39. Ozonation of secondary treated emulsion crumb
carbon effluent at 30°C and pH 5 without U\l
light (type B).
115
-------
is preferred both from the standpoint of reaction rate and from the stand-
point of ozone utilization efficiency.
EXPERIMENTAL RESULTS FOR END-OF-PIPE SOLUTION CRUMB WASTEWATER
Pretreatment Comparison Testing
Filtrate turbidity and column headless are plotted as a function of
operating time in Figure 40 for the processing of raw solution crumb (RSC)
wastewater through the 0.05 m diameter dual-media depth filter. The feed
turbidity was 81 NTU, as received, and averaged 40 NTU following pH
adjustment to pH = 8.3 with concentrated HaSO^. The filtrate turbidity, as
observed in Figure 40, never decreased below 8 NTU. Although the run
extended over a 12-hour period, it was evident after 5 hours that break-
through was beginning to take place. The column headless remained stable
throughout the run indicating that no substantial surface straining occurred.
For ultrafiltration, 0.15 m3 (40 gal) of the end-of-pipe solution crumb
wastewater were charged to the feed tank and were processed in the total-
recycle mode for 46 hours. The permeate flux vs. time curve for this test is
shown in Figure 41. After 5 hours operating time, the permeate flux
stabilized between 1.2 and 1.6 m3/m2-day (30-40 gfd). During the course of
the experiment, the ultrafiltrate turbidity averaged 0.25 NTU.
The high turbidity of the dual-media depth filter effluent precludes the
use of this process for pretreatment of the feed to the hollow-fine-fiber
reverse osmosis module. In contrast, the low turbidity ultrafiltrate was
quite acceptable for reverse osmosis feed. Also, the ultrafiltrate flux
stabilized at economically acceptable levels indicating that ultrafiltra-
tion is an acceptable pretreatment option.
Ultrafiltration Tests
Membrane Flux--
The ultrafiltrate flux vs. time for a 15.6X volumetric concentration
(93.6% conversion) of the RSC wastewater is shown in Figure 42. As is
typical of most batch concentrations, the permeate flux decreased with both
increasing feed concentration and time. The average flux over the 14-hour
batch concentration was 1.77 m3/m2-day (44.3 gfd).
Total recycle ultrafiltration experiments at 5X, 10X, and 20X volumetric
feed concentrations were performed to further detail the HFM membrane flux
characteristics with the raw solution crumb wastewater and to provide pro-
longed membrane exposure to this waste stream. The flux vs. time data for
these total recycle experiments are presented in Figure 43. No severe
membrane fouling with time is observed. The slightly improved flux for the
10X feed sample as compared to the 5X feed after ~9 hours recirculation is
not readily explainable.
116
-------
Q.-P
> I
i- c
-o o
O
OJ -
03 rO
+J -l->
(J
O)
3
(niN)
117
-------
in
0)
ro
o;
e
o
a
a>
S-
S-
Q.
Ol
c
O
CO
n
03
S_
01
o
c
o
o
-o
0)
a;
x
s-
o
<+- $-
a>
O) 4->
-io a>
-4->
t/J
co co
> S
M- S-
U
d)
c c
tO O
E O
CO
c
O O)
i- a.
M -
-------
o
CM
$.
ro
3
s-
ro o
o
CO
Oi
S-
in
0)
to nj
> 2
r- 3
M- t.
O
(!)
C C
(O O
S- -
^D -(->
E 13
£ O
t/)
C
o a>
r- Q-
r O
t T3
03 C
5- (U
CVJ
CM
ro
CO
CM
O
CM
CO
O
o
ai
s_
a>
u_
119
-------
OJ
X
O
CVJ
O)
-M
(O
2
OJ
4->
CO
re
LT> =3
s- 'o
O CO
01
s_
3
o
Q.
i
X CU
4- O
E O
re '*-
S- -(->
jo re
E c
OJ
c o
o c
i- o
»-> o
re
s- o
4-> !-
i- -t->
M- O)
-(-> I
r O
ZD >
CO
O)
S-
120
-------
Membrane Rejection--
Feed, composite permeate, and final concentrate contaminant analyses
are presented in Table 38 for ultrafiltration of the raw solution crumb
wastewater. The HFM membrane rejections for BOD5, COD, and TOC were 65%,
29%, and 15%, respectively. The overall rejection for oil and grease was
61%. A comparison of the composite ultrafiltrate analyses and the effluent
guideline standards is presented in Table 39. It does not appear that
ultrafiltration alone can successfully treat the raw solution crumb waste-
water to the degree required to meet either the BPCTCA or BATEA standards.
TABLE 39. COMPARISON OF RSC ULTRAFILTRATE WITH
EFFLUENT GUIDELINES FOR SOLUTION CRUMB
MANUFACTURING HASTES
Composite
Assay Ultrafiltrate BPCTCA BATEA
COD (mg/Jl)
BOD 5 (mg/£)
Suspended Solids (mg/£)
Oil and Grease (mg/&)
444
30
<4
11
245
25
40
10
130
5
10
5
Membrane Flux Recovery
Table 40 presents the flux recovery data and accumulated operating
time for the ultrafiltration membranes used to treat the raw solution crumb
wastewaters. The data are also summarized for all previous exposures of
these membranes to synthetic rubber manufacturing wastewaters. Throughout
nearly 300 hours of exposure to these wastes, no membrane degradation due
to environmental attack is evident.
Reverse Osmosis Tests
Module Productivity--
The reverse osmosis module employed during the processing of the end-
of-pipe solution crumb ultrafiltrate was a duPont B-9 polyamide hollow-fine-
fiber permeator. This module is well-suited for the range of dissolved
solids (800-1200 mg/J,) in the ultrafiltrate. The B-9 operates with a feed
pressure of 27.6 bar (400 psig).
The B-9 permeate flow rate is shown in Figure 44 as a function of
volumetric feed concentration for batch concentration to 17.3X. As expected,
the productivity of the B-9 module declined gradually with the increase in
volumetric feed concentration. This productivity loss is associated with the
increase in the feed osmotic pressure which results in the reduction of the
net driving pressure across the membrane. The flux level at 10X was, how-
ever, still 6.34 m3/day (1.15 gpm) which is an economically attractive
121
-------
TABLE 38. CONTAMINANT ANALYSES FOR ULTRAFILTRATION OF
END-OF-PIPE SOLUTION CRUMB WASTEUATER
Contaminant
Total Dissolved
Solids (mg/1)
Suspended Solids
(mg/1)
Oil and Grease
(mg/1 )
TOC (mg/1)
COD (mg/1)
BODc (mg/1 )
Surfactants (mg/1 )
Iron (mg/1 )
Lead (mg/1 )
Color (units)
Turbidity (NTU)
pH
Conductivity
(ymhos/cm)
*Removal
pH Adjusted
Feed
1060
123
28
144
625
86
0.66
3.4
<1.0
100
55
8.4
1230
_ concentration
Final
Concentrate
1460
2740
105
1100
2660
200
1.8
58
<1.0
4000
255
8.3
(1900)
Composite
Permeate
1050
<4
11
122
444
30
0.52
<1.0
<1.0
80
0.55
8.3
1300
Removal
Efficiency, %*
...
>96.7
60.7
15.3
23.0
65.1
21.2
>70.6
20.0
99.0
of feed - concentration of composite permeate
Efficiency, ' concentration oT feed
Note: ( ) indicates suspected error in analysis
-------
TABLE 40. FLUX RECOVERY AND ACCUMULATED OPERATING TIME FOR TUBULAR
HFM MEMBRANES OPERATING ON SYNTHETIC RUBBER WASTEWATERS
Wastewater
Description
New Membranes
End-of-pipe
Emulsion Crumb
End-of-pipe
Latex
Accumulated
Operating
Time (hours)
0
76
87
Water Flux
before
"Sponaeball"
(m3/mz-day)
10,4
7.96
2.40
Water Flux
after
"Sponqeball"
(iTH/m^-day)
11.1
10.7
End-of-pipe
Solution Crumb
a) IX Recycle 133
b) Batch
Concentration 147
c) 5X Recycle 192
d) 10X Recycle 220
e) 20X Recycle 292
7.72
5.68
8.44
12.4
8.52
11.8
12.8
14.3
14.6
11.1
123
-------
O
in
en
- o
res
-Q
10 co
r~T co
o
E
r^
CM
QJ
3
O)
^1
(J
QJ
ce
o
r3
-D
O
i-
"O "O
QJ OJ
QJ
U- U_
i i
TO (T3
'r *r~
4-} 4^
*^ *r
C C
ro
i.
cr
OJ
o
c:
o
^_
fc
f
LL.
Q_
QJ
13
o
O
2!
QJ
Dl
ro
i.
Q)
*
X
o
CD
o
cr
o
CJ
QJ
E
O)
O)
X
OJ
CO
+->
c:
o
Q- CD
3 4J
T3 fO
S_
O
£
o s-
03
+->
c
O)
o
o
o
O !-
I- +-)
0} O
i 0)
O 0.
> T-
Q.
O I
<1J 4-
OJ O
<4- I
T3
C
to c:
+-) to
o to
3 OJ
T3 0
O O
S- S-
Q- Q-
O>
CD
X'ep/ iii
124
-------
productivity for this type of module. The average module productivity for
concentration to 17.3X was 10.95 m3/day (1.99 gpm). A productivity of this
magnitude should make reverse osmosis a viable unit process for treatment
of the raw solution crumb ultrafiltrate.
The B-9 module was also exposed to the raw solution crumb ultrafiltrate
for extended time periods during total recycle experiments at IX and 5X
volumetric feed concentrations. The CA spiral-wound module was operated in
series with the B-9 module during the IX recycle test only. Although not
readily explainable, the narrow pH range for the CA module (pH 4-6) could
not be maintained with this wastewater without constant pH adjustment;
therefore, testing of the CA module was discontinued.
The permeate flow rates during these total recycle experiments are
presented as a function of operating time in Figure 45. Except for minor
fluctuations due to a temperature increase following system start-up(s),
the permeate flow levels were stable for the course of the experiments.
This indicates the absence of membrane fouling during these tests.
The results of the standard NaCl performance tests presented in Table
41 also indicate the favorable and consistent performance of the CA and
B-9 modules. Productivity of the CA module increased 8%, while the B-9
module productivity decreased by 2%. Module rejection decreased 1% and
2.5% for the CA and B-9 modules, respectively. These data indicate little
change in module performance after exposure to this waste stream.
Module Rejection--
Feed, point and composite permeate, and final concentrate analyses for
the batch concentration of the RSC ultrafiltrate are presented in Table 42.
Those assays which are in common with the Development Document guidelines
are further summarized in Table 43. The reverse osmosis permeate meets
essentially all BPCTCA and BATEA standards. The oil and grease level in
the permeate is, however, in excess of the BATEA standard by 2 mg/£.
TABLE 43. COMPARISON OF RSC REVERSE OSMOSIS PERMEATE
WITH EFFLUENT GUIDELINES FOR SOLUTION CRUMB
MANUFACTURING WASTES
Composite Reverse
Assay Osmosis Permeate BPCTCA BATEA
COD (mg/fc)
BOD5 (mg/A)
Suspended Solids (mg/£)
Oil and Grease (mg/£)
36
4
nil
7
245
25
40
10
130
5
10
5
125
-------
~
<
C d)
3 +J
O S-
~
!-> CD
t/1 -r-
>> C
^
to
i. S-
OJ
i- S-
4J -p
>,
4-> S-
i- O
>
r- C
-U O
U !-
3 +->
-O 3
So
Q- (/)
0) O)
r D-
3 -i-
O Q-
§1
14-
O
(/) I
c- -a
w c
o QJ
CO JC
O -4->
10
J- in
cu +J
> to
Q) OJ
oc: +->
to
3
CD
ui 'A';iAi;Dnpoa
-------
to +J
CD ,
;g£
?Q =
5 J3
J E >>
S-"
O CO r-
OJ
zj u
' S- >.
U U
' Q.r C r
: I T- - 3 J
: 4- M- X
c *
co u_
Q .
o:
U C
- o
in -f
VO
O
CO
co s:
- ZD
co a:
00
co z
oo
co ZD
LU O
> co
LU
C£ LU
a.
o a.
co o
O Q
10
CO*
o
LU U_
O O
LU
a: cs
r** O
CvJ Csj
ai -6
Q. U- C
I CO
a: co
CO LU
u.8
o o:
Q.
dL CD
O -S^
co o:
DC Q
LU
00
£l <
3 I O
CD O 03
na t- - +J
o CL QJ u
CC to O <
o -
-------
1 1 1
LJ_I
' I
OT «=c
i a:
to i
i
r~~ H-H
Z. 1 1
0 «=C
CX. rV
. i
Q _1
"»' ~ **)
00 CQ
00 Z3
O o:
^>^ C_3
o 2:
o
| , ] J [
oo i
az zo
LU _1
> o
LU 00
o;
LU
o; o-
O 1-1
U- D_
00 U_
LU O
00 1
>- Q
eC LU
2:
c< LJ_
f-^)
h-
§ :i
i i OO
s: oo
et LU
1 0
Z O
O r*V
0 CL.
*
OJ
UJ
_l
GO
<£.
|
*o
"ro >>
> U
O C
E cu
CU T-
o; u
4-
4-
LU
cu
+-> cu
r- +->
on ro
O CU
E 1-
o cu
O 0.
cu
01
E
LO D_
CU
ro
CD
E
S-
X CU
i CL,
cu
ro
i.
c
i CU
ro O
C C
r- O
1 i <_J
, .
CU
ro
CU S-
r- I
LO T-
0 4-
O. ro
0 E S-
cu o +J
cu o .
c
ro
C
r-
E
ro
c
o
^O ^ CO CT>
oo i ro en en
1 t~, O VO
3- i r- ro
LO 1 1 CM O
i 1 1 r- *f
CM 1 1
t^ 1 1 IO LO
LO 1 1
1 1 V
o cn o o
CD 1 .3- O CD
CO 1 CM CO
IO «3-
n i
O *3" i CM =!-
LO %/ i CM «i-
O v i >3-
«t
r
IO
"O "X3
CU -r- CU
> . i CO
r- O ro
O "^ OO CU
CO CO S ^"^ ^~*"
CO E "O C3 i i
r * ^ OJ ^^v ^^.
Q T3 "O CO CO
t *o cu . to . ^ -* ^^
ro *r~ o -^_ ^^
J->r U1CT1 r CO CJ Q
OO 3E -£ CD C5
100 OO' O t <_J
p^ i CO LO CVJ
CO OO 1 1 CO LO 1 C/D
OO CM 1 1 CTi 1 CO
1 1 '
CM
< O O LO r
-^^
-B ^ ,E
CO ^ ^>
r- 4J t r- -r- -r-
-^c-^.-^c>, >E
COfl3COC034-> O
E+JEE^-'r- +J^-
o - ~O cj CO
ro i. -r- 3 O
L04-C-OO.O -O^:
QS-OrOi S. CE
O3S-CUO3^;O3.
CQ OO i i _1 CJ ( Q-
O rO
cj s.
1 C
cu
o o
O) C
cu o
4- CJ
4-
O
c
O
r
4J
^
o
c
CU
r
o
1
4-
LLJ
r^
to
o
E
cu
o:
*
128
-------
Carbon Adsorption Tests
The equilibrium adsorption isotherm at 20°C for the raw solution crumb
ultrafiltrate is shown in Figure 46. The data indicate that the Freundlich
isotherm expression does not hold over the entire range of carbon dosages
investigated. However, two straight lines - one for low carbon doses
(0.02 to 0.2 g/fc), the other for high carbon doses (0.2 to 5 g/fc) - fit
the isotherm data quite well. This suggests that the adsorption of organics
from RSC ultrafiltrate proceeds via the following mechanism: A small
number of strongly adsorbed organics are initially removed by the activated
carbon. If additional adsorption sites are available (i.e., more carbon
present), then a large number of moderately adsorbed organics are removed
and the slope of the isotherm levels out.
The TOC breakthrough curve for carbon column treatment of raw solution
crumb ultrafiltrate is shown in Figure 47. Effluent concentrations from
each of the four columns are shown in this figure as a function of the
volume of waste processed. The curve for column 1 approaches the feed
concentration very quickly following the trend observed in the isotherm.
Thus, column 1 removes the small portion of strongly adsorbed organics,
while columns 2, 3, and 4 adsorb the moderately hydrophobic organics. The
curve for column 4 indicates a TOC of 40 mg/l (i.e., COD of -120 mg/Jl) at
42 liters processed. Processing beyond this point would exceed the BATEA
guideline of 130 mg/fc COD in the effluent.
The required carbon dosage would be 1.22 kg/m3 (10,160 Ibs/MM gal).
This is a relatively high carbon dosage and would result in treatment costs
of approximately $1.06/m3 ($4.00/1000 gal). At this cost, activated carbon
treatment of the raw solution crumb ultrafiltrate may be economically un-
attractive.
Ozonation Tests
Raw Solution Crumb Ultrafiltrate--
Results for ozonation of raw solution crumb ultrafiltrate at high ozone
dosages (Type A tests) are presented in Figures 48 and 49. Both tests were
performed at 30°C and without UV irradiation. The test results plotted in
Figure 48 were obtained at pH 9; the test results of Figure 49 at pH 5.
The dissolved ozone concentration in both runs increased to a plateau
value during the first fifteen minutes and remained fairly constant there-
after. This indicates that the rate of TOC removal was limited by the rate
of reaction between dissolved organics and dissolved ozone (i.e., reaction-
rate limited) and was not mass-transfer limited. Comparison of Figures 48
and 49 indicates that ozonation at the higher pH would be preferred from the
reaction rate point of view. At pH 9, the rate of organic oxidation was
approximately twice that achieved at pH 5. After 90 minutes of ozonation at
pH 9, the product water TOC stabilized at 30 mgA (COD -90 mg/£).
129
-------
400 _
300
200
o 100
03
O
01
o
o
cr>
C7>
r-
TJ
<0
O
O
O
50
10
1 I I I I
x/m at Co = 285 mg TOC/g Carbon
9 10
O
Co = 125 rng/X,
I I I I I I I
50
TOC Concentration, mg/£
100
200
Figure 46. Equilibrium adsorption isotherm at 20°C for end-of-
pipe solution crumb ultrafiltrate.
130
-------
160
c
o
O
c
o
o
o
o
140
120
100
80
60
FEED
I
I
10 20 30 40 50 60
Volume Processed (liters)
70
80
90
Figure 47. TOC breakthrough curves for carbon column treatment
of end of pipe solution crumb ultrafiltrate.
131
-------
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure 48. Ozonation of end-of-pipe solution crumb ultrafiltrate
at 30°Cand pH 9 without UV light.
132
-------
10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure 49. Ozonation of end-of-pipe solution crumb ultra-filtrate
at 30°C and pH 5 wfthout UV light.
133
-------
The results of Type B tests on the RSC ultrafiltrate are shown in
Figures 50 and 51 for pH 9 and 7, respectively. Both of these runs were
conducted at 30°C without UV light. For the run at pH 9, the actual TOC
decreased much more slowly than expected for 100% (assuming an ozone to
TOC ratio of 8:1) utilization of ozone for organics oxidation. Over the
6-hour reaction period, the utilization efficiency at pH 9 was only 25%.
For the run at pH 7, the utilization efficiency is -100% throughout the
first 5 hours of reaction. Because of the higher ozone utilization
efficiency at pH 7, these tests indicate it would be preferred to conduct
the reaction at neutral pH.
Selection of the operating conditions for a full-scale ozone contactor
treating raw solution crumb ultrafiltrate would be based on an economic
tradeoff between ozone generation power requirements (low ozone losses at
lower pH) and multi-stage contactor volume (rapid reaction at higher pH).
Although pH-related trends in ozonation have been established with this and
other synthetic rubber manufacturing wastewaters, further testing would be
necessary to fully develop design criteria.
Raw Solution Crumb Carbon Effluent--
A similar series of ozonation tests were performed on the raw solution
crumb carbon column effluent. The Type A tests at pH 9 and 5 are
presented in Figures Bl and B2, in Appendix B, respectively. Both tests
were conducted at 30°C and without UV light. The rate of organic oxidation
was again higher at the higher pH level. In fact, at pH 5 no further
reduction in TOC occurred after the initial 30 minutes of ozonation.
The results of the Type B tests with the raw solution crumb carbon
effluent are given in Figure B3 for ozonation at pH 9 and in Figure B4
for ozonation at pH 5. In both experiments, the temperature was 30°C and
no UV light was employed. A substantial difference in the initial TOC of
the carbon effluent, 44 mg/fc during the run at pH 9 and 110 mg/fc for the
pH 5 run, is not readily explainable. However, very similar results to
those obtained with the raw solution crumb ultrafiltrate (see above) re-
confirm the conclusion that a compromise between ozonation at pH 9 where
the reaction-rate is high and ozonation at pH 5 where the ozone utilization
efficiency is high must be reached.
EXPERIMENTAL RESULTS FOR SECONDARY TREATED SOLUTION CRUMB WASTEWATER
Proposed BATEA Treatment
Dual-Media Depth Filtration--
Filter Performance--Filtrate turbidity and column headloss are plotted
as functions of operating time in Figure 52 for dual-media depth filtration
of the secondary solution crumb (SSC) wastewater. The data presented in this
Figure are for the 0,05 m diameter column. The feed turbidity was 14 NTU as
received and averaged 23 NTU during processing at a pH of 7.5. The filtrate
134
-------
Expected TOC
for 100%
012345i
Time (hours)
Figure 50. Ozonation of end-of-pipe solution crumb
ultrafiltrate at 30°C and pH 9 without
UV light (type B).
135
-------
Expected TOC for
100% utilization
of Ozone
3 4
Time (hours)
Figure 51. Ozonation of end-of-pipe solution crumb
ultrafiltrate at 30°C and pH 7 without
UV light (type B).
136
-------
(ueq) ssoi
CVJ
o
O
r-
o
CO
O
IO
O
O
o
CO
O
t/1
o
-a
ro
01
>>
r-
«* JO
Ol
S-
a: s_ r
01
3 a.
o E
©
I DC
0.
01 3
Q-i
OJ O
-O O
re S-
i- d)
T3 +->
OJ d)
I re
"re T3
3
T3 E
s- Ln
o o
M-
o
O)
4-)
re
co 3
CO CO
co re
-O E
re 3
OJ S_
c
o
O 3
O i
o
"O O
re O
S- OJ
-p to
^ M-
u. o
CNJ
LO
OJ
3
00
(OLN)
137
-------
300
200 -
I
fO
C7!
O
I
C7>
ft
cr>
T3
(O
O
O
O
100
50
20
10
1 I I I r
X/M at CQ = .260 g TOC/g Carbon
O
CQ = 78 mg/8,
j.
_L
10
Figure 53.
20 50
TOC Concentration, mg/£
70
100
Equilibrium adsorption isotherm at 20°C for secondary
treated solution crumb dual-media filtrate.
138
-------
turbidity throughout the 35-hour run remained below 4 NTU and, for the
most part, was less than 2,5 NTU. As observed in Figure 52, breakthrough
did not occur. A filter service time of this duration exceeds normal
practice indicating depth filtration of SSC wastewater is an economically
viable unit process. The column headless during the SSC processing remained
below 0.55 bar (8 psig) and only minor surface straining was observed.
The filtrate turbidity vs. time plot for processing of the SSC waste-
water through the 0.23 m diameter column is shown in Figure B5 of Appendix
B. An average filtrate turbidity of 3 NTU was achieved. Column headless
was not measured during this run.
Backflush Requirements--The standard backflush operating conditions
[backwash flow rate = 646-881 m3/m2-day (11-15 gpm/ft*); duration = 6-8
minutes] were used to regenerate the columns. Neither column was fully
loaded before backflushing. No difficulties were encountered in removing
those solids which had accumulated.
Contaminant RemovalFeed, filtrate, and backwash-water analyses for
dual-media depth filtration of the secondary solution crumb wastewater are
presented in Table 44. As expected, the only parameters to show significant
removals were suspended solids and turbidity. For the two depth filter runs,
the filtrate averaged 6 mg/£ suspended solids with a turbidity of 2.6 NTU.
A filtrate of this quality would be an acceptable feed to a carbon column
operating in an upflow mode.
Carbon Adsorption--
The adsorption isotherm at 20°C for the SSC dual-media filtrate is
presented in Figure 53. The slope of this isotherm (1.06) is moderate and
indicates good adsorption of organics throughout a range of 20-80 mg/e, TOC.
The TOC breakthrough curves for carbon-column treatment of the SSC
depth filter effluent are shown in Figure 54. The test was terminated
before complete breakthrough occurred in the fourth column because of a
pump failure. Sufficient data were obtained, however, to evaluate the
economics of meeting the BATEA COD guideline. The curve for column 4
indicates a TOC of 40 mgA (i.e., COD of -120 rng/fc) at 85 liters processed.
Processing beyond this point would allow the effluent COD to exceed the
BATEA standard of 130 mg/a.
The required carbon dosage would be 0.61 kg/m3 (5093 Ibs/MM gal). This
represents a carbon replacement cost of $0.53/ms ($2.00/1000 gal) processed.
This replacement cost exceeds the entire operating and maintenance projection
for depth filtration and carbon treatment of the SSC wastewater presented in
the Development Document. The projected treatment costs (corrected to March,
1976, dollars) were $0.23/m3 ($0.87/1000 gal) for the addition of sand filtra-
tion and carbon adsorption to the existing secondary treatment facility.
Overall costs for SSC wastewater processing by primary and secondary
treatment and dual-media filtration and activated carbon are presented
in Section 7.
139
-------
z
^^
co
oc.
et
C£
O UJ
Z^ H
<£. ^
Z UJ
O 1
t-H 00
H~ ^C
*^ -^
1 OQ
_1 2!
H 1 ~~*
u_ o:
0
a. o
UJ I i
Q 1
ef 3
l-H O
Q GO
LU
«ST f- >
^ > i
LU
i I --i
__ i r^
=) LU
Q a:
o >-
| i fy"
=£.
{/} C~*\
uj -z.
oo o
>- 0
1 1 1 1
<: oo
^ u_
o
l_ ._
^? ^^
Z. HH
*" 1 J
^>" f\
rf" Qi
?o
0 Q
0 <
"
^"
«=j-
( | 1
1
CD
*-
O 10 a)
(U 4-> OJ
E -r- 4->
Ti C tO fO
-sos-
C3 j O- 4-*
E O O !-
LT) CJ CJ U-
O
0
O)
+-> CU
E - +->
<- E 0 S-
CU 3 Q.4->
E r
E O O T-
OO «_3C_J U-
CNJ
t-
O)
O-cC LL.
^j
QJ
.,_
XI '-'§
1 ' 1 1 1 1 1 . f». 1 °* to ' 1 1 Isl-CTlO
11-^ 1 <"> «3 I 1 1 1 r LO
1 1 CM 1 1 1 1 CM
^
^^
01 r
^-«* o^ ^^^
E r*~
I/I ^ -
"O ~O Ol ^^
0} » i- v. "o to *~* co z; >>
ooioo ^^^^-,^^"^?^T.'J^"~' tJ'E'
C1~-- XlCZJr r C^.^.C>, >O
r- 01 . ~-» ' IOOlOl^)-l-> -
T3 C C-S->S.^.« "^S-^ 3^
'la.IIcL m<4-cx)O-Ci XJE
-IJOCOr CJQQS-OtO' t- Cp.
OCO 3-r-OOC3 3 S- ^
to
x>
X
r^
CO
S-
o
y-
C_J
o
si-
1 *
ta
-a
ai
i-
o
CO
OJ
Q.
E
to
oo
to
i
en
'to
c
to
c
i.
o
t.
i-
0)
XI
-------
Q
o
C\J
o o o o
O O O CJ
< O
o
o
o
en
o
co
s_
O)
O
s-
QJ
O
to
to
-a
c
o
o
QJ
c/1
0)
03
OJ
S-
I 4J
O r
O -r-
4-
C
O (O
J2 -I-
S- T3
tO O)
O E
S- r
O 03
4- 3
a
IO
O) -Q
3 i-
O O
CD O
3 !-
O -!->
S- 3
.C i
4-> O
-iii CO
ro
oj ~a
i- Q)
<_3 O)
o s-
O
oo
o
ID
O
LD
(3/6111)
301
141
-------
Contaminant analyses for carbon treatment of the SSC wastewater are
given in Table 44. The composite carbon effluent satisfies the BATEA
suspended solids, COD and BOD5 requirements. The oil and grease level of
9 mg/2, is in error since the feed (as received) and the dual-media filtrate
had oil and grease levels of 7 mg/£ and <4 mg/&, respectively. Thus, the
SSC depth filter carbon effluent meets the BATEA standards.
142
-------
SECTION 7
PREFERRED TREATMENT OPTIONS AND ASSOCIATED COSTS
INTRODUCTION
The preferred treatment options presented in this section are based on
limited experimental data. With the exception of tests conducted with in-
process latex manufacturing wastes, tests were conducted with only one
sample of each waste type obtained from one manufacturing site. The
breadth of applicability of these recommendations in the synthetic rubber
industry is therefore uncertain at present and should be verified by on-
site demonstration programs.
The economic analyses for the use of dual -media depth filtration and
carbon adsorption were obtained from the Effluent Guidelines Development
Document. The costs given in the Development Document were based on August,
1971, dollars. These costs were adjusted to March ,1976, dollars by using
the ratio of the Chemical Engineering plant cost indices for 1971 and
March, 1976, as follows:
August, ,97,, dollars x ' ' """ 1976' d°lla"
The capital and operating costs for ul trafiltration and reverse osmosis are
also based on March, 1976, dollars. These costs are based on system estima-
tes provided by Abcor, Inc.
Land costs were excluded from all total capital cost estimates, since
they are dependent upon plant location (4). Smaller land area requirements
are projected, however, when membrane separation systems are employed to
reduce the loading on (or to replace) primary and secondary treatment
operations. This may be a major consideration for plants operating in urban
areas which have limited expansion land available.
System depreciation was not computed into the annual operating costs
for any of the treatment options presented in this report, A short-term
depreciation period of 5 years (straight line) is currently acceptable under
Internal Revenue Service Regulations pertaining to industrial pollution
control equipment (4),
143
-------
LATEX MANUFACTURING WASTEWATER
Preferred Treatment
Two modifications to the present waste stream flow pattern are proposed
for latex producing plants. These changes (dashed lines), along with the
present wastewater flow pattern (solid lines), are shown schematically in
Figure 55. The concentration of the LWW stream for recovery of latex
involves two steps: reuse of rinse waters to build up their latex concen-
tration to a 0.5% solids level and ultrafiltration of the 0.5% latex
stream to a 15% solids (30X) concentration. The benefits of this treatment
scheme include:
- Reduction of overall waste stream flow from 13.3 m3/
metric ton (1600 gal/1000 Ib) product to 5.4 m3/metric
ton (650 gal/1000 Ib) product.
- Reduction of the raw waste stream COD loading from
37 kg/metric ton (lb/1000 Ib) product to 3.1 kg/metric
ton (lb/1000 Ib) product.
- Recovery of 0.09 m3 (24 gal) of 15% latex solids/metric
ton product @ $0.11 per kg of latex (dry weight basis)
resulting in a credit of $7.52/metric ton ($3.42/1000
Ib) product.
It is of interest to compare the anticipated COD concentration of the
remaining end-of-pipe waste for the scheme of Figure 55 to the BPCTCA and
BATEA guideline concentrations:
Predicted Raw Waste
Loading Following
Present Raw Ultrafiltration of BPCTCA BATEA
Assay Haste Loading LWW Stream Guidelines Guidelines
COD (kg/
metric ton) 30-40 2-4 6.85 1.78
The significant reduction in COD loading (-90%) means that the BPCTCA
effluent guideline for COD could be met without any further treatment of
the total wastewater flow. Effluent suspended solids, BOD5, and oil and
grease levels after incorporation of in-process ultrafiltration are difficult
to predict due to the limited data available. It is clear, however, that
virtually no suspended solids will be present in the ultrafiltrate discharge
and that sizable BOD5 reductions will occur. It is predicted that the BATEA
COD guideline would be easily achieved by primary and secondary treatment of
the ultrafiltrate.
The data obtained from the combined ultrafiltration/reverse osmosis
treatment experiment (see page 64) indicate that closed loop operation on
tank and tank car washdown wastewaters is feasible. Therefore, reverse
osmosis treatment of the ultrafiltrate should be considered in lieu of
conventional treatment processes if the plant is located in an urban area
144
-------
£ 5
o
3
o
o
t-
CL
C
o
4->
-^
CO
E
CO
CO
r-»
|
1
1
1
i
i
i--^7--]
UJ Cf. \ ~~^^
' ss1
^ ^ ' 4^.
L j ~^
S s
» i
z: co
r -3
CO
^ CD
"*
2=
<: o
UJ O
C*L m
H- CO
t « a:
_l UJ
S §
O U_ CO
*-« O Ce:
h- UJ
*3 ID 0
o o o
to z:
o
c
!a
o
-j
E
o
OJ
u
&
QJ
ce:
-<4
x
QJ
«
r*v in ro
CO ^ O r Is-*
c^ * J^ *
CO r- CO-1 0-)
\
C\l
c\j ro o *cr
»« ro o ro
o » o
f^.
l>
^
CM O
CM CM O UD
*e . o
Ol »^ LO O
1^
^
co o rx
i r O *£>
Vt O
< O CD O f^
^
*+j"
U
3
k
_. o Q-
^ 3 C
o -a c c
U. £ r- 4-> X
a. *> "^ t
^~ Ol O» C
DC E J^ _
*-> o
o (-> a
i -~- Q a -4-
f*l O O V
« E 00
P i g
fV i t-^ n
I *-> i cr.
fO\ ' csl i
S» |
"1 , 1 a?
« \ J 2
h *-* 1 -^
r- 'c \ | ^
1 . Nj to
| UD .
OJ | +J Q
^^ I c ^ g
' x O-D
i QJ -M O cr>
1 4J "^. S_ £
1 r*^ E O
. O
| fc* CJl in r-
' u-> r-- - N
I * ' Csl
1 O CM r-
\
1
1
A
t
!
1
1
> ^
"t^"
^*
t-n
' -S
:
-T^
1
' ^
fi en
c c
O *»
M C -o
^x. i O r m
CO 4~> -M 4-J O
I E O --~ i
O P
<£ Q. ^ _J
r- 0 **
o **.
>1
> w -o
' E <*- *o
O »-» tn
O *O TJ *«
Ci E ^- ***
r- O i OJ
*/) 4-> +-> O O
QJ C QJ O C
u QJ LU ce: to
S- E Q) en ,
O o rtJ c c: -- CQ
co . E to to E
QJ (- Q. CX 13 t/>
«3 > »-> E E 10 10
!-> Q> VI O O 10 (D
7 Q i C\J CO *3" LO UD~
7
c
o
£
4-) 4->
-*J U O
can
7 o> -o "a
g e e
o a. D.
o
§ §
OJ -t-> +->
M -x. -x.
_j t ^e 2*
1
i/i 3: r^, ro
t/> o co a
QJ p O
O Lu LO CJ ro
i
c
_,
M
'5
OJ
+J
(O
X
O)
ro
S-
O
C
s-
OJ
03
O.
§
O)
-P
to
0)
-l->
in
to
-a
0}
c/>
o
Q.
o
a.
(O
O
-------
where expansion land is limited or if a new plant is being designed.
The proposed BATEA unit process combination, dual-media depth filtra-
tion and carbon adsorption, was not studied for the treatment of raw latex
wastewater during this program. Its effectiveness in producing a BATEA
quality effluent is therefore unknown at present. The instability of the
RLX wastewaters during depth filtration pretreatment experiments indicates,
however, that considerable difficulty would be encountered in filtering
the RLX effluent for subsequent carbon treatment.
Economic Summary
Comparative economic analyses for two latex wastewater treatment
schemes are presented in Table 45. Option 1 entails the continuation of
current primary and secondary treatment operations, followed by dual-media
depth filtration and activated carbon adsorption. This treatment option
was projected in the Development Document as being capable of producing an
effluent of BATEA quality. However, the effectiveness of depth filtration
and activated carbon in treating secondary treated latex wastewaters has not
been demonstrated. The second option for latex wastewater treatment begins
with ultrafiltration of the wastewaters used to wash down reactors, tanks,
and tank cars. This stream contributes 70% to 90% of the total wastewater
flow from a latex manufacturing plant. The remaining 10% to 30% of the
total wastewater flow is passed through primary treatment and the entire
pretreated wastewater (ultrafiltrate and primary treated effluent) receives
secondary treatment. Ultrafiltration of latex washdown wastewaters has
been successfully demonstrated with" a number of different latices. The
costing for both Options 1 and 2 are based on a typical latex plant waste-
water flow rate of 382 m3/day (101,000 gal/day) (4).
A breakdown of the capital and annual operating and maintenance (O&M)
costs for primary and secondary treatment (BPCTCA) and dual-media depth
filtration and carbon treatment (BATEA) costs for Option 1 are detailed in
the Development Document. The costs for primary and secondary treatment in
Option 2 were extrapolated from these data and the treatment cost curves
developed in the "Economic Analysis of Proposed Effluent Guidelines, The
Rubber Processing Industry." (3) A breakdown of the costing for the ultra-
filtration section of Option 2 is presented in Table 46.
The capital investment for Option 1 is twice the investment required
for Option 2, In terms of cost per m3 of total influent water per day
(382 m3), Option 1 is $2,925 compared to a cost of $1,504 for Option 2. The
capital cost for the ultrafiltration system (see Table 46) includes stain-
less steel piping which will allow the system to be cleaned with a variety
of solvents, if necessary.
The annual O&M costs are $1.23/m3 ($4.64/1000 gal) for Options 1 and
$0.68/m3 ($2.28/1000 gal) for Option 2. Also, with Option 2 a credit of
$0.57/m3 ($2.14/1000 gal) is realized from latex recovery and reuse, giving
a net O&M cost for Option 2 of $0.11/m3 ($0.14/1000 gal) of wastewater
influent. Thus, the daily savings for use of Option 2 rather than Option 1
146
-------
oo
a:
-
I
3 -I-
c: T3
C O)
<: s-
0
I
1
I
CD
C CU
i- O
-P c:
ra ro
S- co<*
cu a>
O.-P >
O C to
I- -p
i ro to
(0 S 0
3 0
C TJ
C C
«C rO
O
O
o
A
LO
CO
LO
«/*
i
(O **
P to
r- ^
Q- CO
ro O
O C_>
CM
r--
r--
o
CD
1
1
1 1 1
1 1 1
1 1 1
co
CM
O
O
O
«\
LO
CO
O i
o
O 1
" 1
1 1
r-.
LO
r~-
«d-
CD
o
CM
0 0
O CM
CM CD 1
« f\ |
t-^ oo i
l_
r
t\
"
,_
en
00 I
" 1
00 1
!-»
O O
CO LO
r^. LO
t **
LO CO
*d-
o o
O CO
LO r-.
!-» VD
LO OO
, 1
>>
C
ro 1
^
1
1
,
CD
CO 1
" 1
CO 1
r-.
o
o
rt
CD
co
o
co
CM 1
" 1
si" 1
CD
r-»
CD
"-^
o
CO
CM
r^. ^1-
r-- o
^3- LO
«* 1
r^.
LO
r^
LO
0
00
<-D
O
1
i
^1 "^-"l -^ CD-P
,_
+J
C
CU
E
-p
ra
P O)
c s-
CU h-
E
Q.T3
O CU
i O
CU C
> ra
cu >
Q -0
<
o
C T3
ro CU
P
P 0
C CU
CU -f-3
E 0
p i-
ro O-
CU
i- -
1 to
-p
-P C
E CU
0) E
i- 3
S- 0
3 O
0 Q
ro
P
O
P
4-
O
P
E
CU
E
-p
ro
CU
i_
-P
>,
ro «3"1
-a-
E
O -P
cu a3
10 3
-D 4^
C 4-
ro CU
>>-P
S- E
ro ro
E i
i- Q.
S-
Q.
*
HH
C
o
f"
-p
a.
o
S "
C ^~~^
o E
.a o *f*
S- !-
ro -P «
o in >,
CU ro
o -p 3 -a
cu E cr
P CU S-
ro 3 C CU
>!- Q.
r- 4-
-p If- +J +J
0 CU E C
ro CU CU
P E 3
o c -p i
C ro ro 4-
rO i CU C -faO-
£X S- -
C -P
Oi -P
i- rO 4- ro C
-P -P O -P O) -P
ro O O 3 C
S_ +J ^ +J r- O)
-P to 4- E
r- If- CU 4- C -P
r- O c O -i- rO
4- CU CU
P > 00 CO S-
ro C -i- E E I
r- CU -P
-O E O S- S- T3
a> -P cu cu cu cu
E ro 4- a. a. to
1 CU 4- O
i S- CU -P -P O.
ro -P' to to O
3 O O S-
Q O O Q-
1 1
1 1
c
o
*^~
p
8-
*~^-
p
c
in cu
S- 3
CU i
-P 4-
ro 4-
3 CU
C i
S ro
0 -P
"O O
J= -P
in
ro 4-
S 0
X cfS
CU O
P CM
rO
i 4-
O
4-
O -P
C
c cu
o E
i- -P
-P rO
(O CU
S- S-
-p -p
£- >*>
4- S-
ro ro
S- E
-P T-
i i-
=> a.
5- C
Q.T-
T3
O rO
E O
rO i
CU T3
P C
ra ra
P 5
i O
r- i
4- 4-
ro
s- -o
-P CU
r O
3 3
-a
4- CU
O S-
p
C -P
CU E
E a>
-P 3
ro i
CU 4-
S- 4-
P CU
>*>~o
S- CU
ro -P
"O ro
C CU
O S-
O -P
CU
OO
c
0)
E
.p
ra
cu
S- -faO-
-p
n
>5 >>
C r?
ro T3
T3
C S-
o cu
O Q.
CU
in -p
s-^. S
O ^1- 3
t| !
4-
-P CU E
(/) O -r-
O C
O CU i
S- ro
4- CU -P
O 4- O
CU -P
oo a:
~~, co
i c: E
CU S-
E c cu
3 cu a.
m >
to T- -P
ro CD IO
o
0
t«-
A
p
c
cu
3
C
C
oo
E
S-
cu
a.
-p
10
o
o
S-
cu
>
o
-p
c
cu
B
(O
CU
s_
-p
00
T3 CM
cu ^J-
to-te-
O
Q. "
O -P
i_ c
a. cu
4- -P
O ra
cu
to 5-
CD-P
c
r- 1 l
>
ra c
to o
>r-
>>-p
^ 0-
- o
ro
Q
>>
rO
o
co
£
CM
CO
CO
1
CU
-p
ro
^
O
r
4-
-P
ra
i
a.
,
ro
O
a.
>,
P
147
-------
TABLE 46. BREAKDOWN OF CAPITAL COSTS AND ANNUAL OPERATING COSTS
FOR ULTRAFILTRATION SYSTEM TREATING LATEX WASTEWATERS
(TABLE 45, OPTION II)
A. CAPITAL COST
2
87,500 Estimated UF System Cost; 47 m membrane area,
304 S.S. piping based on projected design
flux of 1.64 rrH/m2-day
43,750 Field Installation, Estimated at 50% of UF
System Cost
131,250
26,250 26,250 Auxiliary Tanks, Pumps, Piping - estimated at
== 20% of UF installation cost
157,500 Total installed cost
B. ANNUAL OPERATING COST
7,850 Pumping power, 22.5 kWh (30 hp), 365 days, 24 hrs/day @
$0.04/kWh
15,330 0 + M Labor, 4 hrs/day over 3 shifts @ $6/hr + 75% Fringe
and Overhead
6,390 Supervisory Labor, 1 hr/day over 3 shifts @ $10/hr + 75%
Fringe and Overhead
3,500 Maintenance Materials - estimated @ 4% UF System Cost
4,420 Cleaning Chemicals - 2 detergent cleanings per week
2,890 Taxes and Insurance - assumed @ 2% of Total Installed Cost
5,400 Membrane Replacement - 2 yr life
45,780
148
-------
are $428 ($156,000/year).
Based on the experimental results of this program and the above
economic analysis, concentration and recovery by ultrafiltration of within-
process latex washdown wastewaters is recommended. Conventional primary
and secondary treatment is recommended for the remaining end-of-pipe latex
wastewater, and secondary treatment is recommended for the ultrafiltrate.
Advanced treatment of the secondary effluent is not believed to be necessary
to meet BATEA guidelines and is not recommended. Reverse osmosis treatment
of the ultrafiltrate for closed-loop recycle of washdown wastewaters is
technically feasible and may be economically attractive in certain special
cases.
EMULSION CRUMB MANUFACTURING WASTEWATER
Preferred Treatment
Continued use of primary and secondary treatment for processing of end-
of-pipe emulsion crumb wastewaters is recommended. None of the alternative
treatment options investigated appear suitable for processing the REC waste-
water because of its high concentrations of dissolved solids and TOC. In-
process ultrafiltration of reactor washdown wastewaters, to lower the COD
loading on the secondary treatment operation, is not feasible because of
the infrequent flow and non-uniform nature of the washdown wastewaters.
These washdown wastewaters differ from those generated in latex plants and
are not amenable to concentration by ultrafiltration.
The use of dual-media depth filtration and carbon adsorption for the
treatment of emulsion crumb secondary effluent appears to be the most
practical approach to meeting BATEA standards. The effluent from the
secondary treatment system presently meets or approaches the BPCTCA stan-
dards. Dual-media depth filtration of the SEC wastewater to lower the
suspended solids loading followed by activated carbon treatment for organics
reduction would produce an effluent of BATEA quality. In emulsion crumb
plants, the backwash waters from the depth filters and carbon columns
would be returned to the secondary treatment operation while the spent
carbon would be regenerated on-site (4).
Economic Summary
An economic analysis of the recommended treatment for emulsion crumb
wastewaters is presented in Table 47. This recommendation is identical to
the Development Document's projected BATEA treatment of primary and
secondary treatment followed by DMDF and carbon treatment. A detailed cost
analysis for these unit processes is found in Reference (4). The data
presented in Table 47 are based on a typical emulsion crumb rubber plant
wastewater flow rate of 5,614 m3/day (1,483,000 gal/day)(4).
The capital expenditure for the entire treatment system is $760/m3
149
-------
oi
o
u_
oo co
z o;
0- CO
1 2:
^"^ T^
3? ~
^5. CJ
o 2:
o co
Z J
c^ ^~^
o s:
UJ LU
*
r^**
^ |^
"
LU
1
CO
L
p-_
O *£t
ro t/1
+^
C 0
i- CJ
ro Ol
V- U
a> c
a. ro
0 C
Ol
ro C
c 10
5s"
r
fO
r- U1
OL tn
ro O
O O
-(-)
c
ai
jj^j 4_>
E ro
O> O
E S-
CLI
o
r -a
O> Oi
> 0
Ol C
Q ro
a -a
c O)
C -4->
Ol U
E 01
+-> -r-3
ra O
Ol C
S- 01
i. E
3 13
O U
O
a
o in
in r~»
CM m
^ *N
IO CTi
in *j-
ro CM
o in
in r~-
00 r
CM CM
in r
cc *
CM i
^
ra
'o. c
0
la i-
I * fO*^^
O ^^*
T3
<»- Ol !->
O +J C
ra at
+J > 3
C -I- i
Ol 4J *4
E 04-
-i-> ra ai
ro
Ol "O 4-*
s- c c
4-> ra ra
^^
>> C CL
s- o
ro -r- r-
a +J ra
C ro -(->
0 1-0
O +J -M
01
*! <^ o
-a
C ro +->
ra -^ -r- c
c -a a>
t- 13 E *-*
ra i l ro
EM- r- 0)
r- 4- ro 1-
S- Ol .3 +->
CL a
in
CM
IO 1
A 1
in i
0
VO
in o
CM to
^3 f*1^
to
to
C\J
»
*^
v^
*
>>
ro
o
S.
#1
-M
C
ai
£
ic
r
ro
E
i.
Ol
Q.
| '
t/>
O
>^
ro
-o
"X,^
ro
E
10
V
in
at
ra
S-
3
O
4_>
C
ra
r
Q.
^^
ro
U
a.
t
150
-------
($2.88/gal) of total influent per day. 67% of this amount has already been
invested at most sites for primary and secondary treatment facilities. The
net investment required to reach BATEA standards is therefore estimated at
1.5 million dollars for the "typical" plant. The major portion of the
operating and maintenance costs are also associated with current treatment
processes. Incorporation of dual-media depth filtration and activated
carbon treatment is expected to raise the O&M costs by $0.12/m3 ($0.46/1000
gal) to $0.30/m3 ($1.12/1000 gal). These cost figures were not verified
during this program since carbon column breakthrough was not obtained.
Further testing is essential to develop an accurate economic profile of
DMDF/ACA treatment of emulsion crumb wastewaters.
SOLUTION CRUMB MANUFACTURING WASTEWATER
Preferred Treatment
For solution polymerization, two alternative wastewater treatment
schemes are recommended. First, for existing facilities which currently
have primary and secondary treatment, the addition of dual-media depth
filtration and carbon adsorption is preferred. Second, for existing
facilities which do not have secondary treatment, and for new sources,
treatment of the raw wastewater by ultrafiltration and reverse osmosis is
recommended. Both methods of treating the solution crumb wastewater are
effective and result in a high quality effluent as shown in Table 48.
TABLE 48. COMPARISON OF REVERSE OSMOSIS AND CARBON ADSORPTION PRODUCT WATERS
WITH BATEA GUIDELINES FOR SOLUTION CRUMB MANUFACTURING WASTES
Reverse OsmosisCarbon ColumnBATEA
Assay Composite Permeate Composite Effluent Guidelines
Suspended Solids (mgA) nil <5 10
Oil and Grease (mgA) 7* 95
COD (mgA) 36 72 130
BOD5 (mgA) 4 45
Dissolved Solids
(mgA) 141 820
Color (units) 5 20
Error suspected in analysis. Actual oil and grease level <4 mgA.
The reverse osmosis product water is superior to the carbon effluent in
all respects and, because of its low dissolved solids and color, is expected
to be reusable within the plant for boiler and cooling tower makeup. It may
also be suitable for the crumb slurrying operation.
The use of ultrafiltration and reverse osmosis treatment is not
151
-------
recommended for those existing facilities with secondary treatment pres-
ently in operation since addition of depth filtration and carbon adsorption
is a more cost-effective alternative. If either upgrading or expansion of
the secondary treatment process at an existing plant is contemplated, then
the alternative of ultrafiltration/reverse osmosis should be considered.
Economic Summary
As stated above, two viable options are available for treatment of
solution crumb wastewaters to meet BATEA standards. Option 1 is primary
and secondary treatment of the raw wastewater followed by dual-media depth
filtration and carbon treatment of the secondary effluent. A detailed
economic analysis of this option is given in the Development Document;
however, the operating costs given there for carbon replacement have been
shown during this program to be quite low. These costs have been
appropriately corrected in the ensuing presentation.
Option 2 consists of ultrafiltration of the raw wastewater followed
by reverse osmosis of the ultrafiltrate. This option is divided into two
segments to facilitate the presentation of the economics. Option 2A
economics are based on the use of tubular ultrafiltration modules, while
Option 2B economics were developed for more compact spiral-wound ultra-
filtration modules.
All experimental work was performed with membranes in the tubular
geometry; however, spiral-wound cartridges are potentially applicable for
treatment of the RSC wastewater. The overall economic analysis for Options
1, 2A, and 2B is presented in Table 49. Breakdown of the capital and O&M
costs for Option 2 are given in Table 50 for tubular ultrafiltration, in
Table 51 for spiral-wound ultrafiltration, and in Table 52 for reverse
osmosis. The economics of all options are based on a typical solution-
crumb-plant daily flow rate of 1336 m3/day (353,000 gal/day)(4).
The capital cost for spiral-wound ultrafiltration/reverse osmosis
treatment of the raw wastewater is $647/m3 ($2.45/gal) of the daily total
influent. This is about 30% lower than the capital cost of tubular ultra-
filtration/reverse osmosis treatment and -50% lower than the capital cost
for Option 1. The annual O&M costs are also lowest for the spiral-wound
ultrafiltration/reverse osmosis treatment combination. The O&M costs for
the various options are $0.91/m3 ($3.46/1000 gal), $0.84/m3 ($3.18/1000 gal),
and $0.62/m3 ($2.35/1000 gal) for Option 1, Option 2A, and Option 2B,
respectively.
Two factors have been omitted from the Option 2 annual operating
costs. These factors are ultrafiltration and reverse osmosis concentrate
disposal costs and credits for reuse of reverse osmosis permeate. The
impact of these two factors on the overall Option 2 operating costs
requires further study of alternative concentrate disposal options (other
than hauling) which is beyond the scope of this report.
152
-------
o
1
1
"~^
1
oo
OL
O
1 1
oo oo
z. cc
o ec
' *
r- i
CL O
O Q
1 «
LU !-»
:> O>
1 i
T""^
LU
a: n:
i < >
LL. cf
O S
00
i Qi
OO LU
>- 1
oo
r- 4J
CL 00
fO O
0 0
O CM
O *J-
to i
CM CO
CD ^J1
i CO
o o
in CD
CM CD
* >
1- O
in co
i in
i~
o
c -a
cu cu
4-> E oo
C 4-> to
CU IO J3
3 CU
. s- -a
M- 4-» CU
CU C M-
i~~ -O ~O
C
cu
Jl
(O
4-> CU
C S-
CU 1
E
o cu
i (J
cu c
cu >
a
c -a
ra CU
4-> U
C CU
CU -0
E 0
4-> S-
n3 Q-
CU
S- -
h- oo
4-> C
C CU
CU E
S- 3
S- 0
3 O
0 Q
ra s- o
M 10 E
o o
4-> 00
X) !->
t- CU 00
O 4-> O
(O U
C *r- *J-
J4_>
(J CU
IO O
cu -a cu
s- c s-
+-> -
>> c cu
s- o or
"O -| ' ^~^
C fO 4-* tO
O S- C 4->
o 4-> cu 4- 4- ,
-o >t- to
C to CU 4->
tO !- C
a t cu
>, cu *o S
C E 4-"-?
fO 1 O &-
E ' 4-> cu
r- to CL
S- 3 If- X
a. a o cu
.
c
O
*f~
CL
0
en
CM O C
<* w
M |
in i o.
"S- ot
"3- Cv
00 C
in u3 c
CO CM 1 *3
3-^-1 C
00 CV
to tr
r
j-
ta
s.
E;
0)
e
S-
ta
3
3
1
S-
CU
oo to
cu s
U 'U
V* i- 4->
3 00
" O tO
>. i-
i- -i- to 4->
CU i
CO CO S- -r- 0-
E E I- 4- a
s- t- -a s- *=
cu cu cu 4-> T:
CL CL 00 I C
O =D E
4-> 4-> a.
OO 00 O
O O S-
<_>o a. <<
I-H
C
o
'l~
CL
O
ej- CM
CO tO
>oo o ooo o
5 to CM 1^ VO CO
} r^ i i to r-~ «3- i
f e. « | «>««.|
j r^ o i to i «3- i
3 CM . 1^ CM CD
J i «J- I i CO
>ooco oo|or~-
3OOO1 OOO^-
i-tncjicoi p^incsjto i
> r^ r^ i oo r*-. 10 i
J I en co r*>. 10
3 LO r CM LO CO
r
-a
c
3
O
3
r
(O
S-
a.
00
S- to
CU CU CU X)
tO fO to (O
s- s s- - e
4J >, CU 4J >,
r- fO 4J i tO IO
r- -a 00 ! "O CO
o>- to t- co
tos- y m i. w
S- CU 4><* L. CU 1
4-> CL S 4-> CL
3 4J4JJ- 3 4->«/*tO
C C C t-
<4- CUCULj- l»_ CU«3
O 33O O 34->O
t i 00 ^- C i
00 0>_4_ CCUOO (+-CUM-
r- C C O r !- C 3
W !- -r- 'r- 3 00 -i I 4->
O 4-> X3 O M C
E COCO tOOE CO CrcJ
oo EE 5-Eoo E-r-r-
O 4-> O O.
!.. S- t CU S- t-
CU CUCU'r-CCU CUCUi
OO CLCLOj-fOOO CLCLfO
i- to s- s- o
CU 4->4-> S QCU 4->4->-r-
> 00004->E> OOOOCL
CU O O r CU CU OO>>
o: CJEQ: cjcji
CO
X)
cu
x>
3
U
c
r-
O
C
cu
s_
cu
s.
o
cu
00
cu
s-
cu
tO
cu
CL
00
OO
O
E
00
o
cu
oo
i-
cu
cu
XI
c
tO
oo
oo
o
o
^
to
OO
o
CL
OO
XI
cu
4-1
IO
4->
cu
(J
o
o
cu
o
5p
153
-------
TABLE 50. BREAKDOWN OF CAPITAL COSTS AND ANNUAL OPERATING COSTS FOR
TUBULAR ULTRAFILTRATION SYSTEM TREATING SOLUTION CRUMB
WASTEWATERS (TABLE 49. OPTION IIA)
A. CAPITAL COST
o
376,000 Estimated UF System Cost 820 m membrane area,
carbon steel and PVC piping, based on projected
design flux of 1.64 m-Vm
188,000 Field Installation, 50% of UF System Cost
564,000
56,400 56.400 Auxiliary Tanks, Pumps, Piping - 10% of UF
installation cost
620,400 Total installed cost
B. ANNUAL OPERATING COST
117,730 Pumping power 336 kWh (450 hp), 365 days, 24 hrs/day @ $0.04/
kWh
15,330 0 + M Labor, 4 hrs/day over 3 shifts @ $6/hr + 75% Fringe
6,390 Supervisory Labor, 1 hr/day over 3 shifts @ $10/hr + 75%
Fringe
22,560 Maintenance Materials - estimated at 4% UF System Cost
17,680 Cleaning Chemicals - 2 detergent cleanings per week
12,400 Taxes and Insurance - assumed @ 2% of Total Installed Cost
90,270 Membrane Replacement - 2 yr life
282,360
154
-------
TABLE 51. BREAKDOWN OF CAPITAL COSTS AND ANNUAL OPERATING COSTS FOR
SPIRAL-WOUND ULTRAFILTRATION SYSTEM TREATING SOLUTION CRUMB
WASTEWATERS (TABLE 49. OPTION IIB)
A. CAPITAL COST
175,000 UF System Cost m membrane area, carbon steel
and PVC piping, based on projected design flux
of 1.64 m3/m2-day
87.000 Field Installation, 50% of UF System Cost
262,500
26.200 26.200 Auxiliary Tanks, Pumps, Piping-10% of UF
installation cost
288,700 Total installed cost
B. ANNUAL OPERATING COST
78,490 Pumping power 224 kWh (300 hp), 365 days, 24 hrs/day @ $0.04 kWh
15,330 0 + M Labor, 4 hrs/day over 3 shifts @ $6/hr + 75% Fringe
6,390 Supervisory Labor, 1 hr/day over 3 shifts @ $10/hr +75% Fringe
10,500 Maintenance Materials - estimated at 4% UF System Cost
17,680 Cleaning Chemicals - 2 detergent cleanings per week
15,780 Taxes and Insurance - assumed @ 2% of Total Installed Cost
32,500 Membrane Replacement - 2 yr life
176,670
155
-------
TABLE 52. BREAKDOWN OF CAPITAL COSTS AND ANNUAL OPERATING COSTS FOR
REVERSE OSMOSIS SYSTEM TREATING SOLUTION CRUMB WASTEWATERS
(TABLE 49. OPTION II)
A. CAPITAL COST
350,000 RO System Cost; 35 0.2m diameter permeates,
carbon steel and PVC piping
175,000 Field Installation, 50% of RO System Cost
525,000
52,500 52,500 Auxiliary Tanks, Pumps, Piping - 10% of RO
-_. installation cost
577,500 Total installed cost
B. ANNUAL OPERATING COST
39,250 Pumping power 112 kWh (150 hp), 365 days, 24 hrs/day @
$0.04/kWh
15,330 0 + M Labor, 4 hrs/day over 3 shifts @ $6/hr + 75% Fringe
6,390 Supervisory Labor, 1 hr/day over 3 shifts @ $10/hr + 75%
Fringe
14,000 Maintenance Materials - estimated @ 4% RO System Cost
11,550 Taxes and Insurance - assumed @ 2% of Total Installed Cost
41,240 Membrane Replacement - 3 yr life
127,760
156
-------
At present, the most cost effective treatment plan would be dual-
media filtration/carbon adsorption at plants with existing secondary treat-
ment and spiral-wound ultrafiltration/reverse osmosis treatment at new
facilities. Both treatment schemes are capable of producing a final
effluent of BATEA quality,
157
-------
REFERENCES
1. Kent, James A. (ed), Riegel's Handbook of Industrial Chemistry, Seventh
Edition, Van Nostrand, New York, 1974.
2. EPA Request for Proposal No. CI-75-0149, March 1975.
3. Economic Analysis of Proposed Effluent Guidelines, The Rubber Processing
Industry, EPA Report No. EPA-230 1-73-024, September 1973.
4. Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Tire and Synthetic Segment of the Rubber
Processing Point Source Category, EPA Report No. EPA 440/1-74-013a,
February 1974.
5. Troppe, F. G. Secondary Treatment of Wastewater from Synthetic Rubber
Production, Rubber Chemistry and Technology, 47 (4), 932 (1974).
6. Hazen and Sawyer, Process Design Manual for Suspended Solids Removal,
EPA Technology Transfer Report No. EPA 625 l-75-003a, January, 1975.
7. Thomas, J. M., and W.J. Thomas. Introduction to the Principles of
Heterogeneous Catalysis, Academic Press, New York, p. 32 (1974).
8. Bryce, C. A., et al., "Final Report on MUST Wastewater Treatment System,"
for USAMRDC, Contract No. DADA 17-17-C-1090, 15 July 1973.
9. Blackley, D. C. High Polymer Latices, Their Science and Technology
Noyes Data Corporation, New Jersey, 1975.
10. Swindell-Dressier Company. Process Design Manual for Carbon Adsorption,
EPA 625-1-71-002a, revised October, 1973.
11. Sittig, Marshall, Pollution Control in the Plastics and Rubber Industry.
Noyes Data Corporation, New Jersey, 1975.
12. Abcor, Inc., Internal Report, June, 1976.
158
-------
APPENDIX A
o
u_
oo
^^
UJ
o:
oo
UJ
I
1 h-
1-1 2
<=c i-
OL 00
1 "a
I _i
ID
DQ
i^^\ ^r*
2Z ^1
-i
LiJ
O 00
o oo
0 LU
1
r^
0 LU
0 I
I- 0
^ * >
O
Z. LU
o Q:
CD
OS |
«^ e^
0 h-
O
01-
»-H
^y ^"
CD
re
$-
-t-5
*
£
re
S-
4-*
r
~*
s;
u.
"T~
o
u_
or
i^£
\i
~T~
o
0)
O)
*^^_
c
o
i- 0)
*-* E
Q.-^
r- 1
S-
O ul
V) Ul
Q) <1)
Q O
o
T3 S-
VI) /^
0)
U,
S-
o>
^^
I
CO
CO
r
O
O
CM
1^5
CM
CM
CO
p«^.
CM
I/)
S-
3
O
1^~
CM
*>
-o
CD
0)
U_
^
to
OL
r-
cr> vo co co
00 CO CM
CM i «y in
VO OO CM CO
vo in i vo
CM CM r-. LO
t/1
S- S-
3 3
0 0
-c ^:
S- i. r- ^J-
33 CM
o o
-C -C > f
^? o &s o
^a- CM i co i o
CM r-~ r- i
II II
» X X
X5 TD "O T3
S S S '£ S 'C
re re re f re 1
oi QC ct; oc
CM
in
r---
CO
CM
vo
^^<
LO
CO
vo
VO
r~
S S-
fQ h*
a:
159
-------
2^
0
I i
ci
i
_i
i i
1 1 l U_
OO «=C
cc c£
LU >- 1
> C£. _l
LU ci ^5
a: z
l i -Z.
2: 2: o
_-j oo
OO LU _J
z. o: ro
O Q- 2:
t i LU
I o:
ej; O LU
OL Li- D_
1 , ,
1^ 1 '
^ 00 D-
m^~ i
^_ i
0 et U_
z: LU o
O Di 1
<_> 1 Q
oo z:
x^ LU
O LU
o 1 in
l-< 1-
LU l l
2: 3
z o:
O LU OO
CO Q. t
a: oo
<; Q LU
o z: I
<
O LU
t i Q 1
z: LU <_>
eC LU >-
CS Lu CJ
a: LU
o oo a;
i i
_J 00 _J
«C O «t
l ^- i
1 ^i- r^
o oo o
1 0 1
CM
i
<
LU
_1
CO
cC
1
S-
cu
_1 \
^J
\
Oi
Vfc-*'
(J
g
1
O)
4J
rO
O)
^
O)
Q_
o
1
CO
en
i
CO
-------
co
co
0
O
*
a:
Q
CO
O LU
o
LU
CO I
co;
O Lu
CO L_
O J-
LU LU
COQ
LU CD
LU I
CXI C£
O
a:
o -zL
Lu O
co co
h- LU
O
LU O
"3 LU
LU 1-
o: <
LU
i o:
co >-
oc
LU<
o a
>- o
o: o
o LU
I CO
CO
. . I
n: o
co
cC
d "3
"S *?
h
,
i> -o
o
Si
i in 0
t?5
O
< "o o
I » S
U?
s; i°-
S o~cV
a.*t*
5 '9-^1
O 0 C O
* * r- M
O O C3 O
r? o « o
en en u>
88
o «
-------
o
CD
O
O
O
10 10 us 10 f
s_ s. s.
e -e -e .. m
in
o
I. U
10 IQ
01
u_ u- o a) Q.
ac z 3: «- E
t- c a>
a o < ° - -
CM
in
00
vo
CO
CM
ro
00
CM
o
CM
CO
C5
-M
(O
3
-M
I/)
X (O
3 2
£ -Q
gl
-r-
i Q-
r- I
M- M-
-a
"
0)
13
CD
162
-------
i r
o
o
o
IT)
0 «>«»*"
10
t-
0}
c
re
JD
I
3
-Q
3
a
CO
s: s: *-> .. PO
re I. *»
I I
U. U- O 0)
3; 3: >- r-
O < o »
O
c o
i
O I
C X
o
(J E
o
-a -u
O) -i-
OJ i-
s- ^z
O )->
E $-
i- O)
t/) 4J
CO
X (O
O) 3
E S-
T-
r Q.
I- I
<+- M-
lO O
S- I
--> -a
0)
D1
Kep- ui/ ui '
163
-------
I I I II I I I I I
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure A-3. Ozonation of end-of-pipe emulsion crumb ultrafiHrate
at 30°C and pH 5 without UV light.
164
-------
K
a
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
H
X.
D
"~
<
0
- 7
- 6
- 5
- 4
- 3
-
- 2
-
- 1
-
In
£
u
1 0
EH
-280
-260
-240
-220
-200<
-
-
>
-180*-
-160
-140
-120
-100
- 80
- 60
- 40
r 20
\
\
-
-
S. 1
-Q-
-B-
PH
-Q.
-Q -
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure A-4. Ozonation of end-of-pipe emulsion crumb ultrafiltrate
at 30°C and pH 9 without UV light.
165
-------
I I I 1 I I I I I I I
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure A-5. Ozonation of end-of-pipe emulsion crumb ultra-filtrate
at 30°C and pH 5 with UV light.
166
-------
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure A-6. Ozonation of end-of-pipe emulsion crumb ultrafiltrate
at 30°C and pH 9 with UV light.
167
-------
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure A-7. Ozonation of end-of-pipe emulsion crumb carbon effluent
at 30°C and pH 5 without UV light.
168
-------
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure A-8. Ozonation of end-of-pipe emulsion crumb carbon effluent
at 30°C and pH 9 without UV light.
169
-------
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure A-9. Ozonation of end-of-pipe emulsion crumb carbon effluent
at 30°C and pH 5 with UV light.
170
-------
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure A-10. Ozonation of end-of-pipe emulsion crumb carbon effluent
at 30°C and pH 9 with UV light.
171
-------
100
90 h-
80 -
70
60
50
40
30
20
10
Expected TOC for
100% Utilization
of Ozone
J_
234
Time (hours)
Figure A-ll.
Ozonation of end-of-pipe emulsion crumb carbon
effluent at 30°C and pH 5 without W light
(type B).
172
-------
80
70
60
p 50
cr.
o
o
40
30
20
10
Expected TOC for
1001 Utilization
of Ozone
3 4
Time (hours)
\
Figure A-12 Ozonation of end-of-pipe emulsion crumb carbon
effluent at 30°C and pH 9 without UV light
(type B).
173
-------
CO
CM
(ueq)
«*
r
o
o
o
Q
o
o
q
-o
I
CM
n 01
a>
3
(1) 4J r
+J ra
«j s- r*.
i. a)
-0 U- I Q.
to
Q) T3 "O T3
DC
I/I
0)
a.
to
(O S-
I S-
0)
(O -)->
3 (O
-O S
O)
S- -4->
O «/)
> C E
O 3
IO -I- I
to to o
o o
.
ro O) O)
QJ «->
-C -O O)
0) E
§-)-> fO
ro T-
3 , o »
4-> O T3
r- OJ O
"D to !-
t- S-
J3 i»_ O)
V. O Q.
3
+-> C CD
O C
-»-> o
Q.
O)
CD
(niN)
174
-------
(ueq) SSOIPB9H
in
n
oo
csi
r
O
l/)
s-
Ol
Q-
O) T3
o c
O
to o
i-
fO
2
(U
S-
O CO
q- «
4J 3
O
t/> C
> E E
O =5
t/1 ! I -
to
3 O> T3
'o 4-> E
o
>>co
-o s- c\j
C CO
(O T3 O
C
>5 O "
4-> U T5
r- O) O
"O CO !-
r- S-
J3 l|_ O)
S- O Q-
3
4-> E 01
O C
O> -r- !-
-i-> »-> to
(O (13
»-> O
i i O
i- -r- S-
U- <+- CL
«a-
I
(U
3
o
o
(fUN)
175
-------
O
< o
t>
>
n < O
o
CvJ
CO
CM i-
r 3
O) £
E»*-
.p OJ
(O
S- E
(O 3
O S-
O
O C
M- O
r
CO (/>
O) i
> 3
s- e
3
3 IO
oo;
O
O
C
O
O
O)
I
«=c
O)
3
C7>
O
UD
O
in
C3
CO
o
CM
(l/6uj)
176
-------
APPENDIX B
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
en
- 14
13
12
11
- 10
- 9
- 8
- 7
- 6
- 5
_ 4
- 3
- 2
- 1
K 70
-60
-50
- [3-
1 i i i r
i i
O
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure B-1. Ozonation of end-of-pipe solution crumb carbon effluent
at 30°C and pH 9 without UV light.
177
-------
c.
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
cr.
£
CO
o
- 14
- 13
- 12
- 11
- 10
- 9
- 7
- 6
r
3
- 4
- 3
- 2
- 1
*!
en
g
-70
M>
-60
-50
40
413
C
- 3C
- 2C
- 10
l i ] ! i i i i i i I i
_
-
-
- -
A
_
.
A ~
\/ * A:
i ' ^ o n rn n 1 1 - n
J (J LJ - LJ LJ 1 1
j TOC
-
1 1 1 1 1 1 1 1 1 1 1 I 1
0 10 20 30 40 50 60 70 80 90 100 110 120
Reaction Time (min)
Figure B-2. Ozonation of end-of-pipe solution crumb carbon effluent
at 30°C and pH 5 without UV light.
178
-------
o
o
100
90
80
70
60
50
40
30
20
10
Expected TOC
for 100%
utilization of
Ozone
Actual TOC
Time (hours)
Figure B-3. Ozonation of end-of-pipe solution crumb
carbon effluent at 30°C and pH 9 without
UV light (type B).
179
-------
100-
90
80
70
60-
50
40
30
20
10
A
Expected TOC
for 100%
utilization
of Ozone
0123456
Time (hours)
Figure B-4 Ozonation of end-of-pipe solution crumb carbon
effluent at 30°C and pH 5 without UV light
(type B).
180
-------
i i i
~
O
G
O
.^
(U
!- *-"
0 nj
XI 4->
*-' Ul
3 HI
-C S- ^
Si
(/I C
>) S-
1/1 a>
o
0
O
- 1
1 **
0 / ~1
/ ^
/ ^
/
/ r~
/ "~
/
/ ^? s
' -r- rtJ
o s ^
-0 O
/s- -
"S
O <"
w u.
1 1
i
vo in -a- n CM
(fUN) /^HP!-cljini ^B-t^-UJ
i
""**
~
c
r-
J
~"
ff
01
i.
-3 i*»
to '
J- f^-
OJ
O-
w ~r*
H- 0.
T3 T3
O) O)
0> OJ
U- U.
1
CO ^j
ro
(U
S-
-p
^
(O
T3
f"~ f^
O
O
01
in
M-
o
(0
-p
1
( =
"» 5 3
O-i
O) o
t3 0
^ 0)
E «
r- S_
-P o;
>, CO
CM -P re
5
i- J2
J3 E
3 J-
P O
OJ E
_ -P O
lO T-
S- -P
-P 3
i~" r"
i- O
u_ to
Lf)
f^t 1
CO
r- O
cu
3
181
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-78-192
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Assessment of Best Available Technology Economically
Achievable for Synthetic Rubber Manufacturing
Wastewater
5. REPORT DATE
August 1978 issuing date
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
M.H. Kleper, A.Z. Gollan, R.L. Goldsmith,
K.J. McNulty
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
Wai den Division of Abcor,
850 Main Street
Wilmington, MA 01887
Inc.
IBB610
11. CONTRACT/GRANT NO.
68-03-2341
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Lab. - Cinn, OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Task Final 7/75-10/76
14. SPONSORING AGENCY CODE
EPA/600/12
15. SUPPLEMENTARY NOTES
lERL-Ci project leader for this report is Ronald J. Turner, 513-684-4481
16. ABSTRACT
An assessment of The Best Available Technology Economically Achievable (BATEA) for
treatment of synthetic rubber manufacturing wastewaters has been conducted. This
assessment was based on feasibility tests with actual wastewater samples, both end-of-
pipe (untreated) and after primary and secondary treatment. The wastewater samples
investigated were collected at representative facilities for manufacture of emulsion
crumb, solution crumb and latex rubbers.
The physical-chemical treatment processes examined included dual-media depth
filtration (DMDF) and ultrafiltration (UF) for suspended solids removal; activated
carbon adsorption (ACA), reverse osmosis (RO) and ozonation for removal of dissolved
contaminants; and ozonation as a polishing step after RO or ACA for removal of re-
fractory organics. The proposed BATEA sequence of DMDF followed by ACA was examined
for treatment of wastewater samples collected after secondary treatment. In addition,
various combinations of the processes identified above were evaluated for both
treatment of end-of-pipe effluents and secondary treated effluents.
Based on an assessment of process technical feasibility and estimates of waste-
water treatment costs preferred options for control technologies were selected.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Water Pollution
Latex
BATEA Reverse Osmosis
BPCTCA Ultrafiltration
Effluent Guidelines
Wastewater Treatment
Synthetic Rubber
Carbon Adsorption
Ozonation
68D
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
198
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
irUSGPO: 1978-657-060/1479 Region 5-11
182
-------