United States
                    Environmental Protection
                    Agency
 Municipal Environmental
 Research Laboratory
 Cincinnati OH 45268
                    Research and Development
 EPA-600/S2-84-047  Apr. 1984
&EPA         Project  Summary

                    Volatilization  of  Organic
                    Pollutants in Wastewater
                    Treatment  -  Model  Studies
                    Paul V. Roberts, Christoph Munz, Paul Daridliker, and Christine Matter-Muller
                      Transfer rates of volatile  organic
                    contaminants from water to air were
                    measured in laboratory experiments
                    simulating surface aeration and bubble
                    aeration. The experiments were con-
                    ducted mainly with the following six
                    compounds, ranked in descending
                    order of volatility: CCI2F2, CCU, CCI2=
                    CCI2, CHCI=CCI2, CCI3CH3, and CHCI3.
                    Additional experiments were conducted
                    with chlorobenzene and dichlorobenzene
                    isomers. Volatility of the compounds
                    differed by nearly three orders of
                    magnitude, as measured by the Henry's
                    coefficient.  In surface aeration, the
                    transfer coefficients of all organic
                    compounds were proportional to that of
                    oxygen, with the coefficient of propor-
                    tionality approximately 0.6 for all but
                    the least volatile compounds (the
                    dichlorobenzenes). The surface aeration
                    data are consistent with a model of
                    liquid-phase-controlled mass transfer,
                    except for the least volatile compounds,
                    which  show evidence of a significant
                    gas phase  resistance. In the bubble
                    aeration column, saturation of the gas
                    bubbles was appreciable for all of the
                    organic compounds except CCI2F2.
                    Under conditions of fine bubble aeration,
                    transfer rate constants were measured
                    for oxygen and CCI2F2,  and  Henry's
                    coefficients were measured  for the
                    remaining compounds. The transfer
                    rate constants agreed within  a factor
                    of two with values predicted from semi-
                    empirical correlations based on bound-
                    ary layer theory- Henry's coefficients
                    measured in the bubble column generally
                    agreed in order of magnitude  with
                    values  predicted from vapor pressure
                    and solubility data. For the six com-
                    pounds listed above,  the removal  by
volatilization is estimated to be 85 to 87
percent for typical surface aeration
conditions,  compared with 35  to 85
percent for a typical submerged aeration
situation.
  This Project Summary was developed
by EPA's Municipal Environmental
Research 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
  The goal of this project was to improve
understanding of the process by  which
volatile organic contaminants are  trans-
ferred to the atmosphere during waste-
water treatment. Attention was focused on
the aeration processes commonly used in
activated sludge treatment—namely,
surface aeration  and bubble aeration.
Experiments were conducted  under
controlled conditions using physical
models in the laboratory. As a  basic
premise, it was  hypothesized that  the
transfer rate coefficients of the organic
contaminants ought to be proportional to
that of  oxygen so long as the organic
compounds are sufficiently volatile to
assure liquid phase control of the transfer
rate.  This hypothesis was tested  in  the
laboratory experiments.

Procedure
  Surface aeration and bubble column
experiments were conducted in the batch
mode in laboratoy-scale contactors with
7 to 24 liters of water. Oxygen absorption
from  air was measured simultaneously
with  stripping of organic solutes. The
organic solutes were chosen to represent

-------
a wide range of volatility; the predicted
coefficients (expressed as a dimensionless
ratio of mass  concentration in  the gas
phase to that in the liquid phase) ranged
from 62 (CCI2F2) to 0.06 (1,4-C6H4CI2).
The  initial concentrations  of  organic
solutes  were on the order  of  1  to  10
mg/l, at least two orders of magnitude
below the respective solubility limits.
  The surface aeration experiments were
carried out in a baffled, agitated vessel.
Mixing  was achieved  by a  compound
impeller consisting of a ring-guarded,
three-blade turbine to agitate the water
surface  and a  flat-blade impeller to mix
the bulk liquid  The impeller speed was
controlled by a rheostat and a stroboscope.
Power input was measured by fitting the
impeller shaft  with a torque meter. The
stirring speed was varied from 65 to 375
rpm, corresponding to power input from
0.8to320W/m3.
  The bubble column experiments were
conducted in a 22.5-cm-diameter glass
column with  appurtenances  for gas
metering, sampling, and dissolved oxygen
monitoring. Air was supplied through a
diffuser constructed  by  adapting  a
sintered glass filter. The liquid volume
and the gas rate were varied to acheive
conditions resulting m different rates of
mass transfer. Additional experiments
were conducted by varying the amount of
methanol, which was added along with
the organic solutes to enhance miscibility.
The  methanol  addition affected  neither
the phase equilibria nor the mass transfer
rates m the relevant concentration range.
  Organic solutes were  determined  by
gas  chromatography  with electron-
capture detection,  following  enrichment
by pentane extraction  in  the  sealed
sample  bottle.  The concentrations were
quantitated by comparing the peak areas
to that of a known quantity of the internal
standard,  1,2-dibromoethane. The dis-
solved  oxygen concentration was mea-
sured by stirred electrodes and recorded
throughout the experiment.

Results
  The experimental results were inter-
preted to obtain estimates of the mass
transfer rate,  Ki_a, and the Henry's
coefficient, Hc. Values of  Ki_a were
determined in  the surface aeration
experiments, whereas either Ki_a or  Hc
could be  determined for a  particular
solute in the bubble column experiment,
depending on the solute's volatility and the
experimental conditions

Surface Aeration
  In  the surface aeration experiments,
the conditions were such that accumula-
tion of organic solutes was inconsequen-
tial. Hence the data were analyzed with
the equation
        -/n(C(t)/C0) =-KLa(t - U      (1)
where   C(t)=concentration at time t,
          Ki_=mass transfer coefficient
             [m-s-1],
           a=specific interfacial surface
              area [m"1], and
       (t - t0)=elapsed time [s].
The experimental data fit Equation 1
excellently, with correlation coefficients
exceeding 0.99 for all experiments. The
data were interpreted in  terms of the
product Ki_a because the specific surface
area, a, could not be estimated accurately,
owing  to the formation of water drops
under conditions of high mixing intensity.
  The values of the  mass transfer rate
constant for oxygen,  KL,o2a, observed in
the surface aeration experiments ranged
from 0.008 mm"1 to 0.35 min"1. The value
of KL,o2a increased with increasing stirrer
speed  and power input.  In the  fully
turbulent range  (Reynolds number >
15,000), the transfer rate  constant was
approximately  proportional to the energy
input,  as  shown in  Figure 1, for the
                                    organic solutes as well as oxygen. Over(
                                    the range investigated, the values of Ki_a
                                    for the organic solutes were approximately
                                    proportional to  that of oxygen. The
                                    coefficient of proportionality for a particu-
                                    lar solute i is defined as
                                                , E (KL,,a)/(KL,o2a)
(2)
                                    The  values  of  ^  from experiments in
                                    clean water are shown in Figure 2. The
                                    proportionality coefficient is constant
                                    within 10 percent for all of the compounds
                                    studied over a range of nearly three
                                    orders of magnitude in KL,o2a. Moreover,
                                    the values of V do not differ greatly for the
                                    compounds studied and average approxi-
                                    mately  0.6.  The  average values  and
                                    standard deviations of the experimental
                                    values for the coefficient W are summa-
                                    rized m Table 1. In all cases, the value of
                                    •* for clean water and filtered wastewater
                                    effluent differed by less than 5 percent.
                                    The  value of ¥ was in all  cases higher in
                                    filtered  effluent than in clean water, but
                                    the  differences were not significant
                                    statistically. However,  the  absolute
                                    values of the transfer rate constants were
                                    consistently 10 to 25  percent  lower in
                     Power Dependence of Mass Transfer Rate Constants
    W  L I  I I I
    ;o-3-
I
I
70-"-
    ro"5l  i i i i
       0.5    J                    10                  WO                 WOO

                                      P/VlW/m3]

Figure 7.    Dependence of the mass transfer rate constants on power input in surface aeratio
           experiments.

-------
H
1.0
0.8
0.6
0.4
0.2
0
t
Figu
	 • 	 rreon
1 	 Trichloroethane
	 Carbon Tetrachloride


1 ' ' KLO a
J 0.1 0.2 0.3 [„,,.„%,
re 2. Linear regressions of the trans-
                                         Table 1.
          Average Values of the Transfer Rate Constant Ratio * in Clean Water and Filtered
          Secondary Effluent
                                                                          Mean + Std Dev
                                                                 Difference
            fer rate proportionality coeffici-
            ent M1 versus the oxygen transfer
            rate constant /CL,o2 a [mirf1].
filtered wastewater effluent than in clean
water.

Bubble Aeration
  The bubble  aeration  experiments
corresponded to conditions of fine-bubble
aeration. The average bubble diameter
was 0.6 to 1.4 mm, depending on the air
flow rate. Under these conditions, the air
bubbles  were  substantially saturated
with most of the organic solutes during
the rise through  the water column,
despite the relatively small depth, < 0.6
m. For conditions of partial saturation, the
following equation was derived to inter-
pret the bubble column data:
/n(  c )= Qs'H° [ 1 - e*2* ] (t - to)   (3)
    Co    VL
where  QG=gas flow rate [mVmin],
        Vi_=liquid volume [ma],
         Zs=diffuser submergence [m],
         =saturation factor = [Ki_a • VJ
            /[Hc • QG • ZJ,  and
        Hc=Henry's coefficient expressed
            as the ratio of mass concen-
            trations in the gas and liquid
            phases.

For values of  4>ZS > 5, the exponential
term in Equation 3 is negligible, and the
gas  bubbles  can  be  assumed  to  be
saturated  with the organic compound.
The value of the Henry's  coefficient, Hc,
was then evaluated by linear  regression
of the experimental values of -/n(C/C0) on
(t -10). For values of 4>ZS<0.1, the value of
the transfer  rate constant,  KLa, was
evaluated from linear regression  of the
experimental values of -/n(C/C0) on (t -10).
For intermediate cases, where 0.1 < ZS<
5, the value of QoHc/VL[1  -  e4>Zs] was
evaluated  by  linear  regression, and  Hc
Compound
CCkF2
CHCh
CH3CC/3
ecu
CHCI=CCI2
CC/2=CC/2
Clean
Water*
0.661 +0.039
0.560 + 0.036
0.607+0.030
0.617 + 0.033
0.615 + 0.029
0.608 + 0.034
Filtered
Secondary Effluent
0.668 + 0.076
0.584 ± 0.035
0.619+0.029
0.627 ± 0.037
0.626 ± 0.026
0.625 ± 0.026
AHJ
0.007
0.024
0012
0.010
0.011
0.017
%
1.1
4.3
2.0
1.6
1.8
2.8
* Milli-Q water

was estimated after substituting a value
of KLa obtained from the experimental
value of  KL,o2a-1!',  assuming an average
value of ¥ = 0.60.,The experimental data
fit closely to Equation 3, and the correla-
tion coefficients generally exceeded 0.99.
  Under  the  conditions of  the  bubble
column experiments,  the air bubbles
were saturated negligibly with CCI2F2and
substantially for the other five organic
compounds. The transfer rate constant,
Ki_a, for  CCI2F2  was  proportional to
KL,o2a. The average coefficient of propor-
tionality, irCCI2F2 = (KL,cci2F2a/KL,o2a) was
0.55 + 0.05. The experimental values of
Hc  demonstrated a slight negative
correlation  with  KL,o2a; this apparent
effect was strongest in  the  case  of the
least volatile compound studied,  chloro-
form. The average values of  the Henry's
coefficients determined in  the  bubble
column experiments are listed in Table 2.
The  values  of Hc were somewhat  (as
much as  50 percent) higher in filtered
secondary effluent than  in clean water.

Discussion
  The results of this work are consistent
with  the hypothesis  that  the  mass
transfer of organic contaminants from
water to  air  is controlled by the liquid
phase resistance so long as  the organic
solutes are sufficiently volatile.  Under
these  conditions, the  transfer rate
constants for the organic solutes  are
proportional  to that of oxygen. The
coefficient of proportionality is approxi-
mately 0.55 to 0.65 for organic solutes in
the molecular weight range of 100  to
200. The proportionality between trans-
fer rates  of  organics and oxygen was
virtually the same in filtered wastewater
effluent as  in  clean water,  but the
absolute  values of the  transfer rate
constants were approximately 20 percent
lower for  the effluent at a given set  of
mixing conditions.
  In surface aeration, the transfer rates
of the organic substances were  nearly
equal because their diffusivities differed
very little and  the volatility (Henry's
coefficient) played no role. But with
bubble aeration, the  saturation  of the
rising air  bubbles  is important, and the
stripping rate depends on Hc as well as
KLa.
  Hence the percentage removal of
organic solutes  by transfer to the
atmosphere is expected  to differ more
widely in bubble  aeration than in surface
aeration. This is illustrated by means  of
example calculations for organics removal
in three standard cases  of activated
sludge treatment: surface aeration, fine-
bubble aeration,  and coarse-bubble
aeration. The results of  these  calcula-
tions are shown in Table  3. The  differ-
ences between the aeration processes  in
terms of organics removal  by volatiliza-
tion may be understood in the following
way. In fine-bubble aeration, the desired
quantity of oxygen is achieved by contact-
ing the wastewater with a  smaller
quantity of air than in the case of surface
aeration.  Coarse-bubble aeration  is
Table 2.    Experimental and Predicted Values of the Henry's Coefficient, Hc

                                     Henry's Coefficient, Hc =
                                 Experimental, Mean + Std. Dev.
Compound
CHCI3
CH3CCI3
ecu
CHCI=CCI2
CCIfCCk
Clean Water]
0.203 ± 0.036
0.594 + 0.029
0.977 + 0.041
0.388 + 0 033
0.605 + 0.040
Filtered Effluent^
0.326 + 0.061
0693+0.041
0.980 + 0.020
0.507 + 0.068
0.698 + 0.053
Predicted*
0.129
0.166
0.948
0.398
0.848
*From vapor pressure and solubility data.
\n = 8.
\n =4.

-------
   Table3.    Results of Example Calculations ofOrganics Removal in Activated Sludge Treatment*

                     /      u      \                Percent Removal
Compound
CCkFi
ecu
CC/2=CC/2
CH3CC/3
CHCI=CC/2
CHCI3
| g/rrf in gas J
\ g/m3 in water /
62.
0.98
0.70
0.69
0.51
0.33
Surface
Aeration
87
86
86
86
86
85
Fine-Bubble
Aeration
86
61
53
53
46
35
Coarse-Bubble
Aeration
85
75
71
71
66
57
   ^Conditions: 80 mg/l oxygen transferred; submergence depth = 4 m; DO = 1 mg/l.
   intermediate between  the  other two
   cases because the lower oxygen transfer
   efficiency  (compared with that of fine-
   bubble aeration) requires  a  greater air
   flow rate to achieve the desired oxygen
   transfer rate.

   Conclusions
     A  relatively  simple  model  of  mass
   transfer can be applied successfully to
   interpret laboratory data for the transfer
   of volatile organic compounds to  the
   atmosphere during wastewater treatment.
   The  percentage removal of a particular
   solute can be estimated approximately
   from  knowledge of the oxygen transfer
   rate  constant under the aeration condi-
   tions of interest, together with  the
   Henry's coefficient for the solute. The
   fundamental difference between surface
   and bubble aeration must be recognized
   with regard to the effect of air saturation
   on organics removal  by volatilization. In
   general, organic solutes will be removed
   by volatilization to a greater extent with
   surface aeration than with bubble aera-
   tion,  and  this  difference  will  increase
   with decreasing solute volatility
     The full  report was submitted in fulfill-
   ment of Cooperative Agreement No. R-
   806631 by Stanford University under the
   sponsorship of  the U.S. Environmental
   Protection Agency.
          Paul V. Roberts, Christoph Munz, Paul Dandliker, and Christine Matter-Muller,
            are with Stanford University, Stanford. CA 94305.
          H. Paul Warner is the EPA Project Officer (see below).
          The complete report, entitled "Volatilization of Organic Pollutants in Wastewater
            Treatment—Model Studies," (Order No. PB 84-158 856; Cost: $16.00, 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:
                  Municipal Environmental Research Laboratory
                  U.S. Environmental Protection Agency
                  Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
     BULK RATE
POSTAGE & FEES PAID
        EPA
   PERMIT No. G-35
Official Business
Penalty for Private Use $300
       LOU  W   TILLtY
       REGION  V  tPA
       CHICAGO  IL  60604
                                                                                       •A U.S. GOVERNMENT PRINTING OFFICE 759-015/7656

-------