NCD980602163_19811105_Warren County PCB Landfill_SERB C_A Simulation Study of the Volatilization of PCBs from Landfill Disposal Sites-OCR

r4i) ,~7 .
I . r
I ~
A Simulation Study
Charles Springer
Louis J. Thibodeaux
Srikrishna Chatrathi
Department of Chemical Engineering
University of Arkansas
Fayetteville, Arkansas
ABSTRACT
biphenyls from landfill disposal sites. The model included mechanisms for
the migration of vapor through the landfill covering (cap) by diffusion w~th-
in the pore spaces, barometric pressure pumping due to fluctuating barometric
pressure and sweeping by flow of gases generated from biodegradation of
other waste materials.
The results indicate that increased cap depth and/or decreased cap
porosity have the expected result on the vapor emission rate in the absence
of biologically generated sweep gas, but that the emission rate is approximately
proportional to the gas generation rate.
The vapor emission rate of polychlorinated biphenyls from landfills may
be expected to be quite low in any event, but the accompanying deposit of
biodegradable materials may have undersirable results caused by the generation
of gases from biological decomposition processes.
INTRODUCTION
Landfill disposal of Polychlorinated Biphenyls (PCB) has been used in
the past for disposal of condensers, transformers, or other electrical
I \ . . . ,
-2-
gear containing PCB as a dielectric fluid. Likewise, landfilling has been
used to dispose of PCB contaminated oils and so forth.
While there are now satisfactory chemical destruction methods for PCB
and combustible oils contaminated with it, there are apparently few alternatives
to landfilling for non-combustible, contaminated materials. Specifically,
contaminated sludge from rivers and estuaries might conceivably be dredged
and landfilled in order to facilitate cleanup of the effected waterways
(Tofflemire and Qui~, 1979).
It was the purpose of this study to estim.ate by computer simulation the
volatilization rate from the landfill disposal of PCB as a "pure" material
and the landfill disposal of a contaminated river bed silt.
The results of simulations of this type can be of material assistance
in planning future action and in establishing protocols for effective and
safe disposal.
MATHEMATICAL MODEL
Thibodeaux, et al (2,3) have recently developed models of the processes
by which vapors can be evolved from landfills.
Three processes are considered in the modelling: (1) Vapor phase
molecular diffusion; (2) Diffusion enhancement by fluctuating barometric
pressure (barometric pumping) and (3) sweeping by biologically generated gas
venting. If little or no biodegradeable material is present with the vola-
tilizing chemical, then the third mechanism will be inoperative.
For a uniform composition of gas within a landfill cell, an equation of
continuity (conservation of mass) can be written:
£ C dP -= dt -..f!!.. + .r h g
C
(1)
....
-3-
where E is the cell porosity, h is the cell depth, cm, pis the cell gas C C
density, g/cm3, vis the superficial outward velocity, cm/s, tis the time,
3 s, and r is the rate of gas generation in the cell, g/cm s. g
An equation to describe the undirectional motion of a gas flowing through
porous media can be obtained from Darcy's law:
K v = µL (P-TT) (2)
Here, K is the permeability of the covering material in darcies
2 -(cm • cp/s.atm), µ is the gas viscosity, cp, L is the thickness of the
covering, cm, Pis the pressure within the cell, atm, and TT is_ the barometric
pressure.
The biological gas generation rate is usually expressed as rb, a gas
generation rate per unit mass of material in the cell, cm3/g.s. This may
be converted to volumetricgas rate, r , for specific cell conditions by: g
r = r p p g b C
(3)
where~ is the bulk density of material in the cell.
Combining equations 1, 2, and 3 while solving for the cell gas density
by the ideal gas law results in an expression for the rate of change of
cell pressure:
KP (P-TT)
e: h Lµ
C C
(4)
Since the barometric pressure, TT, varies in time, it is seen that the
cell pressure P will vary also, but will lag due to the capacity of the cell.
The variation of cell pressure with atmospheric pressure will not be linear,
even if the biogas generation .were zero.
The rate of transfer of a single component, A, of a mixture for
unidirectional flow without reaction or adsorption on the pore S?ace walls, ..
... ~ -
n
.. -· ' .
-4-
can be related to the diffusion coefficient of the component by the following
continuity equation:
(5)
where subscript A refers to component A. Thus, PA !s the concentration of
component A in the vapor phase in the cell, g/cm3• DA,P is the diffusivity
of component A in the pore spaces of the covering material, cm2 /s, and y is
the vertical distance into the cover.
Thibodeaux (3) has shown the integration of Eq. (5) as
(6)
where R =Lv/DA,P' NA is the outward flux of component A, g/cm 2 .sand p * A
is the equilibrium concentration of the component in the cell, g/cm 3 The .
model assumes that the vapor within the cell is saturated with respect to
component A at all times.
The simulation was performed by simultaneously solving equations 2, 4
and 6 to determine the flux of the component of interest. The simulation
was carried out using the IBN Continuous Systems Modeling Program (CSMP).
SIMULATION PARAMETERS
Two PCB disposal scenarios were assumed in this study. The first being
the disposal of a quantity of liquid PCB, perhaps one or two gallons, in
the bottom of an excavation. The liquid is assumed to be in a container or
in a piece of discarded electrical gear, and the excavation is backfilled with
a medium porosity soil or other material including other types of waste. The
backfill constitutes a "cell" and is covered with a layer of controlled
permeability soil for a cap. The excavation walls and floor are assumed to be
of negligible permeability. The concept of such a disposal is shown in Figure 1.
-5-
As long as the container remains intact and does not leak, there will of
course be no volatilization. However, after time, it may perforate due to
corrosion and PCB will vaporize.
The second type disposal considered is that of a soil, such as a river
bed sludge, uniformly contaminated with PCB in the absorbed state. The sludge
layer will be covered with a cap of controlled permeability soil as in the first
case, and the walls and floor of the excavation are again assumed_ to be es-
sentially impermeab}e. Such a disposal is pictured in Figure 2.
The choice for a "typical" PCB was Arclor.1248 which would be mostly a
four (4) chlorine substituted material with a molecular weight of about 292.
-4 The vapor pressure of this material would be expected to be about 9.2 xlO mmHg
,
at 25°c (National Research Council, 1979).
In the second case, that is the PCB absorbed in river bed sludge, it is
assumed that its presence as a physically absorbed material will lower the
equilibruim partial pressure somewhat, but there is little information to suggest
what might be a reasonable value. In this study it is assumed to be approximately
one-half of the vapor pressure, or about 5 x 10-4 mmHg.
In order to asses the importance of the presence of other wastes, varying
rates of biogas generation are assumed. It is assumed that the river bed
sludge would h~ve fairly low biogas generation rates since only the organic
component of the sludge would be -responsible for the generation. However, the
pure material might be in the presence of biodegradable organic wastes in varying
amounts so that biogas generation rates could vary from essentially zero to
nearly the maximum which might be possible. The values of the various para-
meters and the ranges used in this study are shown in Table 1 for the pure PCB
disposal, in Table 2 for the disposal in the presence of biogas generation and
in Table 3 for the disposal of contaminated soil.
In the matter of the equilibria between the sludge and the PCB, it is
known that the PCB's are absorded readily by soils which have a high organic
content (Tofflemire, et al, 1979 and Tofflemire and Shen, 1979). However,
adsorption isotherms are not available, and even if some adsorption data were
available it would be highly specific. Therefore, the concentration of PCB
vapors in a disposal cell is quite speculative.
The barometric pressure fluctuations of a two week period in Fayetteville,
Arkansas were taken-as typical and used for the simulation. The normal pressure
in Fayetteville is about l.4in. (35 to 36 rnmHg) less than at sea level due to
elevation.
For comparison, since PCB properties are quite variable depending upon
the composition, two base case simulations were run assuming the properties of
monochloro biphenyl (MCB) which would represent an upper limit of PCB volatility
and mobility.
The parameters used in the study are summarized in Tables 1 through 4.
SIMULATION RESULTS
The results of the simulations are shown in Tables 5) 6, and 7 respectively,
for the cases of disposal with gas generation, without gas generation and con-
taminated sludge.
As expected, decreased cap porosity, and/or increased cap depth reduces the
level of PCB emissions. However, in the presence of biologically generated sweep
gas, the properties of the cap have much less effect on the emission rates,
and that when there exists some gas generation, the emission will be nearly
proportional to the gas generation rate.
The simulations also suggest some significant pressure build up when
there is biogas generation beneath a relatively tight (low porosity or in-
creased depth) cap. Some of the computed pressures are shown in Table 8.
The effects of the variables are shown graphically in Figures 3 through 7.
,. . .
.. ..
-7-
CONCLUSIONS
Probably the most important result of this study is the finding that the
emissions of PCB from landfills may be expected to be quite small. Perhaps
it could be said that if the environment can tolerate any PCB vapor emission at
all, then any reasonable landfill which confines the material under a meter or
more of soil will provide for adequately low emissions. Typical emissions from
a 100 m by 100 m landfill would be about one kg per year when there is no sweep
gas source.
While it may not be readily apparent that the computed emissions are small,
one can form an idea of the effects of these emissions by considering possible
airborne concentrations which might result in a particular circumstance.
. . 2 .
Given the gas generation base case emission of about 2.4 mg/m. day, it
can be deduced that an airborne concentration of about 0.02 ug/m3 would result
at a location 100 m down wind in a typical one to two meter per second wind. It
may be noted that this concentration is about the lower limit of our ability to
detect PCB in air.
For comparison, the NIOSH standard for airborne PCB specifies that it should
be no more than 0.1 Ug/m3 for 8 hour exposure.
One aspect which does not often receive much attention is the consequences
of biogas gene~ation under a well compacted cap, or perhaps, under a plastic
film barier. The simulation showed that excessive pressure can be generated
under a low porosity and/or deep cap. While no one would expect actual pressures
as indicated in Table 7, the simulation results clearly shown that sufficient
pressures can be generated to crack or otherwise compromise nearly any cap which
might be placed over a landfill. It is, therefore, important th;it not only PCB,
hut any hazardous chemical not be deposited in a landfill along with material
which might biodegrade to form a gas, unless some provision is made for venting
to relieve the pressure.
-8-
SIMULATION LIMITATIONS
The model presented earlier has been confirmed by experimention and
field measurements to only a l .imited extent. Those measurements which have
been made, however, suggest the model is a reasonably accurate description.
In application to an actual site, however, one might have considerable dif-
ficulty evaluating the various parameters which control the emission rate.
Some variables are not accounted for the model, especially any adsorption
of the volatile matt.er in the cap material. If this adsorption were signifi-
cant it could reduce the initial emission rate, not increase it.
Moisture in the soil can only be accounted for by the effect it would
have on porosity. There is little question but what an increased soil moisture
content would reduce the emission rate, although decreased porosity may not be
the only effect.
RECOMMEt-U)ATIONS
Materials of low volatility which are disposed of in properly constructed
landfills will present little hazard from vapor emissions to the air. Care
must be taken however, in the construction of any hazardous chemical landfill,
to limit the amount of biodegradable material in the same site since gas
generated can either greatly increase the emission rates by sweeping action, or
can generate pressure which might rupture the cap and destroy its integrity.
Near surface landfills may not be suitable repositories for relatively
volatile, hazardous wastes. The possible limitations for landfill disposal of
volatile materials require more study since emission hazards depend heavily
upon the properties of the hazardous materials as well as upon the nature and
properties of accompanying materials.
... ..,
-· REFERENCES
1. Tofflemire, T. J. and S. 0. Quinn, "PCB in the Upper Hudson River:
Napping and Sediment Relations" New York State Department of Environ-
mental Conservation, Technical Paper No. 56. (1979)
2. Tofflemire, T. J. and T. T. Shen, "Volatilization of PCB from Sediment
and Water: Experimental and Field Data", Proc. 11th An. Ind. Waste
Conf., Univ. Park, PA (1979).
3. Thibodeaux, L. J., C. Springer and L. M. Riley, "Models of Mechanisms
for Hazardous Chemical Emissions from-Landfills", Symp. on Toxic
Substances Management Programs, 18-1st._ A.C.S. National Mt'g., Atlanta
(March 1981).
4. Thibodeaux, L. J., "Estimating the Air Emissions of Chemicals from
Hazardous Waste Landfills", J. Hazardous Materials 4, 235-244 (1981).
5. National Research. Council~ "Polycnlo-rinated Biphenylsu, Report by the
Committee on the Assesment of Polychlorinated Biphenyls in the Environment,
National Academy of Science, Washington (1979}.
TABLE 1 Properties Assumed for PCB
Molecular weight
. 0
vapor pressure (25 C)
Diffusivity in Air
292
-4 9.8xl0 _6mm Hg
(l.3xl0 atm)
0.031 cm2/s.
(These properties are similar to "Arochlor 1248")
TABLE 2 -Parameters Used in Simulation Study
for "Pure" PCB Disposal -with
No Gas Generation
Constant Values
Cell Porosity,
Cell Depth,
Cell Density,
PCB Partial Pressure,
Varied Parameters
Cap Depth, L, (m)
Cap Porosity, e:
e: C
h C
0.2
3.0 m
2.0 g/cm3
l.3xl0-6 atm
(Base Case) (Other Values)
3.0
0.1
1.0
0.08
0 .5
0.2
TABLE 3 -Parameters Used in Simulation Study
for "Pure" PCB Disposal -
with Accompanying Bio-Gas Generation
Constant Values
Cell Porosity,
Cell Depth,
Cell Density,
PCB Partial Pressure
Varied Parameters
e:
C
h C
pc
0.2
3.0 m
2.0 g/cm 3
-6 l.3x10 atm
(Base Case) (Other Values)
Cap Depth, L, (m)
Cap Porosity, e:
Gas Generation
Rate rb (cm 3/g.s)
3.0
0.1
Second Base Case (very low porosity cap)
Cap Porosity, 0.06
Cap Depth
1.0
0.08
I.Om
0.5
0.2
0.5 m
.. •
_ TABLE 4 -Parameters Used in Simulation Study
of Contaminated Soil Disposal
Constant Values
Cell Porosity, 0.2
Cell Depth, 10.0 m
5x10-S
3
Gas Generation Rate, cm I g.s
PCB Partial Pressure -7 5.3x10 atm
Varied Parameters
Cap Porosity
Cap Depth, L
(Base Case)
0 .1
l.Om
(Other Values)
0.08
2.0 m
0.2
0.5 m
.. •
TABLE 5 -Simulation Results -
-"Pure" PCB Disposal with Biogas Generation
Cap Parameters Gas Rate Flux
(Porosity) (Depth) (PCB) (MCB)
(m) 3 cm/gs 2 (mg/m d)
0.1 3.0 3xl0-7 2.42 95.0
2xl0-7 1:6
lxl0-7 0.8
0.1 3.0 3xl0-7 2.42
1.0 2.46
0.5 2.51
0 .1 3.0 3xl0-7 2.42
0.2 2.65
0.08 2.40
0.06 3.0 2.31
1.0 2.36
0.5 2.39
.. ·· . . ..
TABLE 6 -Simulation Results -
"Pure" Disposal Without Biogas Generation
Cap Parameters Flux
(Porosity) (Depth) (PCB) 2 (m) (mg/m d)
0.1 3.0 0.08
0.08 0.05
0.2 0.33
0 .1 3.0 0.08
1.0 0.24
0.5 0.45
(MCB)
5.2
TABLE 7 -Simulation Results -
Disposal of Contaminated Sludge
Cap Parameters Flu2 (Porosity) (Depth) (mg/m .d)
(m)
0 .1 1.0 0.53
0.08 0.30
0.2 0.91
0.1 1.0 0.53
0.5 0.62
2.0 0.48
(Gas Generation Rate assumed to be 5xl0-B
Cell Pressure
(atm)
1.2
2.0
1.0
1.2
1.1
1.4
3 cm /g.s)
...
TABLE 8 -Simulation Results -
Cell Pressures Developed with Biagas Generation
Cap Parameters
(Porosity) (Depth)
0.1
0 .1
0.1
0.2
0.08
0.06
(m)
3.0
3.0
1.0
0.5
3.0
3.0
1.0
0.5
Gas Generation
rjte
(cm /g.s)
3xl0-7
2xl0-7
lxl0-7
3xl0-7
Cell Pressure
(max)
(atm)
2.1
1.7
1.3
2.1
1.3
1.2
2.1
1.0
9.2
110
36.7
19.0
..
Figure 1. Conceptualization of a typical disposal of "pure" poly-
chlorinated biphenyl.
Figure 2. Conceptualization of the disposal of contaminated soil or
river-bed sludge.
Figure 3. Effect of Cap porosity on the emission rate of PCB from a
type I, or "pure", PCB disposal~ This figure shows the results
of both cases, with and without the accompanying generation
of gases.
Figure 4. Effect of Cap depth on the emission rate of PCB from a
disposal of "pure" PCB. This figure shows the results of
· both cases, with and without the accompanying generation
of gases.
' ..
Figure 5. Effect of Cap depth on the emission rate of PCB from a
disposal of contaminated sludge or soil. The simulation
parameters assume a low rate of accompanying gas generation.
Figure 6. Effect of Cap porosity on the emission of PCB from a disposal
of contaminated sludge or soil. The simulation parameters
assume a low rate of accompanying gas generation.
Figure 7. Effect of gas generation rate on the emission of PCB from
a disposal site.
··· -o///,/s~P//////:i L ',,,,,,,,, ,,, , .. , ,11,11{1,,,,,.l'li,1t,1,, ... , lf,,tl, f,111/~ , \ 11 .. , ..LI ~~~~~~ " CELL ~ "'-I he ~""'~~ ~ ~ PCB · """" ffl ~ llllll)~Contoiner I>""~ "' \, ""~ "" "'' TYPE I DISPOSAL SITE . .'
,,. , , , 11, t, 1 111 ., r 1,., 11 n11,1 rf a, I I 1,,. II I II I r t1~, 1 11 , 111 111 u 1//////C/A,P/////////////// L ~~~~~~~~~~~ T ~c~~~~~Z~\~~ l TYPE II DISPOSAL SITE ·,
.. ' .. 2.5 2.0 "'O c:\i t.5 E ' Cl E I 1,0 X ::::> ...I LL 0.5 -WITH GAS GENERATION (3xlo-7 cm3/g·s) FIGURE 3 PCB FLUX vs CAP POROSITY "PURE" PCB DISPOSAL CAP DEPTH = 3 m I -NO GAS GENERATION . ol -; , 1 I I I I I 0. 06 ----. 0.08 0.10 0.12 0.14 CAP PORO-SITY 0.16 0.18 020
. r , "0 N. 2.5 2.0 E '.1.5 CJ' E I X 1.0 => _J LL 0.5 FIGURE 4 PCB FLUX vs CAP DEPTH "PURE" PCB DISPOSAL CAP POROSITY = 0.1 WITH GAS GENERATION (3x I o-7 cm3/g ·s) NO GAS GENERATION ' o~-~~-~-~~--::J-~-~----::~-----o 0.5 1.0 1.5 2.0 2.5 3.0 CAP DEPTH - m
,. ~ 0.8 -0 '1:0.6 I ' Cir E I Q.4 X ::, ..J LI.. 10.2 FIGURE 5 PCB.FLUX vs CAP DEPTH FOR CONTAMINATED SLUDGE DISPOSAL CAP PORO SI TY =0.1 o __________________________ _ 0 0.5 LO 1.5 2.0 CAP DEPTH - m ~ r •
... r • ' . ' . 1.0 0.8 '0.6 I 0.4 102 FIGURE 6 PCB FLUX vs CAP POROSITY CONTA~1INATED SLUDGE DISPOSAL · CAP DEPTH = 1.0 m 0 '----'-----'-----L---.,__--'----'----"-----0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 . CAP POROSITY