Note: Descriptions are shown in the official language in which they were submitted.
~i~'~~~~
1
Membrane-Assisted Leaching Process and Apparatus for the Removal of
Metals from Soil
FIELD OF THE INVENTION
The present invention relates to a process and apparatus to effectively
and efficiently remove metals from contaminated soil wherein a reactor and
membrane extraction unit are utilized. The use of the subject system provides
improved metal removal from soil over a batch process.
BACKGROUND OF THE INVENTION
For many years now, soils near the earth's surface have been used as a
dumping ground for society's residual chemicals with little concern of the
consequences. Although these activities have been greatly reduced through
government regulation, the damage that has already been done cannot be
ignored. In 1991, in the U. S. alone, the National Priority List contained 498
sites eligible for remediation funding. Of these, more than 60 percent
identified heavy metals as the principal contaminant of concern (Yarlagadda,
P.S., et al. 1995. "Characteristics of Heavy Metals in Contaminated Soils."
Journal of Environmental Engineering. 121(4): 276-285.).
Heavy metals can be described as those that are used or discharged by
industrial enterprises or used by humans in various ways (Schalscha, E.B.
1989. "Heavy Metal Movement in Irrigated Soil. " Encyclopedia of
Environmental Control Technology. Volume 3 - Wastewater Treatment
Technology. Gulf Publishing Company. Houston, Texas: 543-555) including
Cd, Cr, Co, Fe, Hg, Mn, Mo, Pb and Zn. All are toxic at high concentrations
and some, including lead, are toxic at very low concentrations.
Heavy metals can reach toxic levels in soils due to natural processes and
more importantly from human activities. These sources can be categorized as:
1. irrigation with non-treated industrial wastewater;
2. disposal of water-treatment plant sludges on land; and,
3 . dumping of wastes and refuse (Schalscha, 1989) .
The interactions between soil and heavy metals, as well as the
Z176~~~
2
movement of heavy metals through soil, has been the focus of many studies
over the past two decades. Nevertheless, these processes are still not well
understood. Many different types of soil/heavy metal interactions have been
studied. Difficulties arise in understanding these processes due to the
simultaneous existence of more than one type of interaction.
The principal cause of heavy metal movement in soils is advection of
ground water. Molecular diffusion is significant only if ground water
movement is extremely slow. For soils of medium and high permeabilities
such as silts, sands and gravels, the effect of molecular diffusion is very
small.
For very low permeability clays such as clay liners used in modern landfills,
molecular diffusion can become the dominant transport mechanism.
An inevitable consequence of advection in soil is a decrease in downstream
contaminant concentrations and a subsequent increase in the volume of the
contaminated area. This phenomenon will be referred to as dissipation.
Mechanisms responsible for dissipation are molecular diffusion and dispersion.
Molecular diffusion contributes to dissipation independently of advection,
but is only significant when advection is small.
The tortuous paths taken by ground water through the soil medium causes
turbulence and mixing. This is referred to as dispersion and is usually a
significant contributor to dissipation.
Retardation and attenuation are two similar processes that affect the fate of
contaminants in the subsurface.
Retardation slows the transport of contaminants as a result of a reversible
interaction between the soil and the contaminant. This interaction removes
portions of the contaminant from the aqueous solution and can take several
forms. Precipitation and adsorption are~two such forms.
Precipitation is a well understood chemical process and can be modelled
using solubility product expressions. Heavy metals will precipitate with
hydrous oxides, carbonates, sulfides, phosphates and silicates that may be
present in the soil or ground water (Schalscha, 1989). It is important to note
that the solubility of heavy metals is strongly related to pH. Precipitates
can be
immobilized due to the filtration ability of soil and this causes retardation
(LaGrega, M.D., Buckingham, P.L., and Evans, J.C. 1994. Hazardous
3
Waste Management, McGraw-Hill, Inc.).
Processes that cause heavy metals to sorb onto the surface of soil particles
include ion exchange, van der Waals and other electrostatic forces. Ion
exchange of metallic ions with soil is a partially reversible process, in that
saturated ion exchange sites may release cations in response to either a
decrease in the concentration of cations or a change in pH. Thus, ion exchange
is considered a retardation mechanism.
Attenuation reduces the concentration of aqueous phase contaminants in
the plume by an irreversible biological or chemical reaction. Important
attenuation mechanisms are chemical and biological oxidation-reduction
reactions (LaGrega et al. , 1994) .
Studies have shown that contaminants sometimes bind preferentially to
the finer fraction of soil (Griffiths, R.A. 1995. "Soil Washing Technology and
Practice." Journal ofHazardous Materials. 40(2): 175-189).
Existing methods for heavy metal removal from soil include those that
can be carried out in-situ and those that require the contaminated soil to be
excavated and then treated. One conventional method of remediation is to
excavate and landfill the contaminated soil. This method does not actually
treat, but simply relocates the contaminated soil. In-situ methods might be a
cost effective alternative, which reduce land disturbance. Other excavation
methods require the soil to be treated, and the treated soil can then be
replaced.
Some of the potential in-situ methods include those that immobilize the
metals by precipitation and those that solubilize and remove metals from the
system. The methods that solubilize the contaminants do not require long term
monitoring because the metals are removed in the process. One method, which
uses solubilization is soil flushing. The methods that bind the metals to the
soil
will require long term monitoring because the metals may be remobilised by
environmental conditions and/or chemicals. One principle of immobilization of
the metals is stabilization.
This method is similar to a process called "solution mining" used in the
mining industry (Sabatini, D.A., and Knox, R.C. 1992. Transport and
Remediation of Subsicrface Contaminants. ACS Symposium Series American
Chemical Society). An extraction fluid is applied to an undisturbed soil and
is
2i 76864
4
allowed to pass through the soil layer. The metals are solubilized from the
soil
matrix into the liquid phase of the extracting fluid. The extracting fluid
carries
the metals downward in the soil to a drainage or collection system. The
advantage of this method is that large areas of land can be treated while
being
left relatively undisturbed. The disadvantages are: a) in that soil is
anisotropic
and non homogeneous, it results in incomplete exposure of soil to the
extracting fluid, and therefore not all of the soil is treated, b) the
potential for
ground water contamination exists if not all extracting fluid is collected, c)
the
difficulty in having acceptable quality control quality assurance, and d) the
extracting fluid may cause soils to swell or plug the aquifer (Sabatini, et
al.,
1992) .
Stabilization uses reagents to minimize the rate of migration into the
environment and to reduce the toxicity of the contaminant (Lagrega, et al. ,
1994). Fixation refers to the use of additives to improve handling and
physical
characteristics of waste, to decrease surface area over which mass transfer
can
occur, to limit solubility of metals and to reduce toxicity of the
contaminant.
Solidification uses a solidifying material to increase the strength and
decrease the compressibility and permeability of the contaminated soil
(Lagrega, et al., 1994). The specific reagent chosen for a site depends on its
ability to precipitate the contaminant. Some examples of stabilization methods
include vitrification, and the use of cement and other chemicals.
Vitrification is another in-situ method used to stop the heavy metals
from migrating further in the soil phase. Vitrification reduces the soil
volume
by 20-40% of its original size and produces an inert soil structure (O'Brien &
Gere Engineers, Inc. 1995. Innovative Engineering Technologies for
Hazardous Waste Remediation. Van Nostrand Reinhold.). The process
simultaneously reduces the volume, mobility and toxicity of the waste.
The use of cement is well suited for inorganic wastes such as heavy metals
(Lagrega, et al . , 1994) . The most common cement employed is "Portland"
cement which is a mixture of calcium, silicate, aluminum and iron oxides.
Extract and treat techniques include extracting the contaminated soil and
then treating it by different processes. After treatment the cleaned soil can
be
replaced and the contaminated portion can be land filled. In these processes
the bulk of the heavy metals are completely removed from the system and
therefore no long-term monitoring is required. One example of this method is
2176864
_ 5
soil washing.
Soil washing is divided into two classes. One of these classes is referred to
as the fines separation technique (Sabatini, et al., 1992). This process is
based
on the principle that the heavy metals are associated with the fine particles
of
the soil. In this process the clay and silt particles, the fines, are scrubbed
from
the soil, using a water based solution (Griffiths, 1995). The advantages of
this
process are that the volume of contaminated soil to be treated is greatly
reduced and the quality control/quality assurance is excellent (Sabatini, et
al. ,
1992). The disadvantages are that the soil is not completely remediated, the
mixing requires energy, and the equipment requires personnel with a large
level of expertise (Sabatini, D.A., and Knox, R.C. 1992. Transport and
Remediation of Subsurface Contaminants. ACS Symposium Series American
Chemical Society).
The second class is a newer, related technology of the first class. This
process is similar to vat leaching used in the mining industry (Sabatini, et
al.,
1992). The soil is extracted from the contaminated site and placed in an
agitation vessel with an extracting solution. Extracting solutions can be
acids,
bases, chelating agents, alcohols or others. The principle for this process is
that the metals will transfer from the soil to the extracting agent, as long
as a
concentration gradient exists. The contaminant must then be removed from the
liquid phase. This process has good quality control/quality assurance but is
also energy intensive and requires high level of operating skills (Sabatini,
et
al., 1992). This process is well suited for sandy soils while other types of
soil
may present chemical and physical problems.
As discussed above, there are many types of interactions that can bind
metals to soil particles and it is likely that several types of soil metal
interactions exist in the contaminated soil used in this experiment. Only some
interactions are reversible by adjusting the pH of the extracting agent. For
example, saturated ion exchange sites may release cations in response to a
drop in pH. Also, precipitated metals may dissolve when the pH lowered.
Although the metal binding and release mechanisms that occur in a
soil-acid slurry are likely extremely complex, they are assumed to behave as
an equilibrium-partitioning process.
When a finite volume of acid is mixed with a batch of soil, an
equilibrium state will be reached before all of the teachable lead is removed
2~~~86
6
from the solid phase, further limiting the removal of lead. However, the
concentration gradient can be restored by removing metal from the slurry.
Thus, in consideration of the above methods of cleaning soils, there has
been a need for a system which maintains a high concentration gradient
between the contaminant and leaching agent to improve the system efficiency.
Membranes are thin films through which certain substances can pass and
are vital components of filtration systems. Today, membranes are made from a
wide variety of materials including polymers, ceramics, metals and even
linings of animal and vegetable bodies. A useful application of filtering
systems is to separate solids from liquids. With this application, membranes
can be energy efficient since no phase change is required (Raycheba, J.M.T.
1990. Membranes Technology Reference Guide. Ontario Hydro). In the past,
however, membranes may be easily or readily fouled when used with very fine
particles to the extent that a complete loss in permeability results.
Accordingly, in the past, the use of membranes with fine particles has been
rejected.
Filtration systems are designed as either dead-end or cross-flow. In dead-
end filtration, the feed flow direction is perpendicular to the membrane
surface
and there is only one output stream; the permeate. In cross-flow filtration,
the
feed flows parallel to the membrane surface and there are two output streams;
the permeate and the concentrate. Cross-flow filtration reduces membrane
fouling (plugging of the membrane pores) (Raycheba, 1990). In cross-flow
Filtration, the linear speed is the average fluid velocity parallel to the
membrane surface.
Tubular modules are extremely simple membrane configurations that
employ cross-flow filtration. These are relatively easily cleaned when fouling
occurs. Fouling is a problem because it leads to increased maintenance and
operating costs. Fouling can be reduced by increasing linear speed but this
increases operating costs.
Accordingly, in view of the above, there has been a need for a simple
and effective process and apparatus to effect metal removal from soils.
Specifically, there has been a need for a process and apparatus utilizing a
membrane unit in which acceptable permeation rates are obtained without
fouling the membrane.
217684
A review of the prior art reveals that, in the past, membranes have not
been used to effect liquid/solid separation in the treatment of contaminated
soils.
SUNINIARY OF THE INVENTION
In accordance with the invention, an apparatus for removing metals
from soil is provided, the apparatus comprising:
a reactor for containing a slurry of leaching agent and soil;
a membrane separation unit in fluid communication with the reactor, the
membrane unit having a semi-permeable membrane permitting the
permeation of leachate therethrough;
means for circulating the slurry between the extractor and membrane
separation unit.
In alternate embodiments of the invention, the membrane separation unit
is preferably a polymeric or a ceramic cross flow filtration unit and the
means
for circulating includes a peristaltic pump and means for regulating pressure
within the membrane separation unit. In a preferred form, the apparatus
includes means for adding leaching agent to the reactor.
In another aspect of the invention, a process for removing metals from
soil is provided comprising the steps of:
a. forming a soil/leaching agent slurry in a reactor;
b. circulating the soil/leaching agent slurry between the reactor and a
membrane separation unit to effect separation of leachate from the soil;
wherein a constant volume of soil/leaching agent is maintained in the
reactor through the addition of fresh leaching agent.
The invention contemplates the treatment of soils contaminated with a
variety of metals wherein the metals may include any one of or a combination
of cadmium, chromium, cobalt, iron, mercury, manganese, molydinum, lead
or zinc. Still further, the leaching agent may be selected from any one, but
is
not limited to, acid, base, chelating agent or other suitable solvent or
solvent
2176864
s
systems.
In the specific case of treating lead contaminated soil, the preferred
leaching agent is hydrochloric acid and the pH in the reactor is 1.
Still further, it is preferred that the soil/leaching agent ratio
(weight/volume) is approximately 1:10 and that the pressure within the
membrane separation unit is between 2 and 20 psi.
BRIEF DESCRIPTION OF THE DRAWINGS
These and other features of the invention will be more apparent from the
following description in which reference is made to the appended drawings
wherein:
Figure 1 is a schematic view of the apparatus in accordance with the
invention;
Figure 2 is a plot of the lead concentration in the leachate as a function of
time comparing the batch and diafiltration processes;
Figure 3 is a plot of the percent lead removal as a function of time comparing
the batch and diafiltration processes;
Figure 4 is a plot of the percent chromium removal as a function of time
comparing the batch and diafiltration processes.
2~~6864
_ 9
DETAILED DESCRIPTION OF THE INVENTION
In accordance with the invention, an apparatus and process is herein
described for the removal of metals from soils. As described above, the
leaching process transfers metals from the solid to liquid phase of a slurry,
thereby removing the metals from the soil, this transfer being driven by the
concentration gradient of the metal between the solid and liquid phases. A
comparison of the effectiveness of a membrane separation process
(diafiltration) with a batch extraction process was investigated.
In the batch experiment, the soil and acid will come to equilibrium at
which point no further mass transfer will occur. The diafiltration process
overcomes this limitation by maintaining a high concentration gradient to
achieve a cleaner soil by continuously adding fresh acid to the slurry,
removing leachate and adding fresh leaching agent to maintain a constant
volume. The addition of leaching agent reduces the concentration of metal in
the liquid phase, therefore keeping the concentration gradient high. Various
metals suitable for leaching include but are not limited to cadmium,
chromium, cobalt, iron, mercury, manganese, molydinum, lead or zinc. The
apparatus 100 of the diafiltration process is shown in Figure 1.
With reference to Figure 1, the apparatus 100 includes a reactor 1 in
fluid communication with a membrane unit 3. Pump 11 is provided to maintain
fluid flow from the reactor 1 to the membrane unit 3 and from the membrane
unit 3 back to the reactor 1. Pressure valve 6 is used to control the system
pressure and is monitored by pressure gauge 5. The reactor 1 is provided with
stirring device 2 to maintain a soil/liquid slurry. Fresh leaching agent 9 is
pumped to the reactor 1 by pump 4 to maintain a constant reactor volume to
compensate for leachate removal 10 from the membrane unit 3.
The membrane unit 3 is preferably a tubular membrane unit such as an
ENKA''M (Enka, Dusseldorf, Germany) 0.2 ~,m unit (filtration surface area-
0.036 m2). This membrane unit has a braided polypropylene support with a
polypropylene microfiltration membrane. A ceramic MEMBR.ALOX'~ (Alcoa
Separation Technology Inc.) is also suitable. This membrane unit has an
average pore diameter of 0.05 ,um and filtration surface area of 0.0044 m2.
The MEMBRALOX ultrafilter elements have an asymmetric ceramic structure
composed of zirconium oxide. .
At the start of an experiment, a batch of clean leaching agent and
contaminated soil are added to the reactor 1. Constant stirring is required to
2176864
to
maintain a suspended slurry. The peristaltic pump 11 draws slurry la out of
the reactor 1 and passes it through the tubular membrane unit 3. The pressure
on the membrane determines the quantity of leaching agent removed as
leachate 10 wherein the higher the pressure, the higher the rate of permeation
of leachate.
A soil sample, contaminated with lead, was obtained from a location
outside Montreal, Quebec. A sieve analysis was conducted to determine the
particle size distribution of the sample. The results of the sieve analysis
are
presented in Table 1. As stated previously, the lead tends to bind to the
fines
of the soil. Accordingly, an initial digestion was conducted on the different
particle sizes to determine their respective lead concentrations. The results
of
the metal concentration as shown in Table 2 show that the concentration of
lead increased as the particle size decreased.
For experimental purposes, a soil sample with an adequate lead
concentration was required so that the removal of lead could be readily
monitored. From the results of the digestion of the different particle sizes,
fines passing an 80 mesh sieve were selected as containing a satisfactory
amount of lead for testing. The soil was prepared by removing as much rock
as possible as well as any other debris. The remaining soil was passed through
a rock crusher to reduce an adequate amount of soil to the proper size.
Batch Experiments
Batch experiments were conducted as a basis of comparison for the
diafiltration process. These tests were done by placing 100 g of soil with 900
ml solution concentrated of hydrochloric acid (--- lOM) at a pH of 1 (~0.1) in
a beaker. The pH of the slurry was maintained manually by adding strong
hydrochloric acid at each sampling time. The mixture was continuously stirred
and duplicate samples were taken using a syringe at approximate time intervals
of 5 min, 15 min, 30 min, 1 hr, 2 hrs, 5 hrs, 8 hrs and 24 hrs. The samples
were centrifuged to separate the acid and soil. The acid was decanted from the
soil mass to a vial. As a small amount of acid still remained with the soil,
leaching was thought to be still occurring. Accordingly, distilled water was
added to this mixture and stirred to stop or at least reduce further leaching.
This new mixture was centrifuged to separate the soil and water. The water
was decanted and the soil was removed and dried in an oven at 105°C
overnight. The soil was then digested and the liquid sample was analysed using
flame atomic absorption spectroscopy to determine respective concentrations
of the contaminant.
,;:
f
2 ~ 76864
11
Diafiltration Experiments
The diafiltration experiments were conducted by placing 100 g of
contaminated soil with 900m1 of hydrochloric acid at pH 1 into the reactor.
The mixture was continuously stirred to maintain a suspended slurry. As with
the batch experiments, the pH rose slightly between the sampling times. The
pH was monitored and maintained at 1 by adding a solution of hydrochloric
acid at each sampling time using pump 4. Samples were taken at the same
times and in the same fashion as the batch experiments. A number of runs
were conducted at a linear speed of 1.7 - 3.7 m/s and at a pressure between 2
and 8 psi.. The change in pressure altered the permeate flow and therefore the
hydraulic retention time (HRT). Hydraulic retention time is average fluid
residence time in the system and is calculated on the basis of the system
volume and permeate flow rate.
Test results reveal that in case of batch leaching, the concentration of
leached metals grows steadily until it reaches an equilibrium between metal
concentrations in the liquid and solid phases (Figure 2). In same experiments,
a decrease in the metal concentration in the liquid phase was observed after
several hours of leaching. This phenomenon should be attributed to the
resorption of metal ions onto soil particles. In diafiltration tests, metal
concentration first increased then decreased due to the removal of metals with
the permeate. This greatly reduced chances for resorption.
In light of the above, the incorporation of a membrane into the leaching
process is advantageous. When slurry is pumped through the module, heavy
metal ions are continuously removed from the system; therefore, their
accumulation is largely eliminated. Since the metal concentration in the
aqueous phase remains at a relatively low level, the driving force of the
membrane-assisted leaching (MASL) is higher than one of the batch leaching.
Figures 3 and 4 illustrate the concentration of metals remaining in the
soil, as a function of time, for the batch and diafiltration modes. For both
the
lead and chromium, the initial rate of metal removal was substantially higher
in case of diafiltration. After six hours of extraction, soil treated with
MASL
had only 29 % of initial chromium and 30 % of initial lead, compared to 39
and 48 % achieved in the batch process.
As indicated previously, the concentration gradient is the driving force
for the transfer of lead from soil to acid. The larger the concentration
__ 21 T 686
12
gradient, the faster the rate of mass transfer. A large gradient can be
maintained by keeping the concentration of lead in the liquid phase low. This
was accomplished by lower HRTs and it was therefore hypothesized that those
diafiltration experiments run at lower HRTs will have better removal of lead.
A table of the percent of lead removed versus time for a batch experiment and
three diafiltration experiments run at different HRTs is presented in Table 3.
As predicted, the shortest HRT of 28.51 hours had the best removal of lead of
approximately 80 % . The next shortest HRT had a removal of almost 70
while the longest HRT and batch had the lowest removals of about 40 % after
24 hours.
The terms and expressions which have been employed in this
specification are used as terms of description and not of limitations, and
there
is no intention in the use of such terms and expressions to exclude any
equivalents of the features shown and described or portions thereof, but it is
recognized that various modifications are possible within the scope of the
claims.
~.1'~~8~4
13
Table 1- Sieve Analysis of Contaminated Soil
Sieve Size Mass Retained (g) % Pass
4 70.3 ~ 95.56382912
16 602 57.57556635
40 566.4 21.83378557
80 203.8 8.973307251
120 45.4 ~ 6.108411687
200 33.7 3.981826213
P~ 63.1
S~ = 1 S 84.7
2176864
14
Table 2 - Metal Concentration of Different Particle Sizes
Sieve Size Concentration (mg/g)
200-pan 33.3
200-pan 29.9
200 to 80 2g,g
200 to 80 19
80 to 120 18.5
80 to 120 14
40 to 80
40 to 80 g
16 to 40
16 to 40 14
X176864
,S
Table 3
Remaining Lead
Time (hours) Batch processDiafiltration
'
HRT = 28.5 HRT = 40.9 HRT = 50'.5
0 0 0 0 0
0,25 ?2 64 55 70
0.5 70 52 53 65
1.0 61 55 SI 64~
2,0 40 60 47 63
4:0 55 41 45 57
6.0 56 SO 40 42
24 65 20 3 I 32
Note: Initial average lead concentration = 28 mg/g, pH =1, Pressure a 8 psi.
