United States
Environmental Protection
Agency
Water Engineering Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-84-196 Mar. 1985
SERA Project Summary
Sludge Demetalization by the
Union Carbide Corporation
Electrochemical Process
Oscar W. Haas
This study was undertaken to eval-
uate an electrochemical sludge de-
metalization process developed by
Union Carbide Corporation for re-
moving toxic metals such as cadmium,
nickel, and zinc from municipal waste-
water sludges. The process deposits
metals on the surface of an electrode
when slqdge having a pH of about 3 is
placed in an electrochemical cell and a
controlled electric potential is applied.
A laboratory-scale apparatus was
designed and constructed to determine
the most desirable electrode and cell
operating conditions for the process
tests and to investigate the various
phenomena that occur during electroly-
sis. These tests were conducted using
several reference electrode systems in a
solution of distilled water and ions so
that the process could be observed
without the complications attendant to
sludge. Dropping mercury and solid
electrodes were used to generate linear
polarization plots of current versus
applied voltage to establish the plating
potential appropriate for removing
metal ions. This potential was deter-
mined to be -1.45 volts versus the Cu-
CuSGu reference electrode.
An existing bench-scale apparatus
was modified to hold about 17 liters of
fluid, and batch tests were conducted
using both sludge and metal ion solution
as the electrolyte. Efforts centered on
achieving a balance of metals during the
experiments and on improving metal
removal rates and coulombic efficien-
cies. Coulombic efficiencies remained
below 10 percent and metal removal
performance was erratic. The use of the
Cu-CuSO* reference electrode system
to control total cell voltage and thus, in-
directly, the current density did appear
feasible. The quality and quantity of
data obtained did not allow a strong cor-
relation to be made between system
design, operating parameters, and
process performance.
This Project Summary was developed
by EPA's Water Engineering 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 use of sewage sludge as a soil
amendment has been practiced for many
years in certain parts of the world. In the
United States, recent studies reveal that
up to 25 percent of the Nation's sludge
production is disposed of on land. This
statistic does not include disposal in
designated landfills, but rather it applies
to reclamation of marginal or drastically
disturbed land as well as the use of
sludge as a fertilizer for croplands.
As population and the degree of
wastewater treatment sophistication
increases) the quantity of sewage sludge
produced by each community can be
expected to multiply dramatically. Fur-
thermore, economic and environmental
pressure may shift the emphasis from
presently preferred methods of disposal
(ocean dumping, 15 percent; landfill, 25
percent; incineration, 35 percent) to land
application. Naturally, such a shift in
emphasis will be accompanied by an
increased demand for suitable land
disposal sites. This problem is expected
to reach critical proportions in certain
-------
localized areas (such as in the north-
eastern United States), when large
populations are confronted with a relative
scarcity of land. In these areas, the
disadvantages of land application (lack of
public acceptance, odors, pathogens,
toxic metals, and the possible presence of
industrial chemicals) will become greatly
magnified. Concern over the fate of such
pollutants in the soil and groundwater
and their impact on vegetation and the
food chain is well placed and must be
definitively answered in the future.
The purpose of this study was to
investigate an electrochemical process
for the removal of heavy metals directly
from sewage sludge without previous
dewatering. Sludge does contain macro-
nutrients useful to plant growth such as
nitrogen, phosphorus, and potassium.
But it may also contain substantial
quantities of metals, some of which (most
notably Cd, Zn, Ni, Cu, and Cr) have been
found to be toxic to vegetation at elevated
concentrations and may accumulate in
the human food chain. Concern over the
migration of these elements into the food
chain has resulted in the proliferation of
many guidelines that recommend limits
on the amount of sludge applied to
farmland based on the amount and
species of metal contaminants involved.
The objective of the Union Carbide
demetalization process is to reduce the
concentrations of such metals sufficiently
to allow an increase in the amount of
sludge that may be applied to the
available acreage.
Process Description
The Union Carbide demetalization
process involves the direct electrowin-
ning of metals from aerobically digested,
acidified municipal sludge. Work done by
other investigators and in bench-scale
experiments at Union Carbide's Tona-
wanda Laboratories have confirmed that
the bulk of metals in municipal sludge are
locked in the solid phase rather than the
aqueous. These solid phase metals are
held primarily in two forms: insoluble
organometallic complexes and insoluble
hydroxide, carbonate, or sulfide com-
pounds. To reduce their concentration in
sewage sludge, metals must first be
liberated from this solid phase and
dissolved in aqueous solution.
The proposed sludge demetalization
process solubilizes the metal ions in
sewage sludge through a pair of pretreat-
ment steps. First, the sludge is aerobically
digested to oxidize organic material and
convert insoluble heavy metal sulfidesto
the soluble sulfate form. Thermophilic
(greater than 45°C) aerobic digestion
with high purity oxygen for 1 day is
sufficient to convert the metal sulfides to
their soluble forms. The temperature
increase in the sludge is produced by heat
liberated by the oxidation of organic
material in the sludge and is self-
sustaining as long as oxygen and sludge
are fed to the digester in appropriate
amounts.
The second stop involves a shift in the
equilibrium that exists between the
dissolved heavy metals and the insoluble
metallic complexes contained in the
sludge solid phase. This shift is produced
by adding acid, which lowers the solution
pH and causes the displacement of metal
ions by hT ions. The extent of metal
solubilization by this method is a function
of the pH to which the sludge is acidified.
The shift in metal complex equilibrium
occurs relatively rapidly upon acid
addition and is usually complete in 5 to 20
minutes.
At this point the proposed demetaliza-
tion process diverges from other published
metal removal methods for sludge, which
advocate the separation of the solid and
liquid phases by a dewatering step
followed by treatment of the liquid
supernatant by chemical precipitation,
electrolysis, or some other means. This
approach is limited by the efficiency of
sludge dewatering technology and by the
fact that no further metal transfer from
the solid to the liquid phase may occur
after separation. The proposed sludge
demetalization process involves the
continuous stripping of metal ions
directly from the sludge in an electrolytic
cell. This cell may consist of a series of
parallel plates arranged in electrode pairs
across which an electric potential is
applied. A reference electrode situated in
the bulk solution provides the set point for
control of the voltage applied between the
electrodes. The positive electrode of each
pair (the electron acceptor) is defined as
the anode, and the negative (electron
donor) is known as the cathode. Aerated
sludge is circulated between the plates to
enhance mass transfer of metal ions from
the bulk solution to the surface of the
cathodes, where they are removed by
electrodeposition. The equilibrium be-
tween the solid and aqueous phase metal
ion concentration acts to replenish the
soluble metal concentration by releasing
more metals from the solid phase. Thus
through the judicious choice of operating
conditions for the electrolytic cell (voltage,
pH, and temperature), the metal species
of interest may be substantially removed.
The demetalization process provides a
stabilized, pasteurized, metal-depleted
sludge that is ready for disposal exceptfor
lime treatment to neutralize pH. Further-
more, when the useful life of the
cathodes has been reached, they may be
regenerated, sold as scrap, or disposed of
along with any metal-concentrated depo-
sits in an environmentally secure manner.
Laboratory Tests
Experimental Apparatus and
Procedures
Stock sample solutions were prepared
containing 10 to 100 ppm each of
cadmium, zinc, copper, and other metals
in varying combinations. The pH of these
aqueous solutions were depressed by the
addition of H2SCu, and then deaeration
was achieved through agitation with
nitrogen. In some instances, suppressants
such as methyl-red or gelatin were added
to act as maxima suppressants or to
simulate the effect that organic matter
might have on the shape of the current
potential curves. These sample solutions
were placed in an electrochemical cell
(Figure 1) to obtain current-potential
relationships for the various metal ions
being investigated. The apparatus con-
sisted of an Aardvark potentiostat,
potential scanner, current integrator,
reference electrode system, and various
pieces of recording equipment. The cell
can be used with either a dropping
mercury electrode or a stationary elec-
trode.
Polarographic tests were conducted by
varying the potential and measuring the
current that results from the reactions at
the electrodes. The solution pH, metal ion
species, and concentration were varied
between test runs, and the results were
recorded as a series of sigmoidal waves,
each representing a different reaction.
Reference electrodes determined the
potential between the working electrode
(or cathode) and the electrolyte, whereas
potentiostats controlled this potential
during the course of a test. Only a very
small current is tolerated between the
working and reference electrodes.
Samples were analyzed by Union Car-
bide's Analytical Services Laboratory at
Tonawanda. Electrodes were washed
with a solution of nitric acid, and the
metals in the resulting solution were
measured by means of atomic absorption.
A microprocessor was used to analyze
absorption data and determine the metal
species. Electrodes and electrolyte were
analyzed for Cu, Zn, Pb, Ni, Cd, and Cr.
Laboratory Results
Small-scale (about 1 liter) tests werej
conducted to make controlled electro
-------
chemical measurements of electrode
potentials and to investigate the condi-
tions required for metal deposition.
Particular attention was paid to deter-
mining the extent to which unwanted
side reactions might reduce the coulombic
efficiency of the process. Laboratory-
scale tests were grouped in three main
categories: Polarographic, electrode, and
plating studies.
Polargraphic Reduction
Spectrum
These experiments determined the
actual half-wave potentials of metals in
solution. Because of the possibility of
chemical complexation, these potentials
Test Solution
Dropping
Mercury
Cathode
Dropping Mercury
M \\ VWV
*T+
Solid Electrode
Figure 1. Typical electrochemical cells.
could be very different from reference
values of the standard reduction potential.
To obtain the current-potential curves
(polarograms) for the reactions of interest,
a dropping mercury electrode apparatus
was designed and constructed. Tests
were performed with dilute aqueous
solutions of metal sulfates adjusted to a
pH of 4.0 with sulfuric acid. Dilute
solutions (10~5 to 10~2 M) of Cu+2, Cd+2,
and Zn+2 were studied extensively with
some additional testing of Ni*2, Cr+2 and
Fe+2 solutions.
The recorded polarograms were of the
shape anticipated, with currents remain-
ing low until the reduction potential was
approached, whereupon the current then
rose in a sigmoidal wave and reached a
plateau. This level remained reasonably
flat until the next reduction wave or until
Ha evolution occurred. The shape of a
typical test polarogram is illustrated in
Figure 2. The measured half-wave
potentials of Cu+2, Cd+2, and Zn+2 were
close to the values of the standard
reduction potentials calculated for these
ions with respect to HgSCU, Hg2CI2, and
CuS04 reference electrodes, whereas the
reductions of Cr+3, Ni+2, and Fe+2 were
considerably more cathodic than their
standard reduction potentials. Cr+3 gave a
broad wave between Cd+2 and Zn+2, and
Ni+2 gave a very broad wave more
cathodic than Zn+2. With Fe*2, a wave
could just be observed superimposed on
the final hydrogen wave.
The Zn+2 ion, the most difficult to
reduce based on standard hydrogen
electrode (SHE) potentials, can be reduced
to concentrations of 10~7 equivalents
per liter by the following potential/refer-
ence combinations:
Potential (Volts) Reference Electrode
-1.00 normal hydrogen electrode
-1.25 saturated Hg£lz
-1.32 saturated CuSOt
-1.64 saturated Hg^Ot
Hydrogen gas evolution was observed
at approximately -2.1 volts, indicating
that this reaction does not compete in a
major way for metals removed under
conditions suitable for plating zinc. The
concentrations of metal ions more
electronegative than zinc would be
expected to be reduced even further
under these conditions.
The major conclusion of this test series
is that the best cathode potential for
controlled potential deposition is approx-
imately -1.45 volts versus the Cu-CuS04
reference electrode.
Electrochemical Cell
Experiments - Electrode Tests
Attention was shifted in this series of
experiments to the reduction and plating
of metals on solid cathodes. Exploratory
tests were made with 3003-F aluminum
and various steel cathodes plating Cd+2
and Zn+2 ions at a pH of 4. Multiple
potentiometric scans (cyclic voltammetry)
24-1
12-
Figure 2.
0 -0.3 -0.7 -1.1 -1.5
Applied Voltage
Typical polarogram (current versus voltage plot).
-1.9
-2.3
-------
showed a distinct wave and plateau
corresponding to the reduction of Cd*2 to
Cd, followed by a current rise (probably a
result of H2 evolution). On sweeping in the
anodic direction, a peak corresponding to
the stripping of Cd from the cathode was
observed at a potential somewhat below
the reduction potential. The aluminum
cathodes appeared to behave satisfac-
torily, but with the steel cathodes, a
substantial anodic current was observed
at potentials below about -0.7 volts
versus the Cu-CuSC>4 reference because
of the corrosion of iron. Iron or steel are
therefore not satisfactory cathode mate-
rials.
Graphite anodes were used extensively
during these experimental runs. But the
solution was observed to darken as
plating progressed because of the partial
dissolution of the graphite. Since the cost
of using a noble metal as an anode mate-
rial would be prohibitive, aluminum or
sulfated lead anodes were tested as
alternatives to graphite. The aluminum
anode exhibited areas of highly.promoted
general attack and crevicing at its
surface. The sulfated lead anode tests
were marked by very poor mechanical
adherence of metallic deposits on the
aluminum cathode. The rate of metal
removal was also poor for the lead anode.
Electrochemical Cell
Experiments - Plating Studies
Plating studies were initiated using a 1 -
liter test apparatus manufactured by
Princeton Applied Research.* The test
solutions nominally contained 100 ppm
concentrations of cadmium and zinc ions
adjusted to pH 4 by dilute sulf uric acid. All
the solutions were deaerated before
testing with nitrogen and stirred using a
Teflon-coated magnetic stirrer. The
surface area of the aluminum cathodes
was 5 cm2. A pair of high-density graphite
rods served as anodes. The open-circuit
potential of the aluminum in this environ-
ment was -0.76 volts versus Cu-CuSO4
reference.
Previous polarographic tests on 10~3 M
Cd+2 and Zn+2 solutions indicated a
diffusion current density of about 2
amps/m2. Comparable current densities
with the solid aluminum cathodes were
anticipated, with current decreasing as
plating progressed as a result of metal ion
depletion in the solution. The current
measured during the plating studies was
initially in the range expected; but as
electrolysis proceeded, the current
'Mention of trade names or commercial products
does not imply endorsement or recommendation for
use.
greatly increased. The reason is believed
to be a large increase in the effective
cathode surface area. With Cd+2 and Zn+2
ions, a somewhat denser deposit was
formed. In all cases, the deposit was very
porous and apparently has a large surface
area.
When cadmium was selectively plated
at an applied potential of -0.80 volts
versus the Cu-CuSCu reference, the
coatings were not very adherent. Large
portions of the coating would detach
completely or partially from the cathode
surface. The observed current density
fluctuated with these events. The plated
solids that detached from the cathode did
not immediately redissolve, and they
appeared to have a long residence time in
solution. When both cadmium and zinc
were plated from solution at an applied
potential of -1.45 volts versus the Cu-
CuS04 reference, the coatings were more
adherent and uniform. The pores in the
coating surface were approximately
0.076 cm in diameter, and the cathode
substrate was visible through these
pores.
Though the mass balances in the above
tests were greater than 95 percent, the
coulombic efficiencies were relatively
poor at less than 20 percent. These test
results are summarized in Table 1.
Bench Scale Tests
Experimental Apparatus and
Procedures
The demetalization test apparatus
(Figure 3) consisted of a 7-liter, rectang-
ular, Plexiglas vessel equipped with nine
parallel plate electrodes being alternately
made of 1100 aluminum cathodes and
graphite anodes. A 2.5-cm diameter
copper sulfate reference electrode was
placed into solution near the electro rack.
Power was provided by a constant voltage
(or constant current) supply by means of a
network of switches and wires (not
shown) and a Magnar model 4700M re-
search potentiostat and its associated
meters and recorders. A gas sparger was
located at the bottom of the electrolytic
cell to allow oxygen or other gas to be
bubbled into solution. A separate holding
vessel connected to the electrolytic cell by
a common channel at the vessel bottom
contained a mixer and temperature con-
troller. A variable-speed pump forced the
fluid to recirculate from the holding
vessel back to the top of the electrolytic
cell where it again flowed downward be-
tween the parallel plate electrodes. A
Plexiglas manifold distributed the fluid
evenly over the electrode channels. After
initial testing, some modifications were
made to the apparatus to improve per-
formance. The nature of these improve-
ments were as follows:
1) A small Plexiglas shield was con-
structed around the tip of the reference
electrode to prevent interference by
rising gas bubbles.
2) Metal components (such as screws,
gauges, etc.) in the apparatus structure
were identified as a source of metal
contamination and were removed.
3) The stainless steel rotor of the
recirculation pump was replaced by
equivalent drive coated with GE Glyptol
1201-A insulating enamel to prevent
corrosion.
4) Alligator clips that had been used
to make anode connections were found to
be a source of cadmium contamination in
early tests. These clips were replaced by
insulated copper wire adhesive bonded
into a hole drilled into the carbon anode.
A silicone rubber cement was used to
seal the connection from the electrolyte.
5) Additional gas space volume was
added to the electrolytic reactor to
prevent foam overflow. Because of a
desire to operate without chemical
antifoam addition, the apparatus was
provided with a rotating paddle above the
liquid level, which beat down any
foaming that did occur.
Secondary sludge was obtained from
the Lockport, New York, Wastewater
Treatment Plant on a weekly basis and
kept in refrigerated storage. When ready
for use, the sludge was gravity-thickened
and fed to two 14-liter Microferm
fermenters equipped with variable speed
mixers, a gas sparger, and heating coil. In
these reactors, the sludge was aerobically
digested at 50 to 55°C for a period of
about 1 day. The sludge was then
screened through a 0.60-cm mesh and
fed to the demetalization reactor. The
recirculation pump, mixers, and sludge
heaters were adjusted to their desired
points and turned on. The pH of the
sludge was adjusted by the addition of
10N H2SO4. Gas purge from a regulated
cylinder provided agitation between the
electrodes.
After approximately 5 minutes of
operation to allow for homogeneity to
develop throughout the system, a sludge
sample was withdrawn as the initial
sample. The electrical current was then
applied, and the total cell voltage was
adjusted to maintain the desired reference
voltage. Readings and samples were then
taken at intervals during the course of
each test. Total cell voltage, reference
voltage, and total cell amperage we
recorded continuously while the following
-------
Table 1. Electrochemical Cell - Plating Studies
pH
Test ID Initial
1
II
III
IV
V
4
4
4
4
4
PH
Final
2.9
2.9
4.3
2.7
Metal in
Initial
Solution
0. 1337g
0. 1433
0. 1295
0.1410
0. 1260
Metal in
Final
Solution
0.0721 g
0.0099
O.OS67
O.O019
0.0503
Metal
Removed
from
Solution
0.0616g
0. 1334
0.0728*
0.1391
0.0757
Weight
of
Solids
Filtered
0.1810*
0.0760*
0.4199*
0.0694
Metals
Analyzed
from
Filtered
Solids
0.0602g
0. 1380
0.0780
0.0606
Mass
Balance
97.7%
96.7
56.1
80.1
Coulombs to
Total Remove
Time (Cd/Zn)
22.8 hrs
17.7
28.5
24.0
114
308
330
154
Total
Coulombs
2851
1641
1221
Coulombic
Efficiency Anodes
10.8%
20.1
12.6
graphite
graphite
graphite
A1
PbSOt
* includes weight of graphite filtered
* includes weight of aluminum filtered
Voltmeter
Power Source
a B -•
X.
Heating & pH Control
Sludge Recirculation
Figure 3. Demetallization test apparatus.
data were hand recorded: time, gas-flow
rate, sludge recirculation rate, tempera-
ture, pH, reactor volume, and the amount
of acid addition (if any). The volume of
each sludge sample was recorded, and at
the conclusion of each test, the reactor
was emptied, its contents were measured
and any water loss due to evaporation
was made up by the addition of distilled
water. The electrodes were removed, any
observations as to the condition of the
metal deposits were noted, and then each
was sent to the laboratory in a separate
beaker for analysis. All metals were
analyzed on a Perkin-Elmer 460 Atomic
Adsorption Spectrophotometer with
flame technique.
Bench-Scale Results
Upon completion of initial tests, several
important conclusions were made. First,
the total cell voltage was considerably
lower than that normally used in previous
tests, despite reference potentials (versus
the Cu-CuSO4 electrode) in the -1.0 to
1.5-volt range. This observation suggested
that much of the power input of previous
tests may have been wasted on unwanted
side reactions. The second important
conclusion was that the most electro-
negative metal, zinc, could indeed be
plated out at these low cell voltages.
Third, the amount of material plated (as
determined by removing and analyzing
electrode deposits) was substantially
more for certain metals than the amount
of metal removed (as determined by a
metals balance and solution concentra-
tions). This discovery led to an extensive
search for possible contamination sources
in the bench-scale apparatus.
The bulk of the remaining sludge tests
were oriented toward eliminating various
sources of contamination such as reactor
screws, fining, recirculation pump, and
electrode clips. That this program was
fairly successful is shown by a change in
the metal balances from negative to
positive in later tests, indicating contam-
ination-free operation. Unfortunately the
coulombic efficiencies, defined as (cou-
lombs metals plated/coulombs input to
electrodes) x 100, were all extremely low,
with the highest being less than 10
percent. Furthermore, the percentage of
metals removed from solution was also
disappointingly low.
After the eighth demetalization test,
the electrolyte was changed to a synthetic
solution of cadmium, copper, zinc, and
nickel ions in distilled water. The purpose
of this change was to allow a more
accurate determination of a reactor metal
balance and to provide visual monitoring
of the electrodes during testing. Despite
the elimination of any possible inter-
ference from the sludge solids, the metal
removals and plating efficiencies still
remained low. This observation suggested
that a shortcoming of the electrochemical
process itself may prevent the attainment
of better performance rather than any
characteristics of the sludge.
Results obtained from the laboratory-
scale apparatus during this time allowed
hope for the improved, metal removal.
These small-scale tests showed that a
porous, dendritic structure of plated
metals formed on the cathode, and that
the composition and structure of the
deposit depended on the electrode
materials used. During the course of a
test, some pieces of the metal deposit were
found to spall off the cathode because of
agitation of the bath. This spalled
material remained suspended in the
bath solution as particles or chips of
metals and was not counted as plated
metal. Furthermore, these particles acted
as points of high metal concentration in
the solution, making it nonhomogeneous
and increasing the difficulty of withdraw-
ing representative liquid samples for
analysis. This problem was addressed in
Tests 14 through 17. In these tests,
sampling procedures were modified to
account for the plated materials that
spalls off the electrodes and back into
bulk solution. This step was accomplished
-------
by filtering each sample, the final reactor
f I uid, a nd each reactor wash to col lect a ny
particulate matter. The liquid and elec-
trode samples were then analyzed in the
usual way, and the filtered solids were
analyzed for their metal content. Such
procedures did improve metal balances in
the 17-liter reactor to within 25 percent,
which may be a reasonable limit given the
accuracy of the large-scale analysis.
Conclusions
The sludge demetalization process
successfully removed metals from sewage
sludge without previous dewatering.
Copper, zinc, cadmium, nickel, chromium
and (to a smaller extent) lead were
shown to be removed from solution and to
accumulate on the electrodes. Unfortu-
nately, the metal concentration reductions
and the coulombic efficiency of the
plating process were low, resulting in
projected treatment costs that were
higher than most methods of sludge
disposal currently in use. Furthermore,
attempts to correlate metal removal
performance with system operating
parameters were largely unsuccessful.
Thus many improvements and additional
developmental work must be done before
the process can be commercially suc-
cessful. Other important conclusions
developed as a result of this study are as
follows:
1) The Cu-CuS04 reference electrode
was a useful tool in controlling total cell
voltage and thus, indirectly, current
density. Operating voltages in the 2 to 3
volt range were observed at a reference
voltage of 1.4 to 1.45 volts. Such
potentials were shown to be capable of
plating very electronegative ions (such as
zinc) at lower values than the 3 to 4 total
cell volts that were used when operating
without the reference electrode.
2) The identification and elimination of
possible sources of metal contamination
in the reactor design was a critical step in
attaining reasonable metal removals and
approaching a metal balance around the
reactor.
3) Despite the elimination of possible
metal contamination and a change in
process fluid from sludge to synthetic
metal ion solution, process performance
remained below expectations. This
result suggests that obstacles to higher
metal removals and coulombic efficiency
lie in the electrodeposition process itself
rather than in interferences from the
sludge solids.
4) In both laboratory and bench-scale
tests, flakes and particles of deposited
metal can spall off the electrodes and
remain suspended in the agitated electro-
lyte. This agitation is necessary to reduce
the boundary layer thickness at the
electrode surface and to enhance the
transport of metal ion species from the
bulk solution to that boundary layer. Thus
some compromise must be made between
the improvement of mass transport from
bulk agitation and the difficulty of
recovering the spalled-off metal that
results. Bench-scale testing indicates
that this resuspended material can
account for up to 10 percent of the metal
deposited on the electrodes.
5) Gas aeration plays a significant role
in improving bulk agitation, but perform-
ance does not seem to be sensitive to the
quantity of gas used within the range
tested. Nitrogen or air seem to be
interchangeable as the aeration gas.
6) Aluminum is the preferred material
for the cathode, and graphite for the
anode. Stainless steel was also tried out
as a cathode material, but it was deemed
unsuitable because of extensive corrosion
at low voltages. Aluminum was found to
be a poor anode, since it slowly dissolved
into solution. A sulfated lead electrode
with low solubility was also tried out as
an anode material, but the cathode
deposits formed during its use were less
adherent than those experienced with
graphite anodes under the same condi-
tions.
7) Laboratory-scale tests revealed that
deposits formed from cadmium solution
were less adherent than those formed
from a solution containing both cadmium
and zinc ions. Furthermore, as metals
plated out on the cathode, its surface area
increased, resulting in an increase in
current until the metal ions were depleted
from the solution.
8) In the bench-scale tests with
sludge, the solution pH was initially
dropped to the 3 to 4 range by the addition
of 10N H2S04. During the course of the
experimental runs, small amounts of acid
solution often had to be added to maintain
this low pH. Our hypothesis is that the
reduction of metal ions at the cathode
shifts the acid hydrolysis equilibrium,
causing more I-T ions to be consumed.
9) Sludge foam, when it did occur, was
successfully suppressed, either through
the addition of small amounts of anti-
foaming agent or by a flat-bladed impeller
rotating in the gas space above the fluid.
Recommendations
At present, the direct electrochemical
treatment of sewage sludge does not
seem to offer an economical means of
metal removal. But since the removal of
heavy metals from both municipal and
industrial sludges is becoming increasingly
critical in today's environment and since
alternatives are not plentiful, further
investigation of the process to improve its
performance may be warranted. The
relatively low metal removals and low
coulombic efficiency of this process seem
to be tied to the electrolytic cell perform-
ance rather than to any interference from
the sludge. Thus future experiments
should be undertaken to identify and
eliminate these obstacles to improved
performance.
The full report was submitted in
fulfillment of Contract No. 68-03-2968 by
Union Carbide Corporation under the
sponsorship of the U.S. Environmental
Protection Agency.
US GOVERNMENT PRINTING OFFICE 1965 - 559-111/10794
-------
Oscar W. Haas is with Union Carbide Corporation, Tonawanda, NY 14150.
Howard Wall is the EPA Project Officer fsee below).
The complete report, entitled "Sludge Demetalization by the Union Carbide
Corporation Electrochemical Process," (Order No. PB 85-137 347; Cost:
$14.50, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Water Engineering 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 PAI
EPA
PERMIT No G-35
Official Business
Penalty for Private Use $300
LC'J W TILL-Y "
* i „- T 7 N V ^ ° R
' L. (- . r ri, *" " ' ~ ' '
------- |