Electrocatalytic dechlorination of atrazine

Article abstract of DOI:10.1139/v02-008

Atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine), a photosynthetic inhibitor that is used in large quantities for weed control in corn and sorghum, is dechlorinated in aqueous solution upon electrolysis at a reticulated vitreous carbon cathode

Full text of DOI:10.1139/v02-008

200
Electrocatalytic dechlorination of atrazine
Naomi L. Stock and Nigel J. Bunce
Abstract: Atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine), a photosynthetic inhibitor that is used in large
quantities for weed control in corn and sorghum, is dechlorinated in aqueous solution upon electrolysis at a reticulated
vitreous carbon cathode in the presence of noble-metal catalysts. Electrocatalytic hydrogenolysis to 2-ethylamino-4-
isopropylamino-s-triazine occurs in quantitative yield, and is most efficient with palladium-based catalysts. Current effi-
ciency increases with atrazine and catalyst concentration, and decreasing current density. A previously unobserved phe-
nomenon with Pd catalysts is that current must be passed for a certain time before dechlorination commences. This lag
time is explained in terms of the palladium lattice absorbing a finite amount of hydrogen before catalytically active hy-
drogen atoms appear on the catalyst surface.
Key words: electrocatalytic hydrogenoloysis, dechlorination, atrazine, palladium catalysts.
Résumé : Soumise à une électrolyse en solution aqueuse, au niveau d’une cathode de carbone vitreuse et réticulée et
en présence de catalyseurs de métaux nobles, l’atrazine (la 2-chloro-4-éthylamino-6-isopropylamino-s-triazine), un inhi-
biteur de la photosynthèse utilisé en grandes quantités pour contrôler les mauvaises herbes dans le maïs et dans le
shorgo, subit une déchloration. L’hydrogénolyse électrocatalytique en 2-éthylamino-4-isopropylamino-s-triazine se pro-
duit avec un rendement quantitatif et les catalyseurs à base de palladium sont les plus efficaces. L’efficacité de courant
augmente avec les concentrations d’atrazine et de catalyseur et elle diminue avec la densité du courant. Avec les cata-
lyseurs de Pd, il se produit un phénomène qui n’a pas été observé antérieurement; on doit faire passer le courant pour
un certain temps avant que la déchloration ne débute. On explique ce décalage dans le temps en fonction d’une absorp-
tion dans un premier temps d’une quantité finie d’hydrogène par le réseau du palladium qui précéderait l’apparition
d’atomes d’hydrogène catalytiquement actifs à la surface du catalyseur.
Mots clés : hydrogénolyse électrocatalytique; déchloration; atrazine; catalyseurs de palladium.
[Traduit par la Rédaction] Stock and Bunce 206
Introduction
ECH always competes with the hydrogen evolution reac-
tion (HER), which proceeds via the Volmer-Heyrovsky-Tafel
mechanism (6).
Reactions involving electrochemically generated hydrogen
adsorbed to a metal catalyst surface (M) have been known
since the beginning of the twentieth century (1, 2). Two
classes of these reactions are known: electrocatalytic hydro-
genation and electrocatalytic hydrogenolysis (both abbrevi-
ated ECH). The former involves the addition of hydrogen to
π bonds (3–6), whereas the latter involves the reductive
cleavage of σ bonds, as in dechlorination reactions (7–11).
Both types of ECH reactions involve the electrochemical re-
duction of a protic solvent to produce adsorbed hydrogen at-
oms (Hads, eq. [1]) that react chemically with an adsorbed
organic substrate (R-X)ads, eq. [3]): see below, which is
adapted from ref. 6.
[5]
H2O + e– + M → HadsM + OH–
(Volmer reaction)
[6]
[7]
HadsM + H2O + e– → H2 + M
+ OH– (Heyrovsky reaction)
HadsM + HadsM → H2 + 2M (Tafel reaction)
The use of ECH for the remediation of chloroaromatic
compounds is potentially advantageous compared with tradi-
tional techniques such as adsorption on granular activated
carbon (GAC) or incineration. As with other electrochemical
technologies, ECH is useful as a “front-end” technology for
dechlorinating toxic organic compounds prior to biological
treatment (12). Comparing ECH with GAC adsorption, the
chloro compounds are dechlorinated rather than simply trans-
ferred to a different phase; while in comparison with incinera-
tion, ECH avoids the possible formation of toxic byproducts
such as chlorinated dibenzodioxins and dibenzofurans. As an
[1]
2H2O(or H3O+) + 2e– + M
→ 2HadsM + 2OH– (or H2O)
R-X + M → (R-X)ads
[2]
[3]
[4]
M
(R-X)adsM + 2HadsM → (R-H)adsM + HX
(R-H)adsM → R-H + M
Received 16 September 2001. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 21 February 2002.
N.L. Stock and N.J. Bunce.1 Electrochemical Technology Centre, Department of Chemistry and Biochemistry, University of
Guelph, Guelph, ON N1G 2W1, Canada.
1Corresponding author (e-mail: bunce@chembio.uoguelph.ca).
Can. J. Chem. 80: 200–206 (2002)
DOI: 10.1139/V02-008
© 2002 NRC Canada

Stock and Bunce
201
electrochemical process, ECH is a “green” technology, which
employs ambient temperatures, aqueous media, and elec-
trons rather than chemical additives to promote chemical
change. Provided that high current efficiencies can be
achieved, ECH is also cost and energy efficient, technically
flexible, and readily allows for the selective dechlorination
of many different organic compounds (7).
Atrazine (2-chloro-4-ethylamino-6-isopropylamino-s-triazine) is
used in large quantities in North America to control broadleaf
and grassy weeds, primarily in corn and sorghum, and as a
result has become a common contaminant in ground and sur-
face water (13). This is of particular concern, since this her-
bicide has been identified as a possible carcinogen (14) and
endocrine disruptor (15).
ments were conducted on a computer-controlled EG&G
Princeton Applied Research Potentiostat (Model 273). Data
were acquired using Model 270/250 Research Electrochemis-
try Software (version 4.10), converted to ASCII-text format
and plotted in Microsoft Excel 2000 (version 9.0.2720). A
saturated silver–silver chloride electrode, which was cali-
brated to 200 mV, was used as a reference electrode. Refer-
ence potentials were calculated against a standard hydrogen
electrode (SHE).
All electrocatalytic hydrogenations were carried out in a
flow-through batch cell (Fig. 1) designed by Patrick Dubé
(University of Sherbrooke) and fabricated by Yves Savoret
(University of Guelph). In this cell, the anode and cathode
compartments were separated with a Nafion 417 membrane.
In all experiments the cathode was RVC (80 ppi, 5 mm thick
and 36 mm in diameter) with a copper wire lead (enclosed in
glass tubing for stability) and the anode was platinum foil
(16 mm × 12 mm and 0.1 mm thick) with a platinum wire
lead. The anolyte solution was 20 mL of sulphuric acid
(1 M) and the catholyte solution was water–methanol
(75:5 mL), and contained sodium sulphate as supporting
electrolyte to a concentration of 0.15 M. Atrazine was dis-
solved in the methanol aliquot and catalyst powder was
added to the cathode compartment. A cross-shaped stir bar
was placed in the cathode compartment and upon stirring,
the catholyte solution and catalyst powder were forced
through the RVC cathode, resulting in the impregnation of
catalyst powder in the cathode. The catalyst remained im-
pregnated in the RVC cathode throughout the experiment.
When the charge was removed and the magnetic stirrer
turned off, the catalyst powder returned to solution. An
Amel Instrument General Purpose Potentiostat/Galvanostat
(Model 2049) was used for experiments carried out at con-
stant current. A computer controlled EG&G Princeton Ap-
plied Research Potentiostat (Model 273), equipped with
Model 270/250 Research Electrochemistry Software (version
4.10) was used for experiments carried out at constant po-
tential. A saturated silver – silver chloride electrode, which
was calibrated to 200 mV, was used as a reference electrode.
All potentials were calculated against a standard hydrogen
electrode (SHE). Samples of the pesticide solutions were re-
moved for HPLC analysis at appropriate time intervals dur-
ing all experiments.
Cl
N
N
HN
N
NH
In this work we studied electrocatalytic hydrogenolysis
(ECH) as a possible remediation method for the dechlorina-
tion of atrazine, using various noble-metal catalysts at a
reticulated vitreous carbon (RVC) electrode in aqueous solu-
tions. In particular, we examined the influence of substrate
concentration, catalyst concentration and reusability, and
current density on dechlorination efficiency.
Experimental
Materials
Analytical grade atrazine (99% pure) was purchased from
Supelco (Oakville, Ontario). Technical grade atrazine
(Aatrex-Nine-O, 90% pure) was generously donated by Peter
Smith (Plant Agriculture Department, University of Guelph),
and was purified by hot filtration and recrystallization from
methanol. An analytical standard of dechlorinated atrazine
(DCA) was generously donated by Dr. Nelson Johnson
(Novartis, Jonesboro, North Carolina). All catalyst powders
were 5% by weight and were obtained from Aldrich
(Oakville, Ontario). Sulphuric acid (H2SO4), lithium per-
chlorate (LiClO4), sodium sulphate (Na2SO4), and HPLC
grade acetonitrile and methanol were obtained from Fisher
Scientific (Toronto, Ontario). Nafion-417 membranes and
RVC (80 ppi) were obtained from the Electrosynthesis Com-
pany (Lancaster, New York). Platinum foil (99.95% pure),
platinum wire (99.95% pure, 0.368 mm diameter), and cop-
per wire (99.9% pure, 1.0 mm diameter) were purchased
from Alfa Aesar (Ward Hill, Massachusetts). Water used in
all experiments was purified using a Millipore Milli-Q Re-
agent Water System and had a resistivity of not less than 10
MΩ cm.
Degradation of the atrazine solutions was monitored using
a Waters Model 600 HPLC unit, equipped with a Waters
Model 486 variable UV–vis detector, a reverse phase Gene-
sis C18 column (4.6 × 250 mm) and a Waters silica pre-
column guard. For all analyses, the detector was set at λ =
220 nm and a methanol–water (80:20) mobile phase
(1 mL min–1 flow rate) was used. Prior to use, solvents were
filtered using Supelco Nylon 66 membranes (0.2 µm ×
47 mm) to remove microparticles. All injections were done
manually using a Rheodyne syringe (Waters), which injected
25 µL of sample into a 5 µL sample loop. Chromatogram
generation and peak integration was carried out with Waters
Millennium software (version 1.10).
Apparatus and procedures
Cyclic voltammetry was conducted on unstirred atrazine–
acetonitrile solutions (200 ppm) in a 20 mL open-top glass
cell. Both the working and counter electrodes were platinum
(0.94 and 1.26 cm2, respectively). In all experiments, 0.15 M
lithium chloride was used as the supporting electrolyte and a
sweep rate of 200 mV s–1 was employed. All CV experi-
Identification of the dechlorinated product was determined
by Perry Martos (University of Guelph) via LC–MS–MS
analyses on a VG Quattro II equipped with a quadrapole MS
detector and MassLynx software (version 3.5). The samples
were ionized using Atmospheric Pressure Chemical Ioniza-
© 2002 NRC Canada

202
Can. J. Chem. Vol. 80, 2002
Fig. 1. Schematic of cell used for ECH.
E1/2 value for atrazine would have been obtained in the
water–methanol solution used for electrolyses, but the reduc-
tion of the solvent masked the reduction of atrazine, and so
the aprotic solvent acetonitrile was employed instead. In
acetonitrile at a platinum cathode, a cyclic voltammogram of
200 ppm atrazine indicated an irreversible reduction process
with E1/2 = –1.3 V vs. SHE. Upon electrolysis in water–
methanol, no reduction was observed at E = –0.5 V, and the
rate of atrazine reduction was essentially independent of po-
tential at E ≤ –0.8 V (Fig. 3). Since these potentials are
much less negative than E1/2 for reduction of atrazine (albeit
in acetonitrile) and in the same range as for the reduction of
water (–0.83 V (16)), the observed dechlorination of
atrazine cannot be as a result of direct electroreduction.
PLATINUM LEAD
ANODE
COMPARTMENT
(1 M H2SO4)
GLASS-
ENCLOSED
COPPER
LEAD
PLATINUM ANODE
RVC
CATHODE
NAFION
MEMBRANE
Dechlorination efficiency (Table 1)
The influences of current density, catalyst concentration,
variation of the noble-metal catalyst, and initial substrate
concentration on the dechlorination efficiency of atrazine
were examined. In order to study the effect of current den-
sity, a series of 25 ppm atrazine solutions were electrolyzed
in the presence of 50 mg Pd/Al2O3 at 5, 10, 20, 50, 100, and
200 mA. Applied currents ≥20 mA led to ≥93% loss of the
initial atrazine within 30 min (Table 1, entries 1–6). At
10 mA, ECH treatment led to ~65% dechlorination, but at
5 mA dechlorination was negligible. Except for the case of
the 5 mA electrolysis, the current efficiency decreased as the
applied current increased, and was greatest (16%) at 10 mA
(Table 1, entry 2). A decrease in current efficiency with in-
creasing current density was also observed by Yusem and
Pintauro (4) upon ECH of soybean oil. The explanation of
this effect is that eq. [3], namely electrocatalytic hydrogena-
tion or hydrogenolysis, is in competition with hydrogen evo-
lution, eq. [6]. At high-current densities, the rate of the
electrochemical generation of adsorbed hydrogen becomes
faster than the rate at which the substrate can migrate to the
surface and become adsorbed, eq. [2]. Consequently, a
greater proportion of Hads is lost to the unproductive reaction
of hydrogen gas evolution, eq. [6].
The influence of catalyst concentration on the dechlorina-
tion of atrazine was investigated by employing 0, 25, 50,
100, 200, 400, and 600 mg of Pd/Al2O3 catalyst powder in
the ECH treatment of 80 mL of a 25 ppm atrazine solution
at 50 mA (Table 1, entries 4, 7–13). Catalyst concentrations
of 50 mg resulted in the dechlorination of ≥96% of the ini-
tial atrazine within 30 min. The control experiment (atrazine
without Pd/Al2O3) gave negligible reaction, further supporting
ECH over the electronation–protonation mechanism. Current
efficiency increased with catalyst concentration, with the ex-
ception of 600 mg, the maximum catalyst concentration in-
vestigated, and the best efficiency (7%) was obtained with a
catalyst concentration of 400 mg. We suggest that increasing
the amount of catalyst increases the production of adsorbed
hydrogen (note that at constant current the overall production
of hydrogen is independent of the amount of catalyst),
thereby promoting hydrogenolysis. At some point, we would
expect a further increase in the amount of catalyst to be un-
productive, due to clogging the pores of the RVC cathode;
this is presumably the explanation for the lower current effi-
ciency observed with 600 mg of catalyst.
ELECTROLYTE
SOLUTION
(H2O/MeOH/Na2SO4)
CROSS-SHAPED
STIRBAR
tion (APCI). APCI probe temperatures ranged from 200 to
300°C and cone voltages ranged from 25 to 50 V. Since the
samples analyzed by LC–MS contained only single com-
pounds, a column was not required. The mobile phase em-
ployed was acetronitrile–water (75:25) (0.6 mL min–1 flow
rate). The water portion of the mobile phase was spiked with
0.1% formic acid to facilitate ionization of the sample. Prior
to use, solvents were filtered to remove microparticles.
Calculations
The apparent current efficiencies for the dechlorination of
atrazine were calculated from the number of moles of sub-
strate reacted and the charge passed assuming that two elec-
trons were required for the removal of one chlorine. Changes
in volume were taken into consideration due to the removal
of samples for analysis.
Results and discussion
The ECH of a 25 ppm solution of atrazine at 50 mA, us-
ing as catalyst 50 mg Pd/Al2O3, resulted in the rapid produc-
tion of dechlorinated atrazine (DCA) (2-ethylamino-4-
isopropylamino-1,3,5-triazine). The identity of the product
was established as DCA by showing that it had the same
HPLC retention time and mass spectral properties as authen-
tic material. Following 30 min of ECH treatment, greater
than 95% of the atrazine was dechlorinated, and a complete
mass balance was obtained (Fig. 2).
Is it really ECH?
Cyclic voltammetry was employed to determine whether
the dechlorination of atrazine proceeded via ECH or by di-
rect electroreduction followed by protonation. Ideally the
© 2002 NRC Canada

Stock and Bunce
203
Fig. 2. Kinetic curves for the ECH (50 mA, 50 mg Pd/Al2O3) of
Fig. 3. Influence of constant potential on the degradation of
atrazine (25 ppm).
atrazine (25 ppm) using ECH (50 mg Pd/Al2O3).
120
100
80
60
40
20
0
120
100
80
60
40
20
0
0
10
Time (min)
20
30
0
5
10
15
20
25
30
Time (min)
-0.5V (vs SHE)
-1.0V (vs SHE)
-0.7V (vs SHE)
-1.2V (vs SHE)
-0.8V (vs SHE)
-1.5V (vs SHE)
atrazine
dechlorinated atrazine
Table 1. Influence of current density, catalyst concentration, various noble-metal catalysts, and initial substrate concentration on the
ECH of atrazine.
Current
(mA)
Catalyst
concentration (mg)
Initial atrazine
concentration (ppm)
Current
Atrazine
Entry
Catalyst
efficiency (%)a
dechlorinated (%)b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
50
50
50
50
50
50
0
25
50
100
200
400
600
400
400
400
400
400
400
400
400
5
16
6
4
4
4
0
1
3
4
5
7
6
2
4
2
2
7
64
93
97
98
100
0
62
97
96
100
100
100
28
31
31
35
97
97
98
85
5
10
20
50
100
200
50
50
50
50
50
50
50
20
20
20
20
20
20
20
20
Pd/Al2O3
Pd/Al2O3
Pd/Al2O3
Pd/Al2O3
Pd/Al2O3
Pd/Al2O3
Pd/Al2O3
Pd/Al2O3
Pd/Al2O3
Pd/Al2O3
Pd/Al2O3
Pd/Al2O3
Pd/Al2O3
Rh/Al2O3
Pt/Al2O3
Raney Ni
Ru/Al2O3
Pd/Al2O3
Pd/CaCO3
Pd/Al2O3
Pd/Al2O3
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
25
4
6
11
1
83
14
aRepresentative experiments were run in triplicate. In these experiments, the average standard deviation was ±0.4%.
bRepresentative experiments were run in triplicate. In these experiments, the average standard deviation was ±6.3%.
The electrocatalytic properties of various noble-metal cat-
alysts — Raney nickel; palladium, platinum, ruthenium, and
rhodium on aluminia; and palladium on calcium carbon-
ate — were investigated (Table 1, entries 14–19). Catalysts
on carbon powder supports were not included in the study
because atrazine rapidly and strongly adsorbed to carbon
powder resulting in its removal from solution without a
chemical change. Palladium was the most active ECH cata-
lyst (Fig. 4), and current efficiencies of 11 and 6% were
achieved with palladium on calcium carbonate and alumina,
respectively. This result can be attributed to palladium’s abil-
ity to both adsorb and absorb hydrogen (17). Similarly, Dabo
et al. (18) observed that palladium catalysts offered maxi-
mum current efficiencies in the ECH of 4-phenoxyphenol.
Palladized carbon felt was found to be an efficient catalyst
for the dechlorination of 4-chlorophenol (19) and 2,4-
dichlorophenoxyacetic acid (7–9); in the latter studies, cur-
rent efficiencies for dechlorination reaction followed the se-
quence Pd > Rh > Ru catalysts.
The influence of the initial substrate concentration on the
dechlorination of atrazine was examined using initial
atrazine concentrations of 4, 25, and 83 ppm; 50 mA; and
400 mg Pd/Al2O3 (Table 1, entries 18, 20, 21). Within
30 min, >80% of the initial atrazine was dechlorinated. Cur-
rent efficiency increased with increasing initial substrate
concentration, a result that can be explained in terms of in-
© 2002 NRC Canada

204
Can. J. Chem. Vol. 80, 2002
Fig. 4. Influence of various noble-metal catalysts on the ECH of
atrazine (400 mg of catalyst, 20 mA).
Table 2. Current efficiencies and first-order rate constants for the
loss of atrazine during long-term ECH treatments (50 mg
Pd/Al2O3, 20 mA) periodically spiked with concentrated atrazine
solution.
120
100
80
60
40
20
0
Maximum current
efficiency (%)
Rate constant
Hour
(k) (s–1)
1st
5
8
6
4
3
—
2nd
3rd
4th
5th
1.23
0.87
0.53
0.30
0
5
10
15
Time (min)
20
25
30
electrolysis at 50 mA gave almost complete dechlorination
of the starting material, with very similar current efficiency
(10% in run 2 compared with 8% in run 1).
Pd/Al2O3
Rh/Al2O3
Pd/CaCO3
Pt/Al2O3
Ru/Al2O3
Raney Ni:Al
In the second experiment, involving Pd/Al2O3 catalyst
(50 mg), the atrazine solution was electrolyzed continuously
at 20 mA, but every 60 min sufficient atrazine was added to
restore its concentration to the original value of 25 ppm. The
small amount of catalyst for this experiment was chosen so
that all of it was trapped in the pores of the RVC and none
was inadvertently removed when sampling the solution for
analysis. As illustrated in Fig. 5A, the concentration of
atrazine rapidly decreased after each addition, although
gradual loss of catalyst activity was evident. Figure 5B
shows that DCA continuously accumulated in solution dur-
ing the experiment, eventually reaching a concentration >
160 ppm. Maximum current efficiencies and apparent first-
order rate constants for loss of atrazine are listed in Table 2.
A first-order plot is illustrated in Fig. 6.
Fig. 5. Concentration of atrazine (A) and dechlorinated atrazine
(B) during long-term ECH treatment (20 mA, 50 mg Pd/Al2O3)
periodically spiked with concentrated atrazine solution.
30
(A)
25
20
15
10
5
0
0
60
120
180
240
300
360
Time (min)
180
160
140
120
100
80
Lag time in dechlorination
(B)
The ECH reaction of atrazine, catalyzed by Pd/Al2O3, was
characterized by a lag time that decreased with an increase
of both current density and catalyst concentration (Fig. 7).
The lag time (t) was inversely proportional to current density
(i) for a fixed catalyst concentration, suggesting that a fixed
amount of charge (Q) must pass through the solution before
dechlorination of atrazine begins, i.e., Q = it. In support of
this hypothesis, when atrazine (50 mA, 50 mg Pd/Al2O3)
was electrolyzed with preactivation of the cathode for 5 min
(the lag time observed in Fig. 7A under these conditions) be-
fore the introduction of atrazine, the lag time disappeared
and the degradation of atrazine was immediate. Furthermore,
the maximum current efficiency increased from 4 to 26%.
Likewise, in the experiment on long-term catalyst use
(Fig. 5A), a lag time was observed only with the initial solu-
tion; subsequent additions of atrazine were followed by im-
mediate dechlorination.
60
40
20
0
0
60
120
180
240
300
360
Time (min)
creased adsorption of the substrate on the catalyst (eq. [2]),
allowing dechlorination (eq. [3]) to compete more effec-
tively with hydrogen evolution. Previously, Mahdavi et al.
(20) found that the current efficiency for the electrocatalytic
hydrogenation of cyclohexenone increased with increasing
substrate concentration.
We interpret the decreased lag time with higher catalyst
concentration in terms of the proportion of the RVC surface
that is in contact with catalyst. Since current was constant in
the experiments shown in Fig. 7B, the total rate of hydrogen
production was the same, but only that produced at Pd was
catalytically useful. As noted already, the behaviour of Pd
catalysts was unique: no lag time was observed with the
other noble metals studied (although it was difficult to deter-
mine whether or not a lag time occured with Rh or Ru, be-
cause dechlorination was so slow). We therefore hypothesize
Catalyst reusability
Longevity of the catalyst is a critical parameter in any cat-
alytic process. We investigated the reusability of Pd/A2O3
for the ECH of atrazine in two different experiments. In the
first of these, the Pd/A2O3 catalyst (200 mg) was removed
by filtration using glass fiber paper after an ECH experiment
with 25 ppm of atrazine at 50 mA (catalyst recovery 98 ±
3%). After drying at 80°C, the recycled Pd/A2O3 catalyst
was used for a second similar ECH experiment; 30 min of
© 2002 NRC Canada

Stock and Bunce
205
Fig. 6. Superimposed first-order kinetic curves for the degrada-
tion of atrazine each hour during long-term ECH treatment
(20 mg, 50 mg Pd/Al2O3) periodically spiked with concentrated
atrazine solution.
Fig. 7. Occurrence of lag time in the degradation of atrazine. In-
fluence of current density (50 mg Pd/Al2O3) (A) and catalyst
concentration (50 mA) (B).
120
(A)
-8
-10
-12
-14
100
80
60
40
20
0
0
5
10
15
20
25
30
Time (min)
5 mA
10 mA
20 mA
50 mA
100 mA
200 mA
0
20
40
60
80
100
Time (min)
120
100
80
60
40
20
0
(B)
2nd Hour
3rd Hour
4th Hour
5th Hour
that the unique ability of palladium to both adsorb and ad-
sorb hydrogen (17) accounts for both its superior electro-
catalytic properties and the observed lag time. By analogy to
a bucket filling with water and then overflowing, we pro-
pose that the palladium must first absorb the electrochemi-
cally produced hydrogen; once the palladium lattice is
“full,” hydrogen can be adsorbed at the palladium surface,
allowing dechlorination to begin. The following calculation
supports the foregoing concept. Consider the experiments in
Fig. 7B involving 400 mg of catalyst. We assume that under
these conditions the RVC pores are filled with catalyst and
hence all the electrolytic hydrogen is formed at a Pd surface.
The lag time is 3 min, which at 50 mA corresponds to Q = 9
C, or 9 × 10–5 mol H. The mass of Pd (5% on alumina) is
20 mg = 2 × 10–4 mol, i.e., mol H ~ mol Pd. This is consis-
tent with H occupying the interstices of the Pd lattice, but
not with Pd merely covering the surface of Pd particles.
In considering why the lag phenomenon had not previ-
ously been reported, we noted that most ECH reactions are
characterized by reaction times on the order of several hours
with samples taken every hour (8) or only at the beginning
and end of the experiment (4). In our experiments, complete
dechlorination was observed in 30 min and samples were
taken every 2 min, making the lag effect easy to observe.
Related observations have been reported by Tsyganok et al.
(7), in whose dechlorination of 2,4-dichlorophenoxyacetic
acid (2,4-D), the electrocatalytic activity of carbon-
supported palladium was dramatically enhanced 2 h into the
ECH treatment. These authors, who had also noted that pre-
activation of the cathode resulted in accelerated dechlorina-
tion, suggested that preactivation led to the formation of a
hydrogen-saturated palladium phase and a change in spatial
distribution of the supported Pd-clusters.
0
5
10
15
Time (min)
20
25
30
Pd/Al2O3
Pt/Al2O3
Pd/CaCO3
Raney Ni:Al
Ru/Al2O3
Rh/Al2O3
not clear whether it has sufficient efficiency to be useful for
contaminated ground water. The reaction product, DCA, is
formed in quantitative yield. The DCA molecule still retains
the s-triazine moiety, and it is not clear whether this reaction
product will be a photosynthesis inhibitor like the parent
molecule. Plant bioassays on the electrolyzed solutions of
atrazine are in progress and will be reported separately.
Acknowledgments
We thank Dr. H. Menard, Francois LaPlante, and Patrick
Dubé from the University of Sherbrooke for sharing their
design of the electrochemical cell, and the Natural Sciences
and Engineering Research Council of Canada (NSERC) for
financial assistance.
References
1. S. Fokin. Z. Elektrochem. 12, 749 (1906).
2. M.P. Breteau. Bull. Soc. Chim. Fr. Ser. 4, 764 (1911).
3. L.L. Miller and L. Christensen. J. Org. Chem. 43, 2059 (1978).
4. G.J. Yusem and P.N. Pintauro. J. Am. Oil Chemists Soc. 69,
399 (1992).
5. V. Anaatharaman and P.N. Pintauro. J. Electrochem. Soc. 141,
2729 (1994).
6. J.M. Chapuzet, A. Lasia, and J. Lessard. In Frontiers of elec-
trochemistry. Edited by J. Lipkowski and P.N. Ross. VCH Pub-
lishers Inc., New York. 1998. pp. 155–196.
Conclusion
We found that electrocatalytic hydrogenolysis is an effec-
tive method of dechlorinating atrazine solutions at ppm con-
centrations. The method may have application to
remediation, particularly of waste water streams, though it is
7. A.I. Tsyganok, I. Yamanaka, and K. Otsuka. J. Electrochem.
Soc. 145, 3844 (1998).
© 2002 NRC Canada

206
8. A.I. Tsyganok and K. Otsuka. Appl. Catalysis B: Environ. 22,
Can. J. Chem. Vol. 80, 2002
14. United States Environmental Protection Agency (USA EPA). The
triazine pesticides: atrazine, cyanazine, simazine, and propazine.
Government Printing Office, Washington DC. 1999. pp. 1–4.
15. J.C. Eldrige. Biol Reprod. 44, 133 (1991).
16. D. Dobos. Electrochemical data. Elsevier Scientific Publishing
Company, New York. 1975. p. 252.
15 (1999).
9. A.I. Tsyganok, I. Yamanaka, and K. Otsuka. Chemosphere, 39,
1819 (1999).
10. A. Cyr, P. Huot, G. Belot, and J. Lessard. Electrochim. Acta,
35, 147 (1990).
11. P. Dabo, A. Cyr, F. Laplante, F. Jean, H. Menard, and J.
Lessard. Environ. Sci. Technol. 34, 1256 (2000).
12. J.D. Rodgers, W. Jedral, and N.J. Bunce. Environ. Sci Technol.
33, 1453 (1999).
17. F.A. Lewis. The palladium hydrogen system. Academic Press,
New York. 1967. pp. 1–11.
18. P. Dabo, A. Cyr, J. Lessard, L. Brossard, and H.Menard. Can.
J. Chem. 77, 1225 (1999).
13. K.R. Solomon, D.B. Baker, R.P. Richards, K.R. Dixon, S.J.
Klaine, T.W. La Point, R.J. Kendall, C.P. Weisskopf, J.M.
Giddings, J.P. Giesy, L.W. Hall, and W.M. Williams. Environ.
Tox. Chem. 15, 31 (1996).
19. I.F. Cheng, Q. Fernando, and N. Korte. Environ. Sci. Technol.
31, 1074 (1997).
20. B. Mahdavi, P. Chambrion, J. Binette, E. Martel, and J.
Lessard. Can. J. Chem. 73, 846 (1995).
© 2002 NRC Canada