- Email: [email protected]
Electrochemical oxidation of cyanide on 3D Ti–RuO2 anode using a filter-press electrolyzer
Accepted Manuscript Electrochemical oxidation of cyanide on 3D Ti–RuO2 anode using a filter-press electrolyzer Tzayam Pérez, Rosa L. López, José L. Nava, Isabel Lázaro, Guillermo Velasco, Roel Cruz, Israel Rodríguez PII:
S0045-6535(17)30325-9
DOI:
10.1016/j.chemosphere.2017.02.136
Reference:
CHEM 18899
To appear in:
ECSN
Received Date: 26 October 2016 Revised Date:
8 February 2017
Accepted Date: 26 February 2017
Please cite this article as: Pérez, T., López, R.L., Nava, J.L., Lázaro, I., Velasco, G., Cruz, R., Rodríguez, I., Electrochemical oxidation of cyanide on 3D Ti–RuO2 anode using a filter-press electrolyzer, Chemosphere (2017), doi: 10.1016/j.chemosphere.2017.02.136. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Electrochemical oxidation of cyanide on 3D Ti–RuO2 anode using a filter-press electrolyzer
4
Tzayam Péreza, Rosa L. Lópeza, José L. Navaa,*, Isabel Lázarob, Guillermo Velascoc,
5
Roel Cruzb, Israel Rodríguezb
RI PT
1 2 3
6 7
a
Universidad de Guanajuato, Departamento de Ingeniería Geomática e Hidráulica, Av. Juárez 77, Zona Centro, 36000, Guanajuato, Guanajuato, México. E-mail:
9
[email protected]; [email protected]
SC
8
b
11
M AN U
10
Universidad Autónoma de San Luis Potosí, Instituto de Metalurgia-Facultad de
12
Ingeniería, Av. Sierra Leona 550, 78210, San Luis Potosí, San Luis Potosí, México.
13
E-mail: ilazaro@[email protected]; [email protected]; [email protected]
14
19 20 21
TE D
18
Minera San Xavier S.A. de C.V., 78440 Cerro de San Pedro, SLP, México. E-mail: [email protected]
EP
17
c
AC C
15 16
22
*Corresponding author: [email protected].
23
Tel: + 52-4731020100 ext. 2289; fax: + 52-4731020100 ext. 2209.
24
1
ACCEPTED MANUSCRIPT Abstract
26
The novelty of this communication lies in the use of a Ti–RuO2 anode which has not
27
been tested for the oxidation of free cyanide in alkaline media at concentrations similar
28
to those found in wastewater from the Merrill Crowe process (100 mg L-1 KCN and pH
29
11), which is typically used for the recovery of gold and silver. The anode was prepared
30
by the Pechini method and characterized by SEM. Linear sweep voltammetries on a Ti–
31
RuO2 rotating disk electrode (RDE) confirmed that cyanide is oxidized at 0.45 < E < 1.0
32
V vs SHE, while significant oxygen evolution reaction (OER) occurred. Bulk oxidation
33
of free cyanide was investigated on Ti–RuO2 meshes fitted into a filter-press
34
electrolyzer. Bulk electrolyzes were performed at constant potentials of 0.85 V and 0.95
35
V and at different mean linear flow rates ranging between 1.2 and 4.9 cm s−1. The bulk
36
anodic oxidation of cyanide at 0.85 V and 3.7 cm s−1 achieved a degradation of 94%,
37
with current efficiencies of 38% and an energy consumption of 24.6 kWh m−3.
38
Moreover, the degradation sequence of cyanide was also examined by HPLC.
SC
M AN U
TE D
39
RI PT
25
Keywords: cyanide oxidation, ruthenium dioxide anode, mining wastewater treatment,
41
liquid chromatographic analysis, filter-press electrolyzer.
43
AC C
42
EP
40
44
2
ACCEPTED MANUSCRIPT 1. Introduction
46
For more than a century, cyanide has been widely used in mining activities for
47
recovering gold and silver (Palomo-Briones et al., 2016); and the mill tailings produced
48
after the extraction process (Merril-Crowe) have to be treated as they pose a hazard,
49
given that they could still contain concentrations of cyanide as high as 100 mg L-1
50
(Akcil, 2003).
51
Cyanide is well known to be toxic to humans (Shifrin et al., 1996); hence, several
52
methods have been employed for treating cyanide-containing wastewater; among these,
53
biological methods (Akcil, 2003; Dash et al., 2009), physicochemical methods (Golbaz
54
et al., 2014; Hu et al., 2015) and the so-called advanced oxidation processes (AOP’s)
55
(Oturan and Aaron, 2014) have been proposed as suitable. The first, is certainly the
56
cheapest, however might not be applicable as most microorganisms capable of
57
biodegrading cyanide are sensitive to cyanide concentration with biodegradation or
58
growth rate decreasing above specific thresholds for each organism (Dash et al., 2009).
59
In the case of physicochemical methods, there is the issue of low removal efficiency and
60
HCN volatilization at low pH (Golbaz et al., 2014) and the problem of generating
61
intermediates o byproducts such as those of alkaline chlorination (Young and Jordan;
62
1995; Hu et al., 2015).
63
The potential of the AOP´s lies on the formation of highly reactive species that lead to
64
the destruction of cyanide and in the particular case of electrochemical advanced
65
oxidation processes (EAOP’s), there is the advantage of an in situ production of the
66
oxidant (Sirés et al., 2014; Martinez-Huitle et al., 2015). Pineda and Silva (2007) have
67
investigated the oxidation of cyanide with hydrogen peroxide, which is produced by
68
electrolysis on a reticulated vitreous carbon cathode by oxygen reduction. However, the
AC C
EP
TE D
M AN U
SC
RI PT
45
3
ACCEPTED MANUSCRIPT formation of hydrogen peroxide was limited by the pH range of H2O2 production,
70
typically at pH 3. This reaction is dangerous because of the formation of HCN, which
71
volatilizes at atmospheric pressure, producing an environmental hazard. Recently, Tian
72
et al. (2015) investigated cyanide removal with a copper/active carbon fiber cathode via
73
a combined oxidation through a Fenton-like reaction and the in situ generated copper
74
oxides at the anode. Meanwhile, Abdel-Aziz et al. in (2016) performed a study of
75
indirect electrochemical oxidation of free cyanide in an interval of pH between 4 and
76
12, employing graphite anodes to produce active chlorine, which reacts with the cyanide
77
to produce CO2 and N2.
M AN U
SC
RI PT
69
78
On the other hand, free and complexed cyanide have been treated by anodic oxidation
80
using graphite and reticulated vitreous electrodes (Öğütveren et al. 1999; Felix et al.,
81
2011); this method proved to be effective for the treatment of wastewater having
82
various compositions (e.g., heavy metals and organic compounds). The oxidation of
83
cyanide to cyanate, followed by the oxidation of cyanate to carbon dioxide in strong
84
alkaline solutions (pH>12), is described by the following reactions (Hwang et al., 1987;
85
Dhamo, 1996):
87
EP
AC C
86
TE D
79
CN− + 2OH− → CNO− + H2O + 2e−
E0 = −0.97 V
(1)
CNO− + 2OH− → CO2 + ½N2 + H2O + 3e−
E0 = −0.76 V
(2)
88
89
90 4
ACCEPTED MANUSCRIPT 91
Here, E0 is the standard electrode potential. The sum of reactions 1 and 2 results in the
92
following reaction
93
CN− + 4OH− → CO2 + ½N2 + 2H2O + 5e−
E0 = −0.84 V
(3)
RI PT
94
95
In this context, Valiüniené et al. (2013) investigated the oxidation of cyanide employing
97
a Pt and Ti-platinized anodes and stainless steel cathode and demonstrated the
98
convenience of using Platinized anode for the oxidation of cyanide. The same authors in
99
(2015) performed a fundamental study employing a platinized Ti electrode, founding
100
both the direct and indirect oxidation of cyanide through intermediate oxygen/hydroxyl
101
radicals. Worth mentioning that in spite of the proved convenience of using metallic
102
coated titanium electrodes (Valiüniené et al., 2013, 2015), for the anodic oxidation of
103
cyanide, scarce studies have been reported in the literature.
104
The novelty of this communication lies in the use of a 3D Ti–RuO2 anode fitted into a
105
filter-press electrolyzer for oxidizing free cyanide in alkaline media at concentrations
106
similar to those found in the wastewater produced after the recovery of gold and silver
107
(100 mg L-1 KCN and pH 11). Moreover, the degradation sequence of cyanide is also
108
examined by HPLC techniques. The anode was prepared by the Pechini method and
109
characterized by SEM. The effect of anodic potential and hydrodynamics on the
110
electrochemical degradation performance of cyanide was evaluated. The mineralization
111
current efficiency and energy consumption during the electrolysis were also calculated.
112
It is important to mention that 3D electrode is an ensemble of one Ti-RuO2 plate and 4
113
meshes of expanded Ti with a RuO2 film.
AC C
EP
TE D
M AN U
SC
96
5
ACCEPTED MANUSCRIPT 2. Experimental
115
2.1 Chemicals and solutions
116
Cyanide solutions were prepared at 3.85 mM KCN (100 mg L-1) in 45 mM NaNO3 at
117
pH 11; concentrations that are similar to those found in effluents from gold and silver
118
extraction processes. Because oxygen oxidizes cyanide all solutions were deoxygenated
119
by bubbling under a vigorous flow of nitrogen gas for 10 min before each experiment to
120
obtain experimental reproducibility during the electrolysis trials. Potassium cyanide
121
(KCN) was of analytical standard purity purchased from Sigma-Aldrich® and was used
122
as received. Sodium nitrate (NaNO3) was of analytical grade supplied by Fermont®. The
123
KCN solutions were prepared with deionized water and their pH was adjusted with
124
NaOH solution of analytical grade, also supplied by Fermont®. Other chemicals and
125
solvents were either HPLC or analytical grade supplied by Sigma–Aldrich®, Fermont®
126
and Fluka®. Standard solutions and mobile phases were prepared with high-purity water
127
obtained from a Millipore Milli-Q system with resistivity >18 mΩ cm-1 at 25 °C or
128
HPLC chromatography grade water supplied by Sigma-Aldrich®.
129
EP
TE D
M AN U
SC
RI PT
114
2.2 Equipment
131
2.2.1 Microelectrolysis studies
132
A typical three-electrode cell was employed where a Ti-RuO2 rotating disk electrode
133
(RDE) was used as the working electrode with an area of 0.071 cm2, while a graphite
134
rod and saturated calomel electrode (SCE) were used as the counter and reference
135
electrodes, respectively. All electrode potentials reported here were corrected vs the
AC C
130
6
ACCEPTED MANUSCRIPT 136
standard hydrogen electrode (SHE). A potentiostat−galvanostant Biologic® model SP-
137
150 was used. A Pine® variable speed controller was employed for RDE tests.
138
2.2.2 Filter-press electrolyzer
140
Fig. 1a and Fig. 1b show the undivided filter-press reactor, with a single electrolyte
141
compartment and the electrolyte flow circuit, respectively. The reactor consisted of two
142
polypropylene channel distributors with internal dimensions of 8.1 cm × 3.1 cm × 1 cm,
143
through which the electrolyte flows. The graphite felt used as cathode was attached at
144
the end of the channel distributor and glued with carbon cement to a stainless steel plate
145
as current collector, while one Ti-RuO2 plate and 4 meshes of expanded Ti with a RuO2
146
film, 3D Ti–RuO2, were used as anodes and located into one channel distributor, as
147
shown in Fig. 1a. The channel distributor containing the 3D Ti–RuO2 electrode had two
148
holes with a diameter of 0.1 cm for introducing two wires of Ti–RuO2, which were used
149
as current collectors. Two polypropylene end plates were used to complete the filter
150
press cell; in each extreme, a hole with a diameter of 0.55 cm allowed the entry and exit
151
of the electrolyte. A plastic rhomboid mesh supplied by Netlon® was placed within the
152
channel distributor, to promote turbulence between electrodes. The area of the working
153
electrode (4 expanded Ti meshes and the plate coated with RuO2) that was exposed to
154
the electrolyte was approximately 98.5 cm2. Table SM-1 lists the detailed characteristics
155
of the filter-press reactor. For providing a constant flow to the reactor, an Iwaki®
156
magnetic pump model MD-20RZ-115NL with a capacity of 11 L min−1 and a Blue
157
White® flow meter model F440N with a capacity of 0.2–1.0 L min−1 were installed in a
158
hydraulic circuit composed of 0.5 inch diameter chlorinated polyvinyl chloride (CPVC)
159
pipes. The valves and three-way connectors were also made of CPVC.
AC C
EP
TE D
M AN U
SC
RI PT
139
7
ACCEPTED MANUSCRIPT 2.3 Methodology
161
2.3.1 Preparation of the DSA electrodes
162
The titanium pieces were pretreated by dipping in a hot bath at 70 °C containing
163
concentrated hydrochloric acid for approximately 1 h, followed by and immediate
164
submersion in concentrated nitric acid at room temperature for another 15 min. Finally,
165
the titanium pieces were rinsed with distilled water and dried at room temperature. This
166
pretreatment facilitates surface roughness, which increases the adherence of RuO2 on
167
the substrate.
SC
RI PT
160
M AN U
168
A RuO2 film was deposited onto a Ti mesh, which was used in the filter-press reactor,
170
Fig. 1a, and onto a Ti rod, which was used in the RDE tests. Both electrodes were
171
prepared by PPM, using appropriate molar ratios of oxide. The precursor polymer
172
solution consisted of a mixture of citric acid (CA) in ethylene glycol (EG) at 60–70 °C.
173
After CA was completely dissolved, RuCl3×H2O was added to the mixture in a molar
174
ratio of EG:CA:Ru of 16:0.12:0.0296 while maintaining a temperature between 60–70
175
°C for 30 min. This mixture was then applied to the pretreated Ti substrates using a
176
brush. After the coating was applied, the electrodes were heated at 110 °C for 10 min in
177
a furnace to induce the precursor polymerization. Then, the electrodes were maintained
178
at 550 °C for 10 min for the calcination of the polymer and formation of metal oxides.
179
This procedure was repeated eight times; after the final coating, the metallic pieces were
180
maintained in the furnace for 1 h at 550 °C. This sequence was repeated until 32 layers
181
of coating were obtained. Precaution was taken to ensure that the temperature did not
182
exceed 600 °C to prevent the formation of TiO2, which markedly decreases the
AC C
EP
TE D
169
8
ACCEPTED MANUSCRIPT 183
electrocatalytic properties of the Ti–RuO2 coating due to passivation (Pechini, 1967). A
184
microanalysis of the RuO2 film was obtained by means of a scanning electron
185
microscope (SEM) ZEISS, model EVO15-HD.
RI PT
186
2.3.2 Microelectrolysis studies
188
The potential and current densities associated to the region where degradation of
189
cyanide occurs were established through anodic polarization curves on the Ti–RuO2
190
RDE. A typical three-electrode cell was used (see Section 2.2.1). Polarization curves
191
were obtained at a linear potential sweep rate of 5 mV s−1, starting from the open-circuit
192
potential (OCP) towards more positive values until reaching 1.2 V vs SHE. These
193
experiments were conducted at rotation rates of: 100, 200, 300, 400, and 500 rpm. For
194
the background electrolyte (solution without cyanide), a velocity of 300 rpm was
195
employed.
196
TE D
M AN U
SC
187
2.3.3 HPLC analysis
198
Ion chromatography analyses of CN- and CNO- ions were carried out by injecting 100
199
µL aliquots into a Perkin Elmer® Flexar LC coupled to an Adept® Cecil CE4710
200
conductivity detector and using a Hamilton® PRP-X110, 150 mm × 4.1 mm (i.d.),
201
anionic column at room temperature; previous to this analysis, the samples were filtered
202
with 0.45 µm PTFE filters. The chromatograms exhibited peaks related to cyanide and
203
cyanate at retention times of 3.38 and 3.8 min, respectively. The mobile phase was
204
composed of 4 mM p-hidroxibenzoic acid, 0.1 mM NaSCN, and 2.5% methanol at pH 9
205
with a flow rate of 1 mL min-1.
AC C
EP
197
9
ACCEPTED MANUSCRIPT 206
2.3.4 Bulk electrolysis
208
Two set of experiments were performed using the electrolysis system shown in Fig. 1b
209
by applying a constant anodic potential of 0.85 and 0.95 V vs SHE at different flow
210
rates between 1.2−4.9 cm s−1. The cyanide concentration during electrolysis was
211
determined by the HPLC analysis of section 2.3.3 and electrolysis was conducted using
212
the potentiostat–galvanostant of section 2.2.1.
SC
RI PT
207
M AN U
213
3. Results and discussion
215
3.1 Characterization of DSA
216
Fig. 2 shows a typical SEM image of the Ti–RuO2 electrode surface obtained by the
217
Pechini method. A uniformly distributed RuO2 coating with a slightly rough surface was
218
observed. The surface morphology of the layer was characterized by the presence of
219
small cracks and plates. In addition, some very small gaps were observed between the
220
particles, thus revealing the underlying Ti substrate. The presence of plates and small
221
gaps on the surface is probably attributed to the drastic heat treatment, which resulted in
222
the rapid release of CO2 gas formed by the decomposition of the organic polymer
223
(Chaiyont et al., 2013).
224
3.2 Microelectrolysis
225
Figs. 3b–f show the j–E curves for the oxidation of cyanide on the electrode surface in
226
the potential range of 0 < E < 1.2 V vs SHE at rotation rates of the RDE between 100
227
and 500 rpm. At potentials in the range of 0 < E < 0.7 V vs SHE, the current was not
AC C
EP
TE D
214
10
ACCEPTED MANUSCRIPT affected by hydrodynamics, indicating that only charge transfer occurs within such a
229
potential range. Conversely, for E > 0.7 V vs SHE, the j–E curves were significantly
230
affected by hydrodynamics; nevertheless a limiting current plateau was not observed. In
231
fact, at potentials in the range of 0.8 < E < 1.2 V vs SHE, the curves developed a semi-
232
plateau, whose slope raised as the rotation speed increased. This semi-plateau is
233
attributed to the presence of an additional process to the oxidation of cyanide. To
234
elucidate the above mentioned, a polarization curve in the absence of cyanide was also
235
obtained (Fig. 3a); a faradaic current was detected at E > 0.42 V vs SHE, which
236
exhibited a behavior typically associated with the OER in alkaline media (Eq. (4)) (Lee
237
et al., 2012). It is important to highlight that the polarization curve in Fig. 3a at E > 0.95
238
V showed a significant change in slope (1.37 to 4.34 mA cm-2 V-1), implying the
239
predominance of OER; this coincides with typical Tafel analysis where two slopes
240
during OER on DSA® type electrodes appeared (Hu et al., 2004). This latter indicates
241
that the OER inhibits the formation of the limiting current plateau in the cyanide-
242
containing solution. Likewise, the anodic curves show that this side reaction is favored
243
as the rotation rate increases.
246
SC
M AN U
TE D
EP
245
4OH− → O2 + 2H2O + 4e−
AC C
244
RI PT
228
E0 = 0.401 V (4)
247
According to the RDE tests, at pH 11, cyanide might be oxidized to cyanate, followed
248
by the subsequent oxidation of cyanate to CO2 and N2 according to Eqs. (1)–(2) (Hwang
249
et al., 1987; Dhamo, 1996) accompanied by OER. Given that the OER is less
250
quantitative at E ≤ 0.95 V, two sets of bulk electrolysis were performed in the filter11
ACCEPTED MANUSCRIPT 251
press electrolyzer by applying a constant potential of 0.85 and 0.95 V, to favor the
252
oxidation of cyanide.
253
3.3 Bulk electrolysis
255
Fig. 4a-d shows the influence of mean linear flow rate on the normalized decay of the
256
cyanide concentration, cyanate evolution, integral current efficiency and energy
257
consumption, respectively, as a function of electrolysis time for a constant anodic
258
working potential of 0.85 V vs SHE. From the analysis of Fig. 4a the cyanide
259
concentration decreased with time, and this depletion seemed to be in correlation with
260
the increase of flow rate from 1.9 to 3.7 cm s-1; afterwards, at 4.9 cm s-1 (Fig. 4a) the
261
depletion decreased with regard to that obtained at 3.7 cm s-1. This behavior suggests
262
that the OER is favored at 4.9 cm s-1, hence diminishing the degradation of cyanide. On
263
the other hand, Fig. 4b indicates that the cyanide is oxidized to cyanate, and this
264
oxidation increases with hydrodynamics; at the end of the electrolysis the cyanide-
265
cyanate relationship, obtained by HPLC tests, was around 1:1/2. The mass balance
266
analysis of Figs. 4a-b allows proposing the following reaction of cyanide oxidation at a
267
working electrode potential of 0.85 V vs. SHE employing a 3D Ti-RuO2 electrode:
269
SC
M AN U
TE D
EP
AC C
268
RI PT
254
CN + 3OH → CNO + CO + N + H O + 3.5e
(5)
270
271
With the simultaneous OER; this latter is confirmed with the poor current efficiencies
272
(<42%) showed in Fig 4c. In this paper, we considered that the initial N atom of cyanide 12
ACCEPTED MANUSCRIPT was predominantly converted into N2, as it has been reported in the literature (Hwang et
274
al., 1987; Dhamo 1996), although NO3− ion might be formed as well. The use of nitrates
275
(45 mM NaNO3), in the supporting electrolyte employed here, impeded the analysis of
276
NO3− by ion chromatography. Worth mentioning that the simultaneous oxidation of
277
cyanide and OER found here did not allow performing a coherent kinetics analysis of
278
cyanide degradation.
279
From the analysis of Fig. 4c the current efficiency increased with hydrodynamics,
280
although the simultaneous OER consumed part of the supplied current. As shown in
281
Fig. 4d, energy consumption slightly decreases with hydrodynamics. Here, the current
282
efficiency and energy consumption were calculated according to Eqs. (6) and (7),
283
respectively:
M AN U
SC
RI PT
273
285
ϕ =
∆
× 100
286
EP
!"## ∆ .$ϕ
TE D
284
287
=
288
where, ∆C = C0−C(t) in mol mL−1, VR is the reaction volume (1200 mL), and Q is the
289
charge employed during electrolysis in C, n represents the electrons transferred during
290
the electrochemical oxidation of cyanide (i.e., n = 3.5 according to Eq. (5) for the
291
electrolysis shown in Fig. 4a), Ecell is the cell potential in V, and 3.6 is a factor that
292
converts to units of kWh m−3.
(7)
AC C
× 100
(6)
13
ACCEPTED MANUSCRIPT The best electrolysis at a working electrode potential of 0.85 V vs SHE achieved 94% of
294
cyanide oxidation at a flow rate of 3.7 cm s–1, with current efficiency of 38% and energy
295
consumption of 24.6 kWh m−3, while a low limit was reached at a flow rate of 1.2 cm
296
s−1, resulting in a cyanide destruction of only 30% for the same electrolysis time of 480
297
min.
298
A similar analysis was performed for the bulk electrolysis at a working electrode
299
potential of 0.95 V vs SHE, Fig. 5a-d. From the analysis of Fig. 5a, it is observed that
300
the cyanide concentration decreased with time, but in this case with lower kinetics than
301
the obtained at E = 0.85 V. The degradation performance however was more affected in
302
this case by the side OER, reaching only a 70 % of cyanide degradation at the mean
303
linear flow rate of 3.7 cm s-1.
304
Fig. 5b confirms that the cyanide is oxidized to cyanate; however, the cyanate
305
concentration decreased with hydrodynamics, contrary to that obtained at E = 0.85 V.
306
This latter suggests that at E = 0.95 V the OER is favored with hydrodynamics
307
inhibiting the cyanide oxidation to cyanate. The cyanide oxidation to cyanate at the end
308
of electrolysis, Figs. 5a-b, gave the following molar relationships, around 1:1 at u = 1.2
309
cm s-1, 1:3/5 at u = 2.5 cm s-1, and 1:3/10 for u = 3.7. This stochastic pattern is attributed
310
to the massive OER (see Fig. 3a). From the analysis of Figs. 5a-b, in connection with
311
the molar relationships, the cyanide oxidation at u = 1.2 cm s-1 and u = 2.5 cm s-1, can
312
be described by Eq. (1) and (8), respectively. While at u = 3.7 cm s-1the cyanide
313
degradation follows the Eq. (9):
AC C
EP
TE D
M AN U
SC
RI PT
293
314
315
%& +
'
*
() → ' %&( + ' %( + ' & + ' ) ( + 3.2,
(8) 14
ACCEPTED MANUSCRIPT 316
317
%& +
* '
() →
-
%&( +
* -
%( +
*
-
& +
* -
) ( + 4.1,
(9)
318
From the analysis of Fig. 5c, the current efficiency increased with hydrodynamics,
320
although these were slightly lower than the obtained in the set of electrolysis performed
321
at a holding potential of 0.85 V, owing that at 0.95 V, the simultaneous OER consumed
322
most of the supplied current. As shown in Fig. 5d, Econs does not present significant
323
differences between the mean linear flow velocities. It is important to mention that the
324
determination of current efficiency and energy consumption from Fig. 5c-d considered n
325
equal to 2, 3.2 or 4.1 according to Eqs. (1), (8) and (9), respectively. A relatively high-
326
energy consumption was attained for this particular process, which is attributed to the
327
simultaneous OER.
328
The best electrolysis obtained here allowed 94% oxidation of cyanide (initially having
329
100 mg L-1) to yield cyanate and CO2 with a current efficiency an energy consumption
330
of electrolysis of 38% and 24.6 kW h m−3. This anodic oxidation of cyanide is similar to
331
that obtained on graphite Rashing rings (Öğütveren et al., 1999), and reticulated
332
vitreous carbon (Felix-Navarro, 2011) anodes, where the degradation achieved values of
333
92% and 100%. The anodic oxidation of cyanide decreased to 80% on Ti-platinized
334
anode (Valiūnienė et al., 2013), which is attributed to the simultaneous OER. Future
335
work should be done by proving secondary or ternary oxide coated titanium anodes in
336
order to avoid the OER.
AC C
EP
TE D
M AN U
SC
RI PT
319
337
15
ACCEPTED MANUSCRIPT 4. Conclusions
339
A 3D Ti–RuO2 anode fitted into a pre-pilot filter-press electrolyzer, which had not been
340
tested for the oxidation of free cyanide in alkaline media, resembling concentrations
341
similar to those found in wastewater from the recovery of gold and silver, allowed 94%
342
oxidation of cyanide with a current efficiency an energy consumption of electrolysis of
343
38% and 24.6 kW h m−3, respectively. These results were achieved at a flow rate of 3.7
344
cm s−1 and constant applied potential of 0.85 V. The anodic oxidation of cyanide to
345
CNO− and CO2, was confirmed by HPLC analysis; however, the electrochemical
346
cyanide degradation occurred simultaneously with the OER.
M AN U
SC
RI PT
338
347
Acknowledgments
349
The authors would like to thank the following organizations for their financial support:
350
CONACYT and UASLP through the project FORDECYT 190966, CONACYT project
351
240522, and the University of Guanajuato project 869/2016.
352
EP
TE D
348
References
354
Abdel-Aziz, M.H., Bassyouni, M., Gutub, S.A., El-Ashtoukhy, E.S.Z., Abdel-Hamid,
355
S.M.S., Sedahmed, G.H., 2016. Removal of Cyanide from Liquid Waste by
356
Electrochemical Oxidation in a New Cell Design Employing a Graphite Anode. Chem.
357
Eng. Comm. 203, 1045–1052.
358
Akcil, A., 2003. Destruction of cyanide in gold mill effluents: biological versus
359
chemical treatments. Biotechnol. Advances, 21, 501-511.
AC C
353
16
ACCEPTED MANUSCRIPT Chaiyonot, R., Badoe, C., Ponce de León, C., Nava, J. L., Recio, J., Sirés, I., Herrasti,
361
P., Walsh, F.C., 2013. Decolorization of methyl orange dye at IrO2-SnO2-Sb2O5 coated
362
titanium anodes. Chem. Eng. Technol. 36, 123-129.
363
Cui, J., Wang, X., Yuan, Y., Guo, X., Gu, X., Jian, L., 2014. Combined ozone oxidation
364
and biological aerated filter processes for treatment of cyanide containing electroplating
365
wastewater. Chem. Eng. J. 241, 184-189.
366
Dash,R.R. Gaur, A., Balomajumder, C., 2009. Cyanide in industrial wastewaters and its
367
removal: A review on biotreatment. J. Hazard. Mater. 163, 1-11.
368
Dhamo, N., 1996. Electrochemical oxidation of cyanide in the hydrocyclone cell. Waste
369
Manage. 16, 257-261.
370
Felix-Navarro, R.M., Lin, S.W., Violante-Delgadillo, V., Zizumbo-Lopez, A., Perez-
371
Sicairos, S., 2011. Cyanide degradation by direct and indirect electrochemical oxidation
372
in electro-active support electrolyte aqueous solutions. J. Mex. Chem. Soc. 55, 51-56.
373
Golbaz, S., Jafari, A.J., Rafiee, M., Kalantary, R.R., 2014. Separate and simultaneous
374
removal
375
coagulation/precipitation: Mechanisms and theory. Chem. Eng. J. 253, 251-257.
376
Hanela, S., Durán, J., Jacobo, S., 2015. Removal of iron-cyanide complexes from
377
wastewaters by combined UV-ozone and modified zeolite treatment. J. Environ. Chem.
378
Eng. 3, 1794-1801.
379
Hu, C., You, L., Liu, H., Qu, J., 2015. Effective treatment of cadmium-cyanide complex
380
by a reagent with combined function of oxidation and coagulation. Chem. Eng. J. 262,
381
96-100.
TE D
phenol,
chromium,
and
cyanide
from
aqueous
solution
by
AC C
EP
of
M AN U
SC
RI PT
360
17
ACCEPTED MANUSCRIPT Hu, J.M. , Zhang J.Q., Cao, C.N., 2004. Oxygen evolution reaction on IrO2-based
383
DSA® type electrodes: kinetics analysis of Tafel lines and EIS. Int. J. Hydrogen Energy
384
28, 791-797
385
Hwang, J.Y., Wang, Y.Y., Wan, C.C., 1987. Electrolytic oxidation of cuprocyanide
386
electroplating wastewaters under different pH conditions. J. Appl. Electrochem. 17,
387
684-694.
388
Lee, Y., Suntivich, J., May, K.J., Perry, E.E., Shao-Horn, Y., 2012. Synthesis and
389
Activities of Rutile IrO2 and RuO2 Nanoparticles for Oxygen Evolution in Acid and
390
Alkaline Solutions. J. Phys. Chem. Lett. 3, 399−404.
391
Martínez-Huitle, C.A., Rodrigo, M.A., Sirés, I., Scialdone, O., 2015. Single and
392
Coupled Electrochemical Processes and Reactors for the Abatement of Organic Water
393
Pollutants: A Critical Review. Chem. Rev. 115, 13362–13407.
394
Öğütveren, Ü.B., Törü, E., Koparal, S., 1999. Removal of cyanide by anodic oxidation
395
for wastewater treatment. Water Res. 33, 1851-1856.
396
Oturan, M. A., Aaron, J. J., 2014. Advanced oxidation processes in water/wastewater
397
treatment: Principles and applications. A review. Crit. Revi. Environ. Sci. Technol. 44,
398
23, 2577-2641.
399
Palomo-Briones, R., Ovando-Franco, M., Razo-Flores, E., Celis, L. B., Cruz, R.,
400
Lázaro, I., Briones, R., 2016. An overview of reclaimed wastewater reuse in gold heap
401
leaching. Miner. Process. Extractive Metallurgy Rev., 37, 274-285.
402
Pechini, M. P., 1967. US Patent 3330697 A.
AC C
EP
TE D
M AN U
SC
RI PT
382
18
ACCEPTED MANUSCRIPT Pineda-Arellano, C.A., Silva-Martínez, S., 2007. Indirect electrochemical oxidation of
404
cyanide by hydrogen peroxide generated at a carbon cathode. Int. J. Hydrogen Energy
405
32, 3163-3169.
406
Shifrin N.S., Beck B.D., Gauthier T.D., Chapnick S.D., Goodman G., 1996. Chemistry,
407
toxicology, and human health risk of cyanide compounds in soils at former
408
manufactures gas plant sites. Regulatory Toxicol. Pharmacology 23, 106-116.
409
Sirés, I., Brillas, E., Oturan, M., Rodrigo, M. A., Panizza, M., 2014. Eleactrochemical
410
advanced oxidation processes: today and tomorrow. A review. Environ. Sci. Pollut.
411
Res., 21, 8336-8367.
412
Tian, S., Li, Y., Zhao, X., 2015. Cyanide removal with a copper/active carbon fiber
413
Cathode via a combined oxidation of a Fenton-like reaction and in situ generated copper
414
oxides at anode. Electrochim. Acta 180, 746–755.
415
Valiūnienė, A., Baltrūnas, G., Keršulytė, V., Margarian, Ž., Valinčius, G., 2013. The
416
degradation of cyanide by anodic electrooxidation using different anode materials.
417
Process Saf. Environ. Protection 91, 269-274.
418
Valiūnienė, A., Margarian, Ž., Valiūnas, R., 2015. Electrooxidation of cyanide ion on a
419
platinized Ti electrode. Reac. Kinet. Mech. Cat. 115, 449–461.
420
Young, C.A., Jordan, T S., 1995. Cyanide remediation: current and past technologies,
421
in: Erickson, L.E., Tillison, D.L., Grant, S.C., McDonald J.P. (Eds.), Proceedings of the
422
10th Annual Conference on Hazardous Waste Research. pp. 104-129.
AC C
EP
TE D
M AN U
SC
RI PT
403
423
424 19
ACCEPTED MANUSCRIPT Figure captions
426
Fig. 1. (a) Exploded view drawing of the filter-press reactor, (b) electrolysis system
427
depicting the electrical and flow circuits throughout the filter-press reactor.
428
Fig. 2. SEM image of the Ti–RuO2 electrode.
429
Fig. 3. Linear sweep voltammograms at the Ti–RuO2 RDE at rotation rates of: (b) 100
430
rpm, (c) 200 rpm, (d) 300 rpm, (e) 400 rpm, and (f) 500 rpm. The potential scan (5 mV
431
s−1) was initiated from OCP (0 V vs SHE) in positive-going potential sweep.
432
Electrolyte: 3.85 mM (100 mg L-1) CN− in 45 mM NaNO3 at pH = 11 and T = 298 K.
433
These LSVs were compared with a solution in the absence of cyanide at 300 rpm (a).
434
ARDE = 0.071 cm2.
435
Fig. 4. Effect of hydrodynamics on (a) normalized decay of cyanide concentration, (b)
436
cyanate evolution, (c) current efficiency and (d) energy consumption versus electrolysis
437
time in the filter-press electrolyzer at mean linear flow rates of: 1.2 cm s−1 (▲), 2.5 cm
438
s−1 (●), 3.7 cm s−1 (■) and 4.9 cm s−1 (◊) at constant applied potential of 0.85 V vs SHE.
439
Electrolyte: 3.85 mM (100 mg L-1) CN− in 45 mM NaNO3 at pH = 11 and T = 303 K.
440
Area of the Ti–RuO2 electrode, 98.5 cm2.
441
Fig. 5. Effect of hydrodynamics on (a) normalized decay of cyanide concentration, (b)
442
cyanate evolution, (c) current efficiency and (d) energy consumption versus electrolysis
443
time in the filter-press electrolyzer at mean linear flow rates of: 1.2 cm s−1 (▲), 2.5 cm
444
s−1 (●), and 3.7 cm s−1 (■) at constant applied potential of 0.95 V vs SHE. Electrolyte:
445
3.85 mM (100 mg L-1) CN− in 45 mM NaNO3 at pH = 11 and T = 303 K. Area of the
446
Ti–RuO2 electrode, 98.5 cm2.
AC C
EP
TE D
M AN U
SC
RI PT
425
20
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights • Ruthenium dioxide film on a titanium mesh was used as anode. • Anodic oxidation of free cyanide in alkaline media achieved a degradation of 94%.
AC C
EP
TE D
M AN U
SC
• Identification of cyanate as product.
RI PT
• Apparent kinetic rate order of cyanide oxidation depends on the applied potential.
S0045-6535(17)30325-9
DOI:
10.1016/j.chemosphere.2017.02.136
Reference:
CHEM 18899
To appear in:
ECSN
Received Date: 26 October 2016 Revised Date:
8 February 2017
Accepted Date: 26 February 2017
Please cite this article as: Pérez, T., López, R.L., Nava, J.L., Lázaro, I., Velasco, G., Cruz, R., Rodríguez, I., Electrochemical oxidation of cyanide on 3D Ti–RuO2 anode using a filter-press electrolyzer, Chemosphere (2017), doi: 10.1016/j.chemosphere.2017.02.136. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Electrochemical oxidation of cyanide on 3D Ti–RuO2 anode using a filter-press electrolyzer
4
Tzayam Péreza, Rosa L. Lópeza, José L. Navaa,*, Isabel Lázarob, Guillermo Velascoc,
5
Roel Cruzb, Israel Rodríguezb
RI PT
1 2 3
6 7
a
Universidad de Guanajuato, Departamento de Ingeniería Geomática e Hidráulica, Av. Juárez 77, Zona Centro, 36000, Guanajuato, Guanajuato, México. E-mail:
9
[email protected]; [email protected]
SC
8
b
11
M AN U
10
Universidad Autónoma de San Luis Potosí, Instituto de Metalurgia-Facultad de
12
Ingeniería, Av. Sierra Leona 550, 78210, San Luis Potosí, San Luis Potosí, México.
13
E-mail: ilazaro@[email protected]; [email protected]; [email protected]
14
19 20 21
TE D
18
Minera San Xavier S.A. de C.V., 78440 Cerro de San Pedro, SLP, México. E-mail: [email protected]
EP
17
c
AC C
15 16
22
*Corresponding author: [email protected].
23
Tel: + 52-4731020100 ext. 2289; fax: + 52-4731020100 ext. 2209.
24
1
ACCEPTED MANUSCRIPT Abstract
26
The novelty of this communication lies in the use of a Ti–RuO2 anode which has not
27
been tested for the oxidation of free cyanide in alkaline media at concentrations similar
28
to those found in wastewater from the Merrill Crowe process (100 mg L-1 KCN and pH
29
11), which is typically used for the recovery of gold and silver. The anode was prepared
30
by the Pechini method and characterized by SEM. Linear sweep voltammetries on a Ti–
31
RuO2 rotating disk electrode (RDE) confirmed that cyanide is oxidized at 0.45 < E < 1.0
32
V vs SHE, while significant oxygen evolution reaction (OER) occurred. Bulk oxidation
33
of free cyanide was investigated on Ti–RuO2 meshes fitted into a filter-press
34
electrolyzer. Bulk electrolyzes were performed at constant potentials of 0.85 V and 0.95
35
V and at different mean linear flow rates ranging between 1.2 and 4.9 cm s−1. The bulk
36
anodic oxidation of cyanide at 0.85 V and 3.7 cm s−1 achieved a degradation of 94%,
37
with current efficiencies of 38% and an energy consumption of 24.6 kWh m−3.
38
Moreover, the degradation sequence of cyanide was also examined by HPLC.
SC
M AN U
TE D
39
RI PT
25
Keywords: cyanide oxidation, ruthenium dioxide anode, mining wastewater treatment,
41
liquid chromatographic analysis, filter-press electrolyzer.
43
AC C
42
EP
40
44
2
ACCEPTED MANUSCRIPT 1. Introduction
46
For more than a century, cyanide has been widely used in mining activities for
47
recovering gold and silver (Palomo-Briones et al., 2016); and the mill tailings produced
48
after the extraction process (Merril-Crowe) have to be treated as they pose a hazard,
49
given that they could still contain concentrations of cyanide as high as 100 mg L-1
50
(Akcil, 2003).
51
Cyanide is well known to be toxic to humans (Shifrin et al., 1996); hence, several
52
methods have been employed for treating cyanide-containing wastewater; among these,
53
biological methods (Akcil, 2003; Dash et al., 2009), physicochemical methods (Golbaz
54
et al., 2014; Hu et al., 2015) and the so-called advanced oxidation processes (AOP’s)
55
(Oturan and Aaron, 2014) have been proposed as suitable. The first, is certainly the
56
cheapest, however might not be applicable as most microorganisms capable of
57
biodegrading cyanide are sensitive to cyanide concentration with biodegradation or
58
growth rate decreasing above specific thresholds for each organism (Dash et al., 2009).
59
In the case of physicochemical methods, there is the issue of low removal efficiency and
60
HCN volatilization at low pH (Golbaz et al., 2014) and the problem of generating
61
intermediates o byproducts such as those of alkaline chlorination (Young and Jordan;
62
1995; Hu et al., 2015).
63
The potential of the AOP´s lies on the formation of highly reactive species that lead to
64
the destruction of cyanide and in the particular case of electrochemical advanced
65
oxidation processes (EAOP’s), there is the advantage of an in situ production of the
66
oxidant (Sirés et al., 2014; Martinez-Huitle et al., 2015). Pineda and Silva (2007) have
67
investigated the oxidation of cyanide with hydrogen peroxide, which is produced by
68
electrolysis on a reticulated vitreous carbon cathode by oxygen reduction. However, the
AC C
EP
TE D
M AN U
SC
RI PT
45
3
ACCEPTED MANUSCRIPT formation of hydrogen peroxide was limited by the pH range of H2O2 production,
70
typically at pH 3. This reaction is dangerous because of the formation of HCN, which
71
volatilizes at atmospheric pressure, producing an environmental hazard. Recently, Tian
72
et al. (2015) investigated cyanide removal with a copper/active carbon fiber cathode via
73
a combined oxidation through a Fenton-like reaction and the in situ generated copper
74
oxides at the anode. Meanwhile, Abdel-Aziz et al. in (2016) performed a study of
75
indirect electrochemical oxidation of free cyanide in an interval of pH between 4 and
76
12, employing graphite anodes to produce active chlorine, which reacts with the cyanide
77
to produce CO2 and N2.
M AN U
SC
RI PT
69
78
On the other hand, free and complexed cyanide have been treated by anodic oxidation
80
using graphite and reticulated vitreous electrodes (Öğütveren et al. 1999; Felix et al.,
81
2011); this method proved to be effective for the treatment of wastewater having
82
various compositions (e.g., heavy metals and organic compounds). The oxidation of
83
cyanide to cyanate, followed by the oxidation of cyanate to carbon dioxide in strong
84
alkaline solutions (pH>12), is described by the following reactions (Hwang et al., 1987;
85
Dhamo, 1996):
87
EP
AC C
86
TE D
79
CN− + 2OH− → CNO− + H2O + 2e−
E0 = −0.97 V
(1)
CNO− + 2OH− → CO2 + ½N2 + H2O + 3e−
E0 = −0.76 V
(2)
88
89
90 4
ACCEPTED MANUSCRIPT 91
Here, E0 is the standard electrode potential. The sum of reactions 1 and 2 results in the
92
following reaction
93
CN− + 4OH− → CO2 + ½N2 + 2H2O + 5e−
E0 = −0.84 V
(3)
RI PT
94
95
In this context, Valiüniené et al. (2013) investigated the oxidation of cyanide employing
97
a Pt and Ti-platinized anodes and stainless steel cathode and demonstrated the
98
convenience of using Platinized anode for the oxidation of cyanide. The same authors in
99
(2015) performed a fundamental study employing a platinized Ti electrode, founding
100
both the direct and indirect oxidation of cyanide through intermediate oxygen/hydroxyl
101
radicals. Worth mentioning that in spite of the proved convenience of using metallic
102
coated titanium electrodes (Valiüniené et al., 2013, 2015), for the anodic oxidation of
103
cyanide, scarce studies have been reported in the literature.
104
The novelty of this communication lies in the use of a 3D Ti–RuO2 anode fitted into a
105
filter-press electrolyzer for oxidizing free cyanide in alkaline media at concentrations
106
similar to those found in the wastewater produced after the recovery of gold and silver
107
(100 mg L-1 KCN and pH 11). Moreover, the degradation sequence of cyanide is also
108
examined by HPLC techniques. The anode was prepared by the Pechini method and
109
characterized by SEM. The effect of anodic potential and hydrodynamics on the
110
electrochemical degradation performance of cyanide was evaluated. The mineralization
111
current efficiency and energy consumption during the electrolysis were also calculated.
112
It is important to mention that 3D electrode is an ensemble of one Ti-RuO2 plate and 4
113
meshes of expanded Ti with a RuO2 film.
AC C
EP
TE D
M AN U
SC
96
5
ACCEPTED MANUSCRIPT 2. Experimental
115
2.1 Chemicals and solutions
116
Cyanide solutions were prepared at 3.85 mM KCN (100 mg L-1) in 45 mM NaNO3 at
117
pH 11; concentrations that are similar to those found in effluents from gold and silver
118
extraction processes. Because oxygen oxidizes cyanide all solutions were deoxygenated
119
by bubbling under a vigorous flow of nitrogen gas for 10 min before each experiment to
120
obtain experimental reproducibility during the electrolysis trials. Potassium cyanide
121
(KCN) was of analytical standard purity purchased from Sigma-Aldrich® and was used
122
as received. Sodium nitrate (NaNO3) was of analytical grade supplied by Fermont®. The
123
KCN solutions were prepared with deionized water and their pH was adjusted with
124
NaOH solution of analytical grade, also supplied by Fermont®. Other chemicals and
125
solvents were either HPLC or analytical grade supplied by Sigma–Aldrich®, Fermont®
126
and Fluka®. Standard solutions and mobile phases were prepared with high-purity water
127
obtained from a Millipore Milli-Q system with resistivity >18 mΩ cm-1 at 25 °C or
128
HPLC chromatography grade water supplied by Sigma-Aldrich®.
129
EP
TE D
M AN U
SC
RI PT
114
2.2 Equipment
131
2.2.1 Microelectrolysis studies
132
A typical three-electrode cell was employed where a Ti-RuO2 rotating disk electrode
133
(RDE) was used as the working electrode with an area of 0.071 cm2, while a graphite
134
rod and saturated calomel electrode (SCE) were used as the counter and reference
135
electrodes, respectively. All electrode potentials reported here were corrected vs the
AC C
130
6
ACCEPTED MANUSCRIPT 136
standard hydrogen electrode (SHE). A potentiostat−galvanostant Biologic® model SP-
137
150 was used. A Pine® variable speed controller was employed for RDE tests.
138
2.2.2 Filter-press electrolyzer
140
Fig. 1a and Fig. 1b show the undivided filter-press reactor, with a single electrolyte
141
compartment and the electrolyte flow circuit, respectively. The reactor consisted of two
142
polypropylene channel distributors with internal dimensions of 8.1 cm × 3.1 cm × 1 cm,
143
through which the electrolyte flows. The graphite felt used as cathode was attached at
144
the end of the channel distributor and glued with carbon cement to a stainless steel plate
145
as current collector, while one Ti-RuO2 plate and 4 meshes of expanded Ti with a RuO2
146
film, 3D Ti–RuO2, were used as anodes and located into one channel distributor, as
147
shown in Fig. 1a. The channel distributor containing the 3D Ti–RuO2 electrode had two
148
holes with a diameter of 0.1 cm for introducing two wires of Ti–RuO2, which were used
149
as current collectors. Two polypropylene end plates were used to complete the filter
150
press cell; in each extreme, a hole with a diameter of 0.55 cm allowed the entry and exit
151
of the electrolyte. A plastic rhomboid mesh supplied by Netlon® was placed within the
152
channel distributor, to promote turbulence between electrodes. The area of the working
153
electrode (4 expanded Ti meshes and the plate coated with RuO2) that was exposed to
154
the electrolyte was approximately 98.5 cm2. Table SM-1 lists the detailed characteristics
155
of the filter-press reactor. For providing a constant flow to the reactor, an Iwaki®
156
magnetic pump model MD-20RZ-115NL with a capacity of 11 L min−1 and a Blue
157
White® flow meter model F440N with a capacity of 0.2–1.0 L min−1 were installed in a
158
hydraulic circuit composed of 0.5 inch diameter chlorinated polyvinyl chloride (CPVC)
159
pipes. The valves and three-way connectors were also made of CPVC.
AC C
EP
TE D
M AN U
SC
RI PT
139
7
ACCEPTED MANUSCRIPT 2.3 Methodology
161
2.3.1 Preparation of the DSA electrodes
162
The titanium pieces were pretreated by dipping in a hot bath at 70 °C containing
163
concentrated hydrochloric acid for approximately 1 h, followed by and immediate
164
submersion in concentrated nitric acid at room temperature for another 15 min. Finally,
165
the titanium pieces were rinsed with distilled water and dried at room temperature. This
166
pretreatment facilitates surface roughness, which increases the adherence of RuO2 on
167
the substrate.
SC
RI PT
160
M AN U
168
A RuO2 film was deposited onto a Ti mesh, which was used in the filter-press reactor,
170
Fig. 1a, and onto a Ti rod, which was used in the RDE tests. Both electrodes were
171
prepared by PPM, using appropriate molar ratios of oxide. The precursor polymer
172
solution consisted of a mixture of citric acid (CA) in ethylene glycol (EG) at 60–70 °C.
173
After CA was completely dissolved, RuCl3×H2O was added to the mixture in a molar
174
ratio of EG:CA:Ru of 16:0.12:0.0296 while maintaining a temperature between 60–70
175
°C for 30 min. This mixture was then applied to the pretreated Ti substrates using a
176
brush. After the coating was applied, the electrodes were heated at 110 °C for 10 min in
177
a furnace to induce the precursor polymerization. Then, the electrodes were maintained
178
at 550 °C for 10 min for the calcination of the polymer and formation of metal oxides.
179
This procedure was repeated eight times; after the final coating, the metallic pieces were
180
maintained in the furnace for 1 h at 550 °C. This sequence was repeated until 32 layers
181
of coating were obtained. Precaution was taken to ensure that the temperature did not
182
exceed 600 °C to prevent the formation of TiO2, which markedly decreases the
AC C
EP
TE D
169
8
ACCEPTED MANUSCRIPT 183
electrocatalytic properties of the Ti–RuO2 coating due to passivation (Pechini, 1967). A
184
microanalysis of the RuO2 film was obtained by means of a scanning electron
185
microscope (SEM) ZEISS, model EVO15-HD.
RI PT
186
2.3.2 Microelectrolysis studies
188
The potential and current densities associated to the region where degradation of
189
cyanide occurs were established through anodic polarization curves on the Ti–RuO2
190
RDE. A typical three-electrode cell was used (see Section 2.2.1). Polarization curves
191
were obtained at a linear potential sweep rate of 5 mV s−1, starting from the open-circuit
192
potential (OCP) towards more positive values until reaching 1.2 V vs SHE. These
193
experiments were conducted at rotation rates of: 100, 200, 300, 400, and 500 rpm. For
194
the background electrolyte (solution without cyanide), a velocity of 300 rpm was
195
employed.
196
TE D
M AN U
SC
187
2.3.3 HPLC analysis
198
Ion chromatography analyses of CN- and CNO- ions were carried out by injecting 100
199
µL aliquots into a Perkin Elmer® Flexar LC coupled to an Adept® Cecil CE4710
200
conductivity detector and using a Hamilton® PRP-X110, 150 mm × 4.1 mm (i.d.),
201
anionic column at room temperature; previous to this analysis, the samples were filtered
202
with 0.45 µm PTFE filters. The chromatograms exhibited peaks related to cyanide and
203
cyanate at retention times of 3.38 and 3.8 min, respectively. The mobile phase was
204
composed of 4 mM p-hidroxibenzoic acid, 0.1 mM NaSCN, and 2.5% methanol at pH 9
205
with a flow rate of 1 mL min-1.
AC C
EP
197
9
ACCEPTED MANUSCRIPT 206
2.3.4 Bulk electrolysis
208
Two set of experiments were performed using the electrolysis system shown in Fig. 1b
209
by applying a constant anodic potential of 0.85 and 0.95 V vs SHE at different flow
210
rates between 1.2−4.9 cm s−1. The cyanide concentration during electrolysis was
211
determined by the HPLC analysis of section 2.3.3 and electrolysis was conducted using
212
the potentiostat–galvanostant of section 2.2.1.
SC
RI PT
207
M AN U
213
3. Results and discussion
215
3.1 Characterization of DSA
216
Fig. 2 shows a typical SEM image of the Ti–RuO2 electrode surface obtained by the
217
Pechini method. A uniformly distributed RuO2 coating with a slightly rough surface was
218
observed. The surface morphology of the layer was characterized by the presence of
219
small cracks and plates. In addition, some very small gaps were observed between the
220
particles, thus revealing the underlying Ti substrate. The presence of plates and small
221
gaps on the surface is probably attributed to the drastic heat treatment, which resulted in
222
the rapid release of CO2 gas formed by the decomposition of the organic polymer
223
(Chaiyont et al., 2013).
224
3.2 Microelectrolysis
225
Figs. 3b–f show the j–E curves for the oxidation of cyanide on the electrode surface in
226
the potential range of 0 < E < 1.2 V vs SHE at rotation rates of the RDE between 100
227
and 500 rpm. At potentials in the range of 0 < E < 0.7 V vs SHE, the current was not
AC C
EP
TE D
214
10
ACCEPTED MANUSCRIPT affected by hydrodynamics, indicating that only charge transfer occurs within such a
229
potential range. Conversely, for E > 0.7 V vs SHE, the j–E curves were significantly
230
affected by hydrodynamics; nevertheless a limiting current plateau was not observed. In
231
fact, at potentials in the range of 0.8 < E < 1.2 V vs SHE, the curves developed a semi-
232
plateau, whose slope raised as the rotation speed increased. This semi-plateau is
233
attributed to the presence of an additional process to the oxidation of cyanide. To
234
elucidate the above mentioned, a polarization curve in the absence of cyanide was also
235
obtained (Fig. 3a); a faradaic current was detected at E > 0.42 V vs SHE, which
236
exhibited a behavior typically associated with the OER in alkaline media (Eq. (4)) (Lee
237
et al., 2012). It is important to highlight that the polarization curve in Fig. 3a at E > 0.95
238
V showed a significant change in slope (1.37 to 4.34 mA cm-2 V-1), implying the
239
predominance of OER; this coincides with typical Tafel analysis where two slopes
240
during OER on DSA® type electrodes appeared (Hu et al., 2004). This latter indicates
241
that the OER inhibits the formation of the limiting current plateau in the cyanide-
242
containing solution. Likewise, the anodic curves show that this side reaction is favored
243
as the rotation rate increases.
246
SC
M AN U
TE D
EP
245
4OH− → O2 + 2H2O + 4e−
AC C
244
RI PT
228
E0 = 0.401 V (4)
247
According to the RDE tests, at pH 11, cyanide might be oxidized to cyanate, followed
248
by the subsequent oxidation of cyanate to CO2 and N2 according to Eqs. (1)–(2) (Hwang
249
et al., 1987; Dhamo, 1996) accompanied by OER. Given that the OER is less
250
quantitative at E ≤ 0.95 V, two sets of bulk electrolysis were performed in the filter11
ACCEPTED MANUSCRIPT 251
press electrolyzer by applying a constant potential of 0.85 and 0.95 V, to favor the
252
oxidation of cyanide.
253
3.3 Bulk electrolysis
255
Fig. 4a-d shows the influence of mean linear flow rate on the normalized decay of the
256
cyanide concentration, cyanate evolution, integral current efficiency and energy
257
consumption, respectively, as a function of electrolysis time for a constant anodic
258
working potential of 0.85 V vs SHE. From the analysis of Fig. 4a the cyanide
259
concentration decreased with time, and this depletion seemed to be in correlation with
260
the increase of flow rate from 1.9 to 3.7 cm s-1; afterwards, at 4.9 cm s-1 (Fig. 4a) the
261
depletion decreased with regard to that obtained at 3.7 cm s-1. This behavior suggests
262
that the OER is favored at 4.9 cm s-1, hence diminishing the degradation of cyanide. On
263
the other hand, Fig. 4b indicates that the cyanide is oxidized to cyanate, and this
264
oxidation increases with hydrodynamics; at the end of the electrolysis the cyanide-
265
cyanate relationship, obtained by HPLC tests, was around 1:1/2. The mass balance
266
analysis of Figs. 4a-b allows proposing the following reaction of cyanide oxidation at a
267
working electrode potential of 0.85 V vs. SHE employing a 3D Ti-RuO2 electrode:
269
SC
M AN U
TE D
EP
AC C
268
RI PT
254
CN + 3OH → CNO + CO + N + H O + 3.5e
(5)
270
271
With the simultaneous OER; this latter is confirmed with the poor current efficiencies
272
(<42%) showed in Fig 4c. In this paper, we considered that the initial N atom of cyanide 12
ACCEPTED MANUSCRIPT was predominantly converted into N2, as it has been reported in the literature (Hwang et
274
al., 1987; Dhamo 1996), although NO3− ion might be formed as well. The use of nitrates
275
(45 mM NaNO3), in the supporting electrolyte employed here, impeded the analysis of
276
NO3− by ion chromatography. Worth mentioning that the simultaneous oxidation of
277
cyanide and OER found here did not allow performing a coherent kinetics analysis of
278
cyanide degradation.
279
From the analysis of Fig. 4c the current efficiency increased with hydrodynamics,
280
although the simultaneous OER consumed part of the supplied current. As shown in
281
Fig. 4d, energy consumption slightly decreases with hydrodynamics. Here, the current
282
efficiency and energy consumption were calculated according to Eqs. (6) and (7),
283
respectively:
M AN U
SC
RI PT
273
285
ϕ =
∆
× 100
286
EP
!"## ∆ .$ϕ
TE D
284
287
=
288
where, ∆C = C0−C(t) in mol mL−1, VR is the reaction volume (1200 mL), and Q is the
289
charge employed during electrolysis in C, n represents the electrons transferred during
290
the electrochemical oxidation of cyanide (i.e., n = 3.5 according to Eq. (5) for the
291
electrolysis shown in Fig. 4a), Ecell is the cell potential in V, and 3.6 is a factor that
292
converts to units of kWh m−3.
(7)
AC C
× 100
(6)
13
ACCEPTED MANUSCRIPT The best electrolysis at a working electrode potential of 0.85 V vs SHE achieved 94% of
294
cyanide oxidation at a flow rate of 3.7 cm s–1, with current efficiency of 38% and energy
295
consumption of 24.6 kWh m−3, while a low limit was reached at a flow rate of 1.2 cm
296
s−1, resulting in a cyanide destruction of only 30% for the same electrolysis time of 480
297
min.
298
A similar analysis was performed for the bulk electrolysis at a working electrode
299
potential of 0.95 V vs SHE, Fig. 5a-d. From the analysis of Fig. 5a, it is observed that
300
the cyanide concentration decreased with time, but in this case with lower kinetics than
301
the obtained at E = 0.85 V. The degradation performance however was more affected in
302
this case by the side OER, reaching only a 70 % of cyanide degradation at the mean
303
linear flow rate of 3.7 cm s-1.
304
Fig. 5b confirms that the cyanide is oxidized to cyanate; however, the cyanate
305
concentration decreased with hydrodynamics, contrary to that obtained at E = 0.85 V.
306
This latter suggests that at E = 0.95 V the OER is favored with hydrodynamics
307
inhibiting the cyanide oxidation to cyanate. The cyanide oxidation to cyanate at the end
308
of electrolysis, Figs. 5a-b, gave the following molar relationships, around 1:1 at u = 1.2
309
cm s-1, 1:3/5 at u = 2.5 cm s-1, and 1:3/10 for u = 3.7. This stochastic pattern is attributed
310
to the massive OER (see Fig. 3a). From the analysis of Figs. 5a-b, in connection with
311
the molar relationships, the cyanide oxidation at u = 1.2 cm s-1 and u = 2.5 cm s-1, can
312
be described by Eq. (1) and (8), respectively. While at u = 3.7 cm s-1the cyanide
313
degradation follows the Eq. (9):
AC C
EP
TE D
M AN U
SC
RI PT
293
314
315
%& +
'
*
() → ' %&( + ' %( + ' & + ' ) ( + 3.2,
(8) 14
ACCEPTED MANUSCRIPT 316
317
%& +
* '
() →
-
%&( +
* -
%( +
*
-
& +
* -
) ( + 4.1,
(9)
318
From the analysis of Fig. 5c, the current efficiency increased with hydrodynamics,
320
although these were slightly lower than the obtained in the set of electrolysis performed
321
at a holding potential of 0.85 V, owing that at 0.95 V, the simultaneous OER consumed
322
most of the supplied current. As shown in Fig. 5d, Econs does not present significant
323
differences between the mean linear flow velocities. It is important to mention that the
324
determination of current efficiency and energy consumption from Fig. 5c-d considered n
325
equal to 2, 3.2 or 4.1 according to Eqs. (1), (8) and (9), respectively. A relatively high-
326
energy consumption was attained for this particular process, which is attributed to the
327
simultaneous OER.
328
The best electrolysis obtained here allowed 94% oxidation of cyanide (initially having
329
100 mg L-1) to yield cyanate and CO2 with a current efficiency an energy consumption
330
of electrolysis of 38% and 24.6 kW h m−3. This anodic oxidation of cyanide is similar to
331
that obtained on graphite Rashing rings (Öğütveren et al., 1999), and reticulated
332
vitreous carbon (Felix-Navarro, 2011) anodes, where the degradation achieved values of
333
92% and 100%. The anodic oxidation of cyanide decreased to 80% on Ti-platinized
334
anode (Valiūnienė et al., 2013), which is attributed to the simultaneous OER. Future
335
work should be done by proving secondary or ternary oxide coated titanium anodes in
336
order to avoid the OER.
AC C
EP
TE D
M AN U
SC
RI PT
319
337
15
ACCEPTED MANUSCRIPT 4. Conclusions
339
A 3D Ti–RuO2 anode fitted into a pre-pilot filter-press electrolyzer, which had not been
340
tested for the oxidation of free cyanide in alkaline media, resembling concentrations
341
similar to those found in wastewater from the recovery of gold and silver, allowed 94%
342
oxidation of cyanide with a current efficiency an energy consumption of electrolysis of
343
38% and 24.6 kW h m−3, respectively. These results were achieved at a flow rate of 3.7
344
cm s−1 and constant applied potential of 0.85 V. The anodic oxidation of cyanide to
345
CNO− and CO2, was confirmed by HPLC analysis; however, the electrochemical
346
cyanide degradation occurred simultaneously with the OER.
M AN U
SC
RI PT
338
347
Acknowledgments
349
The authors would like to thank the following organizations for their financial support:
350
CONACYT and UASLP through the project FORDECYT 190966, CONACYT project
351
240522, and the University of Guanajuato project 869/2016.
352
EP
TE D
348
References
354
Abdel-Aziz, M.H., Bassyouni, M., Gutub, S.A., El-Ashtoukhy, E.S.Z., Abdel-Hamid,
355
S.M.S., Sedahmed, G.H., 2016. Removal of Cyanide from Liquid Waste by
356
Electrochemical Oxidation in a New Cell Design Employing a Graphite Anode. Chem.
357
Eng. Comm. 203, 1045–1052.
358
Akcil, A., 2003. Destruction of cyanide in gold mill effluents: biological versus
359
chemical treatments. Biotechnol. Advances, 21, 501-511.
AC C
353
16
ACCEPTED MANUSCRIPT Chaiyonot, R., Badoe, C., Ponce de León, C., Nava, J. L., Recio, J., Sirés, I., Herrasti,
361
P., Walsh, F.C., 2013. Decolorization of methyl orange dye at IrO2-SnO2-Sb2O5 coated
362
titanium anodes. Chem. Eng. Technol. 36, 123-129.
363
Cui, J., Wang, X., Yuan, Y., Guo, X., Gu, X., Jian, L., 2014. Combined ozone oxidation
364
and biological aerated filter processes for treatment of cyanide containing electroplating
365
wastewater. Chem. Eng. J. 241, 184-189.
366
Dash,R.R. Gaur, A., Balomajumder, C., 2009. Cyanide in industrial wastewaters and its
367
removal: A review on biotreatment. J. Hazard. Mater. 163, 1-11.
368
Dhamo, N., 1996. Electrochemical oxidation of cyanide in the hydrocyclone cell. Waste
369
Manage. 16, 257-261.
370
Felix-Navarro, R.M., Lin, S.W., Violante-Delgadillo, V., Zizumbo-Lopez, A., Perez-
371
Sicairos, S., 2011. Cyanide degradation by direct and indirect electrochemical oxidation
372
in electro-active support electrolyte aqueous solutions. J. Mex. Chem. Soc. 55, 51-56.
373
Golbaz, S., Jafari, A.J., Rafiee, M., Kalantary, R.R., 2014. Separate and simultaneous
374
removal
375
coagulation/precipitation: Mechanisms and theory. Chem. Eng. J. 253, 251-257.
376
Hanela, S., Durán, J., Jacobo, S., 2015. Removal of iron-cyanide complexes from
377
wastewaters by combined UV-ozone and modified zeolite treatment. J. Environ. Chem.
378
Eng. 3, 1794-1801.
379
Hu, C., You, L., Liu, H., Qu, J., 2015. Effective treatment of cadmium-cyanide complex
380
by a reagent with combined function of oxidation and coagulation. Chem. Eng. J. 262,
381
96-100.
TE D
phenol,
chromium,
and
cyanide
from
aqueous
solution
by
AC C
EP
of
M AN U
SC
RI PT
360
17
ACCEPTED MANUSCRIPT Hu, J.M. , Zhang J.Q., Cao, C.N., 2004. Oxygen evolution reaction on IrO2-based
383
DSA® type electrodes: kinetics analysis of Tafel lines and EIS. Int. J. Hydrogen Energy
384
28, 791-797
385
Hwang, J.Y., Wang, Y.Y., Wan, C.C., 1987. Electrolytic oxidation of cuprocyanide
386
electroplating wastewaters under different pH conditions. J. Appl. Electrochem. 17,
387
684-694.
388
Lee, Y., Suntivich, J., May, K.J., Perry, E.E., Shao-Horn, Y., 2012. Synthesis and
389
Activities of Rutile IrO2 and RuO2 Nanoparticles for Oxygen Evolution in Acid and
390
Alkaline Solutions. J. Phys. Chem. Lett. 3, 399−404.
391
Martínez-Huitle, C.A., Rodrigo, M.A., Sirés, I., Scialdone, O., 2015. Single and
392
Coupled Electrochemical Processes and Reactors for the Abatement of Organic Water
393
Pollutants: A Critical Review. Chem. Rev. 115, 13362–13407.
394
Öğütveren, Ü.B., Törü, E., Koparal, S., 1999. Removal of cyanide by anodic oxidation
395
for wastewater treatment. Water Res. 33, 1851-1856.
396
Oturan, M. A., Aaron, J. J., 2014. Advanced oxidation processes in water/wastewater
397
treatment: Principles and applications. A review. Crit. Revi. Environ. Sci. Technol. 44,
398
23, 2577-2641.
399
Palomo-Briones, R., Ovando-Franco, M., Razo-Flores, E., Celis, L. B., Cruz, R.,
400
Lázaro, I., Briones, R., 2016. An overview of reclaimed wastewater reuse in gold heap
401
leaching. Miner. Process. Extractive Metallurgy Rev., 37, 274-285.
402
Pechini, M. P., 1967. US Patent 3330697 A.
AC C
EP
TE D
M AN U
SC
RI PT
382
18
ACCEPTED MANUSCRIPT Pineda-Arellano, C.A., Silva-Martínez, S., 2007. Indirect electrochemical oxidation of
404
cyanide by hydrogen peroxide generated at a carbon cathode. Int. J. Hydrogen Energy
405
32, 3163-3169.
406
Shifrin N.S., Beck B.D., Gauthier T.D., Chapnick S.D., Goodman G., 1996. Chemistry,
407
toxicology, and human health risk of cyanide compounds in soils at former
408
manufactures gas plant sites. Regulatory Toxicol. Pharmacology 23, 106-116.
409
Sirés, I., Brillas, E., Oturan, M., Rodrigo, M. A., Panizza, M., 2014. Eleactrochemical
410
advanced oxidation processes: today and tomorrow. A review. Environ. Sci. Pollut.
411
Res., 21, 8336-8367.
412
Tian, S., Li, Y., Zhao, X., 2015. Cyanide removal with a copper/active carbon fiber
413
Cathode via a combined oxidation of a Fenton-like reaction and in situ generated copper
414
oxides at anode. Electrochim. Acta 180, 746–755.
415
Valiūnienė, A., Baltrūnas, G., Keršulytė, V., Margarian, Ž., Valinčius, G., 2013. The
416
degradation of cyanide by anodic electrooxidation using different anode materials.
417
Process Saf. Environ. Protection 91, 269-274.
418
Valiūnienė, A., Margarian, Ž., Valiūnas, R., 2015. Electrooxidation of cyanide ion on a
419
platinized Ti electrode. Reac. Kinet. Mech. Cat. 115, 449–461.
420
Young, C.A., Jordan, T S., 1995. Cyanide remediation: current and past technologies,
421
in: Erickson, L.E., Tillison, D.L., Grant, S.C., McDonald J.P. (Eds.), Proceedings of the
422
10th Annual Conference on Hazardous Waste Research. pp. 104-129.
AC C
EP
TE D
M AN U
SC
RI PT
403
423
424 19
ACCEPTED MANUSCRIPT Figure captions
426
Fig. 1. (a) Exploded view drawing of the filter-press reactor, (b) electrolysis system
427
depicting the electrical and flow circuits throughout the filter-press reactor.
428
Fig. 2. SEM image of the Ti–RuO2 electrode.
429
Fig. 3. Linear sweep voltammograms at the Ti–RuO2 RDE at rotation rates of: (b) 100
430
rpm, (c) 200 rpm, (d) 300 rpm, (e) 400 rpm, and (f) 500 rpm. The potential scan (5 mV
431
s−1) was initiated from OCP (0 V vs SHE) in positive-going potential sweep.
432
Electrolyte: 3.85 mM (100 mg L-1) CN− in 45 mM NaNO3 at pH = 11 and T = 298 K.
433
These LSVs were compared with a solution in the absence of cyanide at 300 rpm (a).
434
ARDE = 0.071 cm2.
435
Fig. 4. Effect of hydrodynamics on (a) normalized decay of cyanide concentration, (b)
436
cyanate evolution, (c) current efficiency and (d) energy consumption versus electrolysis
437
time in the filter-press electrolyzer at mean linear flow rates of: 1.2 cm s−1 (▲), 2.5 cm
438
s−1 (●), 3.7 cm s−1 (■) and 4.9 cm s−1 (◊) at constant applied potential of 0.85 V vs SHE.
439
Electrolyte: 3.85 mM (100 mg L-1) CN− in 45 mM NaNO3 at pH = 11 and T = 303 K.
440
Area of the Ti–RuO2 electrode, 98.5 cm2.
441
Fig. 5. Effect of hydrodynamics on (a) normalized decay of cyanide concentration, (b)
442
cyanate evolution, (c) current efficiency and (d) energy consumption versus electrolysis
443
time in the filter-press electrolyzer at mean linear flow rates of: 1.2 cm s−1 (▲), 2.5 cm
444
s−1 (●), and 3.7 cm s−1 (■) at constant applied potential of 0.95 V vs SHE. Electrolyte:
445
3.85 mM (100 mg L-1) CN− in 45 mM NaNO3 at pH = 11 and T = 303 K. Area of the
446
Ti–RuO2 electrode, 98.5 cm2.
AC C
EP
TE D
M AN U
SC
RI PT
425
20
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
Highlights • Ruthenium dioxide film on a titanium mesh was used as anode. • Anodic oxidation of free cyanide in alkaline media achieved a degradation of 94%.
AC C
EP
TE D
M AN U
SC
• Identification of cyanate as product.
RI PT
• Apparent kinetic rate order of cyanide oxidation depends on the applied potential.