Regeneration by membrane electrolysis of an etching solution based on copper chloride

Article abstract of DOI:10.1007/s11167-005-0535-1

Regeneration of an acid solution for copper etching, based on copper(II) chloride, hydrochloric acid, and ammonium chloride, by membrane electrolysis was studied. The concentrations of copper(I, II) ions in the cathode and anode spaces, current efficiency, degree of copper recovery, and specific consumption of electric power at different quantities of electricity passed through the electrolyzer were measured. The influence exerted by the current density on the electric power expenditure for recovery of metallic copper was examined. The anode current efficiency by chlorine was determined with a spent etching solution and an H₂SO₄ solution used as anolyte.

Full text of DOI:10.1007/s11167-005-0535-1

Russian Journal of Applied Chemistry, Vol. 78, No. 9, 2005, pp. 1444 1449. Translated from Zhurnal Prikladnoi Khimii, Vol. 78, No. 9,
2005, pp. 1469 1474.
Original Russian Text Copyright
2005 by Turaev, Kruglikov, Parfenova.
APPLIED ELECTROCHEMISTRY
AND CORROSION PROTECTION OF METALS
Regeneration by Membrane Electrolysis of an Etching Solution
Based on Copper Chloride
D. Yu. Turaev, S. S. Kruglikov, and A. V. Parfenova
Mendeleev Russian University of Chemical Engineering, Moscow, Russia
Received March 24, 2005
Abstract Regeneration of an acid solution for copper etching, based on copper(II) chloride, hydrochloric
acid, and ammonium chloride, by membrane electrolysis was studied. The concentrations of copper(I, II) ions
in the cathode and anode spaces, current efficiency, degree of copper recovery, and specific consumption
of electric power at different quantities of electricity passed through the electrolyzer were measured. The in-
fluence exerted by the current density on the electric power expenditure for recovery of metallic copper was
examined. The anode current efficiency by chlorine was determined with a spent etching solution and an
H2SO4 solution used as anolyte.
One of the main procedures in manufacture of
printed-circuit boards is copper etching, for which an
acid solution based on Cu(II) chloride is frequently
used. The working principle of this etching solution is
based on the reaction
exchange membrane and circulation of the etching
solution between the anode chamber and the etching
machine was considered in [2]. It should be noted that
the amount of ions of bivalent copper(II) formed
at the anode in a process of this kind will always
markedly exceed the amount of copper ions that
passed across the membrane from the anode space
into the cathode chamber. This mismatch leads to a
decrease in the concentration of copper ions and in-
crease in that of hydrogen ions in the catholyte. This,
in turn, gives rise to a side reaction of hydrogen
evolution at the cathode. Thus, circulation of the etch-
ing solution only between the anolyte and the etching
machine leads to limitation of the rate of copper re-
moval from the etching solution by the rate of its
migration across the cation-exchange membrane from
the anode space into the cathode space in the form of
Cu2+ + Cu + 4CI = 2[CuCl2] ,
(1)
whence follows that the etching solution should con-
tain chloride ions in an amount sufficient for binding
of the whole amount of copper(I) into a complex.
The necessary excess of chloride ions is introduced
into solution in the form of HCl, NH Cl, or NaCl. In
4
the course of etching, the concentration of Cu(II) ions
decreases through an increase in the concentration
of Cu(I) ions, which makes slower the etching rate.
Under industrial conditions, such a spent etching
solution can be regenerated [1] by oxidizing Cu(I)
ions with atmospheric oxygen or hydrogen peroxide.
2+
Cu ions. In the steady state, the current efficiency
by copper will be rather low, because it should be
2+
equal to the transport number of Cu ions across
To maintain a constant, nearly optimal composition
of the solution, all its components except Cu(II) ions
should be added to it. As a result, each cycle of etch-
ing and chemical regeneration leads to a substantial
increase in the volume of the etching solution and in-
volves an additional expenditure of chemicals. Fur-
thermore, the appearing excess amount of the solution
requires an appropriate treatment, which consists at
the majority of plants in the reagent method for pro-
duction of metallic copper or of its sparingly soluble
the membrane. The additional supply of copper ions,
which is necessary for maintaining their concentration
in the catholyte at a prescribed level, can be provided
by circulating the etching solution not only through
the anode chamber, but also through the cathode
chamber.
Copper is removed from the surface of printed-cir-
cuit boards in the etching machine by spraying of
the etching solution over a printed-circuit board.
Owing to the contact between the etching solution and
atmospheric oxygen, copper(I) ions are partly oxidized
to Cu(II) already in the etching machine. This reaction
compounds (oxides, CuO CuCl ).
2
Electrochemical regeneration of a spent etching
solution in a two-chamber electrolyzer with cation-
1070-4272/05/7809-1444 2005 Pleiades Publishing, Inc.

REGENERATION BY MEMBRANE ELECTROLYSIS OF AN ETCHING SOLUTION
1445
makes lower the concentration of copper(I) in the ano-
lyte and thereby leads to a decrease in the quantity of
electricity that should be passed through the anolyte
before the onset of evolution of the toxic chlorine.
This quantity of electricity is several times smaller
than that necessary for removal of etched-off copper
from the etching solution. Therefore, the scheme
under consideration cannot balance the processes
at the anode and cathode without formation of a
chlorine oxygen mixture at the anode.
Composition of the solutions used
Concentration, g l
1
Composition
catholyte
anolyte
wash water
Cu2+
NH4Cl
39.3 41.2
24.1 34.6
150
2.54
9.46
3.15
0
150
50
0
HCl
50
Including Cu+
23.8 33.5
Thus, a topical task is to modify the process of
electrochemical regeneration to preclude formation of
ions in the cathode space and 1.0 N H SO solution
in the anode space.
2
4
Cl at the anode in considerable amounts. In this
2
study, two possible ways to solve the problem are
analyzed: (1) regeneration in two stages, with each
portion of a spent etching solution electrochemically
treated first in the cathode chamber and then in the
anode chamber; and (2) electrochemical treatment of
a spent etching solution only in the cathode chamber,
with a sulfuric acid solution used as anolyte.
Scheme 2 was also used in experiments simulating
the recovery of copper ions from wash water in a
trapping bath after etching. In these experiments,
the catholyte simulated the wash water in the trapping
bath and a 1 N H SO solution served as the anolyte.
2
4
The chlorine contained in the anode gas was absorbed
with a 1 M solution of KI. The electrolysis was per-
2
In the first version of the process, oxidation of
formed at current densities of 2 and 4 A dm .
CuCl ions is the predominant anodic reaction,
The starting solution had the following composition
2
1
because nearly the whole amount of copper(II) con-
tained in an etching solution during its residence in
the cathode chamber will be converted either into the
(g l ): CuCl 2H O 106 111, NH Cl 150, and
2
2
4
HCl 50. Metallic copper was dissolved to prepare the
anolyte. The composition of the solutions poured into
the cell before the experiments is listed in the table.
metal or into CuCl ions.
2
In the second version, the rate of discharge of
chloride ions at the anode will be limited by the rate
of their transfer from the catholyte across the mem-
brane, i.e., will be determined solely by the selectivity
of the membrane.
Cathodic and anodic polarization curves were ob-
tained in the potentiodynamic mode at potential sweep
rates of 0.5 and 1 mV s with a PI-50-1.1 potentio-
1
stat. The content of Cu(I) and Cu(II) ions was deter-
mined in the starting solutions and in the anolyte in
the course of electrolysis. In the experiments using an
H SO solution as the anolyte, the content of chloride
EXPERIMENTAL
2
4
ions was determined. The content of Cl in the anode
2
gas was found by titration of the iodine evolved in
passing the anode gas through a KI solution.
A two-chamber cell [3] with a Flemion cation-
exchange membrane [4] was used in the study. The
anode space was hermetically sealed in order to catch
and absorb gases evolved at the anode. For a more
effective removal of Joule heat, the cell was placed in
a vessel with cold water. The current and voltage were
monitored with a digital ammeter and an M-830B
digital voltmeter. The cell was fed from a B5-48 dc
power supply. A titanium plate with a working surface
Cathodic process. The scheme of processes occur-
ring in the two-chamber electrolyzer in electrolysis
of a chloride solution of copper is shown in Fig. 1.
2+
The main cathodic reaction is reduction of Cu
ions, which occurs in two stages:
Cu2+ + 2Cl + e
CuCl2 + e
CuCl2,
(2)
(3)
2
area of 0.25 dm served as a cathode, and a platinum-
plated niobium plate with the same working surface
area, as an anode.
Cu + 2Cl .
The electrolysis was carried out in accordance with
the above-mentioned two schemes of the process.
Scheme 1: an etching solution containing no Cu(I)
ions in the 350-ml cathode chamber and a solution
containing Cu(I) and Cu(II) ions in the 350-ml anode
space; scheme 2: etching solution containing no Cu(I)
In this case, it can be seen from curve 3 (Figs. 2a
2d) that, first, reaction (2) dominates, which is mani-
+
1
fested in accumulation of Cu ions in solution. Only
after approximately 20 A h l of electricity is passed,
their concentration starts to decrease, i.e., the rate of
further reduction becomes higher than the rate of their
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 78 No. 9 2005

1446
TURAEV et al.
Cation-exchange membrane
Catholyte
Anolyte
Fig. 1. Schematic of the processes occurring in a two-chamber cell.
formation. Presumably, the cathode potential is the
main factor determining the predominance of a one-
or two-electron process. This is indicated by the
manner in which the current density affects the run
of curve 3 (Figs. 2a 2d): at a current density of
current density (Figs. 2a 2d, curves 2). The cathode
potential, naturally, depends on the current density
and is steadily shifted in the course of electrolysis in
the negative direction because of a change in the solu-
tion composition, which consists in that the total con-
centration of copper ions gradually decreases. At a
lowered current density and, accordingly, at a more
positive averaged cathode potential, it is possible to
accumulate univalent copper in amounts exceeding the
2
+
2 A dm , a higher concentration of Cu ions is
2
reached than that at a current density of 4 A dm .
This explanation is also supported by the faster de-
2+
crease in the concentration of Cu ions at a lower
2+
content of Cu ions, which results in differential
current efficiencies that exceed 100% (Figs. 3a 3d,
curves 3).
1
1
c, g l
c, g l
(a)
(b)
A considerable decrease in the concentration of
Cu(I) and Cu(II) ions in the course of further elec-
trolysis at an amount of passed electricity exceeding
1
30 40 A h l will diminish the current efficiency
by metallic copper and copper(I) to values that are
markedly lower than 100%. With account of the fact
that the concentrations of Cu(I) and Cu(II) are close,
the decrease in the differential current efficiency can
be only attributed to occurrence of a side process of
hydrogen evolution. The results of polarization meas-
urements confirm this conclusion (Fig. 4, curve 4).
1
1
c, g l
c, g l
(c)
(d)
2
At a cathode current density of 4 A dm , the aver-
0
age current efficiency by Cu is, after the recovery of
62% of the whole amount of copper(I,II) present in
the anolyte at the beginning of an experiment, 66%,
and the average electric power consumption for re-
1
1
Q, A h l
Q, A h l
0
1
2
covery of Cu , 25.3 kW h kg . At 2 A dm , these
values are strongly different: after the recovery of
77.1% of the whole amount of copper(I,II) contained
in the catholyte at the beginning of an experiment, the
Fig. 2. Concentration c of copper ions in the catholyte vs.
the amount Q of passed electricity. Anolyte: (a, b) Cu(I)-
containing copper etching solution and (c, d) 1.0 N H SO
solution. Cathode current density, A dm : (a, c) 4 and
(b, d) 2. Concentration, g l : (1) total, (2) Cu(II) ions, and
2
4
2
0
1
average current efficiency by Cu is 78.1%, and the
0
(3) Cu(I) ions.
average electric power expenditure for Cu recovery,
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 78 No. 9 2005

REGENERATION BY MEMBRANE ELECTROLYSIS OF AN ETCHING SOLUTION
1447
2
1
1
i. A m
CE, %; W, kW h kg
CE, %; W, kW h kg
(b)
(a)
1
1
CEav, %; Wav, kW h kg
(c)
CEav, %; Wav, kW h kg
(d)
E, V
Fig. 4. (1 3) Anodic and (4) cathodic polarization curves.
(i) Current density and (E) polarization. Curves 1 3 were
measured with an anode made of platinum-plated niobium
2
with a working surface area of 6.25 cm (titanium as cath-
1
ode) at a potential sweep rate of 0.5 mV s in a solution of
1
1
1
+
+
1
1
50 g l HCl and 150 g l NH Cl. Content, g l : (1) Cu
Q, A h l
Q, A h l
4
2+
+
2+
7.62, Cu 47.63; (2) Cu 16.51, Cu 43.18; and (3) Cu
Fig. 3. Current efficiency CE by copper, specific electric
power expenditure W, average current efficiency CE , and
average specific electric power expenditure W vs. the
amount Q of passed electricity at cathode current densities
of (1, 2) 4 and (3, 4) 2 A dm . Anolyte: (a, c) Cu(I)-con-
taining copper etching solution and (b, d) 1.0 N H SO
solution. (a, b) (1, 3) CE and (2, 4) W; (c, d) (1, 3) CE
2+
24.135, Cu 40.00. Curve 4 was measured with a copper
cathode with a working area of 5.72 mm (platinum-plated
niobium as anode) in a solution composed of 50 g l HCl,
150 g l NH Cl, and Cu 0.635 g l at a potential sweep
2
av
av
1
1
1
2
4
1
rate of 1 mV s
.
2
4
av
1
1
CE, %
CE, %
c, g l
c, g l
(a)
(b)
and (2, 4) W
.
av
1
11.7 kW h kg (Fig. 3c). Use of a 1.0 N H SO solu-
2
4
tion as anolyte at the same cathode current density
does not lead to any significant changes in the depen-
dence for the cathode space, compared with the ano-
lyte in the form of a solution containing Cu(I,II) ions
(Figs. 2, 3). Minor differences are, in all probability,
2+
due to transfer of Cu ions from the anolyte into the
1
1
Q, A h l
Q, A h l
catholyte, and also to different electrical conductivi-
ties of the anolytes compared.
cCl , M
CE, %
cCl , M
CE, %
(c)
(d)
Anodic processes. The total concentration of
copper in the anode space decreases from 57.5 to
1
50.1 g l (Figs. 5a, 5b; curves 1) via migration of
Cu(II) cations from the anode space into the cathode
chamber. In the process, the concentration of cop-
per(II) in the anode space increases (curve 2), and that
of Cu(I) decreases (curve 3). To prevent evolution of
1
1
the toxic Cl , the anode current density should not
Q, A h l
Fig. 5. Concentration of (a, b) copper ions c and (c, d)
chloride ions c and (a d) current efficiency CE by chlor-
Q, A h l
2
exceed the limiting current of diffusion of Cu(I) ions,
which decreases with its concentration (Fig. 4,
curves 1 3). If a 1.0 N H SO solution is used as
Cl
ine in the anolyte vs. the amount Q of electricity passed.
2
4
Anolyte: (a, b) copper etching solution and (c, d) 1.0 N
anolyte, the anode current efficiency by chlorine
(Figs. 5c, 5d; curves 4) is limited by the concentration
of chloride ions in the anolyte (Fig. 5, curves 5),
whose migration into the anolyte is governed by the
2
H SO solution. Anode current density, A dm : (a, c) 4
2
4
1
and (b, d) 2. Concentration, g l : (1) total copper ions,
(2) Cu(II) ions, and (3) Cu(I) ions; (4) current efficiency by
chlorine; and (5) concentration of chloride ions.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 78 No. 9 2005

1448
TURAEV et al.
CE, %; W, kW h kg
1
1
c, g l
(a)
(b)
1
1
Q, A h l
Q, A h l
CEav, %; Wav, kW h kg
1
1
CE, %; W, kW h kg
(c)
(d)
cCl , M
1
1
Q, A h l
Q, A h l
Fig. 6. Recovery of copper ions from the wash water from a trapping bath by membrane electrolysis. Two-chamber cell with
a perfluorinated cation-exchange membrane; wash water from the trapping bath as catholyte and 1.0 N H SO solution as
2
4
2
anolyte; cathode and anode current density 1 A dm . Dependence on the amount of passed electricity of (a) concentration c
of copper ions in the catholyte: (1) total copper ions, (2) copper(II) ions, and (3) Cu(I) ions; (b) (1) current efficiency CE by
copper and (2) electric power expenditure W; (c) (1) average current efficiency by copper and (2) average electric power expendi-
ture; and (d) (1) concentration of chloride ions in the anolyte and (2) current efficiency by Cl.
selectivity of the cation-exchange membrane. It should
be noted that, in the presence of chloride ions in the
anolyte in a considerable concentration, the anode
current efficiency does not exceed 2.5% at an anode
sealed anode space are shown in Fig. 6. The curve
describing the dependence of the concentration of
copper ions in the catholyte on the amount of passed
electricity show a run similar to that presented above.
0
2
The current efficiency by Cu decreases, which points
current density of 4 A dm and decreases with the
to intense evolution of hydrogen during the entire
experiment. This leads to an increase in the electric
power consumption.
current density (Figs. 5a, 5b; curves 4). The regenera-
tion by scheme (1) is particularly efficient if the con-
centration of Cu(I) ions in the solution being regen-
erated substantially exceeds that of Cu(II) ions. In this
case, virtually complete oxidation of Cu(I) into Cu(II)
is achieved at a simultaneous minor decrease in the
total concentration of copper ions. If it is necessary to
markedly diminish the total concentration of copper
The electrochemical treatment of the wash water
can decrease the concentration of copper ions in the
1
trapping bath from 2.54 to 0.32 g l on passing
1
5.1 A h l , with the average current efficiency by
copper equal to 30.5%, and the average electric power
1
ions, Cl inevitably evolves at the anode. The advant-
expenditure, to 20 kW h kg .
2
age of scheme (2) consists in the possibility of regen-
eration of copper etching solutions with any concen-
CONCLUSIONS
tration ratio of Cu(I) and Cu(II) ions. Cl evolved at
2
the anode is completely absorbed via oxidation of
Cu(I) into Cu(II). In both versions, it is advisable to
use a low current density in order to make lower the
electric power consumption.
(1) Separation of the cathode and anode spaces
with a cation-exchange membrane makes it possible to
prevent oxidation of copper(I) to Cu(II) at the anode,
with its subsequent reduction to Cu(I) at the cathode,
0
to raise the current efficiency by Cu , and to diminish
The results obtained in a study of electrodeposition
0
the current efficiency by Cl .
2
of Cu from the wash water in the cathode space of
the two-chamber electrolyzer with a hermetically
(2) Use of low cathode and anode current densities
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 78 No. 9 2005

REGENERATION BY MEMBRANE ELECTROLYSIS OF AN ETCHING SOLUTION
1449
prevents or markedly diminishes the evolution of H
REFERENCES
2
at the cathode and Cl at the anode, and also reduces
2
1. Kolesnikov, V.A., Kokarev, G.A., Kamynina, L.L., and
Kapustin, Yu.I., Ekologiya i resursosberezhenie v elek-
trokhimicheskikh proizvodstvakh (Ecology and Resource
Saving at Electrochemical Shops), Moscow: Ross.
Khim.-Tekhnol. Univ. im. D.I. Mendeleeva, 1998,
part II.
2. Il’in, V.A., Khimicheskie i elektrokhimicheskie protses-
sy v proizvodstve pechatnykh plat (Chemical and Elec-
trochemical Processes in Manufacture of Printed-
Circuit Bioards), issue 2, Supplement to Gal’vanotekh.
Obrab. Poverkhn., Moscow, 1994.
the electric power expenditure per kilogram of re-
0
covered Cu .
(3) Use of 1.0 N H SO or a copper etching solu-
2
4
tion as anolyte enables regeneration of copper etching
solutions with different Cu(I) and Cu(II) concentra-
tions. However, this requires that the anode space
should be hermetically sealed to trap the evolving
chlorine and use it to oxidize Cu(I) ions.
(4) To decrease the loss of copper ions from the
etching shop to wastewater, it is advisable to intro-
duce a procedure of washing in the trapping bath, with
a low concentration of copper ions maintained in
the bath by means of membrane electrolysis.
3. Kruglikov, S.S., Turaev, D.Yu., and Gulina, V.V.,
Gal’vanotekh. Obrab. Poverkhn., 2003, no. 4.
4. Mazanko, L.F., Kamar’yan, G.M., and Romashin, O.P.,
Promyshlennyi membrannyi elektroliz (Industrial Mem-
brane Electrolysis), Moscow: Khimiya, 1989.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 78 No. 9 2005