Journal of The Electrochemical Society, 148 (1) B51-B57 (2001) B51

S0013-4651/2001/148(1)/B51/7/$7.00 © The Electrochemical Society, Inc.

Electrochemistry of Copper in Aqueous Glycine Solutions

Serdar Aksu and Fiona M. Doyle*,z
Department of Materials Science and Engineering, University of California at Berkeley, Berkeley,
California 94720-1760, USA

Potential-pH equilibria and potentiodynamic polarization studies were used to examine the electrochemical behavior of copper in
aqueous glycine solutions. Potential-pH diagrams for the copper-water-glycine system were derived at different total copper {CuT}
and glycine {LT} activities. The diagrams show that glycine significantly extended the solubility range of copper. Polarization
experiments were conducted in deaerated and aerated aqueous solutions of 10⫺2 M glycine with 10⫺5 M cupric nitrate and 10⫺1
M glycine with 10⫺4 M cupric nitrate at pH values between 9 and 12. The results of these experiments are discussed in terms of
the relevant potential-pH diagrams. Good correlations were observed.
© 2000 The Electrochemical Society. S0013-4651(00)07-032-4. All rights reserved.

Manuscript submitted July 10, 2000; revised manuscript received October 10, 2000.

Copper and its alloys are used extensively in many environments
and applications, because of their relatively high corrosion resistance,
superior electrical and thermal conductivity, ease of fabrication, resis-
tance to biofouling, and other attractive properties. The corrosion of
copper has been studied in various solutions, and the literature on the
subject has been reviewed.1-3 Copper is a relatively noble metal, and
is immune to corrosion in noncomplexing solutions that are free from
oxidizing agents, as predicted by potential-pH diagrams for the Cu-
H2O system.4 It is, however, susceptible to corrosive attack by acidic
or strongly alkaline solutions that contain oxidizing agents.
The corrosion resistance of copper is highly compromised when it
is exposed to aqueous solutions containing complexing agents such
as ammonia and cyanide. Potential-pH diagrams for the Cu-CN⫺-
H2O5 and Cu-NH3-H2O6 systems show that complexation of the
Cu2⫹ and Cu⫹ ions considerably expands the solubility domain of
copper, while significantly diminishing the predominance regions of
cuprous and cupric oxide. The upper limit of the domain of thermo-
dynamic immunity to corrosion drops to significantly lower electrode
potentials. These complexing agents, therefore, exacerbate the corro-
sion of copper, even in the absence of powerful oxidizing agents.
Glycine is an ␣-amino acid with a carboxylic acid and an amine
group. In trace quantities, glycine is used as a leveling agent in acidic
copper plating baths. Although it affects the distribution of copper on
a microscale during plating, at trace concentrations it has little effect
on the overall cathodic kinetics.7 Due to its complexing action,
glycine can enhance the solubility of copper ions in aqueous solu-
tions. Halpern and co-workers8 investigated the kinetics of copper
dissolution in aqueous solutions of different ␣-amino acids. The
behavior in glycine was similar to that with complexing agents such
as ammonia9 and ethylenediamine.10 At low oxygen partial pres-
sures, the rate of copper dissolution was proportional to the oxygen
partial pressure, and independent of the concentration of glycine. At
sufficiently high oxygen pressures, however, the rate was indepen-
dent of the oxygen partial pressure, but dependent on the concentra-
tion of the glycine, whether in the zwitterionic (HL) or anionic (L⫺)
form (discussed in more detail below)
per in glycine solutions. First, the thermodynamics of copper com-
plexation in aqueous glycine solutions are analyzed, and potential-
pH diagrams for the copper-glycine-water system are presented. Ex-
perimental measurements of the polarization behavior of copper in
glycine solutions of different compositions are then reported. This
behavior is discussed in terms of the relevant potential-pH diagrams.
Potential-pH Diagrams
Potential-pH diagrams are conventially plotted with pH on the
horizontal axis, and potential on the vertical axis. Consider a gener-
alized reduction reaction involving n electrons
aA ⫹ mH⫹ ⫹ ne ⫽ bB ⫹ dH2O [2]
The equilibrium electrochemical potential, E, for this reaction is
given by the Nernst equation
2.303 RT  {A}a {H⫹}m 
E ⫽ Eo ⫹ log b d volts [3]
nF  {B} {H 2 O} 
where Eo, R, T, and F are the standard equilibrium electrochemical
potential, gas constant, absolute temperature and Faraday’s constant,
respectively. {i} denotes the thermodynamic activity of i. When plot-
ted on a potential pH diagram, Eq. 3 generates a line with gradient-
2.303RTm/nF V/pH unit. The position of such a line will clearly
depend on the magnitude of {A} and {B}. If these are pure solids, the
activities would be unity. For dissolved or gaseous species, the activi-
ties must be specified; generally values that are appropriate for a given
application are selected. A, the oxidized species in Eq. 1, will pre-
dominate at potentials higher than the equilibrium potential, whereas
B, the reduced species, predominates at potentials lower than the equi-
librium line.
Reactions need not involve a change in oxidation state. For the
general reaction
aA ⫹ mH⫹ ⫽ bB ⫹ dH2O [4]

Dissolution rate (in mg cm⫺2 h⫺1) ⫽ kL[L⫺] ⫹ kHL[HL] [1] the Gibbs free energy of reaction determines the position of equilib-
rium
where square brackets indicate analytical concentrations. The rate {B}b {H 2 O}d
constants kL and kHL were 49 and 31, respectively, demonstrating the ⌬G o ⫽ ⫺RT ln [5]
higher reactivity of the anionic form. Keenan and collaborators11 {A}a {H⫹}m
observed that the corrosion rate of copper was strongly correlated For a specified activity of A and B, equilibrium will lie at a specific
with the concentration of the anionic form of glycine, and was influ- pH
enced only slightly by the zwitterionic form.
The present study aims to expand our fundamental understanding {B}b {H 2 O}d ⌬G o
pH ⫽ ⫺log a ⫺ [6]
of the electrochemical dissolution and passivation behavior of cop- {A} 2.303 RT
* Electrochemical Society Active Member. This pH is independent of potential. Species A is the more basic of
z E-mail: fiona@socrates.berkeley.edu the compounds involved, and will predominate at pH values higher

Downloaded on 2014-12-19 to IP 130.160.4.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 B52 Journal of The Electrochemical Society, 148 (1) B51-B57 (2001)
S0013-4651(00)07-032-4 CCC: $7.00 © The Electrochemical Society, Inc.

than the equilibrium pH. Conversely, species B will predominate at

pH values lower than the equilibrium pH. Table I. Standard Gibbs free energies of formation of species in
Glycine can exist in aqueous solutions in three different forms, the copper-water-glycine system at 25⬚C and 1 atm (from
namely ⫹H3NCH2COOH (cation), ⫹H3NCH2COO⫺ (zwitterion), Ref. 13).
and H2NCH2COO⫺ (anion). These species are denoted as H2L⫹,
HL, and L⫺, respectively, for brevity. The equilibria between these Species State ⌬Gof (kJ/mol)
may be depicted as12
H⫹ Aqueous 0
pKa1 ⫽ 2.350 pKa2 ⫽ 9.778 OH⫺ Aqueous ⫺117.244
⫹H NCH COOH } ⫹H NCH COO⫺ } H NCH COO⫺ [7]
3 2 3 2 2 2 H2O Liquid ⫺237.129
(H2L⫹) (HL) (L⫺) H2 Gas 0
cation zwitterion anion O2 Gas 0
Cu⫹ Aqueous 49.98
As shown in Fig. 1, H2L⫹ predominates at pH values below 2.35 Cu2⫹ Aqueous 65.49
(pKa1), while L⫺ predominates at pH values above 9.778 (pKa2). HL Cu Solid 0
predominates at intermediate pH values. Glycine forms soluble com- CuO Solid ⫺129.7
plexes with both cupric and cuprous ions. The principal copper(II) Cu2O Solid ⫺146.0
glycinate complexes are Cu(H3NCH2COO)2⫹, Cu(H2NCH2COO)⫹, CuO22⫺ Aqueous ⫺183.6
and Cu(H2NCH2COO)2, while the principal Cu(I) species is HCuO2⫺ Aqueous ⫺258.5
H2L⫹ Aqueous ⫺384.061
Cu(H2NCH2COO)⫺ 2⫹ ⫹
2 . These are referred to as CuHL , CuL , CuL2, HL Aqueous ⫺370.647
and CuL⫺ 2 , respectively. L⫺ Aqueous ⫺314.833
Table I lists the standard Gibbs free energies of formation of CuL⫹ Aqueous ⫺298.2
species in the copper-water-glycine system.13 Cupric hydroxide,
Cu(OH)2, was not considered because it is less stable than cupric
oxide.4 Standard Gibbs free energies of formation have not been tab- capable of bidentate chelation, while the latter can only give mon-
ulated for all the copper-glycine complexes identified above. The odentate coordination through the carboxylate group. CuHL2⫹,
stabilities of these were deduced using the association constants pre- which contains the zwitterion, predominates only when the total
sented in Table II.12 Table III shows all thermodynamically mean- glycine activity exceeds 1.94 ⫻ 10⫺2. Anionic hydrolyzed Cu(II)
ingful reactions between the species in the copper-water-glycine sys- species appear at high pH, and their predominance regions are unaf-
tem, along with the equations of each equilibrium line, calculated fected by glycine. A very narrow region appears for HCuO⫺ 2 at
using the data provided in Tables I and II, and Eq. 3 or 6. 10⫺6 M {CuT}, along with CuO2⫺ 2 , whereas at higher {CuT} only
Figure 2 shows potential-pH diagrams plotted from the equilibria CuO22⫺ appears. The stability regions of cupric oxide, CuO, and
in Table III at various combinations of total glycine activity {LT} and cuprous oxide, Cu2O, expand with increasing CuT, and contract with
total dissolved Cu activity {CuT}. The lines identified as 1 to 4 cor- increasing LT.
respond to the equations of the same number in Table III. Figure 2 It must be emphasized that potential-pH diagrams only provide
demonstrates that copper metal is oxidized to form copper(II) glyci- information about thermodynamic equilibrium and infer nothing
nate species, either CuHL2⫹, CuL⫹, or CuL2, over a wide range of about the kinetics of reactions. Moreover, although underlying met-
pH. The copper(I) glycinate complex CuL⫺ 2 only appears when LT is als might be passive in regions where oxides predominate, they need
much higher than CuT (diagrams b and c), such that the line for not be; passivation requires coherent, adherent oxide films with low
reduction of CuL⫺ 2 to elemental copper is below the line for reduc- conductivity. Nevertheless, potential-pH diagrams have proved use-
tion of CuL2 to CuL⫺ 2 . It is notable that glycine appears in the anion- ful in elucidating the electrochemistry of copper in various aqueous
ic form at much lower pH values when complexing copper than it media.1,5,6,14,15
does in the free form. Table II and Fig. 1 show that free glycine only Experimental
predominates as an anion above pH 9.778 and predominates in the
zwitterionic form between pH 2.35 and 9.778. The anionic form of Reagents and solutions.—Glycine (>99%) was obtained from
glycine clearly forms more stable copper complexes than does the Acros Organics Co. Reagent grade cupric nitrate hemipentahydrate
zwitterionic form, which would be expected since the former is was purchased from Fisher Scientific Co. Electrolyte solutions were
prepared using doubly distilled water, and their pH was adjusted to
⫾0.1 units using NaOH or HNO3 (both reagent grade from Fisher
Scientific Co.). The electrolyte was either aerated or deaerated be-
fore use. For aeration, breathing (medical) air was purged into the
electrolyte through a porous gas disperser for 18 h. The pH was ad-
justed with NaOH after aeration, to correct for absorption of CO2.
For deaeration, ultrapure nitrogen gas (99.999% N2) was sparged for
24 h; this did not affect the pH.

Table II. Association constants of glycine and copper glycinate

species at 25⬚C and 1 atm (I is the ionic strength at which the
data were acquired; all from Ref. 12).

Stability constant at 25⬚C/1 atm

Reaction Label log K or log B

L⫺ ⫹ H⫹ ⫽ HL KL1 09.778 (I ⫽ 0).0

HL ⫹ H⫹ ⫽ H2L⫹ KL2 02.350 (I ⫽ 0).0
Cu2⫹ ⫹ L⫺ ⫽ CuL⫹ KML1 08.570 (I ⫽ 0).0
Cu2⫹ ⫹ 2L⫺ ⫽ CuL2 BML2 15.640 (I ⫽ 0).0
Figure 1. Distribution of glycine species as a function of pH in the glycine- Cu⫹ ⫹ 2L⫺ ⫽ CuL⫺
2 BmL2 10.100 (I ⫽ 0.2)
water system. CuL⫹ ⫹ H⫹ ⫽ CuHL2⫹ KML3 02.920 (I ⫽ 0.1)

Downloaded on 2014-12-19 to IP 130.160.4.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 Journal of The Electrochemical Society, 148 (1) B51-B57 (2001) B53
S0013-4651(00)07-032-4 CCC: $7.00 © The Electrochemical Society, Inc.

Table III. Equilibria in the copper-water-glycine system at 25⬚C and 1 atm.

No. Reaction Equilibrium line

01 O2 ⫹ 4H⫹ ⫹ 4e ⫽ 2H2O E ⫽ 1.229 ⫺ 0.0591 pH

02 2H⫹ ⫹ 2e ⫽ H2 E ⫽ ⫺0.0591 pH
03 HL ⫹ H⫹ ⫽ H2L⫹ pH ⫽ 2.350 ⫹ log{HL}/{H2L⫹}
04 L⫺ ⫹ H⫹ ⫽ HL pH ⫽ 9.778 ⫹ log{L⫺}/{HL}
05 Cu2⫹ ⫹ H2L⫹ ⫽ CuHL2⫹ ⫹ H⫹ pH ⫽ 0.649 ⫺ log{H2L⫹} ⫹ log {CuHL2⫹}/{Cu2⫹}
06 CuL⫹ ⫹ H⫹ ⫽ CuHL2⫹ pH ⫽ 2.920 ⫹ log{CuL⫹}/{CuHL2⫹}
07 Cu2⫹ ⫹ HL ⫽ CuL⫹ ⫹ H⫹ pH ⫽ 1.219 ⫺ log{HL}⫹ log{CuL⫹}/{Cu2⫹}
08 CuL⫹ ⫹ HL ⫽ CuL2 ⫹ 2H⫹ pH ⫽ 2.710 ⫺ log{HL} ⫹ log{CuL2}/{CuL⫹}
09 CuL2 ⫹ e ⫽ CuL⫺2 E ⫽ ⫺0.167 ⫹ 0.0591 log{CuL2}/{CuL⫺ 2}
⫹ ⫺ pH ⫽ 13.12 ⫹ log{CuO22⫺}/{HCuO⫺
10 2 ⫹ H ⫽HCuO2
CuO2⫺ 2}
11 ⫹
Cu2O ⫹ 2H ⫹ 2e ⫽ 2Cu ⫹ H2O E ⫽ 0.472 ⫺ 0.0591pH
12 2CuO ⫹2H⫹ ⫹2e ⫽ Cu2O ⫹ 2H2O E ⫽ 0.641 ⫺ 0.0591pH
13 Cu2⫹ ⫹ 2e ⫽ Cu E ⫽ 0.339 ⫹ 0.0295 log{Cu2⫹}
14 CuHL2⫹ ⫹ H⫹ ⫹ 2e ⫽ Cu ⫹ H2L⫹ E ⫽ 0.358 ⫺ 0.0295 pH ⫹ 0.0295 log{CuHL2⫹} ⫺ 0.0295 log{H2L⫹}
15 CuHL2⫹ ⫹ 2e ⫽ Cu ⫹ HL E ⫽ 0.288 ⫺ 0.0295 log{CuHL2⫹} – 0.0295 log{HL}
16 CuL⫹ ⫹ H⫹ ⫹ e ⫽ Cu ⫹ HL E ⫽ 0.375 ⫺ 0.0295 pH ⫺ 0.0295 log{CuL⫹} ⫺ 0.0295 log{HL}
17 CuL2 ⫹ 2H⫹ ⫹ 2e ⫽ Cu ⫹ 2HL E ⫽ 0.455 ⫺ 0.0591 pH ⫹ 0.0295 log{CuL2} ⫺ 0.0591 log{HL}
18 CuL2 ⫹ 2e ⫽ Cu ⫹ 2L⫺ E ⫽ ⫺0.126 ⫹ 0.0295 log{CuL2} ⫺ 0.0591 log{L⫺}
19 CuL⫺
2 ⫹ 2H⫹ ⫹ e ⫽ Cu ⫹ 2HL E ⫽ 1.077 ⫺ 0.1182 pH ⫹ 0.0591 log{CuL⫺ 2 } ⫺ 0.0591 log{HL}
20 CuL⫺
2 ⫹ e ⫽ Cu ⫹ 2L
⫺ E ⫽ ⫺0.080 ⫹ 0.0591 log{CuL⫺ ⫺
2 } ⫺ 0.118 log{L }
21 2Cu2⫹ ⫹ H2O ⫹ 2e ⫽ Cu2O ⫹ 2H⫹ E ⫽ 0.207 ⫹ 0.0591 pH ⫹ 0.0591 log{Cu2⫹}
22 2CuL⫹ ⫹ H2O ⫹ 2e ⫽ Cu2O ⫹ 2HL E ⫽ 0.278 ⫹ 0.0591 log{CuL⫹} ⫺ 0.0591 log{HL}
23 Cu2O ⫹2H⫹ ⫹ 4L⫺ ⫽ 2CuL⫺ 2 ⫹ 2H2O pH ⫽ 9.32 ⫺ log{CuL⫺ ⫺
2 } ⫺ 2 log{L }
24 2CuL2 ⫹ 2H2O ⫹ 2e ⫽ Cu2O ⫹ 2H⫹ ⫹ 4L⫺ E ⫽ -0.722 ⫹ 0.0591 pH ⫹ 0.0591 log{CuL2} ⫺ 0.1182 log{L⫺}
25 2CuL2 ⫹ 2H2O ⫹ 2H⫹ ⫹ 2e ⫽ Cu2O ⫹ 4HL E ⫽ 0.438 ⫺ 0.0591 pH ⫹ 0.0591 log{CuL2} ⫺ 0.1182 log{HL}
26 2CuO22⫺ ⫹ 6H⫹ ⫹2e ⫽ Cu2O ⫹ 3H2O E ⫽ 2.562 ⫺ 0.1773 pH ⫹ 0.0591 log{CuO2⫺ 2 }
27 2CuO22⫺ ⫹ 4H⫹ ⫹2e ⫽ Cu ⫹ 2H2O E ⫽ 1.506 ⫺ 0.1182pH ⫹ 0.0591 log{CuO2⫺ 2 }
28 2HCuO⫺ ⫹
2 ⫹ 4H ⫹ 2e ⫽ Cu2O ⫹ 3H2O E ⫽ 1.783 ⫺ 0.1182 pH ⫹ 0.0591 log{HCuO⫺ 2}
29 CuO ⫹ 2H⫹ ⫹ 2L⫺ ⫽ CuL2 ⫹ H2O pH ⫽ 11.495 ⫺ 0.5 log{CuL2} ⫹ log{L⫺}
CuO ⫹ H2O ⫽ CuO2⫺ ⫹
30 2 ⫹ 2H pH ⫽ 16.05 ⫹ 0.5 log{CuO2⫺2 }
31 CuO ⫹ H2O ⫽ HCuO⫺ 2 ⫹H
⫹ pH ⫽ 18.979 ⫹ log{CuO2⫺ 2 }

Electrochemical cell.—Polarization studies were made using
three-electrode cells. Each cylindrical glass cell was 130 mm high,
with a 120 mm id. The Plexiglas cell lid had openings for a glass
Luggin probe, gas disperser, and working and counter electrodes.
When purging air or nitrogen into the solution, the cell was sealed
except for openings on the lid less than 2 mm in diam, which al-
lowed the discharge of excess gas. The working electrode was a cop-
per disk constructed from a 10 mm diam rod of 99.9999% pure cop-
per (Aldrich Chemical Co.), embedded in a nonconductive, nonre-
active epoxy (Tra-Con, Inc.). The od of the electrode was 25.4 mm.
The electrode was first ground to a 600 grit finish, rinsed in water
and acetone, and dried. It was then mirror-polished by a succession
of diamond pastes from 9 to 0.25 ␮m and finally cleaned with ace-
tone and distilled water. The electrode was rotated at 1000 rpm using
a Pine Instrument Co. electrode rotator. A saturated calomel refer-
ence electrode (0.242 V SHE) was held in a Luggin probe filled with
the same solution as the electrochemical cell. The tip of the probe
was held about 2 mm from the working electrode. A platinum mesh
counter electrode was held 20 mm from, and parallel to, the working
electrode. All potentials are reported with respect to the standard
hydrogen electrode (SHE). They are uncorrected for liquid junction
potentials and ohmic potential drops.
Polarization tests.—DC potentiodynamic polarization studies
were made using a computer-controlled EG&G PAR model 273-A
potentiostat. Tests were conducted in 650 mL of nitrogen- or air-
purged aqueous solutions of either 10⫺2 M glycine with 10⫺5 M
cupric nitrate or 10⫺1 M glycine with 10⫺4 M cupric nitrate at 24 ⫾
1⬚C. Air or nitrogen was purged during the experiments. Before each
potentiodynamic scan, a cathodic potential of ⫺800 mV SHE was
applied to the working electrode for 600 s, to reduce any oxide films.
The copper electrode was then allowed about 900 s to attain a stable
open-circuit potential (EOC). Potentiodynamic polarization curves
for the rotating disk copper electrode were obtained by scanning at
1 mV/s, from 500 mV below EOC to 1500 mV above.
Linear polarization resistance measurements were used to esti-
mate the current densities (iOC) at EOC. In these experiments, the elec-
trode was scanned at 0.2 mV/s from ⫺40 mV below EOC to 40 mV
above. A plot of overpotential (␩) from EOC vs. current density (i)
should be linear near EOC, and the slope of such a plot gives the polar-
ization resistance, Rp. iOC is inversely proportional to Rp16,17
␤a␤c
ioc ⫽ [8]
2.303(␤a ⫹ ␤c ) Rp
where ␤a and ␤c are the absolute values of anodic and cathodic Tafel
constants, respectively. Cathodic Tafel constants could seldom be
determined reliably from the potentiodynamic scans. For copper dis-
solution in aerated solutions, ␤c was assumed to be 0.06 V/decade
while the cathodic reaction in deaerated solutions was assumed to be
limited by concentration polarization (␤c ⫽ ⬁ V/decade). The error
in determining iOC with estimated Tafel constants has been shown to
be within the experimental scatter.18
Results and Discussion
Polarization studies were made in aqueous solutions of 10⫺2 M
glycine with 10⫺5 M cupric nitrate, and 10⫺1 M glycine with 10⫺4 M
cupric nitrate. Figures 2a and b are the potential-pH diagrams for
these solution chemistries. Table IV shows EOC and iOC for each group
of experiments, at each pH.
Figure 3 shows the effect of pH from 9 to 12 on the polarization
behavior of the copper electrode rotating at 1000 rpm in deaerated
10⫺2 M glycine with 10⫺5 M cupric nitrate aqueous solutions. The
polarization behavior given in Fig. 3 is reasonably consistent with
the potential-pH diagram shown in Fig. 2a, which predicts that CuL2
should predominate at pH 9 and 10 above potentials ⫺106 and
⫺154 mV, respectively. On polarizing anodically from EOC, copper
dissolves actively at pH 9 and 10, with no evidence of passivation.
Despite the similarity in rest potential at pH 9 and 10, iOC increased
by a factor of about three on increasing the pH from 9 to 10. As the

Downloaded on 2014-12-19 to IP 130.160.4.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 B54 Journal of The Electrochemical Society, 148 (1) B51-B57 (2001)
S0013-4651(00)07-032-4 CCC: $7.00 © The Electrochemical Society, Inc.

pH of the glycine solution increases from 9 to 10 the fraction of

anion increases from 0.143 to 0.625. Since the bidentate (L⫺) is
more reactive than the monodentate (HL),19 the increased dissolu-
tion rate at pH 10 is not unexpected.
At pH 11, the potential-pH diagram predicts formation of Cu2O
at ⫺178 mV. The rest potential measured at this pH was ⫺45 mV
lower. Although the fraction of L⫺ reaches 0.943 at pH 11, there was
a decrease in iOC, suggesting that some of the surface sites were
occupied at this potential, probably by adsorption of OH⫺ ions or by
hydrolysis of Cu(I). As the potential was scanned in the anodic
direction, a region of active dissolution was observed up to about
850 mV, indicating that any Cu2O or precursor phase that formed in
this potential range was not protective. The passive region seen
above 850 mV suggests the formation of a hydrolyzed Cu(II) phase.
In Fig. 2a no oxide predominates at potentials exceeding ⫺128 mV
at pH 11. However, the increase in the local concentration of Cu(II)
and OH⫺ around the electrode after a reasonable amount of anodic
dissolution was probably high enough to nucleate CuO, Cu(OH)2, or
a related phase.
At pH 12, the potential-pH diagram predicts formation of Cu2O
and CuO at ⫺237 and ⫺68 mV, respectively. At this pH, EOC dropped
to ⫺254 mV, which is 17 mV below the potential for equilibrium
between elemental copper and Cu2O. iOC was slightly smaller than at
pH 11, possibly due to further blockage of active sites on the copper
surface by OH⫺ or by hydrolyzed species. The anodic polarization
curve shows an initial active region, followed by a minor passivation
region at around ⫺80 mV, and a well-developed passive region
between about 300 and 650 mV. Figure 2a suggests that the passiva-
tion in these regions was due to cuprous and cupric oxides on the elec-
trode surface, respectively. Passivation by oxides requires the pres-
ence of continuous protective layers. Cu2O probably nucleated in the
initial active region, and spread and thickened as the potential
increased, eventually forming a continuous film. The first passivation
peak suggests that this transformation was complete at around ⫺80
mV. It is not unusual for passivation potentials to be more noble than
the equilibrium potentials for oxide formation,20-23 due to film for-
mation kinetics,23 including nucleation and growth.22 Here uncom-
pensated ohmic drops could also contribute to the overpotential. Fig-
ure 3b shows that at pH 12 the current density increased with increas-
ing potential above ⫺30 mV to the onset of the second passivation
peak at 300 mV. CuO is stable at the potentials of this active region.
The increase in the current density with increasing potential suggests
that when any surface Cu2O and underlying copper are oxidized to
CuO, the initial product is not protective. The 370 mV displacement
of the potential for passivation from the equilibrium potential for oxi-
dation of Cu2O to CuO probably reflects the kinetics for forming a
passive film. There might also be an increased contribution from
ohmic potential drops, due to the high current densities.
The copper electrode surface was bright and shiny after the polar-
ization tests at pH 11 and 12, with no evidence of pitting. This might
suggest that the transpassive region exhibited at these pH values cor-
responds to the evolution of oxygen from water

2H2O ⫽ O2 ⫹ 4H⫹ ⫹ 4e [9]

Since it is difficult to remove oxygen completely, even after purging

with nitrogen, the polarization behavior at potentials immediately
below EOC reflects the mass-transport controlled reduction of traces
of dissolved oxygen, as well as the reduction of CuL2 to elemental
copper, or at the higher pH values, to Cu2O. At about ⫺550 to ⫺700
mV there is a dramatic increase in the cathodic current density, due
to the evolution of hydrogen

2H2O ⫹ 2e ⫽ H2 ⫹2 OH⫺ [10]

Figure 4 shows the effect of pH from 9 to 12 on the polarization

Figure 2. Potential-pH diagrams for the copper-water-glycine system at 25⬚C behavior of copper in aerated 10⫺2 M glycine with 10⫺5 M cupric
and 1 atm. (a) {LT} ⫽ 10⫺2 and {CuT} ⫽ 10⫺5; (b) {LT} ⫽ 10⫺1 and nitrate aqueous solutions. As expected, the EOC and iOC values shown
{CuT} ⫽ 10⫺4; (c) {LT} ⫽ 10⫺2 and {CuT} ⫽ 10⫺6; and (d) {LT} ⫽ 10⫺4 in Table IV are significantly higher than those for the deaerated sys-
and {CuT} ⫽ 10⫺6. tem at the same pH and composition; typically the potentials are 100

Downloaded on 2014-12-19 to IP 130.160.4.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 Journal of The Electrochemical Society, 148 (1) B51-B57 (2001) B55
S0013-4651(00)07-032-4 CCC: $7.00 © The Electrochemical Society, Inc.

Table IV. Open-circuit potential, current densities, and absolute values of Tafel constants of copper in aqueous Cu(NO3)2-glycine solutions.

EOC iOC ␤A ␤C
Solution pH (mV, SHE) (A/m2) mV/decade mV/decade

10⫺2 M glycine with 09 ⫺131 4.87 ⫻ 10⫺3 60 Large

10⫺5 M Cu(NO3)2, deaerated 10 ⫺137 1.47 ⫻ 10⫺2 40 Large
11 ⫺223 8.56 ⫻ 10⫺3 60 Large
12 ⫺254 6.63 ⫻ 10⫺3 60 Large
10⫺2 M glycine with 09 0⫺18 1.32 ⫻ 10⫺1 50 60
10⫺5 M Cu(NO3)2, aerated 10 0⫺42 3.14 ⫻ 10⫺1 50 60
11 0⫺59 2.43 ⫻ 10⫺1 50 60
12 0⫺63 1.63 ⫻ 10⫺1 50 60
10⫺1 M glycine with 9 ⫺107 3.15 ⫻ 10⫺2 50 Large
10⫺4 M Cu(NO3)2, deaerated 10 ⫺212 4.43 ⫻ 10⫺2 60 Large
11 ⫺217 4.61 ⫻ 10⫺2 60 Large
12 ⫺258 2.78 ⫻ 10⫺2 40 Large
10⫺1 M glycine with 09 0⫺61 5.16 ⫻ 10⫺1 60 60
10⫺4 M Cu(NO3)2, aerated 10 ⫺101 9.75 ⫻ 10⫺1 60 60
11 ⫺128 1.62 60 60
12 ⫺142 9.60 ⫻ 10⫺1 50 60

to 150 mV higher, while the current densities are one to two orders cathodic current densities were about two orders of magnitude high-
of magnitude greater. Over most of the potentials scanned, the er than those observed under deaerated conditions at the same over-

Figure 3. Effect of pH on the polarization behavior of copper in deaerated Figure 4. Effect of pH on the polarization behavior of copper in aerated
10⫺2 M glycine with 10⫺5 M cupric nitrate at 1 mV/s and 1000 rpm. 10⫺2 M glycine with 10⫺5 M cupric nitrate at 1 mV/s and 1000 rpm.

Downloaded on 2014-12-19 to IP 130.160.4.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 B56 Journal of The Electrochemical Society, 148 (1) B51-B57 (2001)
S0013-4651(00)07-032-4 CCC: $7.00 © The Electrochemical Society, Inc.

potentials (Fig. 3). The higher potentials and currents reflect cathod-
ic reduction of oxygen
O2 ⫹ 2H2O ⫹ 4e ⫽ 4 OH⫺
The open-circuit potentials are clearly mixed potentials, with mod-
est anodic dissolution of the copper electrode at the steady state.
Active corrosion of copper by oxygen in aqueous glycine solutions
has been reported earlier.15
At pH 9 and 10, the anodic polarization curves for copper are
very similar in Fig. 3 and 4. The passivation behavior exhibited at the
higher pH values is also similar, except that no minor passive region
was seen at about ⫺80 mV at pH 12, because under aerated condi-
tions the net current was cathodic at this potential.
Figure 2b suggests that the solubility region of copper extends to
higher pH at {CuT} ⫽ 10⫺4 and {LT} ⫽ 10⫺1. No Cu2O is predict-
ed below pH 11.3, and CuO is only expected above pH 12.5. Fig-
ure 5 shows the effect of pH from 9 to 12 on the polarization behav-
ior of a copper electrode rotating at 1000 rpm in deaerated 10⫺1 M
glycine with 10⫺4 M cupric nitrate aqueous solutions. EOC and iOC
values at each pH are shown in Table IV. The polarization curves
given in Fig. 5 agree reasonably well with Fig. 2b, which predicts
that CuL2 would form on oxidation of copper at pH 9, while CuL2-
would form in a narrow range of potentials at pH 10 and 11. On
polarizing anodically from EOC, copper dissolved actively at pH 9,
10, and 11 with no evidence of passivation. iOC increased only slight-
[11] ly when the pH was increased from 9 to 11. With 10⫺1 M glycine,
there was abundant L⫺, even at pH 9, therefore copper dissolution
was insensitive to pH-induced L⫺ concentrations.
At pH 12, the potential-pH diagram predicts that copper should
oxidize to Cu2O at about ⫺237 mV. At this pH iOC was slightly lower
than at pH 11, suggesting blockage of some active sites by adsorbed
OH⫺ or hydrolyzed species. Copper dissolved actively on polarizing
anodically from EOC. An inflection at ⫺160 mV probably corresponds
to the transient formation of Cu2O, which would have then oxidized
to CuL2(aq) at somewhat higher potentials. The passive region seen
above 800 mV suggests the formation of a hydrolyzed Cu(II) oxide.
No oxide predominates at potentials higher than ⫺120 mV under the
given conditions. However, the increase in the concentration of Cu(II)
and OH⫺ in the boundary layer after a reasonable amount of anodic
dissolution might have allowed nucleation of CuO or Cu(OH)2.
Figure 6 shows the effect of pH from 9 to 12 on the polarization
behavior of copper in aerated 10⫺1 M glycine with 10⫺4 M cupric
nitrate aqueous solutions. EOC and iOC values at each pH are also in-
cluded in Table IV. As expected, these values are significantly
greater than those for the deaerated system at the same pH and com-

Figure 5. Effect of pH on the polarization behavior of copper in deaerated Figure 6. Effect of pH on the polarization behavior of copper in aerated
10⫺1 M glycine with 10⫺4 M cupric nitrate at 1 mV/s and 1000 rpm. 10⫺1 M glycine with 10⫺4 M cupric nitrate 1 mV/s and 1000 rpm.

Downloaded on 2014-12-19 to IP 130.160.4.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
 Journal of The Electrochemical Society, 148 (1) B51-B57 (2001)
S0013-4651(00)07-032-4 CCC: $7.00 © The Electrochemical Society, Inc.
position due to the availability of oxygen for reduction. The maxima
in current densities seen immediately below EOC on the cathodic
curves probably reflect reduction of CuL2 to CuL⫺2 and elemental
copper, in conjunction with the reduction of oxygen. At lower poten-
tials, the cathodic behavior largely reflects oxygen reduction.
At pH 9, 10, and 11 the anodic polarization curves for copper
were very similar to those in Fig. 5 for deaerated solutions. The pas-
sivation behavior exhibited at pH 12 was also similar, except that no
transient formation of Cu2O was seen at ⫺160 mV, because in aer-
ated solutions the net current was cathodic at this potential.
Conclusions
The electrochemistry of copper in aqueous glycine solutions was
investigated by comparing the polarization behavior with appropri-
ate potential-pH diagrams. Due to its complexation action, glycine
increased the solubility range of copper to lower potentials and high-
er pH. The stability regions of CuO and Cu2O contracted with in-
creasing amounts of glycine. The polarization behavior agreed well
with the thermodynamic potential-pH diagrams. The anodic dissolu-
tion rate of copper in glycine solutions was controlled by the solu-
tion pH, concentrations of dissolved oxygen, and glycine.
The University of California at Berkeley assisted in meeting the publica-
tion costs of this article.
References
1. J. van Muylder, in Comprehensive Treatise of Electrochemistry, J. O’M. Bockris,
B. E. Conway, E. Yeager, and R. E. White, Editors, Vol. 4, p. 1, Plenum Press, New
York (1981).
Downloaded on 2014-12-19 to IP 130.160.4.77 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).
B57
2. C. A. Sequeira, Br. Corros. J., 30, 137 (1995).
3. M. D. Fallston, in Corrosion Engineering Handbook, P. A. Schweitzer, Editor,
p. 89, Marcel Dekker, Inc., New York (1996).
4. M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, p. 384,
Pergamon Press, London (1965).
5. M. Pourbaix, Lectures on Electrochemical Corrosion, p. 49, Plenum Press, New
York (1973).
6. H. E. Johnson and J. Leja, J. Electrochem. Soc., 112, 638 (1965).
7. D. R. Turner and G. R. Johnson, J. Electrochem. Soc., 109, 798 (1962).
8. J. Halpern, H. Milant, and D. R. Wiles, J. Electrochem. Soc., 106, 647 (1959).
9. J. Halpern, J. Electrochem. Soc., 100, 421 (1953).
10. S. C. Sircar and D. R. Wiles, J. Electrochem. Soc., 107, 164 (1960).
11. A. G. Keenan, C. H. Webb, and D. A. Kramer, J. Electrochem. Soc., 123, 179
(1976).
12. R. M. Smith and A. E. Martell, Critical Solubility Constants, Vol. 6, Plenum Press,
New York (1977).
13. D. D. Wagman, W. H. Evans, V. B. Parker, R. H. Schumm, I. H. Halow, S. M. Bai-
ley, K. L. Churney, and R. L. Nuttall, J. Phys. Chem. Ref. Data, 11, Suppl. No. 2
(1982).
14. D. Tromans and J. C. Silva, Corros. Sci., 53, 16 (1997).
15. D. Tromans, J. Electrochem. Soc., 145, L42 (1998).
16. M. Stern and A. L. Geary, J. Electrochem. Soc., 104, 56 (1957).
17. M. Stern and M. Weisert, Proc. ASTM, 59, 1280 (1959).
18. D. A. Jones, Principles and Prevention of Corrosion, p. 146, Prentice Hall, Engle-
wood Cliffs, NJ (1996).
19. J. Halpern, H. Milant, and D. R. Wiles, J. Electrochem. Soc., 106, p. 647 (1959).
20. H. H. Strehblow and B. Titze, Electrochim. Acta, 25, 839 (1980).
21. S. L. Marchiano, C. I. Elsner, and A. J. Arvia, J. Appl. Electrochem., 10, 365
(1980).
22. M. R. Gennero De Chialvo, S. L. Marchiano, and A. J. Arvia, J. Appl. Electrochem.,
14, 165 (1984).
23. W. Kautek and J. G. Gordon II, J. Electrochem. Soc., 137, 2672 (1990).