Electrochimica Acta 146 (2014) 792–797

Contents lists available at ScienceDirect

Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta

On the relation between onset of bubble nucleation and gas

supersaturation concentration
M.E. Tawfik, F.J. Diez ∗
Rutgers, The State University of New Jersey, Piscataway, NJ, 08854 USA

a r t i c l e i n f o a b s t r a c t

Article history: The time taken for gas bubbles to start forming on an electrode operating in a stagnant aqueous solution
Received 4 June 2014 is driven by the dissolved gas mass-transfer near the electrode. This is the result of two processes that
Received in revised form 28 August 2014 compete with one another. One adds dissolved gas molecules from the chemical reactions at the elec-
Accepted 30 August 2014
trode surface and the other diffuses these molecules towards the bulk. Using this mechanism, a model
Available online 21 September 2014
is proposed that can predict the dissolved gas supersaturation concentration up to the onset of bubble
nucleation at the electrodes (heterogeneous nucleation). Experimental measurement of the bubble onset
Keywords:
nucleation time is incorporated within this model to calculate the critical supersaturation concentration
Supersaturation concentration
Onset bubble nucleation
for dissolved hydrogen and oxygen gas. The results show a strong link between the applied current den-
Diffusion sity and the supersaturation concentration at the electrode surface. A relation was obtained that predict
Hydrogen concentration the onset of bubble nucleation showing an excellent agreement with the measurement.
Electrolysis © 2014 Elsevier Ltd. All rights reserved.

1. Introduction The dissolved gas concentration is governed by the rate at which

During electrochemical reaction processes such as electrolysis
of water, electrons flow from the anode to the cathode where they
are consumed by hydrogen ions to form hydrogen gas. The dis-
solved gas builds up until a critical supersaturation concentration
is reached at the electrode surface which marks the onset of bub-
ble nucleation and undergoes a phase change [1,2]. To balance the
electrical charge, hydroxide ions move through the electrolyte solu-
tion to the anode, where they give away electrons. Recent work
has aimed at preventing or controlling gas bubble generation. For
example, in electro-osmotic pumps, the flow rate is restricted due
to the formation of bubbles at the membrane [3]. In electrophoretic
deposition the bubbles generated at the electrodes may yield dam-
aged and poor quality coatings [4]. One approach at controlling
gas generation has been performing the electrochemical process
under modulated electric fields such as pulsed direct current and
alternating current [5–7]. A better understanding of the relation
between the dissolved gas supersaturation concentration and the
bubble nucleation time at the electrode surface during the electro-
chemical reaction could be used to suppress or eliminate the gas
generation.
∗ Corresponding author.
E-mail address: diez@rutgers.edu (F.J. Diez).
the electroactive species are brought to the electrode surface, by the
process of desorption from the liquid phase to the gaseous phase
and by the diffusion of the dissolved gas towards the bulk [8–10].
At a gas-evolving electrode the gas concentration is dependent on
the mass transfer due to single-phase free convection from the
density gradients and also it is dependent on the two-phase free
convection induced by rising bubbles [11,12]. Many studies have
considered the different mechanisms that affect the mass transfer
at gas-evolving electrodes and the efficiency when bubbles adhere
to the electrode surfaces [13–16]. Present studies model the bub-
ble production using empirical correlations with current density
[17,18] without solving for the concentration distribution. These
models implicitly assume that the nucleation and growth of bub-
bles is very fast. Other models assume that the bubbles nucleation
and growth is governed by a linear rate law which depends on the
change of the dissolved gas supersaturation concentration [19,1].
During electrolysis, the mass transfer of the electroactive species
(radicals) is controlled by their migration, diffusion and convec-
tion and it can be modeled by the Nernst-Planck equation [20]
[9,10,21]. These radicals convert to dissolved gas by losing or
gaining electrons at the electrodes. At the short initial stage of
the bubble nucleation in a stagnant electrolyte (the focus of the
present work), the single-phase free convection mass-transfer can
be neglected, and the dissolved gas concentration at the electrode
is governed by the diffusion equation (Fick’s second law) [22,20].
This simplification required a convectionless liquid. The dissolved

http://dx.doi.org/10.1016/j.electacta.2014.08.147
0013-4686/© 2014 Elsevier Ltd. All rights reserved.
 M.E. Tawfik, F.J. Diez / Electrochimica Acta 146 (2014) 792–797 793

gradient as shown in Fig. 1. The temporal evolution of the dissolved

Fig. 1. Graphical illustration of CH+ and CH2 mass transfer near the electrode.
gas concentration increases with time up to the critical supersatu-
ration concentration when the onset of bubble nucleation occurs.
Since the bubbles do not nucleate infinitely fast, some time elapses
until the appearance of the gas phase [23]. This differs substantially
from what happens in continuous electrolysis where nucleation
time is very small or possibly inexistent. While bubble production
and growth has received much attention, there is a lack of infor-
mation on the prediction of onset of bubble formation. Only the
work by Dinkelacker [24] reported the bubble formation time as a
gas concentration can be written in terms of Fick’s second law as,
2
∂Ci (x, t) ∂ Ci (x, t)
= Di , (2)
∂t ∂x2
where Di is the diffusion coefficient of species i in the working elec-
trolyte. At the electrode boundary, the dissolved gas flux, Ji (x, t), is
proportional to the concentration gradient
∂Ci (x, t)
Ji (x, t) = −Di . (3)
∂x
The analytical solution to Eq. 2 provides the evolution of the con-
centration, Ci (x, t) sketched in Fig. 1 but it can not predict the onset
of bubble nucleation t* . The goal is then to combine the solution
to Eq. 2 with experimental measurements of the bubble formation
time t* to calculate Ci∗ . To solve Eq. 2, the boundary and initial con-
ditions are needed. We can assume that, before the electric field is
applied, the species concentration, Ci , is equal to the bulk satura-
tion concentration of the dissolved gas, Ci∞ . Considering that the
electrolysis cell is large compared to the diffusion thickness, the
concentration far from the electrode reaches the bulk concentra-
tion, Ci∞ . This can be written as,
Ci (x, 0) = Ci∞
(4)
lim Ci (x, t) = Ci∞ .
x→∞

function of current density. The present work relates the onset gas When a constant current, I, is applied across the electrolysis cell,
formation time with the critical supersaturation concentration near the resultant flux across the electrode surface, Js , is constant and
the electrode in a stagnant electrolyte. A supersaturation model is given by
proposed that uses Fick’s law and results from experimental mea-  
surements of the bubble formation time. I ∂Ci (x, t)
−Js = −Ji (0, t) = = Di
zFA ∂x x=0
2. Theoretical model We can now solve Eq. 2 to obtain the concentrations of oxidized
and reduced species at the electrode (Sand [22] and Karaoglanoff
2.1. Chemistry of water electrolysis [26]) giving
 1/2    
Js Di t x2 x
The present study models the increase in the dissolved gas con- Ci (x, t) = Ci∞ + 2 exp − − x erfc (5)
Di  4Di t 2(Di t)
1/2
centration Ci , of species i from basic principles to solve for the
critical supersaturation concentration Ci∗ . The number of dissolved This equation can be written in terms of the non-dimensional
gas moles generated at the electrode, n, is related to the current parameter  as
drawn by Faraday’s law,  
Js x 1
It Ci (x, t) = Ci∞ + √ exp(−2 ) − erfc() (6)
n= (1) Di 
Fz
where x is nondimensionalized
√ by a diffusion
√ thickness ı, with a
where I is the constant current drawn by the electrolyte, z is the
general scaling of ı∼ Dt, so that  = x/ 4Dt. Further details about
valency number of ions of the substance (electrons transferred per
the choice of ı will be given next. We assume that before the onset
ion), and F is the Faraday’s constant.
of bubble formation, this diffusion equation (Eq. 6) dominates the
In our experiments we have chosen to work with simple, low
mass transfer. Once bubbles are formed, other mass transfer mech-
ionic concentration aqueous buffers and inert platinum electrodes
anisms such as single-phase and two-phase convection and gas
so that the electrodes half-reactions (acid) is,
adsorption can dominate. These have been widely considered in
Cathode (Reduction) : 2H+ (l) + 2e− −→ H2 (l) the study of the gas-evolving electrodes [8–12,27,28], but will not
be considered here since our analysis of the problem stops at the
Anode (Oxidation) : 2H2 O(l) −→ O2 (l) + 4H+ (l) + 4e− onset. This diffusion equation is plotted in Fig. 2, showing how the
concentration of the dissolved gas, Ci (x, t), varies as we move away
The dissolved H2 and O2 gas, formed at the electrodes, build from the electrode as well its variation at the electrode Ci (0, t). As
∗ and C ∗ respec-
up until it reaches the critical supersaturation CH
2 O2 time increases, the concentration at the electrode surface increases
tively. Beyond this threshold and in the presence of a nucleation and also the the boundary layer thickness from the surface where
site, crack or cavity in the electrode surface, bubbles are initiated the dissolved gas concentration CH2 is larger than the bulk con-
[25]. centration. This region of influence corresponds  to a diffuse layer
thickness ı which can be shown to scale as ∼ Di t.
2.2. Concentration gradient of dissolved gas The results in Fig. 2, show that the concentration of dissolved
gas at the electrode increases with time. In such an environment
The dissolved gas generated at the electrodes diffuses away rich in excess dissolved gas molecules, the conditions are being set
towards the bulk. This produces a higher concentration near the for bubble nucleation. The process is not well understood with the
electrode than at the bulk solution resulting in a concentration traditional view being that the onset of nucleation is the result of
794 M.E. Tawfik, F.J. Diez / Electrochimica Acta 146 (2014) 792–797

Fig. 2. Concentration profile of H2 (l) before onset of bubble nucleation showing a) Concentration of the dissolved gas as a function of distance from the electrode, and b)
Concentration evolution with time at the surface of the electrode.

this environment reaching some energy state where instantaneous

nucleation occur. Nevertheless, some macroscopic-scale proper-
ties can be extracted from this process. First, higher fluxes at the
electrode surface, Js , will increase the dissolved gas concentration
due to increase in the reaction rate. This will increase interactions,
which suggest that the time it takes to reach t* will be faster with an
increase in the applied current. Experimentally we will show this
dependence with I/A. Second, for the proposed diffusion-controlled
model, a diffuse layer, that is a function of time, grows from the
electrode surface. Physically, the diffuse layer represents the region
where the majority of the supersaturated dissolve gas being gener-
ated at the electrode is being contained and which increases with
time. It is expected that the onset of nucleation will be dependent
on this layer properties as will be shown from the experiments.
There are multiple way
√ of defining the diffusion thickness ı, with
a general scaling of ∼ Dt. Here, it is defined as the distance from
the electrode surface where the mass-flux ratio reaches J/Js ∼ 0.01.
Fig. 3. Average Concentration (
CH2 − CH∞ ) evolution with time for different current
The mass-flux J(x, t) can be calculated by differentiating Eq. 5 and 2
densities. Circles represent onset of bubble nucleation measurements discussed in
substituting it into Eq. 3. The solution can then be written in terms section 4.
of the non- dimensional parameter  as

J(x, t) = Js erfc() (7)

√ layer. To find this relation, first we will evaluate the nucleation
where  = x/ 4Dt. This shows that the mass-transfer and the
mass-flux ratio J/Js depends only on the non- dimensional param-
eter . For the ratio of J/Js ∼ 0.01, it can be shown from Eq. 7 that
 ∼ 1.82. This is the distance from the electrode where the mass-flux
is nearly zero and can be written as,
 The electrochemical measurements are conducted at room tem-
ı  3.64 Di t.
The diffuse layer contains most of the dissolved hydrogen CH2
that is being generated at the electrode. This can be estimated by
calculating the average concentration CH2 within the diffuse layer
as,
ı
1 function generator is used to supply power to the electrochemical
Ci (t) = Ci (x, t) dx
ı 0
0.274 Js √
  t + Ci∞ .
DH2
As the diffuse layer grows with time, the average concentration
CH2 within the diffuse layer also grows as shown in Fig. 3. These are
the results for four different constant current densities, I/A, applied.
The measurements show that for high currents the concentration
increases faster with time as expected. Although this plot resolves
the behavior of the concentration in the diffuse layer, additional
information is needed to determine the onset of bubble nuclea-
tion. The hypothesis is that the onset time, t* , should be related
to the concentration of the gas molecules in the growing diffuse
time experimentally.
3. Experimental setup
(8)
perature (20 ◦ C) in a 10 ml electrochemical cell. The primary
solutions used are deionized ultra-filtered water and 1 mM borate
buffer made with sodium tetraborate(≥99% pure) and boric acid
(≥99% pure) from Sigma Aldrich. Platinum wires (99.9 % pure)
0.5 mm in diameter from Sigma Aldrich are used as the cell elec-
trodes. A Trek model 5/80 amplifier with a Tektronix AFG 3021B
cell. A microscope is used to measure the bubble formation time
(9)
by taking a video of the working electrode while applying a con-
stant current across the electrochemical cell. The bubble nucleation
time depends on the critical supersaturation CH∗ value of hydrogen
2
dissolved gas. Experimentally, this is evaluated by measuring the
time it takes for the first bubbles to form from the time the volt-
age is applied. The time is measured until the first bubble reaches
50 m in diameter. This type of technique has been applied in the
past and there can be small variations in the results depending on
the method used to detect the formation of the first bubbles [23].
There could be a systematic bias on the time measured by different
methods but it would not affect the relation between the nuclea-
tion time and supersaturation concentration. In that relation the
 M.E. Tawfik, F.J. Diez / Electrochimica Acta 146 (2014) 792–797 795

Fig. 4. Onset of bubbles nucleation as a function of current density for a) H2 bubbles, and b) O2 bubbles.

Fig. 5. Average critical supersaturation concentration as a function of a) onset time and b) current density, for both H2 and O2 .

slope would remain the same but the constant could show small
variations.

4. Results

The time t* at which the first hydrogen bubble is observed

forming on the surface of the cathode is measured for different
current densities and compared with published data [24] in Fig. 4a.
Similar measurements are performed at the anode, to resolve the
time needed for the onset of oxygen bubbles for different applied
currents as shown in Fig. 4b. The small variations between the
present work and that by Dinkelacker [24] in Fig. 4a is attributed to
different measurement methods used and different definitions of
the critical supersaturation concentration (Ci∗ ) discussed in the lit-
erature. In Dinkelacker [24] measurements are for Nickel electrodes
in 5 molar KOH solution run at 20 ◦ C, while the present experi- Fig. 6. Diffusion layer thickness as a function of current density at t* .

ment uses Platinum electrodes in 1 mM Borate buffer. Shibata [29]

defines Ci∗ at the electrode surface, while Glas [30] defines it at
the bubble surface. A comparison between these two definitions
has been reported by Vogt [31] which discusses the importance of
each definition. A comparison between the results in Fig. 4a and 4b
shows that for an applied current the onset of H2 bubbles is two to
three times faster than for O2 bubbles.
Once the nucleation time is experimentally measured, the
concentration at the electrode surface Ci (0, t* ) and the average con-
centration within the diffuse layer,
Ci∗ can be obtained from Eqs. 5
and 9 respectively. ∗ and
CH CO∗ are plotted in Fig. 5a showing a log-
2 2
arithmic decay for increasing t* . This suggests that the nucleation
time does not occur at constant average concentration within the
diffuse layer but rather
Ci∗ decreases for increasing t* when the cur-
rent applied decreases. Similarly, Fig. 5b captures the dependence
between the calculated ∗
CH and CH2 (0, t ∗ ) and applied current
2
density. The thickness of the diffusion layer, ı, can also be obtained
at the nucleation time from Eq. 8 for the five different current den-
sities tested and it is plotted in Fig. 6. The results show that at the
onset time t* the diffusion thickness decreases as the current den-
sity is increased. The results are compared
√ to Nefedov [27] which
proposed a diffusion thickness, ı = Dt based on an experimen-
tal analysis of the bubble growth. Although the work in [27] deals
with bubble growth rate and how they affect mass transfer, we can
compare their relation at the onset of bubble formation with Eq. 8
as shown in Fig. 6. The trend is the same for both curves, since both
equations only differ by a constant due to the manner in which they
are obtained. One (Eq. 8) is obtained from the diffusion equation,
while the other (Eq. 9) relates to bubble growth. In summary, this
796 M.E. Tawfik, F.J. Diez / Electrochimica Acta 146 (2014) 792–797
Fig. 7. The onset time dependency on the concentration at the electrode surface
and the current density.
thin diffuse layer surrounding the electrode describes where the
excess CH2 is found before bubble nucleation, and according to [27]
it plays a role during bubble growth.
The results have shown that the nucleation time t* is a function
of the applied current flux. They have further shown that it is also a
function of the concentrations at the electrode surface C(0, t* ) and
the average concentration within the diffuse layer, C ∗ . It should be
* ∗
noted that both C(0, t ) and C can be obtained from the diffusion
equation and are simply related by a constant. The measurements
were performed for two different reactions (i.e.: H2 and O2 gas gen-
eration) with different diffusivities, D, (DH2 = 4.5 × 10−9 m2 /s−1 ,
DO2 = 2.4 × 10−9 m2 /s−1 [32]) with the results showing similar
scaling. This further supports that the nucleation time is diffusion-
controlled. An empirical relation can be obtained that includes the
dependency of t* on C(0, t* ) and I/A which can be written as
C(0, t ∗ )a (I/A) = constant.
b
(10)
By fitting the experimental results for both H2 and O2 with Eq.
10, a value of a = 1 and b = −0.5 are obtained which support the
dependencies observed from the measurements. This relation, mul-
tiplied by t* , is plotted in Fig. 7, and the data fitted by 0.85 t* . By
substituting into Eq. 10 the concentration at the electrode surface
obtained from the diffusion Eq. 5 as
2(I/A) √
C(0, t) − C ∞ = √ t, (11)
zF D
we can solve for t* . This provides a relation between t* and the
applied current density, I/A, as
 I −1
t ∗ = 0.19z 2 F 2 D . (12)
A [11] L. Janssen, E. Barendrecht, The effect of electrolytic gas evolution on mass trans-
This relation can be used to predict the onset of bubble nuclea-
tion. It is plotted in Fig. 4 showing an excellent agreement with the
t* measurements.
5. Conclusion
The mass-transfer mechanism that governs the dissolved gas
concentration at an electrode before the onset of bubble nucleation
is discussed. This is a diffusion-control mechanism where the dis-
solved gas concentration distribution due to electrolysis is obtained
by solving Fick’s second law. The closed form solution is evaluated
to obtain the concentration at the electrode surface as well as the
average concentration within the diffuse layer.
The onset of bubble nucleation, t* , at which bubbles are first
observed forming on the surface of inert electrodes was measured
for an aqueous buffer. Both the onset of H2 and O2 bubble forma-
tion was measured for a range of applied currents. At the calculated
time, t* , the dissolved gas concentration at the electrode surface and
the average concentration within the diffuse layer was predicted
using the diffusion model. The results show that the nucleation
time does not occur at constant Ci∗ , but rather it depends on the cur-
rent applied, with Ci∗ decreasing as I/A is decreased. Similarly, the
results also show that the diffusion thickness decreases at the onset
time t* , as the current density is increased. In summary, the nuclea-
tion time t* is a function of the applied current density, and there
is a strong link between the applied current density and the con-
centration in the diffuse layer. An empirical relation was obtained
that demonstrated this dependency showing an excellent agree-
ment with the measurements. These were used to obtain a relation
between t* and the applied current density that predicts the onset
of bubble nucleation in inert electrodes with only the additional
knowledge of the diffusivity of the dissolved gas. This, in turn, can be
used with the diffusion model to obtain the critical supersaturation
concentration.
Acknowledgment
The authors gratefully acknowledge the financial support pro-
vided for this study by the Office of Naval Research (ONR), Grant No.
N00014-11-1-0019, with Dr. Thomas F. Swean of the Ocean Engi-
neering and Marine Systems Program serving as Program Manager.
References
[1] S. Van Damme, P. Maciel, H. Van Parys, J. Deconinck, A. Hubin, H. Deconinck,
Bubble nucleation algorithm for the simulation of gas evolving electrodes, Elec-
trochemistry Communications 12 (5) (2010) 664–667.
[2] S. Jones, G. Evans, K. Galvin, Bubble nucleation from gas cavities - a review,
Advances in Colloid and Interface Science 80 (1) (1999) 27–50.
[3] D. Piwowar, M. Tawfik, F. Diez, High field asymmetric waveform for ultra-
enhanced electroosmotic pumping of porous anodic alumina membranes,
Microfluidics and Nanofluidics 15 (6) (2013) 1–12.
[4] M. Ammam, Electrophoretic deposition under modulated electric fields: a
review, RSC Advances 2 (20) (2012) 7633–7646.
[5] P. Selvaganapathy, Y. shun Leung Ki, P. Renaud, C. Mastrangelo, Bubble-free
electrokinetic pumping, Journal of Microelectromechanical Systems 11 (5)
(2002) 448–453.
[6] A. Nold, R. Clasen, Bubble-free electrophoretic shaping from aqueous suspen-
sion with micro point-electrode, Journal of the European Ceramic Society 30
(14) (2010) 2971–2975.
[7] B. Neirinck, J. Fransaer, O.V.d. Biest, J. Vleugels, Aqueous electrophoretic
deposition in asymmetric AC electric fields (AC-EPD), Electrochemistry Com-
munications 11 (1) (2009) 57–60.
[8] E.O. Vilar, E.B. Cavalcanti, I.L. Albuquerque, A Mass Transfer Study with elec-
trolytic gas production, Advanced Topics in Mass Transfer 9 (2011), 978-953.
[9] H. Vogt, The role of single-phase free convection in mass transfer at gas evolving
electrodes-I. Theoretical, Electrochimica acta 38 (10) (1993) 1421–1426.
[10] H. Vogt, The role of single-phase free convection in mass transfer at gas evolv-
ing electrodes-II. Experimental verification, Electrochimica acta 38 (10) (1993)
1427–1431.
fer at electrodes, Electrochimica Acta 24 (6) (1979) 693–699.
[12] G.M. Whitney, C.W. Tobias, Mass-transfer effects of bubble streams rising near
vertical electrodes, AIChE Journal 34 (12) (1988) 1981–1995.
[13] L. Janssen, Mass transfer at gas evolving electrodes, Electrochimica Acta 23 (2)
(1978) 81–86.
[14] H. Vogt, The rate of gas evolution of electrodes I. An estimate of the efficiency of
gas evolution from the supersaturation of electrolyte adjacent to a gas-evolving
electrode, Electrochimica Acta 29 (2) (1984) 167–173.
[15] H. Vogt, On the gas-evolution efficiency of electrodes I - Theoretical, Elec-
trochimica Acta 56 (3) (2011) 1409–1416.
[16] H. Vogt, On the gas-evolution efficiency of electrodes. II-Numerical analysis,
Electrochimica Acta 56 (5) (2011) 2404–2410.
[17] H. Vogt, R. Balzer, The bubble coverage of gas-evolving electrodes in stagnant
electrolytes, Electrochimica Acta 50 (10) (2005) 2073–2079.
[18] R. Wuthrich, C. Comninellis, H. Bleuler, Bubble evolution on vertical elec-
trodes under extreme current densities, Electrochimica Acta 50 (25-26) (2005)
5242–5246.
[19] P. Maciel, T. Nierhaus, S.V. Damme, H.V. Parys, J. Deconinck, A. Hubin, New
model for gas evolving electrodes based on supersaturation, Electrochemistry
Communications 11 (4) (2009) 875–877.
 M.E. Tawfik, F.J. Diez / Electrochimica Acta 146 (2014) 792–797
[20] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and Applica-
tions, 2 edn., John Wiley & Sons, 2001.
[21] H. Vogt, Mechanisms of mass transfer of dissolved gas from a gas-evolving
electrode and their effect on mass transfer coefficient and concentration over-
potential, Journal of Applied Electrochemistry 19 (1989) 713–719.
[22] H.J. Sand III, On the concentration at the electrodes in a solution, with special
reference to the liberation of hydrogen by electrolysis of a mixture of cop-
per sulphate and sulphuric acid, Philosophical Magazine Series 6 1 (1) (1901)
45–79.
[23] D. Kashchiev, A. Firoozabadi, Kinetics of the initial stage of isothermal gas phase
formation, The Journal of chemical physics 98 (1993) 4690.
[24] M. Dinkelacker, Zur Gasentwicklung und Blasenbildung an Elektronen, Thesis
Universitat Stuttgart (1989).
[25] B.C. Donose, F. Harnisch, E. Taran, Electrochemically produced hydrogen bub-
ble probes for gas evolution kinetics and force spectroscopy, Electrochemistry
Communications 24 (2012) 21–24.
797
[26] V.Z. Karaoglanoff, Über Oxydations-und Reduktionsvorgänge bei der Elek-
trolyse von Eisensalzlösungen, Zeitschrift für Elektrochemie und angewandte
physikalische Chemie 12 (1) (1906) 5–16.
[27] V.G. Nefedov, Mass transfer to a gas-generating electrode, Russian journal of
electrochemistry 34 (1) (1998) 16–24.
[28] V. Nefedov, O. Artyushenko, E. Kashevarova, Mass transfer to horizontal
gas-generating electrodes, Russian Journal of Electrochemistry 42 (6) (2006)
638–642.
[29] S. Shibata, The concentration of molecular hydrogen on the platinum cathode,
Bulletin of the Chemical Society of Japan 36 (1) (1963) 53–57.
[30] J. Glas, J. Westwater, Measurements of the growth of electrolytic bubbles, Inter-
national Journal of Heat and Mass Transfer 7 (12) (1964) 1427–1443.
[31] H. Vogt, On the supersaturation of gas in the concentration boundary layer of
gas evolving electrodes, Electrochimica Acta 25 (5) (1980) 527–531.
[32] D.R. Lide, CRC Handbook Chemistry and Physics, 85th Edition, CRC Press, 85
edition edn., 2004.