Diffusion properties of bilayer membranes based on MC-40 and MF-4SC modified with silicon and zirconium oxides

Article abstract of DOI:10.1134/S0036023610040017

MC-40 membrane samples modified with a thin MF-4SC layer containing inorganic oxide particles have been synthesized. Deposition of an MF-4SC layer raises the diffusion permeability of the membrane. Insertion of ZrO₂ or SiO₂ nanoparticles into this layer enhances the ion transport selectivity in terms of the cation transport number. The best results are obtained with oxide particles synthesized in the pores of the deposited layer.

Full text of DOI:10.1134/S0036023610040017

ISSN 0036ꢀ0236, Russian Journal of Inorganic Chemistry, 2010, Vol. 55, No. 4, pp. 479–483. © Pleiades Publishing, Ltd., 2010.
Original Russian Text © Yu.A. Karavanova, Z.M. Kas’kova, A.G. Veresov, A.B. Yaroslavtsev, 2010, published in Zhurnal Neorganicheskoi Khimii, 2010, Vol. 55, No. 4, pp. 531–536.
SYNTHESIS AND PROPERTIES
OF INORGANIC COMPOUNDS
Diffusion Properties of Bilayer Membranes Based on MCꢀ40
and MFꢀ4SC Modified with Silicon and Zirconium Oxides
Yu. A. Karavanovaa, Z. M. Kas’kovab, A. G. Veresovc, and A. B. Yaroslavtseva
a Kurnakov Institute of General and Inorganic Chemistry, Russian Academy of Sciences,
Leninskii pr. 31, Moscow, 119991 Russia
b Higher Chemical College of the Russian Academy of Sciences, Miusskaya pl. 9, Moscow, 125047 Russia
c Faculty of Chemistry, Moscow State University, Moscow, 119991 Russia
Received October 3, 2008
Abstract—MCꢀ40 membrane samples modified with a thin MFꢀ4SC layer containing inorganic oxide partiꢀ
cles have been synthesized. Deposition of an MFꢀ4SC layer raises the diffusion permeability of the memꢀ
brane. Insertion of ZrO2 or SiO2 nanoparticles into this layer enhances the ion transport selectivity in terms
of the cation transport number. The best results are obtained with oxide particles synthesized in the pores of
the deposited layer.
DOI: 10.1134/S0036023610040017
Ionꢀexchange and membrane materials are widely
used in water purification, separation processes, alterꢀ
native energy sources, gas sensors, etc. However,
materials possessing the best transport properties are
generally rather expensive, and, as a consequence,
they are of limited use in practice. It is, therefore, of
particular interest to modify cheaper membrane mateꢀ
rials so as to improve their transport properties. The
diffusion characteristics of membranes can be
changed considerably by inserting inorganic particles
into their pores [1, 2]. Unfortunately, this method
proved ineffective for modifying the inexpensive heterꢀ
ogeneous membrane material MCꢀ40, which is based
on sulfated polystyrene [3]. Since the key factor in
most transport processes is ion transfer across the
membrane surface, modifying the nearꢀsurface memꢀ
brane layers seems to be a promising approach to the
problem. For example, the properties of MCꢀ40
membranes can be improved by producing a relief on
their surface [4].
EXPERIMENTAL
Bilayer membranes were prepared as follows. Zirꢀ
conium oxide obtained by the hydrolysis of ZrCl4
(Merck) in aqueous ammonia followed by washing [5]
and commercial aluminum oxide (Aldrich, 200–400
mesh, 60 Å) were annealed at 350°С for 30 min. For
preparing a modifying solution, annealed ZrO2 or SiO2
was introduced into a 6.0% solution of MFꢀ4SC (proꢀ
tonated form) in isopropanol (0.05 g of oxide per gram
of the dry MFꢀ4SC polymer). The resulting mixture
was homogenized by holding it an ultrasound bath for
30 min.
Four layers of the modifying solution were applied
onto the MCꢀ40 membrane surface at 12ꢀh intervals.
Thereafter, the membrane was dried for 24 h in air.
Membrane samples were conditioned successively in
concentrated and dilute sodium chloride solutions.
Next, the membranes were converted into hydrogen
form by holding them in 5% HCl for 5–8 h and were
washed until free of chloride ions. The ulyimate thickꢀ
ness of the deposited layer was 0.01 0.001 mm.
In the other modification technique, the MFꢀ4SC
membrane was used as a kind of nanoreactor for oxide
synthesis. ZrCl4 (Merck; 0.05 g of ZrO2 per gram of the
dry polymer) or Si(OEt)4 (Fluka; 0.02 g of SiO2 per
gram of the dry polymer) was introduced into the
modifying MFꢀ4SC solution. The precursor was
hydrolyzed in an acid medium (for SiO2) or in an alkali
(for ZrO2).
A possible way of modifying the MCꢀ40 surface is
by depositing a thin layer of the homogeneous, sulꢀ
fonic cation exchanger–based, perfluorinated memꢀ
brane material MFꢀ4SC (Russian analogue of the
wellꢀknown membrane material Nafion), which has
better diffusion properties. The deposition of a thin
MFꢀ4SC layer does not make the membrane much
more expensive. Another advantage of this approach is
that the transport properties of the membrane can be
further improved by modifying the MFꢀ4SC layer with
inorganic admixtures.
The ionic conductivity of the resulting membranes
was measured with a 2Vꢀ1 ac bridge in the frequency
range of 10–6000000 Hz between 20 and 100°C.
Ionic conductivity at each temperature point was
determined by extrapolating the impedance
hodograph to the ohmic resistance axis.
Here, we report the synthesis and diffusion properꢀ
ties of MCꢀ40 membranes covered with an MFꢀ4SC
layer modified with ZrO2 or SiO2 particles.
479

480
KARAVANOVA et al.
Table 1. Diffusion permeability and H+/Na+ counterdiffusion coefficient data for the MCꢀ40/MFꢀ4SC bilayer membranes
Initial solution in the cell
MCꢀ40/MFꢀ4SC, modified layer MCꢀ40/MFꢀ4SC + ZrO2, modified layer
MCꢀ40
1
2
1
2
1
2
1 M HCl
H2O
H2O
H2O
H2O
6.46
×
×
×
×
×
10–8
1.72
2.74
1.3
×
10–7
10–7
10–7
10–7
10–5
1.13
2.59
9.68
8.90
1.25
×
×
×
×
×
10–7
10–7
10–8
10–8
10–5
1.23
4.55
1.95
4.79
1.78
×
×
×
×
×
10–7
10–7
10–7
10–8
10–5
9.82
3.77
1.17
4.41
1.09
×
×
×
×
×
10–8
10–7
10–7
10–8
10–5
0.1 M HCl
1 M NaCl
0.1 M NaCl
0.1 M HCl
2.32
7.23
1.16
1.24
10–7
10–8
10–7
10–5
×
×
×
×
1.01
2.25
0.1 M NaCl
The diffusion permeability and ionic conductivity microscope (JEOL, Japan) equipped with a PGT
IMIX Xꢀray microanalysis system (United Kingdom).
The thickness of membranes was measured with a
Mitutoyo digital micrometer with an accuracy of
0.001 mm.
were measured in a membraneꢀseparated twoꢀcomꢀ
partment cell filled with HCl and NaCl solutions at
different concentrations [6]. pH was measured with an
Expertꢀ001 pH meter (EconixꢀExpert). The pH meter
was calibrated against standard buffer solutions (pH
1.68, 4.01, 6.86, 9.18). The NaCl concentration was
determined conductometrically using an Econixꢀ
Expertꢀ002 instrument. The conductometer was caliꢀ
brated against standard NaCl solutions. The run was
terminated once the exchange process had arrived at a
steady state.
RESULTS AND DISCUSSION
Initially, we obtained MCꢀ40/MFꢀ4SC bilayer
membranes with presynthesized oxide particles
included in the MFꢀ4SC layer. These membranes have
a higher diffusion permeability than the unmodified
sample (Table 1).
However, bilyaer membranes may show asymmetry
of transport properties. The samples that we prepared
consist of inequivalent ionꢀexchange materials differꢀ
ing significantly in the concentration of ions being
transported. This situation is similar to the gradient
The microstructure of samples and the distribution
of elements were determined by electron probe Xꢀray
microanalysis using a JSMꢀ840A scanning electron
distribution of charge carriers in a p–n semiconductor
junction, which allows electricity to flow in only one
direction. In membranes, the charge carriers are ions
and their concentration gradient is much lower. As a
consequence, there can be a difference between the
ion transfer rates in opposite directions. The ion diffuꢀ
sion rates in our membranes indeed depend on the
membrane orientation in the cell, differing by 10–
50% (Table 1). Furthermore, the bilayer membranes
are characterized by a markedly higher diffusion transꢀ
fer rate.
However, these membranes are unstable and their
poor operational characteristics hamper their practiꢀ
cal application. The particle size of silicon oxide, parꢀ
ticularly that of commercial silica, is well above the
membrane pore size (Fig. 1). This circumstance, in
combination with the fairly low adhesion of the deposꢀ
ited layer, leads to a low mechanical strength and rapid
aging of this layer. After experiments, the deposited
layer of some membranes is disintegrated and its edges
are flaked away from the MCꢀ40 base.
2 μm
Fig. 1. Electron micrograph of the surface of the MCꢀ
40/MFꢀ4SC bilayer membrane with silicon oxide partiꢀ
cles.
RUSSIAN JOURNAL OF INORGANIC CHEMISTRY Vol. 55 No. 4 2010

DIFFUSION PROPERTIES OF BILAYER MEMBRANES
481
thereby ensuring the thermodynamic stability of the
forming nanoparticles.
This synthesis yields 2ꢀ to 5ꢀnm particles included
in the system of membrane pores and channels. We
demonstrated the formation of these particles in the
membrane by transmission electron microscopy, as
was done in an earlier study [7].
The membranes obtained via this procedure posꢀ
sess improved mechanical characteristics, so their
properties can be studied properly. Figure 3 presents
impedance data for these membranes. The ionic conꢀ
ductivity of the modified MCꢀ40 membrane is much
higher than that of the unmodified membrane and is
nearly equal to that of MFꢀ4SC. The membrane as a
whole has essentially the same properties as the thin
layer of the modifier. The ionic conductivity of the
membranes with zirconium oxide or silicon oxide
inclusions is slightly lower, yet it again far exceeds the
conductivity of the unmodified membrane.
2 μm
Fig. 2. Electron micrograph of the surface of the MCꢀ
40/MFꢀ4SC bilayer membrane with zirconium oxide parꢀ
ticles after diffusion measurements.
For this reason, at the next stage of our study we
prepared oxide particles inside MFꢀ4SC pores from
precursors contained in the modifying membrane
solution. Note that the sulfonic cationꢀexchange
membranes provide a unique matrix for nanoparticle
synthesis. Their nanosized pores [6] can efficiently
sorb one of the initial reactants, such as metal cations.
It is then possible to synthesize nanoparticles inside
these pores, using their walls as the size limiter. Furꢀ
thermore, the pore walls can reliably isolate particles
Tables 2 and 3 list diffusion permeability data and
H+/Na+ counterdiffusion coefficients for the memꢀ
branes examined. For the membranes with a deposited
MFꢀ4SC layer, the diffusion permeability is higher in
all solutions. However, the inorganic inclusions
decrease the diffusion permeability in the HCl soluꢀ
tions. For the membrane containing 5% ZrO2, this
effect is stronger than in the case of the membrane
containing 2% SiO2. On the contrary, the H+/Na+
one from another and reduce the surface tension, counterdiffusion coefficients for the membranes with
–1
log
σ
[Ω
]
d
c
–1.0
–1.2
–1.4
–1.6
b
а
30
40
50
60
70
80
T, °C
Fig. 3. Ionic conductivity of (
a
) MCꢀ40, (
b
) MFꢀ4SC, (c) MCꢀ40 modified with an MFꢀ4SC layer, and (d) MCꢀ40 modified with
an MFꢀ4SC layer with ZrO particles at various temperatures.
2
RUSSIAN JOURNAL OF INORGANIC CHEMISTRY Vol. 55 No. 4 2010

482
KARAVANOVA et al.
Table 2. Diffusion permeability coefficients (cm2/s) for the MCꢀ40 membrane modified with a thin MFꢀ4SC layer (the latꢀ
ter is exposed to the salt/acid solution)
Aqueous solution in conꢀ
tact with the membrane
MCꢀ40
MCꢀ40/MFꢀ4SC
MCꢀ40/MFꢀ4SC + SiO2 MCꢀ40/MFꢀ4SC + ZrO2
0.1 M HCl
1 M HCl
2.59
×
×
×
×
10–7
6.83
1.42
2.13
3.06
×
×
×
×
10–7
10–7
10–7
10–7
3.61
1.98
2.15
3.06
×
×
×
×
10–7
10–7
10–7
10–7
2.45
1.53
2.93
3.21
×
×
×
×
10–7
10–7
10–7
10–7
1.07
1.41
2.08
10–7
10–7
10–7
0.1 M NaCl
1 M NaCl
oxide particles are larger. This effect is explicable in pies a rather large fraction of the pore volume, causing
steric hindrance to mass transfer, particularly to anion
transfer, which takes place only in the solution inside
the pores. Therefore, the insertion of oxide particles
must markedly decrease the anion transport rate,
affecting the cation transfer rate to a lesser extent. This
was indeed observed in our experiments. In addition,
since inorganic sorbents typically show higher sorpꢀ
tion selectivity [9], ion transfer across the modified
membranes can be more selective.
terms of the ion transport mechanism. Because the
membranes examined are cation exchangers, their
cation transfer rate is substantially higher than their
anion transfer rate. The diffusion transport of HCl or
NaCl is limited by anion (Cl–) transfer and is on the
order of 10–7 cm2/s. In the counterdiffusion measureꢀ
ments, only cation countertransport takes place. In
this case, the limiting process is the transfer of the less
mobile cation (protons are usually more mobile in sulꢀ
fonic cationꢀexchange membranes [8]).
The membrane samples examined in this study, as
well as the similar samples obtained earlier, exhibit a
wellꢀdefined ion transfer anisotropy (Table 4), whose
magnitude is as high as 50% in some cases. Note that
this anisotropy is much greater for dilute solutions. In
this case, this effect is pronounced much more clearly
than the same effect for the membranes considered
above. The anisotropy of transport properties, in comꢀ
bination with good mechanical characteristics, makes
these materials promising for, e.g., energy saving in
water purification.
The oxide nanoparticles inserted into the system of
membrane pores and channels significantly change
the diffusion properties of the membrane. These partiꢀ
cles have wellꢀpronounced sorption properties and
can transport cations over their surface. This increases
the cation transport rate. In addition, the oxide occuꢀ
Table 3. H+/Na+ counterdiffusion coefficients (cm2/s) for
the MCꢀ40 membrane modified with a thin MFꢀ4SC layer
and with an MFꢀ4SC layer containing inorganic admixtures
Thus, the MCꢀ40 membranes whose surface is covꢀ
ered with a thin MFꢀ4SC layer has diffusion properties
characteristic of single MFꢀ4SC membranes, but betꢀ
ter mechanical properties. This combination of propꢀ
erties makes it possible to use much thinner and,
accordingly, cheaper MFꢀ4SC layers without impairꢀ
ing their properties. Owing to their ion transport
anisotropy, the bilayer membranes can afford energy
saving in water purification.
Modifying layer exposed to
Membrane
HCl
NaCl
MCꢀ40
1.24
×
10–5
MCꢀ40/MFꢀ4SC
MCꢀ40/MFꢀ4SC + SiO2
9.33
1.44
×
×
×
10–6
10–5
10–5
8.90
1.01
1.50
×
×
×
10–6
10–5
10–5
ACKNOWLEDGMENTS
This work was supported by the federal purposeꢀ
oriented program “Research and Development in Priꢀ
ority Areas of Science and Technology in Russia in
2007–2012” and by the Russian Academy of Sciences
program “Theoretical and Experimental Study of the
MCꢀ40/MFꢀ4SC + ZrO2 1.70
RUSSIAN JOURNAL OF INORGANIC CHEMISTRY Vol. 55 No. 4 2010

DIFFUSION PROPERTIES OF BILAYER MEMBRANES
483
Table 4. Diffusion permeability coefficients (cm2/s) for transport in the two different directions across the MCꢀ40 memꢀ
brane modified with a thin MFꢀ4SC layer
MCꢀ40/MFꢀ4SC
salt/acid water
4.93
10–7
MCꢀ40/MFꢀ4SC + SiO2
salt/acid water
MCꢀ40/MFꢀ4SC + ZrO2
salt/acid water
Diffusing
solution
0.1 M HCl
1 M HCl
6.83
1.42
2.13
×
×
×
10–7
10–7
10–7
×
3.61
×
×
×
10–7
10–7
10–7
2.08
×
×
×
10–7
10–7
10–7
2.45
×
×
×
10–7
10–7
10–7
1.21
×
×
×
10–7
10–7
10–7
1.21
1.53
×
10–7
10–7
1.98
2.15
1.62
1.77
1.53
2.93
1.43
2.73
0.1 M NaCl
×
4. V. I. Zabolotskii, S. A. Loza, and M. V. Sharafan, Elekꢀ
trokhimiya 41 (10), 1185 (2005) [Russ. J. Elektrochem.
41 (10), 1053 2005)].
Nature of the Chemical Bond and the Mechanisms of
the Most Important Chemical Reactions and Proꢀ
cesses.”
5. E. Yu. Voropaeva, I. A. Stenina, A. G. Veresov, et al.,
Zh. Neorg. Khim. 53 (3), 397 (2008) [Russ. J. Inorg.
Chem. 53 (3), 350 (2008)].
REFERENCES
1. B. Bonnet, D. J. Jones, J. Roziere, et al., J. New Mater.
6. W. Y. Hsu and T. D. Gierke, J. Membr. Sci. 13, 307
Electrochem. Syst. 3, 87 (2000).
(1983).
2. D. J. Jones and J. Roziere, Handbook of Fuel Cells—
Fundamentals, Technology and Applications, Ed. by
W. Vielstich, H. A. Gasteiger, and V. A. Lamm (Wiley,
New York, 2003).
3. S. A. Novikova, E. I. Volodina, N. D. Pis’menskaya, et
al., Elektrokhimiya 41, 1205 (2005) [Russ. J. Elektroꢀ
chem. 41, 1070 (2005)].
7. E. Yu. Voropaeva, I. A. Stenina, G. Yu. Yurkov, and
A. B. Yaroslavtsev, Zh. Neorg. Khim. 53 (10), 1637
(2008) [Russ. J. Inorg. Chem. 53 (10), 1531 (2008)].
8. A. B. Yaroslavtsev, V. V. Nikonenko, and V. I. Zaboꢀ
lotskii, Usp. Khim. 72, 438 (2003).
9. A. B. Yaroslavtsev, Usp. Khim. 66 (7), 641 (1997).
RUSSIAN JOURNAL OF INORGANIC CHEMISTRY Vol. 55 No. 4 2010