Chloride ion oxidation in molten alkaline-earth metal, magnesium, and zinc chlorides

Article abstract of DOI:10.1134/S0036023612110058

The interaction of molten alkaline-earth metal, magnesium, and zinc chlorides with oxygen has been investigated. The products of these reactions are molecular chlorine and the respective metal oxides. The reactivity of the chlorides increases linearly wit

Full text of DOI:10.1134/S0036023612110058

ISSN 0036ꢀ0236, Russian Journal of Inorganic Chemistry, 2012, Vol. 57, No. 11, pp. 1432–1435. © Pleiades Publishing, Ltd., 2012.
Original Russian Text © Yu.S. Chekryshkin, A.N. Chudinov, A.A. Fedorov, T.A. Rozdyalovskaya, 2012, published in Zhurnal Neorganicheskoi Khimii, 2012, Vol. 57, No. 11,
pp. 1523–1527.
SYNTHESIS AND PROPERTIES
OF INORGANIC COMPOUNDS
Chloride Ion Oxidation in Molten AlkalineꢀEarth Metal,
Magnesium, and Zinc Chlorides
Yu. S. Chekryshkin, A. N. Chudinov, A. A. Fedorov, and T. A. Rozdyalovskaya
Institute of Technical Chemistry, Ural Branch, Russian Academy of Sciences,
ul. Akademika Koroleva 1, Perm, 614000 Russia
Received July 26, 2011
Abstract—The interaction of molten alkalineꢀearth metal, magnesium, and zinc chlorides with oxygen has
been investigated. The products of these reactions are molecular chlorine and the respective metal oxides.
The reactivity of the chlorides increases linearly with an increasing polarizing power of the cation or with an
increasing effective polarizing power of the mixture of cations. The magnesium oxide and magnesium oxide–
zinc oxide powders resulting from the oxidation of the respective chlorides have been analyzed.
DOI: 10.1134/S0036023612110058
The interaction of metal chlorides with oxygen is of the resulting magnesium oxide and MgO–ZnO powꢀ
interest to those who are engaged in the synthesis of ders.
vanadates [1–3] and salt lubricants for hot rolling of
tubes [4], in the synthesis of metal oxide fine powders
EXPERIMENTAL
or singleꢀcrystal films [5–7], in chlorineꢀcontaining
waste treatment yielding chlorine [8, 9], and in the
design of stable catalytic systems for oxidation of haloꢀ
genated hydrocarbons [10–13]. Production of some
metal oxides and oxychlorides, including their recovꢀ
ery from spent nuclear fuel, also includes the oxidation
of chlorinated metalꢀcontaining mixtures, yielding
molecular chlorine [14–17].
The chloride ion was oxidized by bubbling air or
oxygen (5 L/h) through the chloride melt at the preset
temperature. The experimental procedure and appaꢀ
ratus were the same as in our earlier study [11]. For
isolation of the resulting oxides, the melt was cooled to
room temperature and the cake was dissolved in disꢀ
tilled water and was washed clean of the unreacted
–
chlorides. The residual Cl concentration in the wash
There have been studies of the catalytic effect of
transition metal oxides on chloride ion oxidation in
molten sodium, calcium, zinc, and barium chlorides
water was determined by capillary electrophoresis on
an Agilent G1600 AX system using a proprietary buffer
solution (pH 7.7) at a temperature of 20 С and an
applied voltage of 30 kV. The constants characterizing
the chlorides were taken from a handbook [19].
°
[
5, 11, 12, 18]. This effect is due to the polarizing
action (ionic moment) of the oxide’s cation, ze
where is the charge of the metal ion, = 4.803
10–10 esu
z+ is the radius of the cation.
/
rMz+,
z
e
×
The particle size of the synthesized magnesium
oxide and MgO + ZnO mixture was determined with
an Olympus BX51 light microscope. The composition
of the reaction products was determined by Xꢀray difꢀ
fraction on a Shimadzu XRD 7000 diffractometer
r
is the unit charge, and M
It was demonstrated that the chlorine formation rate
increases with an increasing polarizing power (PP) of
the cation. No chloride ion oxidation is observed as
oxygen or air is bubbled through molten sodium chloꢀ
ride. The same process conducted in the presence of a
variableꢀvalence metal oxide yields chlorine, whose
amount is a linear function of the PP of the oxide’s catꢀ
ion. The catalytic activity of transition metal oxides
decreases in the following order: CrO > V O > MoO >
(Cu
K
radiation, = 0.15406 nm, tube current of 30 mA).
λ
α
The crystallite size of the resulting oxide powders was
derived from Xꢀray diffraction data using the Selyaꢀ
kov–Scherrer formula [20]:
D
=
Kλ/β
cos
θ
HKL,
3
2
5
3
Sb O > CuO > Co O4. This order coincides with the
decreasing order of cation PPs [18].
2
5
3
where
broadening (rad).
K
= 0.94 for Xꢀray diffraction and
β
is peak
Here, we report the oxidation of the chloride ion in
The oxides were also examined under a Hitachiꢀ
molten strontium and magnesium chlorides and in 3400 electron microscope with an energyꢀdispersive
mixtures of MgCl2 with sodium and zinc chlorides by Xꢀray (EDX) spectroscopy attachment. The composiꢀ
gasꢀphase oxygen. We will consider the regularities of tion of the MgO–ZnO powder was determined using a
the oxidation of Group II metal chlorides. We have Flyuorat 2M instrument, and its specific surface area
determined the particle size and phase composition of was measured using an ASAP 2020 analyzer.
1432

CHLORIDE ION OXIDATION
1433
Table 1. Amount of chlorine released by MCl melts and the ionic radii and polarizing powers of the cations
2
Cation
rM2+
,
nm
ze rM2+
,
esu cm–1
T
,
°
C
nCl2, mol
τ
, min
Ba2+ [5]
Sr2+
Ca2+ [5]
Zn2+ [5]
Mg2+
0.138
0.12
0.104
0.083
0.074
–
0.070
0.080
0.092
0.116
0.130
0.085
0.085
0.123
960
890
890
890
730
600
750
600
0.00008
0.00075
0.01284*
0.04050*
0.05949
0.00135
0.00703
0.00292
420
300
300
300
270
300
300
300
2+
+
Mg –Na (eutectic, 1 : 1.28 mol/mol)
2+
+
Mg –Na (eutectic, 1 : 1.28 mol/mol)
–
2+
2+
Mg –Zn (1 : 1 mol/mol)
–
*
Calculated from Ea data for reactions occurring under kinetic control.
RESULTS AND DISCUSSION
lation factor of 0.9841, indicating that the amount of
chlorine evolved by the chloride is proportional to the
PP of its cation.
While the ze value for the Group II metal cations is
the same, the chlorine yield depends on the inverse
ionic radius of the cation. The ionization potential of
It was established that the chlorine and metal oxide
yields depend on the reaction temperature and on the
conditions under which the oxidizer gas is introduced
into the melt and are determined by the oxygen partial
pressure in the gas phase [5]. The chlorine and oxide
formation rate under the given experimental condiꢀ
isoelectronic atoms and cations (
I) is related to their
tions obeys a zerothꢀorder rate equation. Note that ionic radius by the formula
I
= a/r – b [22]. For the
variation of the mass of metal chloride in the reactor Group IIA cations, the ionization potential depends
(
20–50 g) exerts no effect on the chlorine formation
on the inverse radius as I II = 0.1056/r + 2.4438 with
rate, because the molten chloride is in stoichiometric
excess over the oxygen fed into the reactor.
Investigation of chloride anion oxidation in molten
Group II metal chlorides demonstrated that the chloꢀ
rine formation rate depends on the PP of the chloꢀ
ride’s cation. Table 1 lists the conditions and outcomes
of Cl oxidation in the molten metal chlorides and the
ionic radii and PP values of the cations.
In the case of a mixed chloride melt, the effective
PP value of the cations was calculated as follows:
a correlation factor of 0.9994.
Because the amount of chlorine released by the
chlorides and the ionization potential of the metals
depend on the inverse radius of the metal atom, there
is a linear correlation between
n
Cl2
and I II
:
n
Cl2
=
–
0.0892 I II – 0.2906(Fig. 2). The correlation factor in
this case (0.9630) is smaller than in the nCl2
–
ze/
rM2+
relationship (Fig. 1). The linearity of the correlation
breaks down if the amount of chlorine resulting from
zinc chloride oxidation is included in the data array.
n
(
ze r
)
=
N ze r 2+
,
Chloride ion oxidation in MCl –V O systems
M = Mg, Ca, Sr, Ba, Zn) was also studied. The proꢀ
i
2
2
5
eff
∑
i=1
M
(
cess was carried out by passing air at a rate of 5 L/h
where ze
Ni is the mole fraction of this cation in the mixture
containing chlorides. For example, the effective PP
of the cations of an equimolar mixture of magnesium
and zinc chlorides is 0.5 0.130 + 0.5 1.116 = 0.123.
A similar method for calculating the effective cationic
radius for a mixture of alkali metal chlorides was
described by Smirnov [21].
/rM2+ is the PP of the cation of a chloride and
n
0.06
0.04
Mg2+
×
×
Zn2+
calc
0.02
Because of the relatively low boiling point of zinc
2+
calc
Ca
2
Sr2+
chloride (732
result from its oxidation at 890
temperature dependence of the Cl oxidation rate
constant. The same calculation was carried out for calꢀ
cium chloride. Figure 1 plots the amount of chlorine
evolved as a function of the cation PP for alkalineꢀ
earth metal, magnesium, and zinc chlorides. This plot
°
C
), the amount of chlorine that could
Ba
+
°
C
was derived from the
0
0.06
–
0.08
0.10
ze/rM2+, esu cm–1
0.12
0.14
Fig. 1. Amount of chlorine released by the chloride melt as
a function of the polarizing power of the cation.
fits the equation nCl2 = 1.0380
x
– 0.0784 with a correꢀ
RUSSIAN JOURNAL OF INORGANIC CHEMISTRY Vol. 57 No. 11 2012

1434
CHEKRYSHKIN et al.
Mg2+
Table 2 lists the rate constants for chloride ion oxidaꢀ
tion in the MgCl2 and NaCl–MgCl2 melts.
0
.06
.04
.02
0
From the temperature dependence of the rate conꢀ
–
0
0
stant of Cl oxidation in the NaCl–MgCl2 melt, we
–
determined the apparent activation energy of Cl oxiꢀ
dation: Еа = 82 kJ/mol. The small Еа value for magneꢀ
sium chloride oxidation is due to the high rate of the
reaction, so the reaction is dependent on oxygen diffuꢀ
sion across the gas/molten salt interface. It is due to
the oxygen diffusion effect that the activation energy
of chloride ion oxidation in molten calcium and zinc
chlorides decreases with increasing temperature [5].
Under the assumption that the interfacial oxygen difꢀ
fusion rate is the same for all of the melts examined,
the role of diffusion will be greatest in magnesium
chloride oxidation because of the high rate of this
reaction.
Ca2+
Sr2+
Ba2+
3.2
3.0
3.4
3.6
III1/2, eV1/2
3.8
4.0
Fig. 2. Amount of chlorine released by molten MCl (M =
2
+
2+
Mg, Ca, Sr, Ba) as a function of the
potential.
М
→
М
ionization
We investigated the interaction of the magnesium
chloride–zinc chloride mixture (1 : 1 mol/mol) with
oxygen at 600 С. Below this temperature, the system
is solid; above 600°C, there is marked reactant carryꢀ
over from the reaction zone. The chlorine formation
over the surface of the reaction mixture at 800
°
С
(
730 for MgCl2 and 550 for ZnCl2). The initial
°
С
°
С
°
vanadium pentoxide concentration was 10 wt %. For
the MCl –V O (M = Mg, Ca, Sr, Ba) systems, the
2
2
5
amount of chlorine released increases with an increasꢀ
rate constant for the MgCl –ZnCl2 mixture is 1.71
×
2+
5+
2
ing effective PP of the M and
V
cations
–7
–1
1
0 mol s . For zinc chloride, the rate constant under
–7
–1
the same conditions is 1.70
×
10 mol s . In the
In order to elucidate the effect of cations having a
smaller PP value than the Group II cations, we invesꢀ
tigated chloride ion oxidation in molten zinc +
sodium, magnesium + sodium, and magnesium +
zinc chloride mixtures. It was demonstrated earlier
that sodium chloride does not evolve chlorine under
the given experimental conditions [11]. The addition
of sodium chloride to zinc chloride leads to a decrease
in the chlorine yield. This effect is particularly strong
at sodium chloride concentrations of 60 mol % and
above. A study of NaCl–ZnCl2 melts varying in comꢀ
position demonstrated that the Na ZnCl complex
resulting powder, the magnesium oxide : zinc oxide
ratio is 4.65 and the cation ratio is 9.42 (Flyuorat 2M).
–
This cation ratio is close to the ratio of the Cl oxidaꢀ
–
7 –1
tion rate constant for molten MgCl (32.87
×
10 s )
2
to the same rate constant for molten ZnCl (3.33
×
2
–
7 –1
1
0
s ), which is 9.86. This finding is evidence that,
in the oxidation of the chloride mixtures, the observed
–
Cl oxidation rate constants are equal to the oxidation
rate constants of the individual chlorides and the
decreases in the chlorine yield is due to the decrease in
the reactant concentration in the mixture (dilution
effect).
It was demonstrated by Xꢀray diffraction that the
resulting oxide mixture contains only magnesium and
zinc oxides. However, the magnesiumꢀtoꢀzinc atomic
ratio determined by EDX spectroscopy indicates the
presence of hydroxides in the sample.
2
4
forms at sodium chloride mole fractions of 0.7 and
above. This compound can be resistant to oxygen. At
the same time, at zinc chloride concentrations of 40–
1
00 wt %, the chlorine yield depends on the effective
PP value of the sum of the mole fractions of sodium
and zinc cations ( = 0.9612).
r
For determining the particle size of magnesium
A similar inhibiting effect of the sodium cation is oxide and MgO–ZnO powders, the powders were disꢀ
observed in the oxidation of the MgCl –NaCl mixꢀ persed in ethyl acetate and were examined by light
2
ture. The dilution of magnesium chloride with sodium microscopy. Most of the MgO particles were in the 2–
chloride causes a marked decrease in the chlorine
5
µ
m range; most of the particles of the MgO–ZnO
mixture, in the 1–2 m range. Both samples conꢀ
tained <1ꢀ and >5 m particles. This wide range of
yield relative to the yield observed for pure MgCl2
.
µ
µ
particle sizes is explained by the long duration of the
–
Table 2. Rate constants of Cl oxidation in MgCl and euꢀ process and by the comparatively high solubility of the
2
tectic NaCl–MgCl melts
2
metal oxides in the molten chlorides. As a conseꢀ
quence, the particles grew larger via recrystallization.
The mean crystallite size determined using the Selyaꢀ
kov–Scherrer formula [20] was 62 nm for magnesium
oxide (both pure and mixed with ZnO) and 58 nm for zinc
oxide. The specific surface area of the MgO–ZnO powder,
7
Composition
T
,
°
C
k0 × 10 , mol/s
NaCl–MgCl (eutectic)
600
700
750
750
0.79 ± 0.07
2.45 ± 0.24
4.22 ± 0.43
2
NaCl–MgCl (eutectic)
2
NaCl–MgCl (eutectic)
2
–1
2
determined by the BET method, was 20.26 m g , the
3
–1
MgCl2
34.38 ± 4.84
pore volume of the mixture was 0.14 cm g , and the
RUSSIAN JOURNAL OF INORGANIC CHEMISTRY Vol. 57 No. 11 2012

CHLORIDE ION OXIDATION
1435
mean pore diameter was 26 nm. The magnesium oxide
powder obtained under the same condition had a
smaller specific surface e area and a smaller pore volꢀ
4. V. P. Kochergin, M. K. Korshunova, M. S. Ulanova,
and Z. A. Shevrina, Zh. Neorg. Khim. 14, 521 (1969).
5
. Yu. S. Chekryshkin, A. N. Chudinov, T. A. Rozdyꢀ
alovskaya, and A. A. Fedorov, Russ. J. Appl. Chem. 83
(8), 1461 (2010).
2
–1
ume: its specific surface area was 10.85 m g , its pore
3
–1
volume was 0.02 cm g , and its mean pore diameter
was 7 nm.
6. K. Kopalko, M. Godlewski, J. Z. Domagala, et al.,
Chem. Mater. 16 (8), 1447 (2004).
The ionic structure of the metal chlorides and the
dependence of their oxidizability on the PP of the catꢀ
ion suggest that the reaction takes place via an ionic
mechanism. Coming into contact with a molten metal
chloride, dioxygen dissolves in it by occupying
7
. V. N. Gaprindashvili, L. V. Bagaturiya, Ts. G. Sulakadze,
and L. A. Tskalobadze, Izv. Akad. Nauk Gruzii, Ser.
Khim., Nos. 3–4, 332 (2004).
8. US Patent No. 6994636.
“
defects” of the ionic stricture of the melt. This gives
9
. K. Takefumi, S. Toshimitsu, I. Kazunori, et al., Ind.
Eng. Chem. Res. 42 (24), 6040 (2003).
rise to the following chemical reaction [21]:
O2(g) + 4Cl–
2–
10. Yu. S. Chekryshkin, T. A. Rozdyalovskaya, Z. R. Ismagiꢀ
lov, et al., Eurasian Chem. Technol. J. (2), 137 (2003).
→
2O + 2Cl .
2(g)
5
−
2−
1
1. T. A. Rozdyalovskaya, Yu. S. Chekryshkin, V. N. Nekrasov,
Oxygen in the melts can exist as O2 and O2 species
et al., Rasplavy, No. 4, 75 (2004).
[
24, 25]. It was demonstrated that the dissolution of
oxygen in chloride melts brings about three parallel 12. A. N. Chudinov, T. A. Rozdyalovskaya, Zh. A. Vnutskikh,
reactions and, under equilibrium conditions, the
et al., Khim. Tekhnol., No. 11, 524 (2007).
dominant oxygen compounds are superoxides and the 13. H. C. Yang, Y. J. Cho, H. C. Eun, and E. H. Kim,
minor ones are peroxides [26, 27].
Chem. Eng. Sci. 62 (18–20), 5137 (2007).
1
4. R. I. Kraidenko, Khim. Tekhnol., No. 1, 8 (2011).
Thus, the study of chloride ion oxidation in molten
Group II metal chlorides demonstrated that the chloꢀ
rine yield in the MCl –O system is linearly correlated
15. H. C. Eun, H. C. Yang, Y. Z. Cho, et al., J. Hazard.
Mater. 160 (2–3), 634 (2008).
2
2
with the PP of the cation and with the square root of
its ionization potential. The observed dependence of
chloride oxidation on the PP of the cation can be used
in kinetic studies and thermodynamic analysis, as well
as in the estimation of the reactivity of chlorides and
oxides in their chlorination. The mural effect of catꢀ
ions should be taken into account in the oxidation of
metal chloride mixtures for obtaining metals and their
oxides.
1
1
1
6. N. Fu, I. Iwasaki, T. Tamagawa, and M. Kobayashi,
Chin. J. Process Eng. 9 (6), 1080 (2009).
7. D. M. Smith, M. P. Neu, E. Garcia, and V. R. Dole,
J. Alloys Compd. 319 (1–2), 258 (2001).
8. T. A. Rozdyalovskaya, Yu. S. Chekryshkin, A. N. Chuꢀ
dinov, and A. A. Fedorov, Russ. J. Appl. Chem. 82 (9),
1510 (2009).
19. Properties of Inorganic Compounds: A Handbook, Ed. by
A. I. Efimova (Khimiya, Leningrad, 1983) [in Russian].
2
0. N. A. Shabanova, V. V. Popov, and P. D. Sarkisov, Nanoꢀ
disperse Oxide Chemistry and Technology (Akaꢀ
demkniga, Moscow, 2007) [in Russian].
ACKNOWLEDGMENTS
This study was supported by the Russian Foundaꢀ
tion for Basic Research (project 10ꢀ03ꢀ00187a) and by
the Presidium of the Russian Academy of sciences
2
2
2
2
2
2
1. M.V. Smirnov, Electrode Potentials in Molten Chlorides
(Nauka, Moscow, 1973) [in Russian].
2. L. T. Bugaenko, S. M. Ryabykh, and A. L. Bugaenko,
(
program no. 18: “Synthesis and Isolation of Nanoꢀ
Vestn. Mosk. Univ., Ser. 2: Khim. 49 (6), 363 (2008).
sized Powders of Metal Oxides in Chloride Melts”).
3. S. J. Shaw and G. S. Perry, Thermochim. Acta 157 (2),
329 (1990).
REFERENCES
4. M. L. Deanhardt and V. H. Starn, J. Electrochem. Soc.
27 (12), 2600 (1980).
1
1
2
3
. M. Trypuc
Chem. Res. 40 (3), 731 (2001).
’, K. Bia lowicz, and K. Mazurek, Ind. Eng.
5. F. L. Whitting, G. Mamantov, and I. P. Young, J. Inorg.
Nucl. Chem. 34 (8), 2475 (1972).
. M. Trypuc’, Z. Torski, and U. Kie lkowska, Ind. Eng.
6. M. V. Smirnov and O. Yu. Tkacheva, Rasplavy, No. 3,
Chem. Res. 40 (4), 1022 (2001).
5
7 (1991).
owicz, and K. Mazurek, Chem. 27. M. V. Smirnov and O. Yu. Tkacheva, Rasplavy, No. 3,
Eng. Sci. 59 (6), 1241 (2004). 66 (1991).
. M. Trypuc, K. Bia
l
RUSSIAN JOURNAL OF INORGANIC CHEMISTRY Vol. 57 No. 11 2012