Electrochemical preparation of germane

Article abstract of DOI:10.1134/S0020168508100105

Germane has been prepared through the electrochemical reduction of the germanate anion in alkaline solutions with a current efficiency of 40-45%. Active solution circulation in the cathodic zone and the use of an Sn or Cd cathode are shown to raise the germane yield. The current density and initial solution concentration have a weak effect on the reduction process.

Full text of DOI:10.1134/S0020168508100105

ISSN 0020-1685, Inorganic Materials, 2008, Vol. 44, No. 10, pp. 1081–1085. © Pleiades Publishing, Ltd., 2008.
Original Russian Text © V.V. Turygin, M.K. Smirnov, N.N. Shalashova, A.V. Khudenko, S.V. Nikolashin, V.A. Fedorov, A.P. Tomilov, 2008, published in Neorganicheskie Materialy,
2
008, Vol. 44, No. 10, pp. 1208–1212.
Electrochemical Preparation of Germane
a
a
a
a
b
V. V. Turygin , M. K. Smirnov , N. N. Shalashova , A. V. Khudenko , S. V. Nikolashin ,
b
a
V. A. Fedorov , and A. P. Tomilov
a
State Research Institute of Organic Chemistry and Technology, sh. Entuziastov 23, Moscow, 111024 Russia
b
Kurnakov Institute of General and Inorganic Chemistry, Russian Academy of Sciences,
Leninskii pr. 31, Moscow, 119991 Russia
e-mail: fedorov@igic.ras.ru
Received April 14, 2008
Abstract—Germane has been prepared through the electrochemical reduction of the germanate anion in alka-
line solutions with a current efficiency of 40–45%. Active solution circulation in the cathodic zone and the use
of an Sn or Cd cathode are shown to raise the germane yield. The current density and initial solution concen-
tration have a weak effect on the reduction process.
DOI: 10.1134/S0020168508100105
+
–
4H O
2
INTRODUCTION
4–
ads
Ge
GeH4,
2OH–
High-purity germane, GeH , is widely used in the
4
which has been confirmed by current–voltage measure-
ments and appears the most plausible.
Thus, the reduction of GeO in alkaline solutions
may yield two products: germane and elemental germa-
fabrication of semiconductor materials for micro- and
optoelectronics and IR applications. It is well docu-
mented [1–3] that the thermal decomposition of high-
purity germane yields single-crystal germanium con-
taining extremely low levels of harmful microimpuri-
2
nium.
10
–3
ties (at a level of n × 10 cm ), which is suitable for the
fabrication of high-energy-resolution nuclear radiation
detectors. The preparation of high-purity germane by
the hydride process offers significant advantages over
the conventional chloride route [4]. In this context,
there is considerable interest in devising unconven-
tional approaches for the preparation of germane,
including electrochemical processes, which are also
attractive for the synthesis of other inorganic hydrides
Devyatykh et al. [7] examined the possibility of pre-
paring high-purity germane via the electrochemical
reduction of aqueous alkaline solutions of germanium
dioxide. They propose running the process in an H-
shaped apparatus with a large electrode separation,
with no diaphragm. The anode and cathode gases are
only separated by a glass membrane situated in the
upper part of the working space. The recommended
2
current density is up to 6.0 A/cm , which is a factor 20
(arsine and phosphine) and can be run at low tempera-
higher than that in large-scale processes. Judging from
the composition of the cathode gas, the current effi-
ciency for germane under such conditions is within
tures and atmospheric pressure, i.e., are environmen-
tally friendly [5, 6].
One way of producing crude germane to be used as 10%. Clearly, this process cannot be used to produce
a raw material for the preparation of a high-purity ger- germane even on a laboratory scale and is of no com-
mane product is electrolysis of aqueous germanate mercial interest. The current density recommended in
solutions [7, 8]. The reduction of the germanate anion
can be represented by the overall scheme
2
[
9, 10] is 1.0–1.5 A/cm . From indirect data, the current
efficiency for germane can be estimated at 15–18%.
However, since the voltage applied to the electrolyzer
was not indicated, the actual power consumption in this
2
3
–
–
–
GeO + 7H O + 8e
GeH + 10 OH
4
2
Since the transfer of eight electrons in one step is process cannot be evaluated.
unlikely, Devyatykh et al. [7] assumed that the first step
The purpose of this work was to develop physico-
is the attachment of four electrons, resulting in the dep-
osition of elemental germanium on the cathode. The
germanium is then reduced to germane by the released
atomic hydrogen. That assumption, however, has not
been confirmed. Nilov [9] proposed the following
scheme for the reduction of the germanate anion:
chemical principles of the electrochemical synthesis of
germane through the reduction of germanium dioxide
in alkaline aqueous solutions.
EXPERIMENTAL
+ 2e–
+ 2e–
Geads
–2OH–
2
3
–
We used high-purity (99.9999%, 6N) germanium
and germanium dioxide prepared from it via oxidation
GeO + 3H O
– GeO · H O
2
2
ads
–
4OH
1
081

1
082
TURYGIN et al.
Table 1. Effect of GeO concentration and current density j
Electrode materials were tested as follows: the cath-
ode to be tested was pretreated with sulfuric or hydro-
chloric acid, cleaned with soda, washed with distilled
water, and mounted in the electrolyzer, following
which three or four experiments were performed with-
out disassembling the electrolyzer. Each process was
run for 3 h. After the experiment, the catholyte was
poured off, and an appropriate amount of germanium
dioxide was added in order to compensate for the ger-
manium consumption during electrolysis.
In assessing the temporal stability of the germane
preparation process, we also used a filter-press electro-
lyzer, but with a larger cathode: 2.5 × 6.0 cm. As the cat-
ion-exchange membrane, we used an MF4-SK. The
catholyte and anolyte were circulated on a gas lift prin-
ciple. After each experiment, the catholyte was satu-
rated with a germanium dioxide solution, without pour-
ing it off.
2
on the germane yield η
ηGeH4, %
j, A/cm2
5
0 g/l GeO2
150 g/l GeO2
0
0
0
0
0
0
0
.1
.15
.2
.3
.4
.5
.6
7.5
–
–
13.0
31.0
31.8
–
9.1
11.4
9.1
9.8
9.8
30.8
–
Note: Nickel cathode, 3 M KOH, quantity of electricity Q = 3.6 A h,
catholyte volume of 75 ml.
Table 2. Temperature effect on the germane yield
In a series of preliminary experiments, germane was
synthesized under conditions very similar to those
described by Vorotyntsev [11] but at a lower current
density. The cathode used was of nickel. The initial
η
GeH4, %
t, °C
j = 0.5 A/cm2,
j = 0.3 A/cm2,
150 g/l GeO2
GeO concentration in the catholyte was 50 g/l in 3 M
5
0 g/l GeO2
2
KOH. The catholyte was stirred by a gas-lift system. At
2
8–30
0–45
7–70
9.8
9.3
15.2
25.8
26.9
2
a current density of 0.3A/cm , the current efficiency for
4
6
germane was 11.4%.
13.7
Assuming that the efficiency of the process may be
limited by mass exchange, we performed a series of
experiments under the same conditions in which the
catholyte was stirred by a centrifugal pump. The aver-
age germane yield in four successive experiments was
Note: Nickel cathode, 3 M KOH.
by a mixture of extrapure-grade nitric acid and reagent-
grade hydrochloric acid. Catholyte and anolyte solu-
tions were prepared using reagent-grade potassium
hydroxide and sulfuric acid, respectively.
9
.6%. At the same current density but at a GeO con-
2
centration of 150 g/l and with mechanical stirring, the
current efficiency for GeH was 31.0%.
4
Experiments were carried out in a filter-press elec-
trolyzer. The cathodic and anodic compartments were
separated by a Nafion-304 cation-exchange membrane.
The cathode was 2.0 × 2.0 cm in dimensions. In most
experiments, the catholyte solution was circulated by a
centrifugal pump, and the anolyte solution, by a gas-lift
system. The set temperature was maintained by a heat
exchanger in both circulation circuits. The anode was a
platinum spiral, and the cathode was fabricated from a
sheet of the material to be tested.
RESULTS AND DISCUSSION
Table 1 illustrates the effect of current density on the
germane yield in solutions with GeO concentrations of
2
5
0 and 150 g/l. At both concentrations, the germane
yield is independent of the current density starting at
.2 A/cm , indicating that the process is kinetically lim-
ited. Raising the current density from 0.3 to 0.6 A/cm
has an insignificant effect on the germane yield. In view
of this, all subsequent experiments were performed
with the catholyte circulated by a centrifugal pump.
Raising the temperature from 25 to 70°C increases the
germane yield only slightly (Table 2).
2
0
2
The experimental procedure was as follows: Before
each experiment, the cathode and anode circulation cir-
cuits were filled with the corresponding solutions. The
cathode chamber and cathode gas collection system
The germanate concentration in the catholyte is of
were purged with argon, the catholyte circulation sys- considerable interest. Since the germanium dioxide sol-
tem was turned on, and electrolysis was started. The ubility in an aqueous solution depends on the alkali
cathode gas was passed through a drying system and concentration, the germanate content of the solution
then through a pyrolysis furnace, where the germane was raised by increasing the potassium hydroxide con-
was decomposed at 500°C to give germanium metal at centration. The results (Table 3) demonstrate that
the outlet of the hot zone. The hydrogen was collected increasing the GeO concentration from 165 to 250 g/l
2
in a Marriott bottle. The germane yield was evaluated has an insignificant effect on the germane yield. This
from the weight gain of the quartz tube in which the suggests that the reduction of the germanate anion is
germane was decomposed.
also kinetically limited. In all of our experiments, a thin
INORGANIC MATERIALS Vol. 44 No. 10 2008

ELECTROCHEMICAL PREPARATION OF GERMANE
1083
layer of elemental germanium was deposited on the Table 3. Effect of GeO content and KOH concentration on
2
the germane yield
nickel cathode, and some of the fine-particle germa-
nium passed into the catholyte. The germanium deposi-
tion takes no more than 1% of the quantity of electricity
passed; the rest goes to the hydrogen release.
To check the assumption that germane may be
formed via the reduction of the deposited elemental
germanium with the cathode hydrogen, we carried out
experiments on the reduction of metallic germanium in
an H-shaped electrolyzer with the electrode compart-
ments separated by a cation-exchange membrane. As
can be seen from Table 4, the current efficiency for ger-
mane was within 1%. This indicates that germane for-
KOH, mol/l
GeO , g/l
ηGeH4, %*
2
3
4
5
6
.0
.0
.0
.0
150
165
180
24.2
33.8
33.5
31.1
250
2
Note: Nickel cathode, j = 0.6 A/cm , Q = 3.6 A h.
* Average over four experiments.
mation via the hydrogen reduction of the elemental ger- Table 4. Reduction of metallic germanium in aqueous elec-
manium deposited on the cathode during the electroly- trolyte solutions
sis of the germanate anion is unlikely.
Catholyte
10% NaOH
j, A/cm2
ηGeH4, %
Thus, with a nickel cathode in alkaline solutions, the
germane yield is within 35%.
The reduction of the germanate anion is known to
significantly depend on the nature of the cathode mate-
rial. In view of this, we examined different cathode
materials under comparable conditions.
0.05
0.1
0.96
0.5
0.2
0.3
0.3
0.2
0
.5
1
0% H SO
2 4
0.05
0
0
.1
.5
It can be seen from Table 5 that a higher (above
0%) germane yield is reached with cathodes from
4
metals that possess a higher hydrogen overvoltage,
though the difference is not so great as might be
expected.
Note: Germanium cathode.
It is of interest to follow the dynamics of germane, Table 5. Effect of cathode material on the germane yield
hydrogen, and germanium formation in a series of suc-
ηGeH4, %
cessive experiments. Figure 1 illustrates the variation in
Cathode material
germane yield for various cathode materials. In all
cases, the germane yield in the first experiment is lower
than that in subsequent experiments. Occasionally, a
max
min
Pt
Zn
22.9
23.7
27.7
26.7
32.4
34.3
34.3
41.7
19.9
19.2
21.5
24.1
27.2
24.9
29.6
35.1
35.8
slight decrease in GeH yield occurs in the fourth exper-
4
iment. Note also interesting data on the balance of the
current responsible for the germane and hydrogen
release. In the first two experiments, a well-defined
unbalance takes place, which is obviously due to the
formation of a germanium coating on the cathode sur-
face. In subsequent experiments, no current is con-
sumed for germanium deposition, which means that
electrolysis yields only two products: germane and
hydrogen. The relatively weak effect of the cathode
material seems to be due to the fact that germane is
formed on the surface of the metallic germanium
deposited on the cathode. At the same time, in the sur-
Ti
Graphite
Cu
Ni
Cd
Pb
Sn
44.2
2
Note: 3 M KOH, j = 0.3 A/cm , GeO concentration of 150 g/l.
2
face areas free of germanium (where the germanium end, we tested quaternary ammonium salts, -toluene-
particles were washed off by the solution), we observed
hydrogen evolution, and the rate of that process was
determined by its overvoltage. Thus, we are led to con-
clude that the optimal cathode materials are cadmium
and copper, because Pb and Sn can be reduced to vola-
tile hydrides, contaminating the forming germane. In
our opinion, this effect calls for further investigation.
sulfonic acid, and zinc and mercury salts. The results
were surprising (Table 6). In particular, the introduction
of tetraethylammonium fluoroborate significantly sup-
presses germane formation. -Toluenesulfonic acid, an
anionic surfactant, has little or no effect on the reduc-
tion of the germanate anion. In the presence of the zinc
salt, the major product is elemental germanium, which
One way to enhance the current efficiency for ger- accumulates in the electrolyte in the form of a highly
mane is to tune the adsorption of the germanate anion dispersed suspension. The mercury salt was tested with
by introducing substances capable of changing the sur- a graphite electrode for the possible reduction of the
face tension at the cathode–electrolyte interface. To this germanate on the highly active mercurated graphite. We
INORGANIC MATERIALS Vol. 44 No. 10 2008

1
084
η GeH4, %
TURYGIN et al.
η, %
4
3
2
1
5
5
5
5
1
80
1
7
6
5
4
3
0
0
0
0
0
2
3
4
5
6
7
2
8
20
10
0
2
4
6
8
10
12
Run no
5
7
9
11
13
15
Q, A h
Fig. 1. Current efficiency for germane as a function of the
quantity of electricity for various electrode materials in a
series of successive experiments without cathode pretreat-
ment: (1) tin, (2) lead, (3) titanium, (4) copper, (5) cad-
mium, (6) nickel, (7) graphite, (8) platinum.
Fig. 2. Variation in the current efficiency for (1) hydrogen
and (2) germane at a copper cathode in a series of successive
experiments with no cathode pretreatment: 3 M KOH, cur-
2
rent density of 0.3 A/cm , GeO concentration of 150 g/l.
2
observed a substantial decrease in both the germane
Note that the purity of the germanium obtained
and hydrogen yields, and the catholyte remained abso- through the thermal decomposition of electrochemi-
lutely transparent. Metallic germanium deposited on cally synthesized germane corresponds to extrapure
the cathode in the form of a dense coating.
grade. According to spectrochemical analysis data, the
total content of metallic impurities in the germanium is
at a level of several ppmw, which is 1.5–2 orders of
magnitude lower than their concentration in the starting
A copper cathode was subjected to long-term testing
under the optimal conditions (3 M KOH, 150 g/l GeO ,
2
2
current density of 0.2 A/cm ). As seen in Fig. 2, the ger- reagents. These results attest to a selective character of
mane yield gradually decreases with time, and a 0.5- to the electrochemical germane synthesis and agree with
1
-mm-thick layer of elemental germanium forms on the earlier data [7, 8]. Elemental germanium prepared
cathode. It is reasonable to assume that the fine-particle through the thermal decomposition of germane is an
germanium deposit reduces the true current density, attractive material for the preparation of a high-purity
leading to a drop in the current efficiency for germane. product.
Table 6. Effects of surfactant additions and metal ions on the germane yield
η, %
Cathode material
Surfactant
Note
GeH4
Ge
H2
Nickel
–
15.5
10.6
11.4
26.1
22.1
8.5
<1.0
<1.0
<1.0
<1.0
<1.0
30.0
–
79.0
88.3
87.1
65.4
60.0
60.3
74.6
93.1
66.4
(
C H ) NCl
2
5 4
(
C H ) NCl
2
5 4
–
p-PTSA*
Zn(CH CO )
Ge suspension in the catholyte
Ge coating on the cathode
3
2 2
Graphite
–
24.1
4.0
(
C H ) NBr
–
2
5 4
Hg(NO3)2
25.3
15.0
2
Note: 3 M KOH, j = 0.3 A/cm , Q = 3.6 A h, GeO concentration of 150 g/l.
2
*
p-Toluenesulfonic acid.
INORGANIC MATERIALS Vol. 44 No. 10 2008

ELECTROCHEMICAL PREPARATION OF GERMANE
1085
CONCLUSIONS
pared by a Hydride Process, At. Energ., 1985, vol. 58,
no. 4, pp. 281–283.
We developed physicochemical principles of the
electrochemical synthesis of germane through the
reduction of germanium dioxide in alkaline solutions
and optimized the process conditions: cadmium or cop-
per cathode; germanium dioxide concentration in solu-
tion, 150 g/l; alkali (KOH) concentration, 3 M; current
2
3
. Devyatykh, G.G., Andreev, B.A., Balabanov, V.V., et al.,
Impurities in High-Purity Germanium Prepared by a
Hydride Process, Izv. Akad. Nauk SSSR, Neorg. Mater.,
1
986, vol. 22, no. 12, pp. 1957–1961.
. Devyatykh, G.G., Andreev, B.A., Bekov, G.I., et al.,
Determination of Electroactive Aluminum, Boron, and
Lithium Relative to the Total Impurity Content in High-
Purity Germanium, Dokl. Akad. Nauk SSSR, 1990,
vol. 310, no. 4, pp. 901–903.
2
density, 0.3 A/cm .
The current efficiency for germane is 35–40%. A
byproduct of the reduction process is elemental germa-
nium, with a current efficiency within 1%.
Cationic surfactants added to the catholyte suppress
germane formation; anionic surfactants have no effect
on the reduction of the germanate anion. In the pres-
ence of zinc ions, the process ends in the formation of
elemental germanium, which passes into the solution in
the form of a highly dispersed suspension.
4
5
. Devyatykh, G.G., Gusev, A.V., and Vorotyntsev, V.M.,
Preparation of High-Purity Germanium, Vysokochist.
Veshchestva, 1988, no. 1, pp. 6–14.
. Vorotyntsev, V.M., Kozin, L.F., Zhylkamanova, K., et al.,
Electrochemical Synthesis of High-Purity Arsine: II.
Cathode Hydrogenation of Elemental Arsenic,
Vysokochist. Veshchestva, 1995, no. 5, pp. 59–66.
The energy consumption in the preparation of ger-
mane with yields up to 40% at a voltage applied to the
electrolyzer from 5 to 6 V is within 6.5 kW h/kg GeH4.
The purity of the elemental germanium obtained
through the thermal decomposition of electrochemi-
cally synthesized germane corresponds to extrapure
grade, and the total content of metallic impurities in it
is at a level of several ppmw.
6. Smetanin, A.V., Smirnov, M.K., Chernykh, I.N., et al.,
Electrochemical Preparation of Arsenic and Its Com-
pounds, Neorg. Mater., 2003, vol. 39, no. 1, pp. 27–42
[
Inorg. Mater. (Engl. Transl.), vol. 39, no. 1, pp. 22–36].
7
8
. Devyatykh, G.G., Vorotyntsev, V.M., Balabanov, V.V.,
et al., Electrochemical Synthesis of High-Purity Ger-
mane, Vysokochist. Veshchestva, 1988, no. 2, pp. 60–65.
. Vorotyntsev, V.M., Balabanov, V.V., Abdrakhmanov,
R.R., and Shcherbinina, S.Yu., Kinetics of the Electro-
chemical Reduction of Germanate Anions at an Ni Cath-
ode, Vysokochist. Veshchestva, 1991, no. 6, pp. 121–128.
ACKNOWLEDGMENTS
9
. Nilov, A.P., Optimization of Mass Transport and Phase
Formation Processes for the Electrochemical Reduction
of As(III), Sb(III), and Ge(IV) Compounds, Extended
Abstract of Doctoral (Chem.) Dissertation, Moscow,
This work was supported by the Russian Foundation
for Basic Research (grant no. 07-03-00343-a) and the
Support to Leading Scientific Schools National Pro-
gram.
2
001.
1
0. Devyatykh, G.G., Vorotyntsev, V.M., Balabanov, V.V.,
and Abdrakhmanov, R.R., USSR Inventor’s Certificate
no. 1 732 697, 1995.
REFERENCES
. Devyatykh, G.G., Gusev, A.V., Gavva, V.A., et al.,
1
X-Ray Detectors from High-Purity Germanium Pre- 11. Vorotyntsev, V.M., RF Patent 2230830, 2004.
INORGANIC MATERIALS Vol. 44 No. 10 2008