Catalytic reduction of U(VI) with hydrazine on palladium catalysts in acid solutions

Article abstract of DOI:10.1007/s11137-008-1005-4

The stability of finely dispersed palladium supported on silica gel with respect to various acids was studied. It was shown that palladium catalysts can be used in moderately acidic media under reducing conditions. In nitric acid solutions within a wide range of experimental conditions, the palladium catalysts do not initiate reduction of U(VI) with hydrazine. The catalytic properties of palladium catalysts differing in the size of nanocrystallites of the active metal were examined in the reduction of U(VI) with hydrazine in sulfuric acid solutions. The specific activity of Pd/SiO₂ catalysts is determined solely by the size of metal nanocrystals and is independent of the metal content on the support. The negative size effect is observed, i.e., the surface Pd atoms located on large crystallites exhibit higher catalytic activity. The results obtained were interpreted on the basis of the concepts of the energy nonuniformity of the surface atoms and of the mechanism of U(VI) catalytic reduction with hydrazine in the sulfuric acid solutions.

Full text of DOI:10.1007/s11137-008-1005-4

ISSN 1066-3622, Radiochemistry, 2008, Vol. 50, No. 1, pp. 38 45. Pleiades Publishing, Inc., 2008.
Original Russian Text M.Yu. Boltoeva, A.V. Trefilova, A.V. Anan’ev, 2008, published in Radiokhimiya, 2008, Vol. 50, No. 1, pp. 34 40.
Catalytic Reduction of U(VI) with Hydrazine
on Palladium Catalysts in Acid Solutions
M. Yu. Boltoeva, A. V. Trefilova, and A. V. Anan’ev
Frumkin Institute of Physical Chemistry and Electrochemistry, Russian Academy of Sciences, Moscow, Russia
Received November 30, 2006
Abstract The stability of finely dispersed palladium supported on silica gel with respect to various acids was
studied. It was shown that palladium catalysts can be used in moderately acidic media under reducing condi-
tions. In nitric acid solutions within a wide range of experimental conditions, the palladium catalysts do not
initiate reduction of U(VI) with hydrazine. The catalytic properties of palladium catalysts differing in the size
of nanocrystallites of the active metal were examined in the reduction of U(VI) with hydrazine in sulfuric
acid solutions. The specific activity of Pd/SiO2 catalysts is determined solely by the size of metal nanocrystals
and is independent of the metal content on the support. The negative size effect is observed, i.e., the surface
Pd atoms located on large crystallites exhibit higher catalytic activity. The results obtained were interpreted
on the basis of the concepts of the energy nonuniformity of the surface atoms and of the mechanism of U(VI)
catalytic reduction with hydrazine in the sulfuric acid solutions.
PACS numbers: 82.20.Pm, 82.65.+r
DOI: 10.1134/S1066362208010050
Preparation of U, Np, and Pu in the preset oxida-
tion states and many stages of nuclear fuel reprocess-
ing and radioactive waste treatment are sometimes
complicated by low rate of the redox reactions. They
can be accelerated by heterogeneous catalysts based
on platinum-group metals.
various acids to determine the limits of application of
palladium catalysts and evaluated the catalytic proper-
ties of Pd in the reduction of U(VI) into U(IV) with
hydrazine.
EXPERIMENTAL
The set of catalytic materials used is mainly re-
stricted to finely dispersed Pt on inert supports [1]. It
should be noted that, in spite of the efficiency and
evident advantages of the catalytic processes, they
have not found universal use in the existing radio-
chemical technologies. The major factor limiting their
use is probably high cost of platinum required for
production of the catalysts. Therefore, search of plati-
num substitutes exhibiting catalytic activity in redox
processes in solutions is an urgent problem.
Preparation and characterization of catalysts.
Catalysts containing 1 and 4% Pd/SiO , differing in
2
the dispersity of the active metal, were prepared by
impregnation of MSK silica gel grains (granule size
0.8 1.0 mm) with a solution of PdCl in 2 M HCl.
2
The volume ratio of the support to the precursor solu-
tion was about 1 : 1.2. The mixture was dried initially
in a desiccator over NaOH to obtain a friable, slightly
wet product and then in an oven at 100 110 C to
the air-dry state. The support with the precursor ap-
plied was divided in two portions. The first of them
was calcined in air at 350 C for 2 h and then reduced
with hydrogen at 400 C for 2 h (method A). The sec-
ond portion was reduced with hydrogen at 400 C for
2 h without preliminary calcination in air (method B).
Palladium is one of such promising materials. As
a platinoid, palladium is characterized by pronounced
catalytic properties. In addition, Pd is one of the ura-
nium fission products and is accumulated in nuclear
waste. The methods of its concentration and recovery
from radioactive solutions are well known and devel-
oped [2 4]. Therefore Pd, in principle, can be re-
peatedly used within the framework of a nuclear fuel
cycle. The expected disadvantage of Pd is the lower
chemical stability as compared to Pt.
The Pd content on silica gel was determined
spectrophotometrically using a color reaction with
SnCl after washing out Pd from the catalyst sample
2
with aqua regia [5]. The average size of Pd crystallites
In this study, we examined the stability of silica-
supported finely dispersed palladium with respect to
(coherent scattering regions) on SiO was determined
from the broadening of the (111) X-ray diffraction
2
38

CATALYTIC REDUCTION OF U(VI) WITH HYDRAZINE
Table 1. Characteristics of Pd/SiO2 catalysts
39
Catalyst
Actual Pd content, %
ddiffr = / cos , nm
dvs
=
nid3i / nid2i , nm DPd = 1.12/dvs
1% Pd/SiO2 (A)
1% Pd/SiO2 (B)
4% Pd/SiO2 (A)
4% Pd/SiO2 (B)
0.89
0.76
3.92
4.1
11.5
*
11.0
10.6
12.9
8.8
12.7
12.3
0.086
0.130
0.088
0.091
* X-ray amorphous material.
line. The line profiles were recorded on an ADP-10
Philips diffractometer (CuK radiation, graphite
monochromator, diffracted beam) and approximated
by the Gaussian curve. The instrumental linewidth
was determined by nonlinear interpolation of the line-
width of the Si powder. Broadening of the (111) line
The stoichiometry of the catalytic reduction of
U(VI) was determined after keeping the reaction mix-
ture with the catalyst in the reactor at a required tem-
perature until the reaction completion. Then the solu-
tion was separated from the catalyst by filtration,
combined with wash waters, and transferred into
a volumetric flask. A sample for analysis was diluted
by a factor of 3 with water, and NaOH was added to
precipitate uranium hydroxide. The precipitate was
separated by centrifugation, and the solution with
wash water was transferred into a titration flask.
The hydrazine concentration in the solution was deter-
mined by iodometric titration in NH HCO solution.
2
2 1/2
was determined by the formula
= (B
b )
,
where B is observed and b, instrumental linewidth.
The crystallite size (in nanometers) was determined
by the formula d = / cos [6], where is expressed
in radians (2 units). The electron microanalysis of
the catalysts synthesized was performed on a Philips
EM-301 transmission electron microscope. The size
distribution of the metal particles was estimated by
measuring and counting a large number of crystallites.
The average surface-volumetric size of crystallites was
4
3
RESULTS AND DISCUSSION
3
2
determined by the formula d
=
n d / n d , where
vs
i i i i
The results of the chemical analysis of the catalysts
synthesized, size of Pd crystallites determined by dif-
ferent methods, and dispersity characteristics of the
active component are listed in Table 1. The catalyst
n is the number of crystallites with diameter d [7].
i
i
The dispersity of the active component (the ratio of
the number of the surface metal atoms to the total
number of metal atoms on the support) was deter-
mined from the equation D = (6/d)(v /a ) [8],
1% Pd/SiO (B) appeared to be X-ray amorphous,
2
M
M M
which did not allow calculation of the average crystal-
lite size of the active component from broadening of
the X-ray diffraction lines. Therefore, to compare
the dispersity of all catalysts, we used the average sur-
where a is the average effective area occupied by
M
the metal on the surface and v , atomic volume of the
M
metal phase. For Pd, D = 1.12/d.
Pd
Experimental procedure. The kinetic experiments
were performed in a temperature-controlled glass
reactor equipped with a reflux condenser to prevent
evaporation of working solutions. Working solutions
were prepared just before the experiment by mixing
the calculated amounts of H SO , N H , uranyl sul-
face-volumetric size of nanoaggregates d , found
vs
from the data of electron microscopy.
The loss of Pd from the catalyst due to dissolution
in mineral acid solutions was measured for the cata-
lyst 1% Pd/SiO (B) at room temperature and S : L =
2
4
2
4
2
1
fate, and distilled water. The total volume was 20 ml.
After thermostatic control of a mixture at a required
temperature, the catalytic reduction of U(VI) was ini-
tiated by introducing a weighed portion of a dry cata-
lyst into the reactor. The reaction mixture was agitated
by argon bubbling at a constant rate. The agitation
intensity was controlled by a float-type flow meter.
0.2-ml aliquots of the working solutions were taken
for analysis at regular intervals. The current U(IV)
concentration in the solution was monitored spectro-
photometrically at 650 nm. The epectra were recorded
on a Shimadzu 3100 spectrophotometer.
0.1 g ml (Table 2). The data obtained showed that
silica-supported finely dispersed Pd is unstable in sul-
furic and nitric acid solutions. In sulfuric acid, Pd
passes into the solution due to oxidation of the surface
of the finely dispersed metal particles with atmos-
pheric oxygen and dissolution of the oxide formed
in the acid. The observed sharp decrease in the Pd loss
in the H SO solutions deaerated by argon bubbling
2
4
confirms this assumption. The Pd dissolution in nitric
acid solutions decelerates in the presence of antinitrite
agents (urea). The presence in the acid solutions of
a reductant, hydrazine, sharply enhances the stability
RADIOCHEMISTRY Vol. 50 No. 1 2008

40
BOLTOEVA et al.
1
Table 2. Influence of solution composition on Pd loss
from 1% Pd/SiO21 (B) catalyst at room temperature
(S : L = 0.1 g ml ; contact time 7 days)
in the presence of 0.1 g ml of 1% Pd/SiO at a
2
noticeable rate even at 30 C. At 60 C, 0.1 M U(VI)
in 0.5 M H SO is completely reduced in the presence
of 0.2 M N H within less than 2 h.
2
4
2
4
Solution composition, M
Pd loss, %
To determine the stoichiometry of the process, we
performed a series of experiments in which the N H
consumption was determined under various condi-
tions. It was found that, within the 40 70 C tem-
perature range, the stoichiometric coefficient n =
1.0 H2SO4
2.0 H2SO4
3.0 H2SO4
1.0 H2SO4*
23.9
34.5
41.0
5.2
2
4
+
0.5 HNO3
0.7 HNO3
1.0 HNO3
1.5 HNO3
1.0 HNO3 + 0.1 CO(NH2)2
1.5 HNO3 + 0.1 CO(NH2)2
0.5 H2SO4 + 0.2 N2H4**
0.5 HNO3 + 0.2 N2H4**
16.6
20.4
28.2
100
12.1
17.3
0
[N H / [U(IV)] remained constant and equal to
2
5
0.4 0.5, irrespective of whether a fresh or aged cata-
lyst was used. This fact suggests that, in the entire
range of the experimental conditions, the catalytic
reduction of U(VI) with hydrazine is described by the
stoichiometric equation
2.8
2UO2SO4 + N2H5HSO4 + H2SO4 = 2U(SO4)2 + N2
* Argon bubbling.
** Residual hydrazine concentration 0.2 (with H SO ) and
+ 4H2O.
(1)
2
4
0.03 M (with HNO )
3
After the completion of U(VI) reduction, the N H
2
4
concentration did not change for a long time, which
indicates that palladium catalysts do not promote de-
composition of hydrazine in sulfuric acid solution.
of the catalyst. A minor Pd dissolution in the nitric
acid solution in the course of contact for a week is
probably associated with catalytic decomposition of
hydrazine. It was shown in a special experiment that,
The kinetic curves of the catalytic reduction of
0.05 M U(VI) in 0.5 M H SO in the presence of
during 6-h agitation with argon at 70 C of 1 M HNO
3
2
4
or 2 M H SO containing 0.2 M hydrazine and
2
4
0.2 M hydrazine and various amounts of 1% Pd/SiO
1
2
0.1 g ml of 1% Pd/SiO (B), Pd did not pass into
2
(S : L ratio) at 60 C are shown in Fig. 1. The char-
acteristic feature of the process is the fact that the
introduction of the catalyst into the working solution
initially causes rapid reduction of uranium in an
amount approximately proportional to catalyst amount,
after which the reduction decelerates. Such a shape of
the kinetic curves suggests the occurrence of two suc-
cessive processes. The first, fast reaction is probably
stoichiometric rather than catalytic. It is caused by fast
conditioning of the catalyst surface upon its contact
with the reaction medium. The second, slower stage is
a result of the steady-state catalyst action. In a sep-
arate series of experiments, we showed that the mode
of launching the reaction (order of adding reactants)
does not affect the shape of the kinetic curve and rate
of U(VI) reduction.
the solution, i.e., the palladium catalysts can be suc-
cessfully used in moderately acidic media under the
reducing conditions.
Preliminary experiments showed that palladium
catalysts did not initiate U(VI) reduction with hy-
drazine in nitric acid solutions in a wide range of the
1
experimental conditions. Introduction of 0.1 g ml of
1 4% Pd/SiO into 0.05 1.0 M U(VI) solutions in
2
0.5 2.0 M HNO containing 0.1 1.0 M N H at 50
3
2 4
70 C causes decomposition of hydrazine, but U(IV) is
not formed. In sulfuric acid, hydrazine reduces U(VI)
The fast formation of U(IV) in the initial step of
the reaction of the dissolved components with the
catalyst surface suggests that this process is a con-
sequence of the radical reaction of uranyl ions with
active reducing species arising during reaction of
the surface Pd atoms with hydrazine. Taking into
account the tendency of Pd to form hydride-type
Fig. 1. Kinetic curves of reduction of 0.05 M U(VI) in
0.5 M H SO in the presence of 0.2 M N H and various
2
4
2 4
compounds with hydrogen [9], we can suggest the
amounts of 1% Pd/SiO (A) at 60 C. Catalyst amount,
2
.
1
mg l : (1) 0.2, (2) 0.1, (3) 0.05, and (4) 0.02.
formation in the reaction mixture of radicals N H
2
3
RADIOCHEMISTRY Vol. 50 No. 1 2008

CATALYTIC REDUCTION OF U(VI) WITH HYDRAZINE
41
generated in the reaction
Pd + N2H4
.
Pd(H) + N2H3,
(2)
(3)
followed by their reaction with uranyl ions:
.
.
UO22+ + N2H3
UO+2 + H2N N + H+.
The kinetics of such reaction can be described by
the equation
.
d[UO22+]/d = k [UO22+][N2H3].
(4)
Fig. 2. Kinetic curves of reduction of various amounts of
.
Since the amount of N H radicals generated dur-
2
3
[U(VI)] in 0.5 M H SO in the presence of 0.2 M N H
0
2
4
2 4
1
ing the initial conditioning of the catalyst surface is
equal to the number of the surface Pd atoms in the
reaction system, Eq. (4) can be written as follows:
and 0.05 g ml of 1% Pd/SiO (A) at 70 C. [U(VI)] , M:
2
0
(1) 0.2, (2) 0.1, (3) 0.05, (4) 0.02, and (5) 0.01.
stoichiometric reaction is relatively low. This is asso-
ciated with rather arbitrary determination of the
boundary of the fast reaction section in the semilog
kinetic curve. The error in the determination of the
dC/d = k C[(A
C0) + C],
(5)
1
1
where k , l mol min , is a rate constant of the bi-
molecular reaction, and C and C , current and initial
constant k of the slow catalytic reaction does not ex-
0
1
U(VI) concentrations, respectively. The constant A is
equal to MD(S : L)/106.4, where M is a Pd weight
fraction on the support; D, catalyst dispersity; S : L,
ratio of the catalyst weight to solution volume; and
106.4, Pd atomic weight.
ceed 10%, which is typical for kinetic measurements.
The data of Table 3 allow us to analyze the main ki-
netic relationships of the catalytic reduction of U(VI)
with hydrazine in the presence of the palladium cata-
lyst. In spite of substantial scattering in k values, their
analysis allows certain conclusions about the character
of the process. The temperature dependences of k
The mathematical treatment of the initial portions
of the kinetic curves and calculation of k were per-
formed with the Table Curve 2D (Jandel Scientific)
software.
measured for the catalysts 1% Pd/SiO (A) and
2
1% Pd/SiO (B) correspond to apparent activation
2
1
energies of 17 and 20 kJ mol , respectively, which
To determine the order of the slow catalytic reac-
tion with respect to the U(VI) concentration, we
carried out a series of experiments within the [U(VI)]
range from 0.01 to 0.2 M. The analysis of the kinetic
curves (Fig. 2) shows that the reaction order with
respect to [U(VI)] decreases from 1 to 0 with an in-
crease in the U(VI) concentration to 0.1 M, which is
a usual phenomenon for heterogeneous catalytic proc-
esses of the adsorption nature. All further kinetic
experiments were performed at the initial U(VI) con-
centration of 0.05 M, i.e., under the conditions when
the slow reaction follows a first-order law with respect
to [U(VI)]. The first-order rate constants were calcu-
lated by the rate equation
is typical of diffusion-controlled processes (activation
energy 19 kJ mol [10]). As expected, the rate of
1
such a reaction is practically independent of the reduc-
tant (hydrazine) concentration, since the concentration
of the active radicals formed during conditioning the
catalyst surface is determined only by the volumetric
concentration of the active centers of the catalysis.
Certain decrease in k with increasing [H SO ] is
2
4
2+
probably caused by a decrease in the UO mobility
2
in the solution with an increase in the ionic strength,
which is typical of processes controlled by the mass
transfer.
The main kinetic relationships of the U(VI) catalyt-
ic reduction with hydrazine in sulfuric acid solutions
dC/d = k1C.
(6)
were determined in the presence of the 1% Pd/SiO
2
(A) catalyst. An increase in the catalyst content of
the reaction mixture is accompanied by a proportional
increase in the rate of the slow catalytic reaction
(Table 1), which is characteristic for heterogeneous-
catalytic processes. An increase in [N H ] within the
The constants k and k were calculated from the
1
results of three independent kinetic experiments. The
results for various catalysts and experimental condi-
tions are listed in Table 3. It should be noted that the
accuracy in determination of the constant k of the fast
2
4
0.05 0.3 M range causes a monotonic increase in k .
1
RADIOCHEMISTRY Vol. 50 No. 1 2008

42
BOLTOEVA et al.
Table 3. Catalytic reduction of U(VI) with hydrazine in sulfuric acid solutions in the presence of palladium catalysts
1
1
1
S : L, g ml
[N2H4], M
[H2SO4], M
T,
C
k , l mol 1 min
k , min
1
1% Pd/SiO2 (A)
4
0.01
0.02
0.05
0.1
0.2
0.2
0.2
0.2
0.2
0.05
0.1
0.2
0.3
0.4
0.1
0.1
0.1
0.2
0.2
0.2
0.5
0.5
0.5
0.5
0.5
0.3
0.3
0.3
0.3
0.3
0.1
0.5
0.7
0.3
0.3
0.3
60
60
60
60
60
70
70
70
70
70
70
70
70
40
50
60
7.1 10
1.66 10
5.61 10
2.97 10
6.18 10
2.50 10
3.67 10
4.97 10
5.90 10
6.92 10
5.82 10
2.92 10
2.57 10
4.39 10
1.08 10
2.54 10
3
3
2
2
2
2
2
2
2
2
2
2
3
2
2
0.2
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
4.0
3.9
4.3
4.2
4.3
4.0
3.7
3.2
2.5
3.2
4.0
1% Pd/SiO2 (B)
3
3
2
2
0.05
0.05
0.05
0.05
0.2
0.2
0.2
0.2
0.3
0.3
0.3
0.3
40
50
60
70
2.5
3.5
4.3
5.2
2.69 10
5.84 10
1.77 10
4.40 10
4% Pd/SiO2 (A)
2
2
2
1
1
0.05
0.05
0.05
0.05
0.05
0.2
0.2
0.2
0.2
0.2
0.3
0.3
0.3
0.3
0.3
32
40
50
60
70
3.73 10
5.73 10
9.12 10
1.25 10
1.70 10
4% Pd/SiO2 (B)
2
2
1
1
0.05
0.05
0.05
0.05
0.2
0.2
0.2
0.2
0.3
0.3
0.3
0.3
32.5
50
60
3.87 10
7.97 10
1.29 10
1.73 10
70
With a further increase in [N H ] to 0.4 M, the rate
0.5 (Fig. 3), which can be interpreted as the reaction
order with respect to N H [11, 12]. The order of 0.5
2
4
of U(VI) reduction does not increase. In the log log
coordinates, within the [N H ] 0.05 0.3 M range, this
2
4
reflects the adsorption type of the process and follows
from the equation of the Langmuir isotherm for the
reactant chemisorption on two neighboring catalysis
centers [13]. A decrease in the reaction order with
respect to hydrazine from 1/2 to 0 at [N H ] > 0.3 M
2
4
dependence is a straight line with a slope close to
2
4
is caused by saturation of the catalysis centers with
the adsorbed N H molecules, which is typical for
2
4
heterogeneous catalytic reactions in solutions [1].
With an increase in the H SO concentration from
2
4
0.1 to 0.7 M at [U(VI)] = 0.05, [N H H SO ] =
0.1 M, and 70 C in the presence of 0.1 g ml of 1%
0
2
4
2
4
Fig. 3. Rate of catalytic reduction of 0.05 M U(VI) in the
1
1
presence of 0.3 M H SO and 0.05 g ml of 1% Pd/SiO
2
4
2
(A) at 70 C as a function of hydrazine concentration.
Pd/SiO (A), k substantially decreases, which can be
2
1
RADIOCHEMISTRY Vol. 50 No. 1 2008

CATALYTIC REDUCTION OF U(VI) WITH HYDRAZINE
43
attributed to an increase in the yield of U(VI) sulfate
UO22+
UO+2
fast
complexes which are less reactive with respect to the
reductant as compared to hydrated uranyl ion. The ap-
parent reaction order with respect to the H SO con-
.
UO22+
> UO+2
<
> Pd(H)
Pd
N2H3
<
<
N2H4
2
4
slow
centration, calculated from the slope of the straight
Fig. 4. Mechanism of U(VI) catalytic reduction with hy-
drazine on the palladium catalyst.
line in the ln[H SO ] lnk coordinates, is 0.42. It
2
4
1
should be noted that the intervals of H SO and N H
2
4
2 4
concentrations examined in this study are relatively
narrow, since at [H SO ] < 0.1 M U(IV) undergoes
involving the surface Pd atoms is realized (Fig. 4). In
this cycle, the reaction of chemisorbed hydrogen
atoms with uranyl ions is the slow, rate-determining
stage.
2
4
hydrolysis, and the upper limit of the H SO and
2
4
N H concentrations is determined by the solubility of
2
4
hydrazine sulfate, decreasing with an increase in
[H SO ].
Summation of the equations of intermediate reac-
tions (7), (3), and (8) (10) with regard to stoichiomet-
ric coefficients gives the total stoichiometric equation
of the process:
2
4
The experimental data on the kinetics and stoi-
chiometry of the catalytic reduction of U(VI) with
hydrazine in sulfuric acid in the presence of palladium
catalyst allow certain assumptions concerning the
reaction mechanism. Uranyl ions are reduced in reac-
2UO22+ + N2H4 + 4H+ = 2U4+ + N2 + 4H2O, (11)
tions with two kinds of reducing species. These are
which is identical to experimental Eq. (1), confirming
the adequacy of the suggested mechanism.
.
radicals N H and hydrogen atoms H chemisorbed
2
3
by Pd atoms. The formation of these species can be
schematically presented by Eq. (2) or, more accurate-
ly, by the following scheme taking into account the
reaction order of 1/2 with respect to hydrazine concen-
tration:
The comparison of the catalytic activity of various
palladium catalysts (Table 1) shows that the catalyst
1% Pd/SiO (A) prepared by calcination of the silica
2
support with the precursor applied, followed by reduc-
tion with hydrogen, exhibits substantially higher
activity in the reduction of U(VI) than the catalyst 1%
2Pd+ N2H4
Pd H NH NH H Pd
Pd/SiO (B) prepared without calcination. At the same
2
.
Pd(H)+ Pd(N2H3).
(7)
time, the catalysts 4% Pd/SiO (A) and 4% Pd/SiO
(B) exhibit virtually the same activity. It is of interest
2
2
In view of the above-noted tendency of Pd atoms
to form with hydrogen fairly strong hydride-type
compounds [9], we can suggest that H atoms are less
reactive than the adsorbed radicals N H with respect
to uranyl ions. The latter radicals rapidly react with
U(VI) following Eq. (3), with the subsequent recom-
bination of H N N radicals in the reaction chain:
that the catalyst 1% Pd/SiO (B), in spite of higher
2
degree of dispersion of the active metal, is substantial-
ly less active at the same S : L ratio than the 1%
.
2
3
Pd/SiO (A) catalyst. To quantitatively estimate the
2
specific activity of catalysts, we used the number of
1
turnovers r, min , i.e., the number of catalytic events
.
2
per surface atom of the catalyst in unit time. The
quantity r was calculated by the formula r = V/N,
2H2N N
H2N N=N NH2
N2 + N2H4.
(8)
1
1
where
is the reaction rate in mol l min units;
V, volume of the reaction mixture in liters; and
N, number of moles of the surface Pd atoms. The
number of moles of the surface Pd atoms involved in
the heterogeneous-catalytic process was calculated
from the catalyst weight, Pd percent content on the
support, and metal dispersity found by electron mi-
croscopy. The U(VI) reduction follows the first-order
Less reactive reducing Pd(H) species slowly react
with U(VI):
Pd(H) + UO22+
UO+2 + H+ + Pd.
(9)
Uranium(V) rapidly disproportionates into U(VI)
and U(IV):
2+
law with respect to [UO ]. The reaction rate de-
2
2UO+2 + 4H+
UO22+ + U4+ + 2H2O,
(10)
creases with the progress of the reaction. Therefore, as
parameter we took the change in hydrazine concen-
tration in the half-conversion time
= ln2/k (k is
and the released surface Pd atoms restore their ability
to chemisorb new hydrazine molecules and continue
the catalytic reaction. In other words, in the system
considered, the heterogeneous-catalytic redox cycle
1/2
1 1
the first-order rate constant), normalized with respect
to , i.e., = C [C is the U(VI) initial
concentration]. The turnover numbers calculated for
1/2
0 1/2
0
RADIOCHEMISTRY Vol. 50 No. 1 2008

44
BOLTOEVA et al.
Table 4. Influence of the size of Pd nanoaggregates on
the specific activity of Pd/SiO2 catalysts in the reduction
of U(VI) with hydrazine in sulfuric acid solutions at
[U(VI)]0 = 0.05, [N2H4] = 0.2, [H2SO4] = 0.3 M, and
T = 60 C
solutions were measured in the presence of various
palladium catalysts. In Arrhenius coordinates, these
dependences are straight lines within the 30 70 C
range. The corresponding values of the apparent ac-
tivation energy are listed in Table 5. As can be seen,
under equal other conditions, the activation energies
depend on the size of Pd particles and on the metal
content on the support. The observed decrease in
the apparent activation energy of U(VI) reduction
1
Catalyst
dvs, nm
D
r, min
1% Pd/SiO2 (A)
1% Pd/SiO2 (B)
4% Pd/SiO2 (A)
4% Pd/SiO2 (B)
12.9
8.8
12.7
12.3
0.086
0.130
0.088
0.091
2.24
1.06
2.18
2.17
on 1% Pd/SiO catalysts with increasing size of Pd
2
nanoaggregates reflects the negative size effect and is
determined by a decrease in the Pd H bond energy in
the intermediates on the catalyst surface. Since the
experimental apparent activation energy of hetero-
geneous-catalytic process is related to the true activa-
Table 5. Apparent activation energies E* of U(VI) reduc-
tion with hydrazine in sulfuric acid solutions in the pres-
ence of various palladium catalysts at [U(VI)]0 = 0.05,
tion energy by the formula E* = E
(
is the
tr
1
[N2H4] = 0.2, [H2SO4] = 0.3 M, and S : L = 0.05 g ml
integral heat of adsorption) [15], it is clear that an in-
crease in the metal content on the support at a con-
stant composition of the reaction mixture should lead
to a decrease in E* because of an increase in the in-
tegral heat of reactant adsorption with a decrease in
the coverage of the catalyst surface.
1
Catalyst
dvs, nm
E*, kJ mol
1% Pd/SiO2 (A)
1% Pd/SiO2 (B)
4% Pd/SiO2 (A)
4% Pd/SiO2 (B)
12.9
8.8
12.7
12.3
77.5 0.8
85.2 5.3
34.5 1.5
35.4 1.6
Unfortunately, we cannot unambiguously answer
why palladium catalysts effectively initiating U(VI)
reduction with hydrazine in sulfuric acid solutions are
inactive in nitric acid media, in contrast to platinum
catalysts [1]. The possible reason of this phenomenon
is different reactivity of reducing intermediates formed
by sorption of hydrazine molecules on the active
various catalysts at [U(VI)] = 0.05, [N H ] =
0
2 4
0.2, [H SO ] = 0.3 M, and T = 60 C are listed in
2
4
Table 4. The data obtained show that the specific (per
surface atom) activity of the palladium catalysts is
determined only by the size of nanoaggregates of the
active metal and is independent of the metal content
on the support, i.e., the surface Pd atoms located on
large crystallites exhibit higher catalytic activity.
Thus, the catalytic reduction of U(VI) with hydrazine
in sulfuric acid solutions belongs to structure-sensi-
tive catalytic processes. The negative size effect is
observed. These results can be interpreted on the bases
of the above-described mechanism of the catalytic
process. Indeed, a decrease in the nanoaggregate size
should increase the fraction of the apical and edge
low-coordinate atoms [14]. It is reasonable to expect
that the energy of the chemical bond between Pd and
chemisorbed H atom on such coordination-unsaturated
centers of the catalysis is higher than on Pd atoms
located on the crystallite faces. The chemisorbed H
atom is probably the active intermediate that reduces
uranyl. Its activity is determined by the energy of
the Pd H bond: the higher this energy, the more dif-
ficult is the oxidation and hence the lower is the cata-
lyst activity.
centers of the catalyst with respect to the competing
2+
oxidizing species, UO
and HNO . Presumably,
2
3
.
N H radicals and H atoms chemisorbed on the sur-
face Pd atoms react mainly with HNO molecules and
nitronium ions NO . This leads to progressive oxida-
2
3
3
+
2
tive decomposition of hydrazine in the presence of
palladium catalyst. In the process, U(VI) is not
reduced. A different pattern is observed with platinum
catalysts, when the reducing species are N H mole-
2
4
cules adsorbed on the surface Pt atoms. In this case,
the U(VI) reduction probably proceeds via electron
transition through a Pt Pt metal bridge between hy-
drazine molecule and hydrated uranyl ion adsorbed on
.
the neighboring catalytic centers [16]. The N H
2
3
radical formed by one-electron transfer probably does
not interact with the catalyst surface, and its reactivity
.
in the solution can differ from that of N H radical
2
3
chemisorbed on surface Pd atoms. The free radicals
can react in the solution with both HNO and U(VI),
3
which results in complex stoichiomerty of the reduc-
tion of U(VI) with hydrazine in HNO solution in the
3
presence of platinized catalysts [16, 17]. The nature
of this phenomenon is not fully understood and may
The temperature dependences of rates of catalytic
reduction of U(VI) with hydrazine in sulfuric acid
RADIOCHEMISTRY Vol. 50 No. 1 2008

CATALYTIC REDUCTION OF U(VI) WITH HYDRAZINE
45
be determined by both thermodynamic and kinetic
factors.
niques Physiques d’Etude des Catalyseurs, Paris:
Technip, 1988, pp. 655 724.
8. Anderson, J.R., Structure of Metallic Catalysts, Lon-
don: Academic, 1975. Translated under the title Struk-
tura metallicheskikh katalizatorov, Moscow: Mir,
1978, p. 364.
REFERENCES
1. Anan’ev, A.V., Tananaev, I.G., and Shilov, V.P., Usp.
Khim., 2005, vol. 47, no. 11, pp. 1132 1155.
9. Lewis, F.H., The Palladium Hydrogen System, New
York: Academic, 1967.
2. Chepiri, R.K. Rao and Trivedi, D.C., Coord. Chem.
Rev., 2005, vol. 249, no. 5 6, pp. 613 631.
10. Moelwyn-Hughes, E.A., The Chemical Statics and
Kinetics of Solutions, London: Academic, 1971.
Translated under the title Ravnovesiya i kinetika
reaktsii v rastvorakh, Moscow: Khimiya, 1975, p. 114.
11. Davis, K.M.C. and Sayer, C.F., J. Chem. Soc., Fara-
day Trans., 1972, vol. 68, pp. 1884 1889.
3. Davydov, V.A., Kirshin, M.Yu., Korolev, V.A., et al.,
Abstracts of Papers, Chetvertaya rossiiskaya konferen-
tsiya po radiokhimii Radiokhimiya-2003 (Fourth
Russian Conf. on Radiochemistry Radiochemistry-
2003 ), Ozersk, October 20 25, 2003, pp. 146 147.
12. Tananaev, I.G. and Shilov, V.P., Radiokhimiya, 1989,
4. Marchenko, V.I., Dvoeglazov, K.N., and Renard, E.V.,
Abstracts of Papers, Chetvertaya rossiiskaya konferen-
tsiya po radiokhimii Radiokhimiya-2003 (Fourth
Russian Conf. on Radiochemistry Radiochemistry-
2003 ), Ozersk, October 20 25, 2003, p. 153.
vol. 31, no. 6, pp. 52 55.
13. Boreskov, G.K., Geterogennyi kataliz (Heterogeneous
Catalysis), Moscow: Nauka, 1986, p. 100.
14. Bukhtiyarov, V.I. and Slin’ko, M.G., Usp. Khim.,
2001, vol. 70, no. 2, pp. 168 181.
5. Ginsburg, S.I., Ezerskaya, N.A., et al., Analitiche-
skaya khimiya elementov. Platinovye metally (Analyt-
ical Chemistry of Elements. Platinum Metals), Mos-
cow: Nauka, 1972, p. 321.
15. Panchenkov, G.M. and Lebedev, V.P., Khimicheskaya
kinetika i kataliz (Chemical Kinetics and Catalysis),
Moscow: Nauka, 1974, p. 412.
16. Anan’ev, A.V., Shilov, V.P., Afonas’eva, T.V., et al.,
6. Kovba, L.M., Rentgenografiya v neorganicheskoi khi-
mii (X-ray Diffraction Analysis in Inorganic Chemis-
try), Moscow: Mosk. Gos. Univ., 1991, pp. 230 238.
Radiokhimiya, 2001, vol. 43, no. 1, pp. 37 40.
17. Abdunnabi, H.M. and Ananiev, A.V., Proc. Int. Conf.
on Fast Reactors and Related Nuclear Fuel Cycles,
Kyoto (Japan), 1991, vol. IV, pp. 6.9 6.16.
7. Dexpert, H., Gallezot, P., and Leclerq, C., in Les Tech-
RADIOCHEMISTRY Vol. 50 No. 1 2008