CHARACTERISTIC FEATURES OF THE SELECTIVE ACTION
113
gen absorption (~2.3 mol, which makes up one-third of ized inside a pore, resulting in equalization of the coor-
the whole amount of hydrogen needed for complete dination factor for all nitro groups.
reduction of the three nitro groups), the reaction mix-
For experimental verification of this hypothesis, we
ture contains only 4HADNT, 2ADNT, and 4ADNT.
prepared GWC samples with developed surfaces, 5 and
15 m2/g. The kinetics of TNT hydrogenation on these
catalyst samples is shown by curves 2 and 3 (Figs. 1, 2).
The experiment confirmed qualitatively the assump-
tion; i.e., the selectivity coefficients decrease with an
increase in the specific surface area. On passing from
the 1 m2/g GWC sample to the 15 m2/g GWC, the selec-
tivity coefficients for the second nitro group decrease
from 6–7 to 2–3, while those for the third nitro group
decrease from 60–70 to 25–35. However, the porous
GWC still remains highly selective compared to the tra-
ditional powder catalysts, which have much more
developed surfaces (500 m2/g).
This study represents the first stage in investigating the
mechanism of hyperselectivity discovered in hydrogena-
tion of polyfunctional nitroaromatic compounds on glass
fiber woven catalysts doped with platinum group metals.
It is necessary to study the specific features of action of
GWCs activated by various metals and metal composi-
tions for different contents of these metals in the glass tem-
plate (Pt, Pd, Rh, Ru). It is important to study in more
detail the role of the specific surface area of the GWC in
the hydrogenation of nitroaromatic compounds. Neverthe-
less, the practical value of this study, which opens the way
for one-step synthesis of complex nitroamines, is already
obvious. Moreover, this process should be regarded as the
foundation for the development of industrial processes for
utilization of nitroaromatic explosives to give useful prod-
ucts for civilian purposes.
The compounds identified after the second hydrogena-
tion step (~6.11 mol of hydrogen) were 2,4DANT,
2A4HANT, 2HA4ANT, 2,4DANST, and 2,4DAHAT.
The third step (~8.4 mol of hydrogen absorbed) yields
TAT and 2,4DANT. These results indicate that the reduc-
tion of trinitrotoluene on platinum-activated GWCs pro-
ceeds by a complex stepwise mechanism involving suc-
cessive hydrogenation of the nitro groups.
This conclusion becomes more evident when the
results are presented in the coordinates current transfor-
mation rate–specific fraction of hydrogen absorbed
(Fig. 2). It can be seen that the GWC sample with a spe-
cific surface area of 1 m2/g (curve 1) reduces the first
nitro group at a rate 6–7 times higher than it reduces the
second nitro group and 60–70 times higher than the
third nitro group is reduced to an amino group. These
data indicate unambiguously that the selectivity of TNT
hydrogenation on GWC systems is extremely high and
allows one-step preparation of any combination of
amines and nitroamines according to a specified task.
For comparison, Figs. 1 and 2 (curves 4) show the
kinetics of reduction of TNT nitro groups on traditional
5% Pd/C powder catalysts (the experimental conditions
and the specific amount of the catalytic metal in the
experiments shown by curves 1 and 4 are identical).
The pattern of TNT transformation on 5% Pd/C and the
data of NMR analysis of the reduction products on this
catalytic system demonstrate substantial differences:
the powder catalyst accomplishes an almost nonselec-
tive reduction of TNT, the three nitro groups being
hydrogenated almost synchronously up to complete
transformation into amino groups.
ACKNOWLEDGMENTS
The authors are grateful to the Polotsk Steklovolo-
kno Production Association for providing the glass
fiber fabric materials used to prepare GWC samples.
This work was supported by the NATO scientific
foundation Science for Peace, project no. SfP 971897.
When analyzing the possible mechanism of the
hyperselectivity discovered for the GWC system in
TNT hydrogenation, we put forward a hypothesis relat-
ing this effect to the presence of two different valence
states of platinum, Pt0 and Pt2+, in the GWC. There are
grounds for assuming that this composition of active
sites can ensure selective coordination of trinitrotolu-
ene adsorbed on the GWC surface. In the case of tradi-
tional powder catalysts, only Pt0 participates in the pro-
cess. This results in planar TNT adsorption, which
ensures equal accessibility of all three nitro groups for
the reaction with hydrogen. A role of the substantial
difference between the specific surface areas of the new
and the traditional catalytic systems (1 m2/g for the
GWC and 500 m2/g for the powder catalyst) in the high
selectivity of TNT hydrogenation also cannot be ruled
out. This version implies that the steric factor is more
pronounced on the catalyst with the small surface area,
and, hence, the coordination of the TNT nitro groups in
the adsorbed state is highly differentiated. In the cata-
lyst sample with the developed surface, the TNT mole-
cule penetrates into the near-surface layer and is local-
REFERENCES
1. Dorokhov, V.G., Barelko, V.V., Bal’zhinimaev, B.S.,
et al., Khim. Prom-st., 1999, no. 8, pp. 44–48.
2. Barelko, V., Dorokhov, V., Fomin, A., et al., Abstracts of
Papers, IV Eur. Congr. on Catalysis (EUROCAT-4), Flo-
rence, 1999, P/II/290, p. 828.
3. Dorokhov, V.G., Barelko, V.V., and Balzhinimaev, B.S.,
Abstracts of Papers, III Int. Symp. on Catalysis in Mul-
tiphase Reactors, Naples, 2000, pp. 367–373.
4. Simonova, L.G., Barelko, V.V., Bal’zhinimaev, B.S.,
et al., Kinet. Katal., 2001, vol. 42, no. 5, pp. 762–772.
5. Simonova, L.G., Barelko, V.V., Bal’zhinimaev, B.S.,
et al., Kinet. Katal., 2001, vol. 42, no. 6, pp. 907–916.
6. Bal’zhinimaev, B.S., Simonova, L.G., Barelko, V.V.,
et al., Kinet. Katal., 2001, vol. 42, no. 6, pp. 917–927.
7. RF Patent No. 2156654, Byull. Izobret., No. 27 (2000).
DOKLADY CHEMISTRY Vol. 402 Part 2 2005