Catalysis and Inhibition of the Thermal Decomposition of Certain Azides
Abstract Generated abstract
This paper examines how homophase and heterophase additives regulate the thermal decomposition of several inorganic azides, with emphasis on silver, lead, barium, and copper azides. Using conductivity observations, prior kinetic data, and experiments with metal ions and oxide or sulfide additives classified as electron donors or acceptors, the authors compare changes in decomposition rate, flash behavior, and activation energy. The results indicate two contrasting groups of azides: electron acceptors accelerate decomposition of silver and lead azides but retard barium azide, while electron donors can retard lead azide and accelerate copper azide. The authors interpret these differences through a multistage electronic and ionic decomposition scheme, arguing that the apparent rate-controlling stage and relevant equilibrium stages vary with the azide and determine whether additives act as catalysts or inhibitors.
Full Text
Catalysis and Inhibition of the Thermal Decomposition of Certain Azides
Yu. A. Zakharov, G. G. Savel'ev, E. B. Chernov
(Presented by the scientific seminar of the Department of Radiation Chemistry)
The problem of directed change in the rate of thermal decomposition of azides is important from both the scientific and practical points of view. The practical significance of regulating the decomposition rate is determined by the use of heavy-metal azides in explosives. On the other hand, the regularities found for azides can be transferred to other objects as well. From the scientific point of view, the relations between the physical and electrophysical properties of azides and the mechanism of their thermal decomposition are also of interest, as is the relation between the electrophysical properties of catalysts and inhibitors and their influence on the rate of decomposition.
For regulating the rate of thermal decomposition, as well as for studying its mechanism, several methods may be used, as is known [1]: these are methods involving homophase and heterophase additives, the action of radiation, electric and, possibly, magnetic fields, and also mechanical treatment, etc.
Since the decomposition mechanism is a sequence of electronic and ionic stages, important information for clarifying the mechanism of decomposition can be obtained from the study of the magnitude and type of conductivity, photoconductivity, thermoelectromotive force, dielectric permittivity, optical and other properties of the solid compound.
According to current concepts [1, 2], the thermal decomposition of most inorganic azides proceeds according to the following scheme.
Thermal decomposition begins with rupture or excitation of the bond between the cation and the anion, which is equivalent to detachment of an electron from the anion and its transition into the conduction band of the crystal:
\[ N_3^- \rightleftarrows N_3^0 + e, \tag{1} \]
or to its excitation to an exciton level with subsequent decay of the exciton:
\[ N_3^- \rightleftarrows N_3^{-*} \overset{\tau}{\rightleftarrows} N_3^0 + \text{Metal}. \tag{2} \]
The holes formed decompose bimolecularly [2]
\[ N_3^0 + N_3^0 \to 3N_2. \tag{3} \]
In the solid substance there also occurs a process of disordering in the lattice, for example,
\[ M^+ \rightleftarrows M_z^+, \tag{4} \]
The ionic defects that are formed can interact with electrons, giving intermediate or final products, for example:
\[ M_z^{+}+e\to \text{Metal}. \tag{5} \]
The possibility, direction, and methods of regulating the decomposition reaction will depend on which of stages 1–5 limit the rate of thermal decomposition.
Earlier, in this direction [4], we carried out experiments on the influence of homophase additives on the thermolysis of silver azide and showed that additions of \( \mathrm{Pb}^{++} \) and \( \mathrm{CO}_3^{--} \), which change the concentration of interstitial cations \((\mathrm{Ag}_z^{+})\), do not affect the rate of thermal decomposition. Measurements of ionic conductivity showed that the concentration of ionic defects does indeed change.
We then studied the influence of heterophase impurities on the thermal decomposition of \( \mathrm{AgN}_3 \). These data are discussed in detail in [6]. Here it is of interest only to note that strong catalysts of thermal decomposition are semiconductors with a large electron work function, i.e., good electron acceptors, such as CoO, \( \mathrm{Co}_2\mathrm{O}_3 \), and others; oxides with a small work function (electron donors)—ZnO, \( \mathrm{ThO}_2 \), CdO—have almost no effect on thermolysis (CdO somewhat decreases the rate of decomposition), which agrees well with the values of the contact potential difference measured by us under conditions close to those of the thermal-decomposition experiments. During decomposition in the presence of catalysts, the activation energy is greatly decreased in comparison with pure \( \mathrm{AgN}_3 \).
Similar studies were also carried out by us with lead azide. As homophase impurities, we introduced into the azide lattice, by coprecipitation, the ions \( \mathrm{Ag}^{+} \) and \( \mathrm{Cu}^{++} \). Measurements of dark conductivity showed that silver increases it, while copper decreases it. The influence of these additives on thermal decomposition is shown in Fig. 1, from which it is seen that \( \mathrm{Ag}^{+} \) increases, while \( \mathrm{Cu}^{++} \) decreases, the rate of thermolysis.
Fig. 1. Influence of additions of \( \mathrm{Ag}^{+} \) (3) and \( \mathrm{Cu}^{++} \) (1)—2.5 mol.%—on the rate of thermal decomposition of lead azide.
The influence of heterophase impurities on the thermal decomposition of \( \mathrm{PbN}_6 \) is similar to their influence on \( \mathrm{AgN}_3 \). Figure 2 shows the results of experiments on the slow thermal decomposition of mechanical mixtures of CoO, \( \mathrm{Co}_2\mathrm{O}_3 \), NiO, ZnO with lead azide. It is seen from the figure that mixtures of \( \mathrm{PbN}_6 \)
with CoO and Co₂O₃ decompose explosively; the addition of Ni₂O₃ accelerates it somewhat, while ZnO retards the thermal decomposition, whereas the electron work function of the oxides varies in the series:
\[ \varphi_{\mathrm{Co_2O_3}} > \varphi_{\mathrm{CoO}} > \varphi_{\mathrm{Ni_2O_3}} > \varphi_{\mathrm{ZnO}} . \]
In this case, strong catalysts also greatly reduce the activation energy of thermolysis. Additives have the same effect on the delay time of the flash of lead azide.
The first data obtained by us on the effect of Ni₂O₃, NiO, ZnO, and Ag₂S on the thermal decomposition of barium azide are shown in Fig. 3. It is evident that Ag₂S and ZnO change the rate of decomposition only slightly, whereas substances with a high electron work function—NiO, Co₂O₃—substantially reduce the decomposition rate of BaN₆.
Fig. 2. Effect of heterogeneous impurities on the rate of thermal decomposition of lead azide; the additives were taken in an amount of 10% by weight.
Fig. 3. Effect of heterogeneous impurities on the rate of thermal decomposition of barium azide; the additives were taken in an amount of 10% by weight.
The flash temperature of monovalent copper azide is lowered upon addition of ZnO and Ag₂S.
All the experimental data can be presented in the form of a table.
It is evident from the table that there are two groups of azides with opposite properties.
Table
| Additive | AgN₃ | PbN₆ | BaN₆ | CuN₃ |
|---|---|---|---|---|
| Me\(^{++}\) | does not affect | retards | — | — |
| Me\(^{+}\) | — | accelerates | — | — |
| Electron donors | do not affect | retard | do not affect | accelerate |
| Electron acceptors | accelerate | accelerate | retard | — |
Discussion
As shown by electron-microscopic observations [4], the thermal decomposition of AgN₃ proceeds in the volume of disordered “precenters”; this can explain the lack of effect of homophase additives on the thermolysis of silver azide. The indicated experimental fact excludes from the possible rate-limiting stages the disordering reaction [4]. If stage (5) limited the rate of the process, then the influence of electron donors and acceptors would be opposite to that observed experimentally. Thus, stages 1 and 3 may limit the rate of the process.
A number of works have shown that for silver azide the Mott relation [5] is fulfilled between the optical activation energy of electron excitation and the activation energy of thermolysis. On this basis it was concluded that the rate of decomposition is limited by stage 1. The kinetics of the chemical stage proper will then be described by the equation:
\[ \frac{d[N_3]}{dt}=-K_1[N_3^-]. \]
If one accepts this assumption, then the observed influence of acceptors can be explained only by a change in the activation energy of this stage, or by a change in the rate-limiting stage itself.
Then the acceleration of the process by acceptors can be explained by the appearance of a new, bypass route for stage 1, for example:
\[ \mathrm{N_3^- + catal. \rightarrow catal.^- + N_3^0,} \]
\[ \mathrm{catal.^- + M^+ \rightarrow M^0 + catal.,} \]
where
\[
\mathrm{M^+}
\]
is a cation, and
\[
\mathrm{M^0}
\]
is a metal atom.
In this case, retardation by donors can be imagined as an increase in the electron concentration to such a degree that the reverse reaction of stage 1 can compete with reaction 3; however, the latter condition is apparently too stringent.
Our data can be explained best if the rate of decomposition is limiting (stage 3). Then, when the concentration of electrons is increased by donors, the rate of the process should decrease because of the shift of stage 1 from right to left; conversely, electron acceptors will promote an increase in the concentration of holes, shifting reaction 1 to the right and thereby increasing the rate of thermolysis. The kinetic stage itself will then be described by the equation:
\[ \frac{1}{3}\cdot \frac{dN_2}{dt}=K_1\cdot k_3[N_3^-], \]
where \(K_1\) is the equilibrium constant of stage 1, \(K_3\) is the rate constant of stage 3, and \([N_3^-]\) is the concentration of undecomposed substance.
It is evident from the equation that the temperature dependence of the thermolysis rate will be determined by the activation energies of reactions 1 and 3.
There is an opinion in the literature [6] that the activation energy of stage 3 is small (since it proceeds on the surface) in comparison with reaction 1. The activation energy of this reaction (1) is equal to the width of the forbidden band. Thus, the activation energy of the entire process in this case is equal to the width of the forbidden band.
It is interesting to note that if stage 3 were monomolecular, the rate of the process would be proportional to \(\sqrt{K_1}\), and its activation energy would be equal to \(1/2\) of the forbidden-band width, which is also in agreement with the statistics of electrons in a solid developed in the works of Mott and Gurney.
The results obtained in the case of \(\mathrm{PbN_6}\) are, on the whole, analogous to those described for \(\mathrm{AgN_3}\), and here analogous arguments may be given concerning the limiting stage. There are, however, differences in the influence of homophase impurities. These differences may be connected with the fact that in lead azide, as compared with \(\mathrm{AgN_3}\), the concentration of electron acceptors (interstitial \(\mathrm{Pb_z^{++}}\) ions or anion vacancies) is much smaller, and a change in their concentration leads to a change in the rate of decomposition.
As is seen from the table, \(\mathrm{BaN_6}\) decomposes noticeably more slowly in the presence of acceptors, i.e., the effect is opposite to that observed in the case of \(\mathrm{PbN_6}\) and \(\mathrm{AgN_3}\). This result can be explained if one assumes that the rate of decomposition depends strongly on stage 5. In this case the temperature dependence will be determined by the equilibrium constants of stages 4 and 1. A distinct acceleration of thermal decomposition is not observed in this case, possibly because the work function of electrons from \(\mathrm{BaN_6}\) is small, while \(\mathrm{Ag_2S}\), \(\mathrm{ZnO}\) are not good electron donors.
The first results on the flash of \(\mathrm{CuN_3}\) can be explained in the same way, since in this case also electron donors accelerate the decomposition of the azide.
From the examples considered it is clear that, despite the commonality of the decomposition mechanism, the methods of regulating the rate of thermolysis of azides may be different. These methods depend on the nature of the limiting stage.
In the case of thermal decomposition, it makes sense to speak only of a formal limiting stage, since the rate of the process will also depend on the equilibrium constants of other stages, and the activation energy of decomposition will not be equal to the activation energy of the limiting stage. This opens one more possibility for regulating the rate of decomposition—by changing the activation energies of the equilibrium stages whose constants enter into the kinetic equation of the process.
References
-
V. V. Boldyrev. The effect of defects in crystals on the rate of thermal decomposition of solids, Tomsk, 1963.
-
F. Bowden and A. Yoffe. Fast Reactions in Solids, IL, Moscow, 1962.
-
Yu. A. Zakharov, G. G. Savel'ev, V. V. Boldyrev, L. A. Votinova. Kinetics and Catalysis, 5, 5, 1964.
-
F. P. Bowden, J. Mc. Auslan, Nature, 178, 408, 1956.
-
M. Gurney, N. Mott. Electronic Processes in Ionic Crystals, IL, 1949.
-
Yu. A. Zakharov, E. S. Kurochkin, G. G. Savel'ev. Kinetics and Catalysis (in press), 1966.
-
P. Gray, T. Waddington, Proc. Roy. Soc., A 241, 100, 1957.