MASS-SPECTROMETRIC STUDY OF THE GASEOUS DECOMPOSITION PRODUCTS OF GUANIDINE PERCHLORATE

R. N. ISAEV, V. N. MELENEVSKII, A. T. TSIKUNOV, Yu. A. ZAKHAROV,
I. A. TIKHOMIROV

(Presented by the scientific seminar of the Department of Radiation Chemistry and the CTF Laboratory)

The mass-spectrometric method, previously used mainly for the purposes of precise determination of nuclear masses and analysis of isotopic composition, has recently become widely used as a method for analyzing the chemical composition of complex mixtures and, in particular, in the analysis of gaseous mixtures [1, 3—5].

The purpose of the present work is to investigate the qualitative composition of the gaseous products of thermal decomposition of pure guanidine perchlorate and the effect of additions of Fe$_2$O$_3$ and ZnO on the spectrum of the evolved gases.

Experimental procedure

Technical-grade guanidine perchlorate, recrystallized from distilled water, was used for the study. The amount of added impurities was 10 percent by weight. The impurities (reagent grade) were introduced by simple mixing with grinding of the mixtures in a porcelain mortar for 15 min, and in the case of Fe$_2$O$_3$, until the composition was uniformly colored by the additive. Metallized fuel was prepared analogously; in this case the mixture consisted of 50% CN$_3$H$_5\cdot$HClO$_4$ and 50% metallic Mg. Also prepared was 50% (CN$_3$H$_5\cdot$HClO$_4$ + 10% ZnO) + 50% Mg. A sample charge of 100 mg was placed in a glass ampoule, evacuated to a residual pressure of $\approx 10^{-3}$ mm Hg, and sealed. The ampoule containing the substance was then heated in a metal bath at a specified constant temperature, which was regulated by means of a contact thermometer with an accuracy of $\pm 0.5^\circ$C.

After decomposition to a specified degree of conversion, the ampoule was removed and placed in an ampoule-opening system connected to the inlet system of an MI-1305 mass spectrometer (Fig. 1). Use of the inlet system of this instrument, intended for isotopic analysis of gases, is impossible when carrying out chemical analysis of complex gas mixtures. Therefore, the first stage of the work was the creation of a system for admitting the analyzed gas mixture into the ion source, operating in the molecular-flow regime. Fulfillment of this requirement is known [1] to be one of the basic conditions of gas mass-spectrometric analysis, since in the case of molecular

Fig. 1. Schematic of the installation for introducing gas into the mass spectrometer

... leakage, the partial pressure of each component of the mixture in the region where ionization takes place does not depend on the presence of the other components and is directly proportional to the pressure of this component in the initial mixture. The inlet system is shown schematically in Fig. 1. The ampoule containing the mixture under study was placed in a removable reservoir (1), made from a metal tube 20 cm long and 15 mm in diameter. A bellows for breaking the ampoules was tightly connected to the reservoir. The reservoir (1) was attached to the shutoff valve (2) by means of a union nut and a rubber sealing ring.

A U-shaped mercury manometer (8) made it possible to measure pressures up to 80 mm with an accuracy of 0.05 mm Hg (the reading was made with the aid of a microscope (9)); in Fig. 1, (6) is a glass bulb with a capacity of 5 liters; (5) is a two-way stopcock for evacuating the inlet system. Molecular leakage was carried out through an aluminum diaphragm (10) of thickness 8 μ, in which two holes 10–15 μ in diameter had been pierced with a needle. The inlet system was first pumped down to a pressure of \(10^{-5}\) mm Hg; then the ampoule was broken by means of the bellows, and with valve (2) the analyzed mixture from the reservoir was admitted into the “standard” (or intermediate) volume formed by the connecting tubes between valves (2) and (3), (3) and (4), and by the cavities of these valves. When valve (4) was opened, the pressure in the intermediate and inlet (6) volumes became equalized and amounted to a value of the order of \(10^{-1}\)—\(10^{-2}\) mm Hg (depending on the pressure in the standard volume).

Experimental results and discussion

The operating mode of the mass spectrometer was chosen as follows: the pressure in the region of the ion source was \((8—9)\cdot 10^{-7}\), and in the analyzer region \((3—4)\cdot 10^{-7}\) mm Hg; the ionizing voltage was 80 V, the emission current 2 mA, and the accelerating voltage 4 kV.

Before carrying out the analyses, the conditions for the leakage of gases from the inlet cylinder into the ion source were studied. The maintenance of proportionality...

…of proportionality between the intensity of the mass-spectral lines and the gas pressure in the intermediate volume (Fig. 2) indicates that the flow over the entire working pressure range is molecular.

Fig. 2. Dependence of the magnitude of the ion current \(N_2^+\) on the pressure in the intermediate volume.

A preliminary analysis of the mass spectra of GP decomposition showed that the spectrum contains lines with \(m/e\) equal to 12, 14, 15, 16, 17, 18, 22, 28, 29, 30, 32, 44, 45, 46. In accordance with the proposed mechanism of the GP decomposition reaction [6–7], it may be assumed that these lines are due to the ions:

\[ {}^{12}C^+,\quad {}^{14}N^+,\quad NH^+,\quad NH_2^+,\quad NH_3^+,\quad H_2O^+,\quad OH^+, \]

\[ N_2O^{++},\quad CO_2^{++},\quad {}^{14}N_2^+,\quad {}^{14}N{}^{15}N^+,\quad {}^{12}C{}^{16}O^+,\quad NO^+,\quad O_2^+,\quad N_2O^+,\quad {}^{13}C{}^{16}O_2^+, \]

\[ {}^{13}C{}^{16}O_2^+,\quad {}^{12}C{}^{16}O{}^{18}O^+. \]

According to the above, the total intensities of the observed lines are written as:

\[ I_{14}=I_{14}^{N_2}+I_{14}^{N_2O}+I_{14}^{NO}; \]

\[ I_{12}=I_{12}^{CO}+I_{12}^{CO_2}; \]

\[ I_{15}=I_{15}^{NH_3}. \]

\[ I_{16}=I_{16}^{O_2}+I_{16}^{NH_3}+I_{16}^{NO}+I_{16}^{CO}+I_{16}^{N_2O}+I_{16}^{CO_2}; \]

\[ I_{17}=I_{17}^{NH_3}+I_{17}^{H_2O}; \]

\[ I_{18}=I_{18}^{NH_3}+I_{18}^{H_2O}; \]

\[ I_{22}=I_{22}^{N_2O}+I_{22}^{CO_2}; \]

\[ I_{28}=I_{28}^{N_2}+I_{28}^{CO_2}+I_{28}^{N_2O}+I_{28}^{CO}; \]

\[ I_{29}=I_{29}^{N_2(^{14}N^{15}N)}+I_{29}^{13CO}+I_{29}^{CO_2}+I_{29}^{N_2O}; \]

\[ I_{30}=I_{30}^{NO}+I_{30}^{N_2O}; \]

\[ I_{44}=I_{44}^{CO_2}+I_{44}^{N_2O}; \]

\[ I_{45}=I_{45}^{Cl_2(^{13}C^{16}O_2)}+I_{45}^{N_2O(^{14}N^{15}N^{16}O)}; \]

\[ I_{46}=I_{45}^{CO_2}+I_{46}^{N_2O}; \]

To identify the spectral lines, the instrument was calibrated with pure gases whose presence was assumed in the samples under study, i.e., \(CO_2\), \(N_2\), \(N_2O\), \(O_2\), \(NH_3\). The calibration-data values obtained by us proved to lie within the limits reported, for example, in [2]. Further interpretation of the mass spectra presented no special difficulties and was carried out by the usual method [3, 4]. The main results of the work are summarized in Table 1. The intensity of the \(CO_2\) line was taken as one hundred (100); the intensities of the remaining lines were referred to the intensity of \(CO_2\).

Table 1

As is evident from the data presented in Table 1, the principal components of the gaseous decomposition products of GP are nitrogen, carbon monoxide, carbon dioxide, and nitrous oxide; NO, H₂O, and NH₃ are also present, and their relative content varies depending on the conditions under which decomposition is carried out and on the type of additives introduced. The results of the analysis show that chlorine and chlorine-containing compounds (HCl, ClO, etc.) are absent from the mass spectra of the decomposition products of GP.

It should also be noted that the data on the content of water and ammonia in the products also, in all probability, do not reflect their true ratio in the samples investigated, since H₂O and NH₃ may be sorbed by the instrument system.

The results of the analysis of the products of the thermal decomposition of pure CH₅N₃HClO₄ show that the composition of the products changes depending on temperature, which agrees with literature data [6, 7]. Below 300°C the principal products are CO₂, N₂O, N₂, NO, O₂, NH₃, and H₂O. At 325°C, CO appears in the composition of the products, N₂O and H₂O are absent, and the content of O₂ increases noticeably. As the degree of decomposition increases, the content of the products decreases, especially N₂O and H₂O. The absence of chlorine-containing compounds in the products can apparently be explained by their being bound by the solid decomposition residue, giving solid nonvolatile products, possibly NH₄Cl, whose melting point is 520°C [8]. On this basis, the decomposition equation for CH₅N₃HClO₄ may be represented in the following form:

\[ \mathrm{CH_5N_3HClO_4} \;\xrightarrow{<\,300^\circ\mathrm{C}}\; \mathrm{NH_4Cl + CO_2 + N_2 + O_2 + N_2O + NO + NH_3 + H_2O} \]

\[ \mathrm{CH_5N_3HClO_4} \;\xrightarrow{>\,300^\circ\mathrm{C}}\; \mathrm{NH_4Cl + CO_2 + CO + N_2 + NO + O_2 + NH_3 + H_2O}. \]

In the case of thermolysis of GP with an admixture of Fe₂O₃, which accelerates the thermal decomposition reaction of CH₅N₃HClO₄, CO₂, N₂, CO, and NO were detected as the principal gaseous products. The presence of NH₃, O₂, and H₂O was also established, but in amounts considerably smaller than in the thermolysis of pure GP, and N₂O was completely absent from the products. During thermal decomposition of the mixture CH₅N₃HClO₄ + 10% ZnO, the composition of the decomposition products is the same as in the case of the mixture with Fe₂O₃, with the only difference that the relative content of the components N₂, CO, NO, and NH₃ decreased considerably. Apparently, this can be explained by the presence of chemical interaction of ZnO either with the gaseous products of decomposition of guanidinium perchlorate, or by a solid-phase reaction of ZnO with CH₅N₃HClO₄ with formation of a solid solution—reaction product in the form of zinc perchlorate, chloride, or nitrate. The formation of zinc perchlorate and chloride is evidently most probable if, following the authors of [6], it is assumed that thermolysis of GP includes, as an obligatory stage, the formation of NH₄ClO₄, whose decomposition limits the overall thermolysis process of GP. The formation of zinc perchlorate in the NH₄ClO₄ + ZnO system has been convincingly demonstrated in a number of works [10, 11].

REFERENCES

1. D. Barnard. Modern Mass Spectrometry. Izd. Inostrannoi Literatury, Moscow, 1957.
2. S. M. Rock, Anal. Chem., 23, No. 2, 261, 1951.
3. N. N. Shumilovskii, R. I. Stakhovskii. Mass-Spectral Methods, Energiya, Moscow–Leningrad, 1966.

1. V. L. Tal’roze. Molecular Mass-Spectral Analysis, Zav. lab. 20, No. 3, 292, 1954.

2. D. Beynon. Mass Spectrometry and Its Application in Organic Chemistry. Moscow, 1964.

3. A. Glasner, Makovky A., J. Chem. Soc., 182, 1953.

4. A. Glasner, Makovky A., J. Chem. Soc. 1606—10, 1954.

5. Brief Handbook of the Chemist, vol. I, Goskhimizdat, p. 609, 1962.

6. A. E. Simchen, J. Appl. Chem., 13, No. 9, 1963.

7. A. V. Boldyreva, V. N. Mozzhova. Kinetics and Analysis, 7, issue 4, 734—6, 1966.

8. R. N. Isaev, Yu. A. Zakharov, V. V. Bordachev. Present collection, p. 84.

ERRATA NOTED