Electron-stimulated condensation of carbon dioxide at a electronegative impurity E. M. Abornev, V. P. Zhukovskaya, O. A. Nerushev, S. A. Novopashin, A. L. Perepelkin, and V. V. Radchenko Institute of Heat Physics, Siberian Branch of the Russian Academy of Sciences, Novosibirsk

~Submitted July 31, 1997! Pis’ma Zh. Tekh. Fiz. 24, 21–25 ~January 12, 1998!

An experimental investigation has been made of the influence of electrons and electronegative impurities on the condensation process observed when CO2, whose molecules possess no electron affinity, expands into a vacuum. © 1998 American Institute of Physics. @S1063-7850~98!00401-7#

Adiabatic cooling accompanying the free expansion of a gas into a vacuum leads to cooling of this gas below the saturation temperature. However, the formation of a condensed phase ~clusters! in the stream may not be observed if the number of collisions following the establishment of saturation conditions is insufficient for the formation of criticalsize clusters. The limiting stage for homogeneous condensation is the formation of small clusters.1 The artificial introduction of nuclei into the stream may produce a condensation process under conditions where no homogeneous condensation occurs. These condensation nuclei may be charged particles ~condensation at charges was first observed by Wilson2!. From the thermodynamics point of view,3 this phenomenon occurs because the free energy of a charged cluster has a minimum for a certain size. For saturation conditions, the radius of an equilibrium cluster with charge e is: r5 $ ~ e/16p a !~ e 21 ! / e % 1/3,

~1!

where e is the permittivity of the condensed phase and a is the surface tension. Condensation accompanying the expansion of a stream to which free charges had been added was observed at ions formed as a result of the positive energy of

FIG. 1. Intensity of scattered light versus initial pressure. 10

Tech. Phys. Lett. 24 (1), January 1998

affinity between neutrals and charged particles: the proton affinity of water molecules,4,5 and the electron affinity of chlorine atoms.6 An estimate of the equilibrium CO2 cluster using formula ~1! gives a value of the order of 102 molecules. Here we study the influence of electrons and electronegative impurities on the condensation process observed when CO2, whose molecules possess no electron affinity,7 expand into a vacuum. The experiments were carried out using a low-pressure gasdynamic system. Inside the vacuum chamber a gas source—an acoustic nozzle ~2 mm diameter! with a thermally stabilized forechamber—was located on a threeaxis stage. The flow parameters were determined by measuring the pressure and temperature in the forechamber. The gas temperature in the forechamber was monitored with a thermocouple and was 295 K in all the experiments described below. A thermionic electron emission source was located in the nozzle forechamber to generate charged clusters. The electron emitter was a lanthanum hexaboride pellet. Under operating conditions the emitter is heated by passing a current through a tungsten filament clamped to it. The temperature of the pellet was varied between room temperature and 800 K. Thermal decoupling from a molybdenum thin-walled holder, a ceramic insert, and additional liquid nitrogen cool-

FIG. 2. Intensity of scattered light versus temperature of electron emitter. 1063-7850/98/010010-02$15.00

© 1998 American Institute of Physics

10

ing of the gas in the forechamber were used to achieve thermal conditions. A method based on Rayleigh light scattering was used for diagnostics.8,9 The scattered radiation intensity J for a condensed-phase stream is given by: J5AI 0

( N ii 2,

~2!

where I 0 is the intensity of the reference radiation, N i is the concentration of clusters containing i molecules, and A is a geometric factor which is determined from calibration experiments using a gas of known concentration. Figure 1 gives the scattered light intensity as a function of the initial pressure. The measurements were made on the axis of the stream 6 mm from the exit cross section. The intensity was normalized to the scattering intensity J 0 corresponding to the gas concentration in the nozzle forechamber. The horizontal section corresponds to expansion of the gas without condensation, while the increase in signal indicates the formation of clusters in the stream. The introduction of free electrons when the lanthanum hexaboride pellet was heated to the maximum emitter temperature did not lead to any significant change in the dependence plotted in Fig. 1. This result indicates that negative CO2 cluster ions do not form under these experimental conditions, evidently because the CO2 molecules do not possess electron affinity. An electronegative impurity gas was introduced into the nozzle forechamber in order to observe charge-stimulated condensation of CO2. For this purpose, F-4 Teflon was placed near the heater element, which resulted in the formation of gas-phase electronegative fluorine and fluorocarbon radicals. Figure 2 gives the experimental dependence of the light scattering in the stream as a function of the pellet temperature. The measurements were made at the same distance from the nozzle edge as for the data plotted in Fig. 1. The pressure in the forechamber is 63103 Pa. Under these conditions, no homogeneous condensation is observed ~this pressure is indi-

11

Tech. Phys. Lett. 24 (1), January 1998

cated by the arrow in Fig. 1!. As the temperature increases above 600 K, the intensity of the scattered signal increases by two orders of magnitude. Control experiments in which the lanthanum hexaboride pellet was replaced with copper indicated that the observed stimulated condensation effect is caused by the presence of both electrons and electronegative impurities. To sum up, two main results have been obtained. 1. The introduction of free electrons in the forechamber does not influence the condensation of CO2 as it expands in a supersonic stream. 2. Electron-stimulated condensation of CO2 is observed at electronegative impurities. To conclude, we note that charge-stimulated processes can be used to obtain cluster streams with a narrow size distribution function which is important for studying the properties of clusters of a particular size and also for various technological applications. This work was partially financed by the Russian Fund for Fundamental Research, Grant No. 96-02-19045. 1

D. Colomb, R. E. Good, A. B. Balley et al., J. Chem. Phys. 57, 3844 ~1972!. 2 C. T. R. Wilson, Philos. Trans. R. Soc. London, Ser. A 189, 265 ~1897!. 3 Yu. V. Rumer and M. Sh. Ryvkin, Thermodynamics, Statistical Physics, and Kinetics @in Russian#, Nauka, Moscow ~1977!. 4 J. Q. Searcy and J. B. Fenn, J. Chem. Phys. 61, 5282 ~1974!. 5 R. J. Benhler and L. Friedman, J. Chem. Phys. 77, 2549 ~1982!. 6 H. Haberland, H. Langosch, H.-G. Schindler, and D. R. Worsnop, Book of Abstracts of the Sixth International Symposium on Molecular Beams, Freiburg, 1983, pp. 123–125. 7 B. M. Smirnov, Complex Ions @in Russian#, Nauka, Moscow ~1983!. 8 S. A. Novopashin, A. L. Perepelkin, and V. N. Yarygin, Prib. Tekh. E´ksp. No. 5, 158 ~1986!. 9 E. M. Abornev, O. A. Nerushev, S. A. Novopashin et al., Pis’ma Zh. Tekh. Fiz. 22~21!, 84 ~1996! @Tech. Phys. Lett. 22, 900 ~1996!#. Translated by R. M. Durham

Abornev et al.

11