(235a) Nanoparticle Production in Cold Atmospheric Pressure Plasmas

In cold or non-equilibrium plasma, free electrons gain sufficient energy form the applied electric field for ionisation and molecular dissociation, while ions and molecules remain at almost the initial gas temperature. Thus these processes are inherently more economical than flame or hot plasma processes for nanoparticle production. When applied to nanoparticle production, the free electrons dissociate a volatile precursor (for instance a metal organic complex or a hydrocarbon). Thus a supersaturated metal or carbon vapor is formed in the cold plasma, condensing to form nanoparticles. Under certain conditions, cold plasmas have produced uniformly sized particles. We attribute this to the unipolar charge of the growing particles in the plasma, by which particle-particle collisions are suppressed, and particle growth is governed by atom/molecule attachment.

So far, cold plasma nanoparticle production has been done using microwave and radio frequency plasmas, operated at pressures of 40 mbar or [1]. This requires a vacuum system and makes analysis of the particles as well as scale-up of production rather difficult. Instead, we used cold atmospheric pressure plasma sources to produce nanoparticles. We tested these sources by the production of carbon nanoparticles from acetylene, and iron particles from ferrocene (Fe(C5H5)2).

The precursors were introduced into the plasma by mass controlled flow (acetylene), or by passing a carrier gas over the sublimating solid (ferrocene). Hydrogen could also be added to the plasma. Two different atmospheric pressure plasma sources were used: an atmospheric pressure plasma jet (APPJ), and a dielectric barrier (DB) source. These two differ mainly in the way the non-equilibrium nature of the plasma is maintained at atmospheric pressure.

The APPJ, in which plasma is generated with a 13.56 MHz RF source [1]. uses a high flow of helium (20 L/min) to literally blow out the charged species before arcing can occur. The second source has a dielectric barrier placed between the electrodes to prevent streamer-to-arc transition. A 75 kHz source is used for the DB. It has been used with helium, argon and nitrogen as carrier gases. For both sources the discharge structure (homogeneous glow or filamentary streamer) depends strongly on gas composition.

A differential mobility analyser (DMA) [3] and an aerosol electrometer (AEM) [3] or condensation particle counter (CPC) [3] were used to determine particle size distributions. Particles were also collected on grids for transmission or scanning electron microscopy (TEM or SEM) analysis.

Results: Atmospheric Pressure Plasma Jet: Amorphous carbon nanoparticles with diameters ranging from 15 to 30 nm were produced from acetylene. From ferrocene, iron particles of approximately 20 nm were created. Upon exposure of the TEM samples to air, these iron particles were immediately oxidised to Fe2O3. Analysis of samples not exposed to air showed pure iron particles. The particle size was quite uniform at small output particle concentrations. Increasing the precursor flow only results in an increase of the output concentration an a broadening of the size distribution. A significant increase in particle count is observed for an increase in plasma power, coinciding with a slight drop in the average particle size.

Dielectric Barrier Source: When helium is used as a carrier gas in conjunction with acetylene, the DMA/ CPC combination showed a large amount of particles, while SEM analysis did not show any. We speculate that the plasma only results in acetylene oligomerization into heavier hydrocarbons. These cannot be detected by the electron microscope. When argon is used as a carrier gas, large amounts of amorphous carbon particles and agglomerates, with a primary particle size of 20-30 nm, were seen.

Preliminary experiments confirm iron particle formation from ferrocene. The primary particle size (~ 8-16 nm) is smaller than for particles created with APPJ. The carrier gas type, ferrocene and hydrogen concentration apparently do not influence it, but the total concentration is affected.

References: [1] D. Vollath, D. Vinga Szabó, in Innovative Processing of Films and Nanocrystalline Powders, ed. K.-L. Choy, p. 219, Imperial College Press, London, 2002. [2] A. Schütze, J.Y. Jeong, S.E. Babayan, J. Park, G.S. Selwyn, and R.F. Hicks - IEEE Transactions on Plasma Science. 26, 6 (1998) [3] K. Willeke, P. A. Baron, Aerosol Measurement, Van Nostrand Reinhold, New York, 1993.