Zane Harvey

An alternative method for gas temperature determination in nitrogen plasmas: Fits of the bands of the first positive system (B Π3g→A Σ3u+)

A method of gas temperature determination in nitrogen or nitrogen doped discharges is presented. The method employs fits of numerically generated spectra of the 0-0, 1-0, and 2-0 bands of the first positive system (B Π3g→A Σ3u+) of nitrogen to experimental measurements. Excellent agreement between gas temperature values inferred by using this method and by using the 3-0 band peak ratio method [M. Simek and S. De Benedictis, Plasma Chem. Plasma Proc. 15, 451 (1995)] is demonstrated for a helicon plasma. The spectral model is available for use by the plasma spectroscopy community. The model, along with user instructions, can be downloaded from Electronic Physics Auxiliary Publication Service of American Institute of Physics. The model includes the line positions, Honl-London factors, and provides rapid determination of gas temperature if one or more of the aforementioned emission rovibrational band spectra are available.

Determination of rotational and vibrational temperatures of a nitrogen helicon plasma

We present a method of rotational and vibrational temperature estimation in nitrogen plasma by simulation of the emission spectrum of the first positive system (BΠg3→AΣu+3) of nitrogen. The gas kinetic temperature, assumed to be in equilibrium with the rotational temperature, was obtained from a fit to the experimentally obtained 0→0 at 1051nm, 1→0 at 872.23nm, and 2→0 at 775.32nm rovibrational band spectra. Excellent agreement between values inferred from each band [(499±29)K] as well as inferred by using the 3→0 band peak ratio method [(473±25)K] was obtained for a 600W, 850G, and 10mTorr helicon N2 plasma. The vibrational temperature was determined from a Boltzmann plot of the integrated band intensities [(6040±225)K] and by numerical fit to the Δv=+1, +2, and +3 sequences after accounting for Franck-Condon factors and electronic-vibrational transition moments [(6200±400)K].

Comparison of gridded energy analyzer and laser induced fluorescence measurements of a two-component ion distribution

We present ion velocity distribution function (IVDF) measurements obtained with a five grid retarding field energy analyzer (RFEA) and IVDF measurements obtained with laser induced fluorescence (LIF) for an expanding helicon plasma. The ion population consists of a background population and an energetic ion beam. When the RFEA measurements are corrected for acceleration due to the electric potential difference across the plasma sheath, we find that the RFEA measurements indicate a smaller background to beam density ratio and a much larger parallel ion temperature than the LIF. The energy of the ion beam is the same in both measurements. These results suggest that ion heating occurs during the transit of the background ions through the sheath and that LIF cannot detect the fraction of the ion beam whose metastable population has been eliminated by collisions.

Observation of increased upstream ionization resulting from spontaneous formation of a double layer in an expanding argon helicon plasma

We report observations that confirm the theoretical prediction1 that formation of a current-free double layer in a plasma expanding into a chamber of larger diameter is accompanied by an increase in ionization upstream of the double layer. The increased ionization is caused by electrons accelerated to the upstream side of the double layer. The theoretical model argues that the increased ionization is needed to balance the difference in diffusive losses upstream and downstream of the expansion region. We use both Laser Induced Fluorescence [LIF] and a Retarded Field Energy Analyzer [RFEA] to measure the ion velocity distribution functions [ivdf] upstream and downstream of the double layer. In our expanding helicon source experiments, measurements of ion-beam energy and density indicate that a stable ion beam appears for rf frequencies above 11.5 MHz. The upstream plasma density increases sharply at the same threshold rf frequency. Below 11.5 MHz, large electrostatic instabilities are observed and appear to suppress formation of the double layer - simultaneously reducing the upstream ionization rate. Further studies of the electrostatic noise suggest a beam driven instability at ∼ 18 kHz.

Vibrational population distribution and the gas temperature in the compact helicon nitrogen plasma source chewie

Summary form only given. We report preliminary results on the development of a helicon plasma source with variable activated nitrogen composition for plasma assisted molecular beam epitaxy of III/V-nitrides. The main idea is to alter the population of specific reactive nitrogen species in a helicon plasma source by modifying the electron energy distribution function through the resonant wave-particle interaction arising from electrons traveling at same velocity as the phase velocity of the helicon wave. The high plasma density and high ion exit flow speed (ne = 1013 cm-3 and v i = 8,000 m/s for argon) should yield significantly higher fluxes at the substrate surface and consequently an improved deposition rate over existing MBE plasma sources. Epilayer quality could also be improved by lowering kinetic energy of reactive species. The active nitrogen source is a steady state, high density, helicon plasma source CHEWIE (Compact HElicon Waves and Instabilities Experiment). The helicon vacuum chamber is a 12 cm long, Pyrex tube, 6 cm in diameter, connected to a stainless steel diffusion chamber, 30 cm long, 15 cm in diameter. Three magnetic field coils surround the source and are capable of generating an axial magnetic field up to 1200 G in the source and about 100 G at the end of the expansion chamber. A 7 cm long, water cooled, Boswell saddle type antenna couples the RF energy into the plasma. RF power of up to 600 W over a frequency range of 3-28 MHz is used to create the steady state plasma in the source which expands away into a region of decreasing magnetic field. Optical emission spectroscopy investigations in the plasma source show that under certain working conditions, the N2 first positive system (B3Pi g rarr A3 Sigmau +) are the dominant transitions in nitrogen, helicon-generated plasma. From band head intensities a Boltzmann relative vibrational population distribution is obtained. From the fit of the Deltav=+1 (at 891.24 nm), Deltav=+2 (at 891.24 nm) and Deltav=+3 (at 687.50 nm) bands, a gas temperature of ~350 K for an input power of 300 W, a magnetic field of 800 G and N2 gas pressure of 20 mtorr is inferred

LIF measurement of nitrogen metastable molecular state in a compact helicon plasma source

Summary form only given. There is a need for moderate energy fluxes of metastable nitrogen molecules in the growth of thin films of GaN, a wide band gap semiconductor. A laser induced fluorescence technique (LIF) has been developed for the measurement of N2 (A3 u + S , v=4) relative population density and flow speed in a steady state, helicon discharge at a pressure of 20 mtorr. This particular transition is part of the first positive system of N2 (3 3 A B u g + S reg P ). A coherent 899 tunable ring dye laser is scanned in the range 596.5-597.0 nm to pump the P12(9) rotational line of the (8,4) band of the A + Su 3 state to J=8 of the B3 g P state which then fluoresces by emitting a photon at 645.96 nm (the Q11(8) line of the (8,5) band). A Stanford Research SR830 lock-in amplifier and a 1 nm band pass interference filter in conjunction with optical chopping of the laser are used to distinguish fluorescent emission from spontaneously emitted background light at the same wavelength. A small portion of the laser light is passed through an iodine cell for a consistent zero-velocity reference measurement. With a calibration obtained by previous optical emission spectroscopy measurements and a model to account for reaction pathways in generation and destruction of 3 u A + S , it is possible to infer the density of this state. From the full width at half maximum (FWHM) of the Doppler broadened fluorescence line, the temperature of N2 molecules can be also inferred and compared with values of the gas temperature obtained by rotational analysis of ro-vibrational emission spectra of 1-0 an d 2-0 bands