In-situ optical analysis of the gas phase during the formation of carbon nanotubes.

A reactor has been developed at ONERA to investigate the gas phase during carbon nanotube formation by laser-induced fluorescence (LIF), Laser-induced incandescence (LII), coherent anti-Stokes Raman Scattering (CARS), and emission spectroscopy. Continuous vaporization is achieved with a continuous wave CO2 laser. Optimized conditions are used for single-walled nanotube growth, that is, a graphite target doped with 2 atom % Ni and 2 atom % Co, helium as buffer gas at a flow rate of 50 ml/s, and a pressure of 300 hPa. Temperature profiles are measured by CARS on H2, and soot images are obtained by LII in the hot carbonaceous flow. LIF and spontaneous emission of the C2 radical and Ni and Co atoms are presented. Spectral investigations are conducted at 3100 and 3200 K to have an optimized pair of excitation/detection wavelengths. Spatial investigations of the relative concentrations in the hot carbonaceous flow are performed up to 3500 K. The concentrations are measured as a function of target temperature. Two regimes of vaporization are observed. Vaporization is slow up to 3350 K and becomes much faster above this temperature. The fast regime in the 3350-3500 K range corresponds to the observed spatial extent of the metal vapors region. At 3500 K, the C2 profiles obtained with and without catalysts are very different as a result of carbon coalescence as well as carbon dissolution into the metal nanoparticles when these are present in the gas phase. The shape of the C2 profile can be related to nanotube formation and growth at a target temperature of 3500 K.

Simultaneous temperature and sensitive two-species concentration measurements by single-shot CARS.

Simultaneous spatially and temporally resolved measurements of N(2) and O(2) mole fractions and of temperature are performed using coherent anti-Stokes Raman scattering (CARS). The CARS setup is used with the crossed-beam arrangement (BOXCARS) and nonresonant-background suppression. The technique employs two Stokes lasers, broadband and narrowband, in combination with a frequency-doubled Nd:YAG laser. Temperature and N(2) mole fractions are obtained by single-shot multiplex CARS spectra of N(2) using the broadband laser; O(2) mole fractions are deduced from a particular rovibrational Q-line of O(2) using the narrowband dye laser. The single-shot detectivity limit is better than 0.4% for oxygen at 2200 K and atmospheric pressure, i.e., 10(16) molecules x cm(-3). The capability of the technique for measuring 2-D probability density functions is demonstrated in the simple cases of an isothermal jet and a laminar premixed flame of air and ethylene. The experimental work reveals grave difficulties in using CARS for precise measurements of mole fractions: appreciable signals can be created very far from the geometrical focus; beam disruption by turbulence and the Stark effect cause large mole fraction measurement errors. These problems are discussed. Referencing the mole fraction of the second species by nitrogen mole fraction is demonstrated to be a solution for the turbulence effect in premixed flames.

Spatial averaging and multiplex coherent anti-Stokes Raman scattering temperature-measurement error.

We discuss the effect of spatial averaging within the probe volume on multiplex coherent anti-Stokes Raman scattering temperature measurements. A numerical simulation and an experimental verification on a small laminar flame burner were performed. Appreciable measurement errors can be made in reactive laminar or turbulent media with large gradients. The errors, which are caused by biasing in favor of cold gases, increase with the temperature difference.