Complete calibration of a Thomson scattering spectrometer system by rotational Raman scattering in H_2

We describe the complete calibration of a ruby laser-based multipoint Thomson scattering system, performed only by rotational Raman scattering from H(2) and the measurement of the absolute Raman cross section of some rotational transitions suitable for such calibrations. The absolute calibration factors (for n(e) measurements by Thomson scattering) have been measured by observing the Stokes J = 1 ? 3 rotational transition at Deltanu = -587 cm(-1) (lambda = 723.8 nm), with a congruent with4.5% accuracy. The relative calibration (for T(e) measurements) has also been performed by Raman scattering, observing the same line in all the spectral channels. The results agree to within congruent with10% with those obtained by a tungsten lamp and a lock-in detection technique. From a comparison of the results of the two methods, systematic errors on the measured T(e) and n(e) of less than ~10% (for T(e) up to 200 eV) are inferred. The absolute cross section of the rotational Raman transitions at Deltaupsilon = -587 cm(-1) (lambda = 723.8 nm), Deltanu = -354 cm(-1) (lambda = 711.8 nm), and Deltanu = 354 cm(-1) (lambda = 677.6 nm) has also been measured, by comparison with the intensity of the Rayleigh line. Significant deviations from thermal equilibrium population in Raman levels, expected from the possible occurrence of stimulated Raman emission, are not observed. From the measured data the anisotropic polarizability of H(2) at lambda = 694.3 nm was estimated to be gamma = 2.93 x 10(-25) cm(3) +/- 3%. The absolute cross section of the strongest rotational lines of H(2) is also given.