Measurements of the gas temperature in an oxygen plasma by spectroscopy of the O2(b 1Σ+g) → O2(X 3Σ−g) transition

A method for measuring the gas temperature in an oxygen plasma by spectroscopy of the electronic transition from the O2(b 1Σ g + , v = 0) metastable state of molecular oxygen into the O2(X 3Σ g − , v = 0) ground state is considered in detail. The method is verified experimentally for the plasma of dc glow discharge in pure oxygen. It is shown that the gas temperature can be determined by analyzing high-resolution spectra of the P branch of this transition, no matter whether its fine structure (P P and P Q branches) is resolved or masked, provided that the rotational structure of the spectrum is resolved. The feasibility of the method proposed in 1999 by P. Maco and P. Veis for determining the gas temperature from the ratio between the intensity maxima of the R and P branches of the O2(b 1Σ g + , v = 0) → O2(X 3Σ g − , v = 0) transition in a poorly resolved spectrum was studied experimentally. It is shown that, in order to use this method, it is necessary to know the spectrograph instrumental function. The effect of the spatial inhomogeneity of the temperature and concentration of O2(b 1Σ g + ) molecules on the accuracy of integral (over the plasma volume) measurements of the gas temperature is investigated using spatially resolved spectroscopy of the O2(b 1Σ g + , v = 0) → O2(X 3Σ g − , v = 0) transition. It is shown that precise measurements of the temperature require that the optical measurement system be thoroughly adjusted in order for the temperature and concentration of the emitting particles to vary insignificantly over the optically selected volume.