Date of this Version
We have found an error in this paper -- [Phys Rev. A 71, 063811 (2005)]. In our calculation of the intensity threshold for breakdown in CS2 we erroneously used the ionization energy of molecular CS2 (10.08 eV). The correct value should be 3.3 eV, the band gap energy of liquid CS2 . Using the band gap energy, nonlinear absorption will be a three-photon process, and the calculated breakdown threshold becomes 8.5 × 1011 W/cm2 (as opposed to the value of 2×1013 W/cm2 reported previously). Multiphoton absorption and plasma defocusing could then become significant for the intensity of the filaments, Ifil = (5.9 ± 2.4) × 1011 W/cm2, that was measured experimentally. ...
In conclusion, our numerical results show that the effect of plasma alone cannot reproduce the experimental results, while including a fifth order nonlinearity (with or without plasma) generates good agreement with the experiments. The main conclusion of the numerical section of the original paper then remains unchanged in that the fifth order nonlinearity provides good agreement with the experiment while plasma effects alone do not.