“…The results, which are given in table 2 and figure 1, indicate that s 12 was a rapidly increasing function of temperature. At T = 170 K, where s 12 and s 11 are similar in magnitude, the experimental half widths served to define s 12 q n with an uncertainty of around 15 ls AE mol AE m À3 and gave a value b = 1 AE 10 5 similar to that found in non-relaxing gases [10]. At higher temperatures, as s 11 declines and s 12 appears to increase, the sensitivity of the analysis declined rapidly so that at T = 250 K, the uncertainty of s 12 was about 100 ls AE mol AE m À3 .…”