OCR Text |
Show The only molccule that we have successfull y observed u ing LIF is 2 CN (although 1 eN can also in this value by optim izing the LIF setup (more effi cient light collcction using elliptical mirrors, better scattered light rejection with a monochromator, etc.) however as we have secn more highly chlorinated species are not amenable to LIF. Rather than dwell on LIF we tum to the more promising REMPI technique. For 1,4 DCB we arrive at a detection limit of -8 ppm for the present molecular beamrrOF MS apparatus operated in the normal TOF mode. However this apparatus is not truly optimized for sensitivity and several changes can be made to improve this detection limit For example the present ionization region is -830 nozzle diameters downstream of the pulsed valve. Since the density in the molecular beam scales as (XlDy2, where X=downstream distance and D=nozzle diameter, a great improvement in detection limit can be made simply by increasing D and decreasing X. Increasing D requires a somewhat larger pumping capability but D=I mm and X=IOO mm (for XID=IOO) are certainly feasible. Another improvement would be to use larger ion extraction and deflection plates in the TOF MS and to add ion optics. The former would allow us to ionize more of the molecular beam and the latter would improve ion collection efficiency. We estimate the detection limit for 1,4 DCB in such an optimized instrument to be -10 ppb. As noted previously the sensitivity of REMPI to 1,2 DCB is not as great and so the detection limit for 1,2 DCB would probably be higher. It is difficult to establish REMPI detection limits for the DCN's because little vapor pressure data is available for these species. Assuming a vapor pressure of -0.1 torr at our operating temperatures implies a detection limit of -1 ppm for 1,4 DCN using the small TOF instrument. Again this number could be easily brought into the ppb range by simple improvements in the apparatus. As noted above jet-cooling and subsequent molecular beam formation results in a dramatic decrease in molecular number density; even at (XID)=IOO the density has dropped to -2xl0-5 of its initial value in the pulsed valve. However; this drop in density is compensated to a degree by the spectral simplification that can be readily seen in comparison of the top and central panels of Fig. 5. The laser bandwidth is small enough (-0.0015 nm) that the number of molecules per unit bandwidth is vastly increased in the jetcooled sample and this partially offsets the overall density decrease. Spectral simplification becomes increasingly important as molecular size increases because of the greater number of vibrational degrees of freedom. In the previous section we saw how increasing the size of the molecule and/or the degree of chlorination can greatly increase the intersystem crossing rate. This makes LIP very unattractive for more complex species such as PCB's, dioxins, and furans. In these molecules the intersystem crossing should be very fast, for example 2,3,7,8 TCDD has a UV absorption maximum near 310 nm when dissolved in acetonitrile, but the molecule is apparently non-fluorescent.16 Thus REMPI is a more attractive technique for the more complicated (and toxic) species. However, if the intersystem crossing is extremely fast even REMPI will prove to be difficult because it may not be possible to ionize enough molecules from S I 14 ) |