Comment on "closing the loop on bond selective chemistry using tailored strong field laser pulses".

Our research on time-of-flight mass spectrometry (TOFMS) coupled with controlled molecular fragmentation and ionization using shaped femtosecond pulses as a means for molecular identification has led us to uncover general trends that describe how molecules behave under intense laser radiation.1 One of the reviewers of that work suggested that we include in the study a molecule with a known behavior under shaped laser pulse irradiation. Among the many different molecules evaluated, we explored acetophenone which was the subject of the most important laser control experiment.2-4 The work by Levis and Rabitz in the field of laser control achieved great importance due to their result of having controlled a complex bond rearrangement reaction (Scheme 2, in ref 2; Scheme 3c in ref 3), specifically the production of toluene starting from acetophenone (see Figure 1 insert). Their original experiment was reported to Science,2 a more extensive discussion on bond selective chemistry using tailored strong field laser pulses was published as a feature article in this journal,3 and finally, an investigation of the effects of experimental parameters was published in Spectrochimica Acta.4 To our surprise, we were not able to reproduce their results. Their mass spectrum for acetophenone (10-5 Torr) taken with 800 nm, 60 fs transform limited pulses at an intensity of ∼1013 W/cm2 according to their caption shows a prominent peak assigned to m/z 92 identified as toluene.2-4 We attempted to reproduce their experimental conditions as closely as possible (sample pressure at 1.1 × 10-5 Torr, 800 nm, 60 fs transform limited pulses at a peak intensity of ∼1013 W/cm2). Focusing the 10 mm diameter beam with a 300 mm focal length lens and using an extraction plate with a ∼0.5 mm aperture see Figure S7 in Supporting Information where details of the instrumentation and experimental conditions can be found. We used acetophenone (Fluka, purity >99.5%) and performed a conventional GC-MS analysis to confirm the identity and purity. In order to match the relative intensity of peaks m/z 77 and 105 to those found in Figure 1 ref 4, we had to use 600 fs pulses (see Figure 1). Every line in our spectrum matches a corresponding line in their spectrum with an accuracy that is better than the line width. The distribution pattern of the fragment ions is quite similar in both cases, indicating we matched the experimental conditions. However, we disagree with their assignment. Their spectrum indicates the abscissa is mass to charge, but we find that it corresponds to time-of-flight. Because mass to charge depends on time-of-flight squared, their spectral assignment was incorrect. The peak assigned to toluene (m/z 92) was the expected benzyl radical ion C6H5 (m/z 77). We believe all other peaks (except for m/z 105) were erroneously assigned. Figure 1 show the correct m/z assignments. The excellent agreement at all m/z values together with the similar intensity distribution pattern between their data and ours indicates that the experimental conditions were sufficiently similar. We find no evidence for toluene production in this experiment, and suggest their data in ref 4 show no evidence for toluene either.4 Next, we attempted to reproduce the data published in their original Science paper (Figure 5a in ref 2; Figure 18 in ref 3). The experimental conditions given for their measurements were 60 fs pulses and a peak intensity of 5 × 1013 W/cm2 (according to Figure 18 caption, in ref 3). Differences between the spectrum shown in refs 2 and 3 with that published in ref 4 indicate the conditions stated cannot be correct. We contacted Levis to make sure that every detail of their experimental setup was reproduced. He indicated pulses may have been as long as 200 fs. We reduced the bandwidth of our laser to 10 nm and increased the duration of the pulses to match as closely as possible the experimental conditions indicated. We were not able to reproduce their results. In our efforts to reproduce their results we explored a wide range of laser pulse durations and intensities using two different experimental setup. By using a long focal length lens 300 mm or a short focal length lens 50 mm (2.2 and 0.066 mm Rayleigh length, respectively, as shown in Figure S2), we were able to achieve peak intensities in the range 1013 to 1016 W/cm2. Experiments were carried out with an extraction mesh, a pinhole, or a slit at the extractor plate; see ref 1 for a full description of our setup. We also explored changing the location of the laser focus with respect to the position of the extraction pinhole, following a suggestion by Levis that toluene was observed when the focus of the laser was 1 cm beyond the extraction pinhole (email communication). We explored longer pulses (up to 1 ps) and shorter pulses (35 fs); we also explored a number of different pulse shaping strategies. None of these experimental * Corresponding author. E-mail: [email protected]. Figure 1. Mass spectrum of acetophenone under intense 800 nm excitation with 600 fs pulse that is comparable with that of Figure 1 in ref 4. J. Phys. Chem. A 2009, 113, 5264–5266 5264