SIMBA

K. Y. Marcaccio, K. HiUstrom, C. Tomassini, and G. E. Turecki, eds., and M. E. WiUiams, founding ed.. Computer Readable Databases, A Directory ana Data Sourcebook, 8th ed.. Gale Research Inc., Detroit, Mich., 1992 Online Services 1992 Review, Trends oNForecast, SIMBA Information Inc., Wilton, Conn., 1992. 

Following the description of the non-gradient SIMBA (selective inverse multiple bond analysis) experiment by one of the authors in 1991 [176], a number of reports describing related methods appeared [177—183]. There have been a few successful applications of SIMBA experiments reported in the literature [70, 184—186] used to establish specific correlations to complete structure assignments. 

The initial demonstration of the SIMBA experiment utilized the simple alkaloid norharmane (5) as a model compound. Comparison spectra showing... 

Fig. 15A-D. Comparisons of the HMBC and SIMBA spectra for the simple model alkaloid harmane (5). Panel A shows the HMBC spectrum of an 0.09 M solution in d -DMSO acquisition time was about 2h. Panel B shows the traces from the HMBC spectrum showing proton responses to each of the quaternary carbon resonances. Panel C shows the SIMBA traces for the same quaternary carbons. Some of the additional weak responses in the traces in this panel were a result of our initial attempt to illustrate the technique using square pulses, which can be implemented on a broader range of instruments than the more recent shaped pulses. Panel D shows, in more detail, the proton multiplets for C4a and Cl, illustrating the appearance of the long-range heteronuclear coupling superimposed over the normal proton multiplet structure (used with the permission of Academic Press)...

As noted above, one implicit impediment to the utilization of SIMBA is the requirement of an accurate chemical shift for the resonance to be pulsed selectively. The problem is exacerbated rather than ameliorated when Gaussian 270° pulses are employed. The need for an accurate chemical shift is circumvented by utilizing the recently reported E-BURP-2 pulse of Geen and Freeman (1991) for the final 90° pulse of the SIMBA experiment. Since the E-BURP-2 pulse gives uniform phase and excitation across a relatively wide bandwidth, in contrast to the Gaussian 270° pulse which provides uniform phase and excitation across only a relative narrow... 

The application of SIMBA to cryptospirolepine (1) highlights another very important usage of selective one-dimensional experiments. While it... 

An experiment that actually preceded SIMBA, but which probably has somewhat lower utility, was the selective ID HMQC-TOCSY experiment described by the authors (Crouch et al. 1990c). The selective experiment differs from the HMQC-TOCSY pulse sequence only in the replacement of the final C 90° pulse by a selective pulse. Initially, we utilized a square pulse in this application but, obviously, the improvement in the SIMBA experiment by using a Gaussian 270° or an E-BURP-2 pulse would be realized in this experiment as well. To the best of our knowledge, there have been no applications of this experiment to alkaloids reported in the literature. 

Practically, SIMBA spectra of C5a and C4a confirmed the correlation from the N-methyl resonance that was excluded from the F2 spectral window, in about 10 min. A series of SIMBA spectra using a range of five optima were also acquired for the ClOa resonance did not exhibit a Jch correlation to the Hll resonance of cryptolepine. Unfortunately, the SIMBA spectra likewise failed to confirm this correlation pathway, but allowed the complete study to be undertaken in less than one half hour. Acquisition of the corresponding set of HMBC spectra to look for the connectivity from Hll—>C10a would have been time prohibitive. 

Finally, cryptospirolepine also provided the first real-world application of the SIMBA experiment in the elucidation of the structure of a new alkaloid. A SIMBA experiment was acquired in which the C13b carbon resonating at 120.14ppm was selectively pulsed, allowing the observation of a weak " Jch coupling to the H4 resonance and confirming the presence of an indolobenzazepine component to the structure. The particulars of this experiment have been discussed above in the experimental section of this chapter and the spectrum is shown in Fig. 16. 

We thank Messrs. David Wliite and Simba Matondo, and Drs. Heinz-Josef Kiippers and Pliil Blower for their invaluable efforts on this project on the crown thioether chemistry, and Piofessors S.W.J. Lamberts and E.P. Krenning for their continuing counsel. 

---