Figure 2. The error sum as a function of 8: (*) 1; ( ) 2; (O) 3.were recorded on a Varían VXR-500 (500 MHz) instrument at 30 ± 1 °C. A normal NOESY pulse sequence was employed. The mixing time was 0.1 s. Additional experimental details have been reported previously.7Computational Procedures. Equation 2 contains three unknown parameters (rs, t(, S), but rs is equal for all protons in the surfactant. When s is chosen, the experimental 7, values can be fitted to eq 2. Variation of ts produces a shallow minimum in the error sum (Figure 2). The error sum is defined20 as the sum over all squares of the differences between the observed and calculated Tf^s, divided by the square of the observed value. The ts values obtained in this way are of moderate accuracy (±1 ns).The errors in rf and 5 were estimated from a data set of 100 7,'s.These 7,'s were calculated from the experimental 7,'s and the experimental errors in 7, using an inverse Gaussian function and a confidence interval of 80%.eter), 200 (Nicolet NT-200 spectrometer), and 300 MHz (Varían VXR-300 spectrometer). Typical parameters in the experiments were 10 pulse interval times from 0.005 to 15 s. The delay time between the pulse sequences was at least 5 times the longest 7,.Good experimental fits were obtained. Using a series of different samples, the 7, values were found to be accurate to within 4% at 200 and 300 MHz, and to within 6% at 90 MHz. All experiments were performed at 35 ± 1 °C. The 2D NOE spectra Acknowledgment. The investigations were supported by the Netherlands Foundation for Chemical Research (SON) with financial aid from the Netherlands Foundation for Scientific Research (NWO). We thank Dr. R. M. Scheek and K. Dijkstra for expert assistance with the Varían VXR-500 spectrometer and with the data analysis of the NOESY experiments. Stimulating discussions with these scientists and with Professor H. J. C. Berendsen are also gratefully acknowledged.

No abstract

The recent interest in the application of density functional theory (DFT) has prompted us to test several functions in molecular geometries of methyl pheophorbides-a (MPa), an important starting material in photodynamic therapy (PDT). In this study, we report on tests for three popular DFT methods: M06-2X, B3LYP, and LSDA. Based on the standard deviation and the mean value, and by using the difference between optimized calculated value and experimental value in geometries, we drew the following conclusions: M06-2X/6-311+G(d,p) attained the smallest standard deviation of difference among the tested DFT methods in terms of bond length, whereas the standard deviation of bond angle in LSDA/6-311+G(d,p) was the smallest. In terms of absolute value, the mean value of LSDA/6-311+G(d,p) calculation was larger than that of M06-2X/6-311+G(d,p). We found that M06-2X/6-311+G(d,p) gave the best performance for MPa in the molecular geometries. The UV-visible spectrum was calculated with time-dependent density-functional theory (TD-DFT). Time-dependent M06-2X/6-311+G(d,p) gave the best performance for MPa in CH2Cl2 solution. In general, TD-DFT calculations in CH2Cl2 solution were more red-shifted compared with those in the solid state.

No abstract

No abstract