4. Acquisition

Time requirement: 40 min Sample: 3% strychnine in CDCl₃ Record normal ¹H and ¹³C NMR spectra of the sample and optimize the spectral widths. Change to the 2D mode of the spectrometer and load the HMSC pulse program. You have to set:

td2: 2 k data points in F, td1: 256 data points in F; sw2: 9.5 ppm swl: 180 ppm offset of ¹H frequency: middle of ¹H NMR spectrum [4.4 ppm] offset of ¹³C frequency: middle of ¹³C NMR spectrum [100 ppm] p1, p2, p4: 90° ¹H transmitter pulse p3, p5: 180° ¹H transmitter pulse p6, p8, p9, p11, p12, p13: 90° ¹³C decoupler pulse p7, p10: 180° ¹³C decoupler pulse dl:2s d2: 1/[4(°J(C,H)] for long-range coupling (8 Hz), set to 31 ms d3: 1/[2('J(C,H)] (145 Hz), set to 3.45 ms initial value for ¢, evolution: 3 pas increment for ¢, evolution: 1/[2-sw1] g1–g3: sinusoidal-shaped field gradients, 1 ms duration and ca. 0.1 T/m strength, ring-down delays (50 μs), lock blanking and gradient coil blanking switches according to actual instrumentation used. Gradient strength

5. Processing

Apply zero-filling in F; to 1 k words in order to have a matrix of 1kx1k real data points. The data set has to be split, which can be achieved for Bruker instruments by a program provided on the authors’ homepage (http://www.nmr.unibe.ch/). Oddnumbered rows form dataset #1 and even-numbered rows form data set #2. It is best to duplicate these two new sets. First apply in both sets a Fourier transformation only in F, after using an unshifted sinusoidal window in F. Addition of set #1 and set #2 will form a set #3, whereas subtraction of set #1 from set #2 will form set #4. In the data sets #3 and #4 apply an unshifted sinusoidal window and Fourier transformation in F,.

Phase correction is not necessary, since magnitude processing is applied.