ID Experiments

Since there is some contribution of the receiver dc to the signal, this needs to be removed. In ID experiments, the FIDs decay substantially, and the last portion of the FID gives a reasonably good estimate of the dc level. 

There are actually two independent time periods involved, t and t. The time period ti after the application of the first pulse is incremented systematically, and separate FIDs are obtained at each value of t. The second time period, represents the detection period and it is kept constant. The first set of Fourier transformations (of rows) yields frequency-domain spectra, as in the ID experiment. When these frequency-domain spectra are stacked together (data transposition), a new data matrix, or pseudo-FID, is obtained, S(absorption-mode signals are modulated in amplitude as a function of t. It is therefore necessary to carry out second Fourier transformation to convert this pseudo FID to frequency domain spectra. The second set of Fourier transformations (across columns) on S (/j, F. produces a two-dimensional spectrum S F, F ). This represents a general procedure for obtaining 2D spectra. 

We can see three sets of cross peaks methyl/methylene and aromatic CH as in COSY, and in addition a clear interaction between the methine proton and the aromatic protons closest to it. This interaction is naturally not visible in the COSY spectrum, as the protons are separated by five bonds. A look back to Section 1.1.6 shows that this NOE was (as must be the case) also visible in the ID experiment. 

Fig. 14.2 displays the fraction of bound ligand as a function of the dissociation constant for two different ligand concentrations (A) 100 pM, about the detection limit for ID experiments by use of cryo-probes, and (B) 1 mM, a concentration for standard equipment. At these concentrations, typical measurement times are in the range of several minutes. From the figure the following conclusion can be drawn for ligand-observe techniques ... 

We have implemented the principle of multiple selective excitation (pulse sequence II in fig. 1) thereby replacing the low-power CW irradiation in the preparation period of the basic ID experiment by a series of selective 180° pulses. The whole series of selective pulses at frequencies /i, /2, , / is applied for several times in the NOE build-up period to achieve sequential saturation of the selected protons. Compared with the basic heteronuclear ID experiment, in this new variant the sensitivity is improved by the combined application of sequential, selective pulses and the more efficient data accumulation scheme. Quantitation of NOEs is no longer straightforward since neither pure steady-state nor pure transient effects are measured and since cross-relaxation in a multi-spin system after perturbation of a single proton (as in the basic experiment) or of several protons (as in the proposed variant) differs. These attributes make this modified experiment most suitable for the qualitative recognition of heteronuclear dipole-dipole interactions rather than for a quantitative evaluation of the corresponding effects. 

In a move in the opposite direction, the overlaps resulting from concatenation of different polarization transfer mechanisms in combined 2D experiments can be eliminated by reducing the dimensionality of an experiment. Similarly to the successful transformation of basic 2D NMR techniques into their ID counterparts [32-34], a conversion of combined 2D NMR techniques into their ID analogs is feasible and has been explored by several groups [35-40]. From a different perspective this process can be seen as a twofold reduction of the dimensionality in a 3D experiment. Equally, concatenation of three polarization transfer steps in a single ID experiment represents transformation of a possible 4D homonuclear experiment into its ID analog. 

One-dimensional spectra obtained in these experiments can be compared to ID traces of nD NMR spectra but offering much better digital resolution and shorter acquisition times. On the negative side each trace needs to be acquired separately and thus, if several sites are to be inspected, a series of ID experiments must be performed. In practice, this exercise is preceded by careful inspection of standard two-dimensional COSY, TOCSY, NOESY or ROESY spectra and only the ambiguous assignments are tackled by combined ID techniques. 

In this chapter, the discussion will be focused on the ID TOCSY (TO-tal Correlation SpectroscopY) [2] experiment, which, together with ID NOESY, is probably the most frequently and routinely used selective ID experiment for elucidating the spin-spin coupling network, and obtaining homonuclear coupling constants. We will first review the development of this technique and the essential features of the pulse sequence. In the second section, we will discuss the practical aspects of this experiment, including the choice of the selective (shaped) pulse, the phase difference of the hard and soft pulses, and the use of the z-filter. The application of the ID TOCSY pulse sequence will be illustrated by examples in oligosaccharides, peptides and mixtures in Section 3. Finally, modifications and extensions of the basic ID TOCSY experiment and their applications will be reviewed briefly in Section 4. 

The simple one pulse as well as more sophisticated multiple pulse ID experiments for the measurement of the corresponding H or C spectra. 

Non-selective ID experiments as well as selective ID experiments including either selective weak pulses (ID ROESY, ID TOCSY) or selective continuous radiofrequency irradiation (ID homonuclear decoupling, ID NOE). 

All spectra were measured on a Bruker DRX 500 spectrometer, at 500.13MHz ( H) and at 125.76MHz ( C). Sample spinning (20Hz) was used for all ID experiments with the exception of the ID NOE and ID ROESY experiments. As is now common, all 2D experiments were performed with the sample static. All experiments were performed at ambient magnet temperature without any special temperature control. 

The central step of any NMR data processing is the Fourier transformation (FT) which transforms the time domain signal s(t) - the raw data - of a ID experiment into a frequency domain signal S(f) - the spectrum ... 

For FIDs obtained with ID experiments, the signal has usually decayed close to zero, since the acquisition time is large compared to the relaxation time T,. However the FIDs (both rows and columns) obtained in a 2D experiment have usually not completely decayed within the available acquisition times in t2 and tl respectively, since the acquisition times are short compared to the relaxation time T,. 

The subtraction of FIDs obtained with selective ID experiments (ID NOE, ID ROE,... 

Assign signals in your spectra only if an assignment is possible and reliable. Be careful and do not hesitate to label an assignment as tentative if you have any doubts. Designate potential target signals for subsequent selective ID experiments, if such experiments are planned. 

---