Shift-Correlation Experiments

The object of this family of experiments is to correlate the chemical shifts of mutually coupled spins. Thus they replace tedious multiple decoupling experiments, and have the additional advantage of high specificity and an absence of Bloch-Siegert frequency distortions. 

The most useful experiment of all is the homonuclear COSY (two-dimensional homonuclear Correlated SpectroscopY) experiment. This yields a square matrix of data. The diagonal, projected onto either frequency axis, is the normal one-dimensional spectrum, and off-diagonal multiplets appear with the x coordinate of resonance A and the y coordinate of resonance B or vice versa, when A and B are mutually spin-coupled. A two-dimensional spectrum of this type is most conveniently presented as a contour diagram, as viewed from above, even though such maps seem at first strange to one-dimensional spectroscopists. 

In the COSY experiment the mutual transfer of polarization between A and B is accomplished by an INEPT-type method, simplified by omission of the two 180° pulses. One component of a spin multiplet (say of A) is inverted relative to the other after a time t which ideally equals and this alters the B magnetization as 

A comparable range of heteronuclear shift correlation experiments is also available, with and without multiplet decoupling. Again, the spins are correlated by polarization transfer. A more elaborate homonuclear shift correlation experiment, INADEQUATE, may be performed to detect C- C connectivities and hence identify carbon frameworks. The elaborations involve double-quantum processes to eliminate the otherwise dominant signals from C bound to C. These and other methods are fully discussed by Sanders and by Ernst et a/.  

The NOESY experiment gives data similar to the above SECSY or COSY experiment, but now the off-diagonal peaks reveal dipolar-coupled resonances, such as might be studied one-dimensionally via NOE difference spectroscopy. The sequence in its simplest form is 

---

In homonuclear-shift-correlated experiments, the Ft domain corresponds to the nucleus under observation in heteronuclear-shift-correlated experiments. Ft relates to the unobserved or decoupled nucleus. It is therefore necessary to set the spectral width SW, after considering the ID spectrum of the nucleus corresponding to the Ft domain. In 2D /-resolved spectra, the value of SW depends on the magnitude of the coupling constants and the type of experiment. In both homonuclear and heteronuclear experiments, the size of the largest multiplet structure, in hertz, determines... 

The matrix obtained after the F Fourier transformation and rearrangement of the data set contains a number of spectra. If we look down the columns of these spectra parallel to h, we can see the variation of signal intensities with different evolution periods. Subdivision of the data matrix parallel to gives columns of data containing both the real and the imaginary parts of each spectrum. An equal number of zeros is now added and the data sets subjected to Fourier transformation along I,. This Fourier transformation may be either a Redfield transform, if the h data are acquired alternately (as on the Bruker instruments), or a complex Fourier transform, if the <2 data are collected as simultaneous A and B quadrature pairs (as on the Varian instruments). Window multiplication for may be with the same function as that employed for (e.g., in COSY), or it may be with a different function (e.g., in 2D /-resolved or heteronuclear-shift-correlation experiments). 

A more useful type of 2D NMR spectroscopy is shift-correlated spectroscopy (COSY), in which both axes describe the chemical shifts of the coupled nuclei, and the cross-peaks obtained tell us which nuclei are coupled to which other nuclei. The coupled nuclei may be of the same type—e.g., protons coupled to protons, as in homonuclear 2D shift-correlated experiments—or of different types—e.g., protons coupled to C nuclei, as in heteronuclear 2D shift-correlated spectroscopy. Thus, in contrast to /-resolved spectroscopy, in which the nuclei were being modulated (i.e., undergoing... 

Another 2D homonuclear shift-correlation experiment that provides the coupling information in a different format is known as SECSY (spin-echo correlation spectroscopy). It is of particular use when the coupled nuclei lie in a narrow chemical shift range and nuclei with large chemical shift differences are not coupled to one another. The experiment differs... 

In homonudear shift-correlation experiments like COSY we were concerned with the correlation of chemical shifts between nuclei of the same nuclear species, e.g., H with H. In heteronuclear shift-correlation experiments, however, the chemical shifts of nuclei belonging to different nuclear species are determined (e.g., H with C). These may be one-bond chemical shift correlations, e.g., between directly bound H and C nuclei, or they may be long-range chemical shift correlations, in which the interactions... 

Figure 5.40 (A) Pulse sequence for the 2D heteronuclear shift-correlation experiment. (B) Effect of the pulse sequence in (A) on H magnetization vectors of CH.

The SELINCOR experiment is a selective ID inverse heteronuclear shift-correlation experiment i.e., ID H,C-COSYinverse experiment) (Berger, 1989). The last C pulse of the HMQC experiment is in this case substituted by a selective 90° Gaussian pulse. Thus the soft pulse is used for coherence transfer and not for excitation at the beginning of the sequence, as is usual for other pulse sequences. The BIRD pulse and the A-i delay are optimized to suppress protons bound to nuclei As is adjusted to correspond to the direct H,C couplings. The soft pulse at the end of the pulse sequence (Fig. 7.8) serves to transfer the heteronuclear double-quantum coherence into the antiphase magnetization of the protons attached to the selectively excited C nuclei. 

This is a variation of the proton-detected shift-correlation experiment via long-range couplings proposed by Bax and Summers (Bax and Summers, 1986), with the difference that the first C pulse is substituted by a frequency selective pulse (Fig. 7.14) (Bermel et al., 1989 Kessler et al., 1989b,1990). This significantly increases resolution in the F dimension. For example, this can be used to remove the overlap between the cross-peaks of the carbonyl resonances of peptide bonds in proteins that all occur within a... 

Inverse experiments Heteronuclear shift-correlation experiments in which magnetization of the less sensitive heteronucleus (e.g., C) is detected through the more sensitive magnetization (e.g., H). 

In the solid, dynamics occurring within the kHz frequency scale can be examined by line-shape analysis of 2H or 13C (or 15N) NMR spectra by respective quadrupolar and CSA interactions, isotropic peaks16,59-62 or dipolar couplings based on dipolar chemical shift correlation experiments.63-65 In the former, tyrosine or phenylalanine dynamics of Leu-enkephalin are examined at frequencies of 103-104 Hz by 2H NMR of deuterated samples and at 1.3 x 102 Hz by 13C CPMAS, respectively.60-62 In the latter, dipolar interactions between the 1H-1H and 1H-13C (or 3H-15N) pairs are determined by a 2D-MAS SLF technique such as wide-line separation (WISE)63 and dipolar chemical shift separation (DIP-SHIFT)64,65 or Lee-Goldburg CP (LGCP) NMR,66 respectively. In the WISE experiment, the XH wide-line spectrum of the blend polymers consists of a rather featureless superposition of components with different dipolar widths which can be separated in the second frequency dimension and related to structural units according to their 13C chemical shifts.63... 

Unlike HMBC /GHMBC and related long-range heteronuclear chemical shift correlation experiments, which have hundreds of reported applications in the published literature, there are considerably fewer reported applications of 1,1-ADEQUATE and related long-range variants in the literature. In part, the dearth of reported applications can be attributed to the considerably lower sensitivity of these experiments relative to, for example HMBC/GHMBC. Sensitivity concerns are largely ameliorated,... 

In order to determine the structure and stereochemistry of compounds 36 and 37, standard H- HCOSY, H- C shift correlation (HETCOR), DEPT, and H NOE NMR experiments were performed. The 13C NMR spectrum of compound 36 is based on DEPT, HETCOR, and long-range H- C shift correlation experiments, which allowed the correct assignment of most of the carbon signals <1997MI328>. 

Fig. 2 (a) DRAMA pulse sequence (using % = t/2 = rr/4 in the text) and a representative calculated dipolar recoupled frequency domain spectrum (reproduced from [23] with permission), (b) RFDR pulse sequence inserted as mixing block in a 2D 13C-13C chemical shift correlation experiment, along with an experimental spectrum of 13C-labeled alanine (reproduced from [24] with permission), (c) Rotational resonance inversion sequence along with an n = 3 rotational resonance differential dephasing curve for 13C-labeled alanine (reproduced from [21] with permission), (d) Double-quantum HORROR experiment along with a 2D HORROR nutation spectrum of 13C2-2,3-L-alanine (reproduced from [26] with permission)... 

Fig. 5 Symmetry-based dipolar recoupling illustrated in terms of pulse sequences for the CN (a) and RNvn (b) pulse sequences, a spin-space selection diagram for the Cl symmetry (c) (reproduced from [118] with permission). Application of POST-CVj [31] as an element in a H- H double-quantum vs 13C chemical shift correlation experiment (d) used as elements (B panel) in a study of water binding to polycrystalline proteins (reproduced from [119] with permission)...

NMR probes are designed with the X-coil closest to the sample for improved sensitivity of rare nuclei. Inverse detection NMR probes have the proton coil inside the X-coil to afford better proton sensitivity, with the X-coil largely relegated to the task of broadband X-nucleus decoupling. These proton optimized probes are often used for heteronuclear shift correlation experiments. 

Smaller diameter probes reduce sample volumes from 500 to 600 pi typical with a 5 mm probe down to 120-160 pi with a 3 mm tube. By reducing the sample volume, the relative concentration of the sample can be correspondingly increased for non-solubility limited samples. This dramatically reduces data acquisition times when more abundant samples are available or sample quantity requirements when dealing with scarce samples. At present, the smallest commercially available NMR tubes have a diameter of 1.0 mm and allow the acquisition of heteronuclear shift correlation experiments on samples as small as 1 pg of material, for example in the case of the small drug molecule, ibu-profen [5]. In addition to conventional tube-based NMR probes, there are also a number of other types of small volume NMR probes and flow probes commercially available [6]. Here again, the primary application of these probes is the reduction of sample requirements to facilitate the structural characterization of mass limited samples. Overall, many probe options are available to optimize the NMR hardware configuration for the type and amount of sample, its solubility, the nucleus to be detected as well as the type and number of experiments to be run. 

The earliest of the magnetization transfer experiments is the spin population inversion (SPI) experiment [27]. By selectively irradiating and inverting one of the 13C satellites of a proton resonance, the recorded proton spectrum is correspondingly perturbed and enhanced. Experiments of this type have been successfully utilized to solve complex structural assignments. They also form the basis for 2D-heteronuclear chemical shift correlation experiments that are discussed in more detail later in this chapter. 

Fig. 10.13. 2D J-resolved NMR spectrum of santonin (4). The data were acquired using the pulse sequence shown in Fig. 10.12. Chemical shifts are sorted along the F2 axis with heteronuclear coupling constant information displayed orthogonally in F . Coupling constants are scaled as J/2, since they evolve only during the second half of the evolution period, t /2. 13C signals are amplitude modulated during the evolution period as opposed to being phase modulated as in other 13C-detected heteronuclear shift correlation experiments.

Direct heteronuclear chemical shift correlation Conceptually, the 2D J-resolved experiments lay the groundwork for heteronuclear chemical shift correlation experiments. For molecules with highly congested 13C spectra, 13C rather than XH detection is desirable due to high resolution in the F% dimension [40]. Otherwise, much more sensitive and time-efficient proton or so-called inverse -detected heteronuclear chemical shift correlation experiments are preferable [41]. 

Fig. 10.15. Pulse sequence for the multiplicity-edited gradient HSQC experiment. Heteronuclear single quantum coherence is created by the first INEPT step within the pulse sequence, followed by the evolution period, t . Following evolution, the heteronuclear single quantum coherence is reconverted to observable proton magnetization by the reverse INEPT step. The simultaneous 180° XH and 13C pulses flanked by the delays, A = l/2( 1 edits magnetization inverting signals for methylene resonances, while leaving methine and methyl signals with positive phase (Fig. 16A). Eliminating this pulse sequence element affords a heteronuclear shift correlation experiment in which all resonances have the same phase (Fig. 16B).

Using strychnine (1) as a model compound, a pair of HSQC spectra are shown in Fig. 10.16. The top panel shows the HSQC spectrum of strychnine without multiplicity editing. All resonances have positive phase. The pulse sequence used is that shown in Fig. 10.15 with the pulse sequence operator enclosed in the box eliminated. In contrast, the multiplicity-edited variant of the experiment is shown in the bottom panel. The pulse sequence operator is comprised of a pair of 180° pulses simultaneously applied to both H and 13C. These pulses are flanked by the delays, A = l/2(xJcii), which invert the magnetization for the methylene signals (red contours in Fig. 10.16B), while leaving methine and methyl resonances (positive phase, black contours) unaffected. Other less commonly used direct heteronuclear shift correlation experiments have been described in the literature [47]. 

To provide the means of sampling a wider range of potential long-range heteronuclear coupling constants, an alternative version of this heteronuclear shift correlation experiment is called the accordion-optimized HMBC, or ACCORD-HMBC [50]. Additional modifications of this experiment are also available [51]. 

Fig. 11.6 Pulse sequence for two-dimensional homonuclear chemical-shift correlation experiments. The gray box indicates the mixing sequence, with some examples shown in more detail a RFDR, b RIL, c C7, d DRAWS, e DREAM.

Concatenation of Polarization Transfer Steps into Homonuclear Chemical Shift Correlated Experiments. Application to Oligo- and Polysaccharides... 

Homonuclear as well as heteronuclear 2D shift correlation experiments ( H/ H-COSY, H/ C-COSY, H/C COSY- H/ H-TOCSY), involving the perturbation of either one or two types of nuclei respectively and in the heteronuclear case including both the conventional, direct C detection, as well as the more sensitive, indirect ( inverse or reverse ) H-detection. 

Torres, A.M., Nakashima, T.T, and Mcclung R.E.D. 1993. J-compensated proton-detected heteronuclear shift-correlation experiments. J. Magn. Reson. Ser. A 102 219-227. 

SCHEME 8. Connectivities derived from II-1II and 11-1 C shift correlation experiments... 

Variable temperature 13C H NMR studies on 45 show large coupling of the a-carbon atoms of the metallacyclopentane ring to the 183W nucleus (70 and 86 Hz), in contrast to the smaller coupling associated with the coordinated ethene (33 and 37 Hz). A two-dimensional 13C NMR shift correlation study confirmed the connectivity observed in the X-ray study, while 13C 111 chemical shift correlation experiments identified the H NMR resonances. 

---