Inversion of 1JCC correlations in 1,n‐ADEQUATE spectra

ADEQUATE experiments provide an alternative to the more commonly employed GHMBC experiment for the establishment of long‐range heteronuclear connectivities. The 1,1‐ADEQUATE experiment allows the unequivocal identification of both protonated and non‐protonated carbon resonances adjacent to a protonated carbon. The 1,n‐ADEQUATE experiment establishes correlations via an initial ¹J_CH heteronuclear transfer followed by an ⁿJ_CC out‐and‐back transfer, most typically, via three carbon–carbon bonds. Hence, the 1,n‐ADEQUATE experiment allows the equivalent of ⁴J_CH heteronuclear correlations to be probed when they are not observed in a GHMBC spectrum. Aside from the lower sensitivity of the 1,n‐ADEQUATE experiment relative to GHMBC experiments, the interpretation of the former is also complicated by the ‘leakage’ of ¹J_CC correlations into the spectrum that must be identified. A method for the inversion of ¹J_CC correlations to facilitate the interpretation of 1,n‐ADEQUATE spectra is presented that allows a single experiment to be performed to access ¹J_CC and ⁿJ_CC correlation information. Copyright © 2012 John Wiley & Sons, Ltd.     

HMBC‐1,n‐ADEQUATE spectra calculated from HMBC and 1,n‐ADEQUATE spectra

Unsymmetrical and generalized indirect covariance processing methods provide a means of mathematically combining pairs of 2D NMR spectra that share a common frequency domain to facilitate the extraction of correlation information. Previous reports have focused on the combination of HSQC spectra with 1,1‐, 1,n‐, and inverted ¹J_CC 1,n‐ADEQUATE spectra to afford carbon–carbon correlation spectra that allow the extraction of direct (¹J_CC), long‐range (ⁿJ_CC, where n ≥ 2), and ¹J_CC‐edited long‐range correlation data, respectively. Covariance processing of HMBC and 1,1‐ADEQUATE spectra has also recently been reported, allowing convenient, high‐sensitivity access to ⁿJ_CC correlation data equivalent to the much lower sensitivity n,1‐ADEQUATE experiment. Furthermore, HMBC‐1,1‐ADEQUATE correlations are observed in the F₁ frequency domain at the intrinsic chemical shift of the ¹³C resonance in question rather than at the double‐quantum frequency of the pair of correlated carbons, as visualized by the n,1, and m,n‐ADEQUATE experiments, greatly simplifying data interpretation. In an extension of previous work, the covariance processing of HMBC and 1,n‐ADEQUATE spectra is now reported. The resulting HMBC‐1,n‐ADEQUATE spectrum affords long‐range carbon–carbon correlation data equivalent to the very low sensitivity m,n‐ADEQUATE experiment. In addition to the significantly higher sensitivity of the covariance calculated spectrum, correlations in the HMBC‐1,n‐ADEQUATE spectrum are again detected at the intrinsic ¹³C chemical shifts of the correlated carbons rather than at the double‐quantum frequency of the pair of correlated carbons. HMBC‐1,n‐ADEQUATE spectra can provide correlations ranging from diagonal (⁰J_CC or diagonal correlations) to ⁴J_CC under normal circumstances to as much as ⁶J_CC in rare instances. The experiment affords the potential means of establishing the structures of severely proton‐deficient molecules. Copyright © 2013 John Wiley & Sons, Ltd.     

Additional pitfalls of using 1,1‐ADEQUATE for structure elucidation

1,1‐ADEQUATE is a powerful and robust NMR experiment to establish carbon–carbon connectivities using modest sample quantities when cryogenic probe technology is available. Yet potential pitfalls of applying this method are not widely appreciated, such as weak or missing ¹J_CC correlations in strongly coupled ¹³C‐¹³C AB spin systems and unusually large multi‐bond (ⁿJ_CC) correlations associated with particular functional groups. These large ⁿJ_CC correlations observed in 1,1‐ADEQUATE spectra could be mistaken for ¹J_CC correlations. Copyright © 2016 John Wiley & Sons, Ltd.     

Observation of untoward 3Jcc correlations in 1,1‐ADEQUATE spectra of pyrimidine analogs: Avoiding potential interpretation pitfalls

Recently, it has been reported that large ⁿJ_CC correlations can sometimes be observed in 1,1‐ADEQUATE spectra with significant intensity, which opens the possibility of structural misassignment. In this work, we have focused on pyrimidine‐based compounds, which exhibit multiple bond correlations in the 1,1‐ADEQUATE experiment as a consequence of ³J_CC coupling constants greater than 10 Hz. Results are supported by both the experimental measurement of ³J_CC coupling constants in question using J‐modulated‐ADEQUATE and density functional theory calculations.     

Observation of potentially troublesome 2JCC correlations in 1,1‐ADEQUATE spectra

Despite the tremendous usage of HMBC to establish long‐range ¹H–¹³C and ¹H–¹⁵N heteronuclear correlations, an inherent drawback of the experiment is the indeterminate nature of the ⁿJ_XH correlations afforded by the experiment. A priori there is no reliable way of determining whether a given ⁿJ_CH correlation is, for example, via two‐, three‐, or sometimes even four‐bonds. This limitation of the HMBC experiment spurred the development of the ADEQUATE family of NMR experiments that rely on, in the case of 1,1‐ADEQUATE, an out‐and‐back transfer of magnetization via the ¹J_CC homonuclear coupling constant, which is significantly larger than ⁿJ_CC (where n = 2–4) couplings in most cases. Hence, the 1,1‐ADEQUATE experiment has generally been assumed to unequivocally provide the equivalent of ²J_CH correlations. The recent development of the 1,1‐ and 1,n‐HD‐ADEQUATE experiments that can provide homodecoupling for certain ¹J_CC and ⁿJ_CC correlations has increased the sensitivity of the ADEQUATE experiments significantly and can allow acquisition of these data in a fraction of the time required for the original iterations of this pulse sequence. With these gains in sensitivity, however, there occasionally come unanticipated consequences. We have observed that the collapse of proton multiplets, in addition to providing better s/n for the desired ¹J_CC correlations can facilitate the observation of typically weaker ²J_CC correlations across intervening carbonyl resonances in 1,1‐HD‐ADEQUATE spectra. Several examples are shown, with the results supported by the measurement of the ²J_CC coupling constants in question using J‐modulated‐HD‐ADEQUATE and DFT calculations. Copyright © 2016 John Wiley & Sons, Ltd.     

STRONG INADEQUATE,an experiment for detection of small J(C,C) couplings in symmetrical molecules

A new version of the two-dimensional INADEQUATE experiment was designed for detection of small couplings between equivalent carbon atoms separated in the molecule by several bonds, where other techniques fail due to rich line splitting and mutual peak cancellation in many molecules. As the proposed method is suitable for detection of couplings in strongly coupled systems in general, we propose the name STRONG INADEQUATE in the paper. Similar to other methods for detection of couplings between equivalent carbons, the STRONG INADEQUATE experiment utilizes one-bond carbon–proton coupling for creation of the effective chemical shift differences. The STRONG INADEQUATE experiment works superbly for ⁿJ_CC, where n ≥ 3. Then the F1 pattern is reduced to a simple antiphase doublet with ⁿJ_CC separation, and this pattern is also preserved when a symmetrical HC···C′H′ system is coupled to other protons. Even in the measurement of ²J_CC couplings, the STRONG INADEQUATE experiment generates a much simpler pattern than the original pulse sequences for measurement of couplings between equivalent carbons.     

HMBC‐1,1‐ADEQUATE via generalized indirect covariance: a high sensitivity alternative to n,1‐ADEQUATE

1,1‐ADEQUATE and the related long‐range 1,n‐ and n,1‐ADEQUATE variants were developed to provide an unequivocal means of establishing ²J_CH and the equivalent of ⁿJ_CH correlations where n = 3,4. Whereas the 1,1‐ and 1,n‐ADEQUATE experiments have two simultaneous evolution periods that refocus the chemical shift and afford net single quantum evolution for the carbon spins, the n,1‐variant has a single evolution period that leaves the carbon spin to be observed at the double quantum frequency. The n,1‐ADEQUATE experiment begins with an HMBC‐type ⁿJ_CH magnetization transfer, which leads to inherently lower sensitivity than the 1,1‐ and 1,n‐ADEQUATE experiments that begin with a ¹J_CH transfer. These attributes, in tandem, serve to render the n,1‐ADEQUATE experiment less generally applicable and more difficult to interpret than the 1,n‐ADEQUATE experiment, which can in principle afford the same structural information. Unsymmetrical and generalized indirect covariance processing methods can complement and enhance the structural information encoded in combinations of experiments e.g. HSQC‐1,1‐ or ?1,n‐ADEQUATE. Another benefit is that covariance processing methods offer the possibility of mathematically combining a higher sensitivity 2D NMR spectrum with for example 1,1‐ or 1,n‐ADEQUATE to improve access to the information content of lower sensitivity congeners. The covariance spectrum also provides a significant enhancement in the F₁ digital resolution. The combination of HMBC and 1,1‐ADEQUATE spectra is shown here using strychnine as a model compound to derive structural information inherent to an n,1‐ADEQUATE spectrum with higher sensitivity and in a more convenient to interpret single quantum presentation. Copyright © 2012 John Wiley & Sons, Ltd.     

1JCC‐edited HSQC‐1,n‐ADEQUATE: a new paradigm for simultaneous direct and long‐range carbon–carbon correlation

Establishing the carbon skeleton of a molecule greatly facilitates the process of structure elucidation, leaving only heteroatoms to be inserted, heterocyclic rings to be closed, and stereochemical features to be defined. INADEQUATE, and more recently PANACEA, have been the only means of coming close to the goal of totally defining the carbon skeleton of a molecule. Unfortunately, the extremely low sensitivity and prodigious sample requirements of these experiments and the multiple receiver requirement for the latter experiment have severely restricted the usage of these experiments. Proton‐detected ADEQUATE experiments, in contrast, have considerably higher sensitivity and more modest sample requirements. By combining experiments such as 1,1‐ADEQUATE and 1,n‐ADEQUATE with higher sensitivity experiments such as GHSQC through covariance processing, sample requirements can be further reduced with a commensurate improvement in the s/n ratio and F₁ resolution of the covariance processed spectrum. We now wish to report the covariance processing of an inverted ¹J_CC 1,n‐ADEQUATE experiment with a non‐edited GHSQC spectrum to afford a spectrum that can trace the carbon skeleton of a molecule with the exception of correlations between quaternary carbons. Copyright © 2012 John Wiley & Sons, Ltd.     

NMR structural characterization from one‐bond 13C?13C couplings: Complete assignment of a hydrogen‐poor depsidone

The connectivity, conformation, tautomeric form, and dynamics of a new depsidone (perisalazinic acid) were characterized using one‐bond ¹³C? ¹³C NMR scalar couplings (¹J_CC) obtained from the INADEQUATE experiment. Characterization of perisalazinic acid using more conventional NMR techniques is problematic due to the extremely limited number of C? H protons present. In the present study, 81 candidate structures were considered and a best fit structure was selected by comparing computed ¹J_CC values for each candidate to 15 experimental values. Of the six flexible moieties in perisalazinic acid, three are adequately represented by a single orientation stabilized by intramolecular hydrogen bonding. The three remaining groups are present as mixtures of conformers with two sites consisting of a pair of conformations and another disordered over six orientations. This study demonstrates the feasibility of complete three‐dimensional structural characterization of an unknown using only theoretical and experimental ¹J_CC values.     

Posaconazole: Application of HSQC‐ADEQUATE from General Indirect Covariance Processing

Posaconazole is a structurally complex triazole antifungal agent that, by virtue of its structural complexity, provides a good test molecule for the evaluation of NMR structure elucidation methodologies. Although GHMBC and related long‐range ¹H–¹³C heteronuclear shift correlation techniques are extremely powerful, at the same time, when dealing with unknowns, they can be problematic in that there is no way to readily differentiate adjacent (² J_CH) correlations from longer range correlations, e.g., ³J_CH and ⁿJ_CH, n > 3. The 1,1‐ADEQUATE experiment, in contrast, provides unequivocal experimental access to adjacent carbon–carbon correlation information, albeit with a sensitivity penalty, as the experiment involves an adjacent ¹³C–¹³C out‐and‐back magnetization transfer. In part, the sensitivity penalty can be overcome by using unsymmetrical indirect covariance or general indirect covariance processing methods. The application of these methods through the coprocessing of multiplicity‐edited GHSQC and 1,1‐ADEQUATE data to generate an HSQC‐ADEQUATE correlation plot is demonstrated for posaconazole.     

Coniothyrione: anatomy of a structure revision

Coniothyrione is a xanthone‐derived antibiotic reported several years ago by researchers at Merck & Co. Inc. Revision of the position of the chloro substitution was recently proposed on the basis of empirical reinterpretation of the carbon chemical shift data and a hypothetical biosynthetic argument without the acquisition of any new spectral data to support the postulated change in substituent location. The originally published HMBC data lead to an equivocal assignment of the structure and do not provide a solid basis of support for either structure. Neural network ¹³C chemical shift calculations and density functional theory calculations also led to undifferentiated structures. Definitive confirmation of the structure of coniothyrione based on the acquisition and interpretation of 1,1‐ADEQUATE and inverted ¹J_CC 1,n‐ADEQUATE data is now reported. Copyright © 2013 John Wiley & Sons, Ltd.     

Evaluating the accuracy of theoretical one‐bond 13C─13C scalar couplings and their ability to predict structure in a natural product

This study explores the feasibility of using a combination of experimental and theoretical 1‐bond ¹³C─¹³C scalar couplings (¹J_CC) to establish structure in organic compounds, including unknowns. Historically, ⁿJ_CC and ⁿJ_CH studies have emphasized 2 and 3‐bond couplings, yet ¹J_CC couplings exhibit significantly larger variations. Moreover, recent improvements in experimental measurement and data processing methods have made ¹J_CC data more available. Herein, an approach is evaluated in which a collection of theoretical structures is created from a partial nuclear magnetic resonance structural characterization. Computed ¹J_CC values are compared to experimental data to identify candidates giving the best agreement. This process requires knowledge of the error in theoretical methods, thus the B3LYP, B3PW91, and PBE0 functionals are evaluated by comparing to 27 experimental values from INADEQUATE. Respective errors of ±1.2, ±3.8, and ±2.3 Hz are observed. An initial test of this methodology involves the natural product 5‐methylmellein. In this case, only a single candidate matches experimental data with high statistical confidence. This analysis establishes the intramolecular hydrogen‐bonding arrangement, ring heteroatom identity, and conformation at one position. This approach is then extended to hydroheptelidic acid, a natural product not fully characterized in prior studies. The experimental/theoretical approach proposed herein identifies a single best‐fit structure from among 26 candidates and establishes, for the first time, 1 configuration and 3 conformations to complete the characterization. These results suggest that accurate and complete structural characterizations of many moderately sized organic structures (<800 Da) may be possible using only ¹J_CC data.     

A study of NMR parameters of cyclobutenones and benzocyclobutenones

An NMR study of ketones 5–12 was undertaken to gain insight into the low electrophilicity of the carbonyl moiety of butenones 9–11. Initial IR studies on compounds 9–12 indicated that there is relatively strong double bond character (and hence low electrophilicity) in the carbonyl of saturated and unsaturated cyclobutyl ketones. The ¹³C chemical shifts confirm that the carbonyl moiety is highly conjugated with the fused benzene ring in 9, and with the olefinic linkage in 10 and 11. Partial positive charge is distributed away from the carbonyl carbon, which is also expected to lower the electrophilicity of the carbonyl carbon atoms of 9–11. One‐bond carbon–proton coupling constants (¹J_CH) depend directly on carbon hybridization. In the four‐membered ring ketones 9–12 the experimental values are larger than in cyclobutane, probably as a result of the additional strain of the extra trigonal centers in the ring. A similar trend is seen in the case of the olefinic CH in 10 and 11 (ca 175 Hz), for which the coupling constant is larger than for the corresponding carbon in cyclobutene. ¹J_CC values between the ring carbon atoms of the cyclobutenones are some 20% lower than in the models—a bigger difference than in cyclobutanes, again indicative of the increased ring strain. The very low 42.4 Hz coupling between C‐1 and C‐2 in 9 might well indicate a measure of bond localization. ²J_CC and ³J_CC values are also discussed. Copyright © 2003 John Wiley & Sons, Ltd.     

Long‐range 19F–15N heteronuclear shift correlation: examination of J‐modulations associated with broad range accordion excitation

The first demonstrated example of ¹⁹F–¹⁵N long‐range heteronuclear shift correlation spectroscopy at natural abundance is reported. Because of the very large variation in the size of ²J(N,F) vs ³J(N,F) long‐range heteronuclear couplings, the utilization of one of the new accordion‐optimized long‐range heteronuclear shift correlations experiments is essential if all possible correlations are to be observed in a single experiment. A modified IMPEACH‐MBC pulse sequence was used in conjunction with an optimization range from 4 to 50 Hz to demonstrate the technique using a mixture of 2‐ and 3‐fluoropyridine, which had ²J(N,F) and ³J(N,F) long‐range couplings of ?52 and 3.6 Hz, respectively. Because of the size of the ²J(N,F) long‐range coupling constant, a J‐modulation of the long‐range correlation response is observed in the spectrum resulting in a ‘doublet’ in F₁ due to amplitude modulation. The size of the ‘doublet’ is shown to be a function of the parameter selection (t_1max,T_max,T_min and spectral width in F₁). This behavior is similar to F₁ ‘skew’ associated with long‐range correlation responses in ACCORD‐HMBC spectra which has been analyzed in detail previously. Copyright © 2002 John Wiley & Sons, Ltd.     

D-HMBC versus LR-HSQMBC: Which experiment provides theoretically and experimentally the best results?

The long-range heteronuclear single quantum multiple bond correlation (LR-HSQMBC) experiment is the experiment of choice for visualizing heteronuclear long-range coupling interactions ⁿJ_CH across 4–6-bonds and is experimentally superior to the decoupled heteronuclear multiple-bond correlation (D-HMBC) experiment. Yet, the exact reasons have not been fully understood and established. On the basis of our recent investigation of the nonrefocused variants LR-HSQC and HMBC, we have extended a J_HH′-dedicated investigation to the D-HMBC and LR-HSQMBC experiments. Unlike the nonrefocused variants, the influence of homonuclear couplings J_HH′ on the intensity of long-range ⁿJ_CH cross-peaks is not easily predictable and may be summarized as follows: (a) irrespective of the magnitude and number of J_HH′ interactions long-range ⁿJ_CH cross-peaks are more intense in D-HMBC spectra as long as the evolution delay Δ is not too large, because in contrast to LR-HSQMBC no J_HH′-caused intensity zeroes will occur. (b) If J_HH′ is small and Δ large, the intensity of cross peaks in D-HMBC spectra may be weakened or may even vanish at Δ = (0.25+0.5k)/J_HH′, whereas for the LR-HSQMBC this unwanted effect occurs at Δ = k + 0.5/J_HH′. Consequently, when Δ is adjusted to visualize weak ⁿJ_CH long-range correlations, our findings corroborate that there are potentially more cross-peaks expected to show up in a LR-HSQMBC spectrum compared with a D-HMBC spectrum. This has been indeed noticed experimentally, even though the intensity of a many long-range ⁿJ_CH cross-peaks may still be higher in the spectra of the D-HMBC experiment correspondingly adjusted for detecting weak ⁿJ_CH correlations.     

Developing nonuniform sampling strategies to improve sensitivity and resolution in 1,1-ADEQUATE experiments

Nonuniform sampling (NUS) strategies are developed for acquiring highly resolved 1,1-ADEQUATE spectra, in both conventional and homodecoupled (HD) variants with improved sensitivity. Specifically, the quantile-directed and Poisson gap methods were critically compared for distributing the samples nonuniformly, and the quantile schedules were further optimized for weighting. Both maximum entropy and iterative soft thresholding spectral estimation algorithms were evaluated. All NUS approaches were robust when the degree of data reduction is moderate, on the order of a 50% reduction of sampling points. Further sampling reduction by NUS is facilitated by using weighted schedules designed by the quantile method, which also suppresses sampling noise well. Seed independence and the ability to specify the sample weighting in quantile scheduling are important in optimizing NUS for 1,1-ADEQUATE data acquisition. Using NUS yields an improvement in sensitivity, while also making longer evolution times accessible that would be difficult or impractical to attain by uniform sampling. Theoretical predictions for the sensitivity enhancements in these experiments are in the range of 5–20%; NUS is shown to disambiguate weak signals, reveal some ⁿJ_CC correlations obscured by noise, and improve signal strength relative to uniform sampling in the same experimental time. This work presents sample schedule development for applying NUS to challenging experiments. The schedules developed here are made available for general use and should facilitate the broader utilization of ADEQUATE experiments (including 1,1-, 1,n-, and HD- variants) for challenging structure elucidation problems.     

Compensating for variation in 1JCH coupling constants in HSQC spectra acquired on small organic molecules

The HSQC sequence provides a sensitive way of determining the ¹³C chemical shift of protonated carbons. It uses INEPT elements for magnetization transfer, which can only be optimized for one value of ¹J_CH, but small organic molecules contain a wide range of ¹J_CH values. One popular method of compensating for ¹J_CH variation is to incorporate adiabatic pulses into the INEPT elements. This article shows that this method fails for a significant subset of functional groups. It also shows that the effects of this failure can be reduced by avoiding refocusing delays and by using a J‐compensated excitation element. Copyright © 2010 John Wiley & Sons, Ltd.     

Some INDO calculations of 1J(PC) and 1J(PF) values

INDO parameterized calculations, employing phosphorus s, p and d valence orbitals, are reported for values of ¹J(PC) and ¹J(PF) relating to phosphorus in formal tri- and pentavalent states. The ¹J(PC), interactions are mainly controlled by the contact term. Thus, trivalent phosphorus compounds have negative values for ¹J(PC), whereas those for pentavalent phosphorus are positive due to the s lone-pair effect. The inclusion of phosphorus 3d orbitals is shown to be important for an understanding of the processes contributing to ¹J(PF) interactions. ¹J(PF) values are shown to be negative for both tri- and pentavalent phosphorus compounds. The contact and orbital interactions are significant for the trivalent phosphorus molecules, whereas in the pentavalent phosphorus case ¹J(PF) is dominated by the orbital term.     

Using LR-HSQMBC to observe long-range H–N correlations

The recently reported LR-HSQMBC experiment has been optimized for ¹H–¹⁵N long-range heteronuclear couplings. Several previously unreported four-bond correlations, consistent with the predicted by DFT calculations (0.2–0.3 Hz ⁴J_NH couplings), have been observed for strychnine using 2 Hz optimization of the LR-HSQMBC experiment. This experiment offers an advantage over accordion-optimized experiments such as IMPEACH and CIGAR for the observation of long-range ¹H–¹⁵N correlations in that the experiment is refocused and employs a CLIP pulse sequence element to bring the long-range correlations into phase, allowing broadband X-decoupling to be employed during acquisition.     

Substituent electronegativities

It is shown that theoretical calculations (ab initio 6-31g^* basis) of the charge on the hydrogen in molecules HX and ecperimental values for J_CC (ipso-ortho) in monosubstituted benzenes both provid measures of substituent electronegativity.