Correlation of 1JCH spin–spin coupling constants and their solvent sensitivities
Graphical abstract
Highlights
► Solvent induced changes of 1JCH coupling constants depend on solvent polarity. ► Solvent polarity dependencies of 1JCH are linear in reaction field coordinates. ► 1JCH obtained by extrapolation to zero reaction field is close to gas phase value. ► Slope of 1JCH dependency is interpreted as its solvent sensitivity. ► Solvent sensitivities and zero field values of 1JCH are linearly correlated.
Introduction
Spin–spin coupling constants (SSCC), discovered sixty years ago, have soon become one of the most important NMR parameters characterizing molecular properties. Since NMR parameters are mostly being measured in macroscopic dense samples, they are always altered by intermolecular interactions in solid, liquid and, to a lesser extent, in gaseous states. Solvent dependence of SSCC was known from early studies in NMR spectroscopy [1], [2] with the experimental and theoretical aspects discussed in many comprehensive reviews [3], [4], [5], [6].
Nowadays, SSCCs can be measured with a very high accuracy, and, in many cases, the errors do not exceed 0.1 Hz, which is less than 0.1% for many one-bond SSCCs. However, the presence of solvent effects decreases the significance of precise SSCC measurements and the importance of SSCCs as individual molecular parameters in general. In some cases, variation of one-bond C–H SSCC, measured in different solvents, can reach up to 10% of its value. For instance, 1JCH of chloroform varies by 9.6 Hz if comparing the measurements in cyclohexane and DMSO.
Solvent effects have two main origins: bulk or non-specific effects arising from the medium, mostly due to its electrostatic properties, and molecular effects arising from specific interactions. They act simultaneously and it is often difficult to separate them, especially for complex multi-component solvents. However, for many small and simple solute molecules and solvents, it is rather possible to simplify solvent effects, reducing them into a single dominating origin. Moreover, an appropriate choice of used solutes and solvents has a potential of allowing to successfully investigate certain types of solvent effects and to reveal solute parameters that are not disturbed by the medium. Here we follow this strategy and try to isolate and consider one particular origin of solvent effects on SSCCs, namely solvent polarity.
There are many attempts to describe solvent effects on SSCC using different approaches of modeling bulk solvent properties and solute–solvent molecular interactions. In many of those, the significant influence of dielectric properties of solvent on one-bond SSCCs has been noted [3], [4], [5], [6], [7], [8] and references therein]. Following Barfield and Johnston [3], we will consider solvent effects on molecules where the relative spatial locations of coupled nuclei are constrained by the structural features of molecules (first category molecules). The second category includes flexible molecules, which have two or more conformations and the net solvent effect on these systems is very complicated, since, in addition to the first category effects, they involve solvent induced changes in the relative populations of conformers, each having different SSCCs.
The spectral parameters for an isolated molecule cannot be measured directly, but they can be determined by combining advanced calculations and measurements in the conditions that approximate the isolated state. Today, such combined estimations of NMR chemical shifts and SSCCs are known for dozens of molecules. The most successful experimental method to measure NMR parameters of nearly isolated molecules is the gas phase NMR spectroscopy [9], [10], [11], [12]. In such studies, the investigated compound is diluted in gaseous solvents, mainly xenon (Xe), carbon dioxide (CO2), sulfur hexafluoride (SF6), or their mixtures. Due to the dependence of the measured parameters on gas density, the measurements are made at various gas pressures, and, the NMR parameter of interest is obtained by extrapolation of the measured data to the zero density. It is assumed that the obtained parameter is very close to its value in isolated molecule, which is confirmed by theoretical calculations [10]. Gas phase NMR studies are done for about two dozens of molecules, mainly various methanes, particularly for acetonitrile [13], [14], bromomethane [15], fluoromethanes [16], [17], [18], iodomethane [19], methane and other tetrahedral molecules (SiH4 and GeH4) [20]. Unfortunately, the gas phase NMR technique is very complicated and is established only at several laboratories. To this end, it is important to find the possibility to obtain pure NMR data for isolated molecules using measurements in liquid phase.
Here we have studied acetonitirile and halogen substituted methanes to understand whether it is possible to estimate the values of NMR parameters (namely one-bond SSCCs) of isolated molecules using only the data measured in solution NMR. Even more preferable opportunity would be to do so by just a single measurement in any solvent, rather than having to perform an array of experiments in different solvents.
The measured SSCC in solvents, in the first approximation, can be represented as the sum of two terms: the ‘pure’ J0-coupling of isolated molecule in the vacuum and the ΔJ modulation induced by the solvent. In general, the ΔJ is formed by contributions reflecting different inter- and intramolecular interactions. Simple dependences of solvent properties and NMR parameters were often searched for, and the most significant was found to be the dielectric constant, ε, of solvent. In several studies, distinctive correlations joining NMR parameters and ε or ε-dependent functions were noted [6], [7]. Recently, we have also shown the strong ε-dependence of one-bond SSCC of acetonitrile in various solvents [8], [21]. Here we have tried to further illuminate the peculiarities of this dependence and exploit the possibilities it opens.
Section snippets
Methods
Experiments were done on Varian Mercury 300VX NMR spectrometer equipped with standard broadband probe. Dielectric constants of used solvents and their mixtures are taken from literature [22], [23], [24].
Acetonitrile and iodomethane were 13C-enriched, while SSCCs of other probe molecules were measured from 1H spectra using the satellite lines observed from compounds with natural abundance of 13C nuclei. Linewidths did not exceed 0.3 Hz for all spectra, with the digital resolution being better
Results and discussion
Solvents can be differentiated into several groups, such as acidic, basic, aromatic, hydrogen bonding, etc., in accordance to different dominating types of interactions with solute molecules. Most of those groups, however, have the polarity of the solvent molecules as one of the major factors defining the solvent effects. Furthermore, there are at least three other properties (polarizability, basicity, and acidity) also shown to be important for some systems [36].
Here we conduct a systematic
Discussion
The existence of solvent effects makes large amount of data of SSCC measurements that are available in the literature difficult for use and/or compare. To this end, an array of SSCC data, measured at different experimental conditions, requires systematization for each molecule. One way of doing this could be the use of SSCC values from isolated molecule (or solvent effect-free ‘pure’ values) together with certain factors that reflect upon the sensitivity of SSCCs to different types of solvent
Conclusion
We have demonstrated on methylic solute molecules that the ‘pure’, solvent effect-free, values of SSCCs of the isolated molecules fully determine the behavior of solvent sensitivities of one-bond SSCCs. SSCCs strongly depend on dielectric constant of the medium and can be described by simple linear regression equation in reaction field coordinates, the slope of which depends on the value of SSCC in the isolated state of the molecule.
The found interdependence can be used to investigate the
Acknowledgments
We would like to thank Armenian State Committee of Science for Research Grant 11-1d153 and Armenian National Science and Education Fund (ANSEF) for financial support. A.B.S. is grateful to the Herchel Smith Fund for generous support.
References (41)
- et al.
- et al.
Annu. Rep. NMR Spectrosc.
(2000) J. Mol. Struct.
(2006)- et al.
Chem. Phys. Lett.
(2002) - et al.
J. Mol. Stuct.
(2003) - et al.
Chem. Phys. Lett.
(2007) - et al.
J. Mol. Struct.
(2002) - et al.
J. Mol. Struct.
(2004) - et al.
J. Mol. Struct.
(2005)
J. Mol. Struct.
Chem. Phys.
Chem. Phys. Lett.
Chem. Rev.
J. Chem. Theory Comput.
Mol. Phys.
Magn. Reson. Chem.
Chem. Educ.
Int. J. Mol. Sci.
Cited by (9)
The influence of superbasic conditions and solvent effects on NMR spectra and structural parameters of acetylene in solution
2023, Journal of Molecular LiquidsRevealing the specific solute-solvent interactions via the measurements of the NMR spin-spin coupling constants
2015, Journal of Molecular StructureCitation Excerpt :In Table 1, the data of linear regression analysis for the one-bond SSCCs and solvent sensitivities for isolated molecules of some model compounds are presented for the solvents of the first subset. The results for the non-specific type of solute–solvent interactions have a better correlation (more than 0.94 of Pearson correlation coefficient for all studied molecules) with their solvent sensitivities (see [3]) and are close to the experimental values of “pure” SSCCs. The second subset involves solvents that specifically interact with the solute molecules, or possibly facilitate the formation of the solute–solute specific association.
Revisiting the influence of pH on 1 J C α H and chemical shifts of glycine and alanine short oligopeptides
2023, Royal Society Open ScienceSolvent Effects on Spin-Spin Coupling Constants in Organomercury Compounds
2021, Journal of Contemporary Physics<sup>1</sup>J<inf>CH</inf> couplings in Group 14/IVA tetramethyls from the gas-phase NMR and DFT structural study: A search for the best computational protocol
2014, Physical Chemistry Chemical Physics