An NMR, IR and theoretical investigation of ¹H Chemical Shifts and hydrogen bonding in phenols

Abstract

The change in (1)H NMR chemical shifts upon hydrogen bonding was investigated using both experimental and theoretical methods. The (1)H NMR spectra of a number of phenols were recorded in CDCl(3) and DMSO solvents. For phenol, 2- and 4-cyanophenol and 2-nitrophenol the OH chemical shifts were measured as a function of concentration in CDCl(3). The plots were all linear with concentration, the gradients varying from 0.940 (phenol) to 7.85 (4-cyanophenol) ppm/M because of competing inter- and intramolecular hydrogen bonding. Ab initio calculations of a model acetone/phenol system showed that the OH shielding was linear with the H...O=C distance (R) for R < 2.1 A with a shielding coefficient of - 7.8 ppm/A and proportional to cos(2)phi where phi is the H...O=C--C dihedral angle. Other geometrical parameters had little effect. It was also found that the nuclear shielding profile is unrelated to the hydrogen bonding energy profile. The dependence of the OH chemical shift on the pi density on the oxygen atom was determined as ca 40 ppm/pi electron. This factor is similar to that for NH but four times the value for sp(2) hybridized carbon atoms. The introduction of these effects into the CHARGE programme allowed the calculation of the (1)H chemical shifts of the compounds studied. The CHARGE calculations were compared with those from the ACD database and from GIAO calculations. The CHARGE calculations were more accurate than other calculations both when all the shifts were considered and also when the OH shifts were excluded. The calculations from the ACD and GIAO approaches were reasonable when the OH shifts were excluded but not as good when all the shifts were considered. The poor treatment of the OH shifts in the GIAO calculations is very likely due to the lack of explicit solvent effects in these calculations.

Accurate ab initio calculations of O–H⋯O and O–H⋯−O proton chemical shifts: towards elucidation of the nature of the hydrogen bond and prediction of hydrogen bond distances

“…The resulting coefficient of linear regression R 2 demonstrates very poor correlation with minimization of the structures both at the M06-2X/6-31+G(d) and the B3LYP/6-31+G(d) level of theory. 85 Taking into consideration that the accuracy in estimating the slope in eqn (4) and (5) is 2 to 3%, it is evident that hydrogen bond distances with an accuracy of ±0.02 Å to ±0.03 Å can be estimated for (O)H⋯O and (O)H⋯ − O hydrogen bonds in the range of 1.24 Å to 1.85 Å. Furthermore, the anions 36 to 43 seem to present a separate class of compounds with an intercept of linear regression nearly parallel to that of compounds 5 to 35.…”

1D and 2D NMR of nanocellulose in aqueous colloidal suspensions

“…Another two peaks at ı 4.04 and 3.30 coupled with C at ı 62.6 were also thought to be from C6. The shifting of these two C6 and H6 peaks of CNF 1.5 was thought to be due to the strong hydrogen bond between the non-oxidized C6 OH and the C6 carbonyl O of partially oxidized CNF 1.5 (Abraham & Mobli, 2007;Ajami et al, 2011;Grzesiek, Cordier, Jaravine, & Barfield, 2004;Moulthrop, Swatloski, Moyna, & Rogers, 2005). Other peaks present in CNF1.5 were assigned as follows: H3 at ı 3.75, C3 at ı 75.7, H4 at ı 3.66, C4 at ı 80.1, H5 at ı 3.52; C5 at ı 74.2.…”

Growth, structural and optical characterization of L-histidine 4-nitrophenolate (LHPNP) single crystals for NLO applications

“…The ortho hydrogen present in the nitro phenol can see each other as aligned (Parallel) or opposed (antiparallel) and come to resonance twice. Hence the ortho hydrogens appear as distorted doublet and give their signals at 8.05 and 6.82 ppm [11,12]. The multiple splitting observed in these signals is due to the presence of amine groups in the histidine molecules.…”