Distinguishing weak hydrogen bonding with NMR Mikhail Kibalchenko, TCM Group, Cambridge University Results published: M. Kibalchenko, D. Lee, L. Shao, M. C. Payne, J. J. Titman, J. R. Yates, (2010). Distinguishing hydrogen bonding networks in alpha-d-galactose using NMR experiments and first principles calculations. Chem. Phys. Lett. In press doi:10.1016/j.cplett.2010.08.077
Aim: distinguishing hydrogen bonds hydrogen bonding is the most important of all directional intermolecular interactions determines: molecular conformations molecular aggregations function of many chemical systems from inorganic to biological 1 Steiner, T. Angew. Chem., Int. Ed. 2002, 41, 48-76
Aim: distinguishing hydrogen bonds Can I use NMR to distinguish hydrogen bonding networks? How to do it? - which nuclei do we need to measure?
Introduction: nuclear magnetic resonance Energy m=-1/2 E 0 E = γ B 0 m=1/2 B 0 Magnetic field
Introduction: nuclear magnetic resonance B induced =-σb 0 Chemical shielding tensor δ iso Isotropic shift Δ cs Chemical shift anisotropy
Introduction: nuclear magnetic resonance Principle components of chemical shielding tensor B induced =-σb 0 Chemical shielding tensor δ iso Isotropic shift Δ cs Chemical shift anisotropy
Introduction: nuclear magnetic resonance B induced =-σb 0 Chemical shielding tensor δ iso Isotropic shift Δ cs Chemical shift anisotropy
System of interest: alpha-d-galactose three X-ray derived molecular crystal structures presenting three differing hydrogen bonding networks after hydrogen atom geometry optimisation: two differing hydrogen bonding networks - which is the true one? "-,(/ "&&-)1!")$%*' %& "-.&/ "&.-#1!,$%#' %( %# 2* 2# 2" "-,#/ "&*-.1!""$%(' 2( 20 2&!"#$%&' "-.)/ "&*-#1 %" $4'!")$%*' "-+,/ ".(-*1 %( 2* 2# "-+./ %# "&*-)1!""$%(' 2" 2( 20 2& "-.,/ "0&-,1 %#!"#$%&' $3' %*!+$%"' #-"(/ "."-,1 Sheldrick >?<@A=9BC %*!,$%#' "-&,/ ".*-#1!+$%"' "-&&/ ".*-(1 %& Kouwijzer 56789:;<=
System of interest: alpha-d-galactose monosaccharide, usually bound to a non-carbohydrate molecule acts as an intermediate in numerous biological processes occurs naturally in foods popular food additive used in pharmaceutical R&D as the core for new drugs
System of interest: alpha-d-galactose three X-ray derived molecular crystal structures presenting three differing hydrogen bonding networks after hydrogen atom geometry optimisation: two differing hydrogen bonding networks - which is the true one? O6 "-,(/ "&&-)1!")$%*' %& "-.&/ "&.-#1!,$%#' %( %# 2* 2# 2" "-,#/ "&*-.1!""$%(' 2( 20 2&!"#$%&' "-.)/ "&*-#1 %" O1 $4'!")$%*' "-+,/ ".(-*1 %( 2* 2# "-+./ %# "&*-)1!""$%(' 2" 2( 20 2& "-.,/ "0&-,1 %#!"#$%&' $3' O2 O3 %*!+$%"' #-"(/ "."-,1 Sheldrick >?<@A=9BC O3 %*!,$%#' "-&,/ ".*-#1!+$%"' "-&&/ ".*-(1 O6 %& Kouwijzer 56789:;<=
Available experimental data - Carbon atom NMR Mean Absolute Error (ppm) Max Absolute Error (ppm) Kouwijzer Sheldrick δiso cs δiso cs 0.9 1.1 1.4 1.5 2.5 2.5 4.1 2.1 alpha-d-galactose C6H12O6 Mean Absolute Error (ppm) Max Absolute Error (ppm) Kanters δiso cs 0.9 0.8 2.7 1.3 beta-d-fructose C6H12O6
What about Hydrogens and Oxygens? identified hydrogens could show a difference even for low resolution NMR experiment. Prediction: Kouwijzer Sheldrick 10 9 8 7 6 5 δ iso (ppm) 4 3 2 1 010 9 8 7 6 5 δ iso (ppm) 4 3 2 1 0
What about Hydrogens and Oxygens? for oxygen, isotropic shifts differ by 12.8 ppm maximum between equivalent atomic sites however, oxygen isotropic shifts are known to span hundreds of ppm therefore, unclear how accurately experiment could differentiate 12.8 ppm
Recent hydrogen nmr results: =>+56',5?@ 0)12'3456 7859:6';< *.- δiso (ppm) black = experiment alpha-d-galactose C6H12O6
Recent hydrogen nmr results: Kanters 8 7 6 5 4 3 2 1 δiso (ppm) black = experiment beta-d-fructose C6H12O6
Recent hydrogen nmr results: Kouwijzer Sheldrick 8 7 6 5 4 3 2 1 8 7 6 5 4 3 2 1 δiso (ppm) black = experiment alpha-d-galactose C6H12O6
some correlations + A%'(@!"# ;<2!==># - ' %+ * %+ %% %, %% %, )?%'(@ ( ) ) * B2CD;EF6G H9654G;IJ K;36"G1L;8 +&% +&+ O-H --- O bond length (Å)
some correlations ) )!$#!==># * %+ %% %, * %+ B2CD;EF6G H9654G;IJ K;36"G1L;8 ) %&' %&( %&) %&* + +&% +&+ %% %,./01/1/1.1$23415637891!:# O-H --- O bond length (Å)
Conclusion hydrogen bonding network in alpha-d-galactose is established showed that combination of NMR experiment and calculations from first principles is extremely powerful for structure determination - even weak interactions such as hydrogen bonds hydrogens are obviously good candidates for studying hydrogen bonding, but even carbon NMR shows a small preference usefulness of oxygen NMR is unclear, but could potentially be more useful than carbons since oxygen atoms are quadrupolar and experiment could measure electric field gradient interactions results provide justification for the study of larger biological systems
Acknowledgements Jonathan Yates, Mike Payne Jeremy Titman, Daniel Lee, Limin Shao, Nottingham University, UK
Thank you for listening
back up slides