NMR- Bridge across the Styx between the solid and solution
Especially in the area of reactive intermediates it is essential to get information about the involved species, in the solid-state but, even more important, in solution, because structural changes in solution like solvation and aggregation determine the reactivity and selec- tivity and hence the product range in organic syntheses and the materials profile. On this issue we closely collaborate with Dr. Michael John heading the NMR-department of the Inorganic Chem- istry at Göttingen and Dr. Adam Lange, heading the solid-state NMR-department of the Max-Planck Institut für biophysikalische Chemie at Göttingen.
Besides the well-established issues of solvation and aggregation, valid for any lithiumorganics, the region for an electrophilic attack and the stereochemical course of the reaction needs to be identified and monitored (Fig. 1).
However, X-ray crystallography provides little insight into the ther- modynamics of aggregation and solvation[1].This is because the crystal structure is commonly believed to represent the least so- luble derivative in the pot and not necessarily the most abundant, let alone the most reactive species. Moreover, the least populated species might represent the eye of the needle in the equilibrium the whole reaction goes through anyway on the course towards the overall product.
Recently we tied in with this issue by a study on 2-thienyllithium coordinated to various donor bases and at different crystallo- graphically assured aggregation states. We synthesised, crystall- ized and structurally characterized various 2-thienyllithium aggre- gates and studied their behaviour in solution by 1/2-D hetero- nuclear NMR experiments to start from firm ground and explore their constitutionand behaviour in solution[2].
The determined structures of
[(Et2O)Li(C4H3S)]4 (1),
[(THF)2Li(C4H3S)]2 (2),
[(DME)Li(C4H3S)]2 (3),
[(TMEDA)Li(C4H3S)]2 (4),
[(PMDETA)Li(C4H3S)] (5)
were solved in non-donating toluene and investigated by Diffusion Ordered NMR spectroscopy (DOSY) as well as Heteronuclear Over- hauser Enhancement NMR spectroscopy (HOESY, Fig. 2)[3]. The distance relation of NOEs with a factor r-6 was employed to gain further insight in the aggregation degree of 1-5 in solution. Com- parison of the slope provided by the linear region of the build-up curves and of the ∑r-6 calculated distances from the crystal struc- tures offers a handle to judge on structure retention versus con- version in solution. The structures of 3-5 are maintained in toluene solution. The data of 2, however, indicates a partly dissociation or a rapid exchange between the vertices of a tetrameric core and free THF molecules. Auxiliary EXSY investigations showed that the nitro- gen-donor base containing compounds 4 and 5 exchange to the signals of a small amount of hydrolysed non-lithiated thiophene. Additionally a slow deuteration of the thienyl-H5-position of 4 and 5 is observed, concomitant with the protonation of the solvent CD3 group. This exchange is explained by temporary and intermediate lithiation of toluene molecules.
Until now the chemistry of mixed lithiumorganics still is in its infancy. Little is known about the species that are present in solution, and only a few studies have been reported so far. Enhanced reactivity of the mixed lithiumorganics compared to the parent material is sometimes reported. This fuelled the idea to study mixed lithiumorganics in more detail and we set out to synthesize and characterize the starting material unambiguously from X-ray diffraction[4].
Interestingly, the unit cell of [tBuLi]4·4[Me2NC6H4Li]4, really contains two different lithium organic molecules. One is the [Me2NC6H4Li]4 tetramer, the second is a [tBuLi]4 molecule. Such separated lithiumorganic aggregates in one-and-the-same single crystal are quite unique. In accordance with the solid state structure the 7Li NMR spectrum should show two signals representing the two different tetrameric homoleptic species [Me2NC6H4Li]4 and [tBuLi]4, respectively. However, the 7Li NMR spectrum in toluene-d8 is unexpectedly complicated and shows five relatively sharp dis- tinguishable signals over a range of nearly 2.5 ppm among further unidentified broad peaks (Fig. 3).
To gain further reasoning of the 7Li NMR spectrum we performed 7Li diffusion ordered spectroscopy (DOSY). The monotonically de- creasing diffusion coefficient D (vertical axis) identifies the five peaks A-E as the five conceivable tetramers
[Me2NC6H4Li]4 (A),
[(Me2NC6H4Li)3(tBuLi)] (B),
[(Me2NC6H4Li)2(tBuLi)2] (C),
[(Me2NC6H4Li)(tBuLi)3] (D),
[tBuLi]4 (E).
Therefore, each of the narrow peaks in the 7Li NMR spectrum is due to a consecutive substitution of one ortho-lithiumanilide by one tBuLi moiety in the tetrameric aggregate (Fig. 4).
______________________________
[1] D. B. Collum Acc. Chem. Res. 1992, 25, 448.
[2] M.Granitzka, A.-C. Pöppler, E. K. Schwarze, D. Stern, T. Schulz, M. John, R. Herbst-Irmer, S. K. Pandey, D. Stalke J. Am. Chem. Soc. 2012, 134, 1344.
[3] F. T. Edelmann, F. Knösel, F. Pauer, D. Stalke, W. Bauer J. Organomet. Chem. 1992, 438, 1.
[4] A.-C. Pöppler, M. M. Meinholz, H. Faßhuber, A. Lange, M. John, D. Stalke Organometallics 2012, 31, 42.