Metal Hydrides Form Halogen Bonds: Measurement of Energetics of Binding

The formation of halogen bonds from iodopentafluorobenzene and 1-iodoperfluorohexane to a series of bis(η5-cyclopentadienyl)metal hydrides (Cp2TaH3, 1; Cp2MH2, M = Mo, 2, M = W, 3; Cp2ReH, 4; Cp2Ta(H)CO, 5; Cp = η5-cyclopentadienyl) is demonstrated by 1H NMR spectroscopy. Interaction enthalpies and entropies for complex 1 with C6F5I and C6F13I are reported (ΔH° = −10.9 ± 0.4 and −11.8 ± 0.3 kJ/mol; ΔS° = −38 ± 2 and −34 ± 2 J/(mol·K), respectively) and found to be stronger than those for 1 with the hydrogen-bond donor indole (ΔH° = −7.3 ± 0.1 kJ/mol, ΔS° = −24 ± 1 J/(mol·K)). For the more reactive complexes 2–5, measurements are limited to determination of their low-temperature (212 K) association constants with C6F5I as 2.9 ± 0.2, 2.5 ± 0.1, <1.5, and 12.5 ± 0.3 M–1, respectively.

NMR data for Cp 2 Ta(H) 13 CO S2

Standard method for titrations S2
Additional methodology adopted for measuring highly reactive host-guest systems S2

Composition of stock solutions S3
Fitting of titration curves S4

Titration curves and van't Hoff plots S5
Stack plot of 1 H NMR spectra for the titration of Cp 2 TaH 3 vs Indole at 212K S10 Interaction of Cp 2 Ta(H) 13 CO with AlMe 3 S11 Calculations S12 Cp 2 WH 2 and Cp 2 TaH 3 S13 Cp 2 Ta(H)CO S16 Summary of calculated energies S18 Comparison of calculated metal-hydride distances and angles upon binding C 6 F 5 I S19 Electrostatic potential plots for Cp 2 TaH 3 , Cp 2 WH 2 and Cp 2 Ta(H)CO S20

Reference 24 of text S21
Cartesian coordinates and total energies for computed stationary points S21 Introduction of C 6 F 5 I into a solution of Cp 2 Ta(H)CO at RT resulted in immediate decomposition with no hydride resonances apparent by 1 H NMR spectroscopy and therefore the following procedure was used to prepare samples for low temperature ( Chem. 1985, 284, 229. in all tubes totalled 600 μL. The mass of the sealed tube was then recorded and removed from the glove box. The tube was placed into a Schlenk-line adapter possessing both an inlet for inert gas and an open top to allow removal of the Young's cap. After cooling the samples to -78 °C the guest solution was introduced by removal of the Young's cap under a strong flow of argon and injection by microsyringe. In all cases the samples were held below -50 °C and the NMR experiments conducted within an hour of sample preparation. During the titration the tubes were inverted before insertion into the NMR machine to ensure mixing and checked for solubility of both host and guest components. After the measurement the total mass of the tube was recorded and from this subtracted the mass of the tube prior to guest addition therefore allowing the mass of guest added to be determined retrospectively. With these precautions only small quantities (1-2%) of decomposition were observed for 2, 3 and 5 in the NMR spectra by the presence of another cyclopentadienyl resonance. Titration samples of 4 were prepared analogously; however a greater degree of iodination of the metal hydride was observed (~10%) and consequently this was accounted for in the fitting analysis.

Fitting of titration curves
The system involves the formation of the 1:1 adduct between the guest and the respective metal hydride, which takes place by R−I···H−M halogen bonding or R−H···H−M hydrogen bonding. In the scheme X-R is either a halogen or a hydrogen bond donor.
There are two parameters to be fitted: the equilibrium constant K and the downfield shift from the signal of free metal hydride for the coordinated hydride of the adduct, Δδ H . The two parameters can be fitted for the whole range of temperatures without any restraints by using a Microsoft Excel macro programmed by Professor Christopher Hunter (University of Sheffield). ΔH 0 and ΔS 0 were determined from the van't Hoff plots of the equilibrium constants by linear regression.     ratio of molar concentrations of C 6 F 5 I and Cp 2 WH 2 (concentration of Cp 2 WH 2 17 mmol dm -3 ).   Organometallics 1988, 7, 1.

Calculations
Calculations were performed by use of the Gaussian 09 series of programs 4 at the DFT level using the BHandHLYP functional. This functional has been shown to provide accurate energetics for noncovalently bonded systems. 5 The SDD effective core potential and associated basis sets was used for Ta, W and I. The 6-31G** basis set was used for C, H, F and O and diffuse functions were added to O and the hydrides. All the geometries were optimized without restraint and used an ultrafine grid together with tight optimization criteria. Energies and geometries were optimised for the separate components and for the adducts. Electrostatic potentials were plotted for the metallocene hydrides 1, 3 and 5 using the computational package GaussView 5. Interaction enthalpies were corrected for basis set superposition error (BSSE) using the counterpoise method. 6 In addition, the interaction energies were first calculated at the 6-31G* level with diffuse functions added to H and O and the differences between this level of theory and that with the additional polarisation function were ≤ 0.2 kJ mol -1 .

Cp 2 WH 2 and Cp 2 TaH 3
For the complexes Cp 2 WH 2 and Cp 2 TaH 3 two possible modes of binding with C 6 F 5 I were explored; a bifurcated interaction bridging two hydrides and side-on interaction with a single hydride (Figure 1).
In order to investigate these modes, three starting geometries for Cp 2 WH 2 ⋅IC 6 F 5 adducts were investigated (a) an approach of C 6 F 5 I along the bisector of the W-H bonds and (b) W-H⋅⋅⋅I configurations with angles in the metal hydride plane of ~170° and ~190°. The initial orientation of the plane of the arene was perpendicular to the WH 2 plane. Optimisations gave minima for Cp 2 WH 2 ⋅IC 6 F 5 adducts both for a bifurcated ( Figure S14a) and a side-on ( Figure S14b) mode with the bifurcated mode found to be more stable by 1.3 kJ mol -1 . The counterpoise corrected binding energies for the bifurcated adduct are calculated to be -13.4 kJ mol -1 and for the side on adduct -12.1 kJ mol -1 . The optimised geometries are shown in Figure S15 and Figure S16. For the bifurcated interaction, optimizations were also started from co-planar and perpendicular orientations of the arene relative to the WH 2 plane and converged to minima without rotation of the arene. The perpendicular geometry was calculated to be more stable by 0.3 kJ mol -1 .
The starting geometries for Cp 2 TaH 3 ⋅IC 6 F 5 adducts were from a linear Ta-H central -I configuration and from Ta-H lateral -I configurations with angles in the metal hydride plane of ~170° and ~190°. The initial orientation of the plane of the arene was perpendicular to the TaH 3 plane. Optimisations of these Cp 2 TaH 3 ⋅IC 6 F 5 adducts led to the location of a minimum for a bifurcated interaction ( Figure S14c) with a counterpoise-corrected binding energy of -14.3 kJ mol -1 . Attempts to locate a minimum of a side-on interaction (Figure S14d) using tight convergence criteria were unsuccessful. The optimised geometries are shown in Figure S17. Figure S14: Investigated binding modes of the halogen bond donor C 6 F 5 I with Cp 2 TaH 3 and Cp 2 WH 2 where only the metal hydride plane is shown. S14 Figure S15:

Cp 2 Ta(H)CO
The geometries for the calculations on Cp 2 Ta(H)CO⋅IC 6 F 5 adducts were started from linear Ta-H⋅⋅⋅I and C≡O⋅⋅⋅I configurations. The initial orientation of the plane of the arene was perpendicular to the Ta(H)CO plane. Optimisation led to the location of two minima of interaction: (a) iodine bound to hydride and (b) iodine bound to oxygen ( Figure S18). The interaction energy with the hydride site was greater than that of the oxygen site by 6.0 kJ mol -1 . The counterpoise-corrected energy calculated for binding the hydride was -14.3 kJ mol -1 compared with the energy calculated for binding the carbonyl oxygen of -8.3 kJ mol -1 . The optimised geometries are shown in Figure S19 and Figure S20.