Solid-state NMR investigations of the immobilization of a BF4(-) salt of a palladium(II) complex on silica.

The structure of the silica supported palladium(II) complex [Pd(dppp)(S2C-NEt2)]BF4 (abbreviated as [Pd(dppp)(dtc)]BF4, where dppp is Ph2P(CH2)3PPh2) and interactions between the [Pd(dppp)(dtc)]+ cation, the BF4(-) anion, and the silica surface are studied using solid-state NMR spectroscopy. The unsupported, crystalline form of [Pd(dppp)(dtc)]BF4 is also investigated, both by X-ray diffraction and NMR. The structures of the cation and anion are found to be essentially the same in both unsupported and supported complex. The [Pd(dppp)(dtc)]BF4 loading has been determined by quantitative measurements of 11B, 19F, and 31P intensities, whereas the arrangement of anions and cations on the surface of silica has been established by two-dimensional heteronuclear correlation experiments involving 1H, 11B, 13C, 19F, 29Si, and 31P nuclei. At low coverages, the [Pd(dppp)(dtc)]+ cations are located near the BF4(-) anions, which in turn are immobilized directly on the surface near the Q4 sites. At higher loadings, which in this study corresponded to 0.06-0.15 mmol/g, the complexes stack on top of each other, despite the fact that the directly adsorbed molecules take up less than 10% of the silica surface. The relevance of these findings to heterogeneous catalysis is discussed.

Solid-state NMR spectroscopic study of coordination compounds of XeF(2) with metal cations and the crystal structure of [Ba(XeF(2))(5)][AsF(6)](2).

The coordination compounds [Mg(XeF(2))(2)][AsF(6)](2), [Mg(XeF(2))(4)][AsF(6)](2), [Ca(XeF(2))(2.5)][AsF(6)](2), [Ba(XeF(2))(3)][AsF(6)](2), and [Ba(XeF(2))(5)][AsF(60](2) were characterized by solid-state (19)F and (129)Xe magic-angle spinning NMR spectroscopy. The (19)F and (129)Xe NMR data of [Mg(XeF(2))(2)][AsF(6)](2), [Mg(XeF(2)(4)][AsF(6)](2), and [Ca(XeF(2))(2.5)][AsF(6)](2) were correlated with the previously determined crystal structures. The isotropic (19)F chemical shifts and (1)J((129)Xe-(19)F) coupling constants were used to distinguish the terminal and bridging coordination modes of XeF(2). Chemical-shift and coupling-constant calculations for [Mg(XeF(2))(4)][AsF(6)](2) confirmed the assignment of terminal and bridging chemical-shift and coupling-constant ranges. The NMR spectroscopic data of [Ba(XeF(2))(3)][AsF(6)](2) and [Ba(XeF(2))(5)][AsF(6)](2) indicate the absence of any terminal XeF(2) ligands, which was verified for [Ba(XeF(2))(5)][AsF(6)](2) by its X-ray crystal structure. The adduct [Ba(XeF(2))(5)][AsF(6)](2) crystallizes in the space group Fmmm, with a = 11.6604(14) Angstrom, b = 13.658(2) Angstrom, c = 13.7802(17) Angstrom, V = 2194.5(5) Angstrom(3) at -73 degrees C, Z = 4, and R = 0.0350 and contains two crystallographically independent bridging XeF(2) molecules and one nonligating XeF(2) molecule. The AsF(6-) anions in [Mg(XeF(2))(4)][AsF(6)](2), [Ca(XeF(2))(2.5)][AsF(6)](2), [Ba(XeF(2))(3)][AsF(6)](2), and [Ba(XeF(2))(5)][AsF(6)](2) were shown to be fluxional with the fluorines-on-arsenic being equivalent on the NMR time scale, emulating perfectly octahedral anion symmetry.

Observation of satellite signals due to scalar coupling to spin-1/2 isotopes in solid-state nuclear magnetic resonance spectroscopy.

A method is introduced to select the signal from a spin-1/2 nucleus I specifically bound to another spin-1/2 nucleus S for solid-state magic angle spinning nuclear magnetic resonance (NMR) spectroscopy via correlation through the heteronuclear J coupling. This experiment is analogous to the bilinear rotation decoupling (BIRD) sequence in liquid-state NMR spectroscopy which selects for signals from 1H directly bound to 13C. The spin dynamics of this modified BIRD experiment is described using the product-operator formalism, where experimental considerations such as rotor synchronization and the effect of large chemical shielding anisotropies on I and S are discussed. Two experiments are proposed that accommodate large chemical shielding anisotropies on S: (1) by stepping the inversion pulse frequency through the entire S spectral range or (2) by adiabatically inverting the S spins. Both these experiments are shown to successfully select the signal of 19F bound to 129Xe in XeF+ salts, removing the contributions from isotopomers containing non-spin-1/2 Xe isotopes. The feasibility in obtaining isotope-selective 19F spectra of inorganic fluoride compounds is discussed, and further modifications are proposed to expand the application to other chemical systems.

Are the NF4+ cations in NF4BF4 really nontetrahedral?

The crystal structure of NF4BF4 has been reexamined. The low-temperature X-ray structure and solid-state 19F MAS NMR spectra are in agreement with the conclusions reached from the vibrational spectra, that solid NF4+ salts contain only tetrahedral NF4+ cations. The alleged observation of two kind of nontetrahedral NF4+ cations in several previous crystal structures is attributed to incorrectly solved structures and, possibly, problems caused by disorder or twinning. It is further evidence for the dangers of over-reliance on crystal structures. Flawed crystal structures can give rise to either bad or unwarranted theory.

Rapid assignment of solution 31P NMR spectra of large proteins by solid-state spectroscopy.

The application of the (31)P NMR spectroscopy to large proteins or protein complexes in solution is hampered by a relatively low intrinsic sensitivity coupled with large line widths. Therefore, the assignment of the phosphorus signals by two-dimensional NMR methods in solution is often extremely time consuming. In contrast, the quality of solid-state NMR spectra is not dependent on the molecular mass and the solubility of the protein. For the complex of Ras with the GTP-analogue GppCH(2)p we show solid-state (31)P NMR methods to be more sensitive by almost one order of magnitude than liquid-state NMR. Thus, solid-state NMR seems to be the method of choice for obtaining the resonance assignment of the phosphorus signals of protein complexes in solution. Experiments on Ras.GDP complexes show that the microcrystalline sample can be substituted by a precipitate of the sample and that unexpectedly the two structural states observed earlier in solution are present in crystals as well.

Solid-state 31P NMR spectroscopy of microcrystals of the Ras protein and its effector loop mutants: comparison between crystalline and solution state.

Cycling between a GTP bound "on" state and a GDP bound "off" state, guanine nucleotide-binding (GNB) proteins act as molecular switches. The switching process and the interaction with effectors, GTPase-activating proteins, and guanosine nucleotide-exchange factors is accompanied by pronounced conformational changes of the switch regions of the GNB proteins. The aim of the present contribution is to correlate conformational changes observed by liquid-state NMR with solid-state (31)P NMR data and with the results of X-ray crystallography. Crystalline wild-type Ras complexed with GTP analogs such as GppCH(2)p and GppNHp could be prepared. At low temperatures, two different signals were found for the gamma-phosphate group of GppNHp bound to wild-type Ras. This behavior indicates the existence of two different conformations of the molecule in the crystalline state as it is found in solution but not by X-ray crystallography. In contrast to the GppNHp complex, the two separate gamma-phosphate signals could not be observed for GppCH(2)p bound to wild-type Ras. However, an increasing linewidth at low temperature indicates the presence of an exchange process. The results obtained for the wild-type protein are compared with the behavior of GppNHp complexes of the effector loop mutants Ras(T35S) and Ras(T35A). These mutants prefer a conformation similar to the GDP bound "off" state.

Phosphorylated amino acids: model compounds for solid-state 31P NMR spectroscopic studies of proteins.

Solid-state 31P NMR spectroscopy was applied to measure the isotropic chemical shifts, chemical shift anisotropies and asymmetry parameters of three phosphorylated amino acids, O-phospho-L-serine, O-phospho-L-threonine and O-phospho-L-tyrosine. The cross-polarization buildup rates and longitudinal relaxation times of 31P and 1H were-determined and compared with the values measured for a triphosphate (GppCH2p) bound to a crystalline protein (Ras). It is shown that the phosphorylated amino acids are well-suited model compounds, e.g. for the optimization of experiments on crystalline proteins. Two-dimensional exchange experiments on O-phospho-L-tyrosine indicate the existence of an exchange between the two different conformations of the molecule.