Page images
PDF
EPUB

LIGAND DYNAMICS IN Mn[N(CN)2]2-PYRAZINE

ne of the major themes of solid state chemistry is the develop

One state with

physical properties for applications such as chemical separation, catalysis, and magnetic devices. Compounds consisting of transition metal ions linked together by polydentate organic ligands are of particular interest because their properties can be tailored by judicious choice of the components. For example, changing the transition metal can alter both the bonding motif and magnetic properties of these systems. The properties of these materials can also be controlled by introducing ancillary π-conjugated ligands such as pyrazine (pyz), 4,4'-bipyridine and 2,2'-bipyridine. These molecular building blocks not only affect the spatial separation of the transition metal cations and the dimensionality of the crystal, but also modulate the superexchange interactions. For instance, of the many Mn[N(CN)2]2 L materials examined to date, only L = pyrazine exhibits long-range magnetic order above 2 K.

The structure of Mn[N(CN)2]2.pyz can be described as an inter-penetrating ReO,-like network with axially elongated Mn2+ octahedral and edges made-up of u-bonded [N(CN)2] anions and

2

neutral pyrazine ligands (Fig. 1) [1]. Upon heating above = 200 K, Rietveld refinements of neutron powder diffraction (NPD) data indicate a marked increase in the Debye-Waller factor for the midnitrogen in the cyanamide ligand and a concomitant appearance of thermal diffuse scattering. Further heating to≈ 400 K results in a phase transition to an unknown structure.

While the structural and magnetic behaviors of these materials have been rather well characterized, very little information has been obtained concerning the interactions that underlie the interesting bonding motifs. Due to its unique sensitivity to hydrogen and the possibility of covering a wide range of timescales, neutron spectroscopy is particularly well suited to probe ligand dynamics that directly reflect the bonding interactions. Quasielastic neutron scattering (QENS) provides information on the geometry and timescale of diffusive motions. For "localized" motions, the geometry is embodied in the elastic incoherent structure factor (EISF); the ratio of the elastic scattering to the total scattered intensity.

For all temperatures studied, a quasielastic signal due to the paramagnetic Mn2+ ions and the dynamics of the cyanamide ligand

[graphic][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]
[blocks in formation]

was observed. However, above≈ 425 K, the quasielastic signal in the pyrazine compound is significantly larger. Figure 2 shows the Q-dependence of the EISF for the protonated compound and the same data after subtracting the measured quasielastic scattering from a deuterated material. These data are very well described by a twofold proton jump model where the only variable parameter, the jump distance, is found to be 4.17(1) Å, consistent with the analysis of the NPD data which gives the D-D distance across the pyrazine ring as ≈ 4.16(2) Å. Further, the width of the quasielastic peak is constant over the entire Q-range, as expected for a simple, localized jump motion (the correlation time is≈ 70 ps at 425 K). Thus, the pyrazine ligands must be performing л-jumps about the axis defined by the coordinating nitrogens.

2

The phonon density of states for Mn[N(CN)2]2.pyz at selected temperatures covering all structural phases is shown in Fig. 3. In addition to the general broadening and softening of the spectral features with temperature, the intense peak assigned to the libration of the pyrazine ring at≈ 11.2 meV is strongly attenuated and, assuming a two-fold cosine potential, suggests an activation energy

[blocks in formation]

of≈ 1.2 eV. Because this activation energy is much larger than k,T at the temperature where one observes fully dynamic pyrazines (425 K) and the quasielastic scattering appears abruptly at the structural transition, the transition must result in a greatly reduced rotational barrier. Unfortunately, the activation energy for this process was not measurable due the proximity of the material's decomposition temperature.

Vibrational spectra at higher energy transfers were recorded on the FANS spectrometer (Fig. 4). Our calculations agree well with observation and show that the spectrum is dominated by the normal modes of the hydrogen containing pyrazine. Current investigations are aimed at understanding how the transition metal affects the dynamics of the ligands.

REFERENCES

[1] J. L. Manson, Q.-Z. Huang, J. W. Lynn, H.-J. Koo, M.-H. Whangbo, R. Bateman, T. Otsuka, N. Wada, D. N. Argyriou, and J. S. Miller. Submitted to Journal of the American Chemical Society.

[blocks in formation]

PROBING DISORDER IN CONFINED METHYL IODIDE

The presence of structural disorder in many materials can have a dramatic effect on macroscopic properties. Typically, structural disorder is described in terms of molecular spatial distributions determined from diffraction measurements. A complementary view is to quantify disorder by determining the distribution of potential energies experienced by the molecules. For a highly ordered system the distribution is sharp while it is broad for a disordered system. To explore this approach, high-resolution neutron inelastic scattering has been used to examine the low-energy rotational dynamics of simple symmetric top molecules confined to extremely tiny pores. In effect, the porous substrate furnishes a static impurity distributed throughout the molecular solid that can be thought of as disorder quenched into the molecular matrix. The modification to the rotational spectrum of the confined molecules measured using neutron scattering can then be related to the distribution of potential energies. This measurement technique, known as rotational tunneling spectroscopy [1], is extremely sensitive to the environment experienced by the molecules. By modeling the observed spectra one can quantitatively extract the potential energy distribution.

Rotational tunneling spectroscopy is rooted firmly in quantum mechanics. In a simple picture the potential barrier to reorientation

that, to a good approximation, has three minima determines the motion of a methyl group (a pyramid with a base of three hydrogen atoms having a carbon atom at its apex seen in Fig. 1). In the limit of a small barrier to reorientation, the methyl group can undergo nearly free rotation about the C-I axis. In the limit of a very high barrier, the methyl group can oscillate (librate) within minima. When the temperature is high enough to provide sufficient kinetic energy, the molecule can reorient stochastically by jumping over the barrier, a process known as rotational diffusion. Some molecules such as methyl iodide (CHI) possess finite barriers in which the rules of quantum mechanics allow the methyl group to reorient via tunneling through the barrier. A schematic illustration of these processes in terms of the potential energy is shown in Fig. 1.

Measurements of the rotational tunneling of methyl iodide (CH,I) confined to a porous glass with a very narrow pore size distribution (diam≈ 58 Å) were performed using the NIST backscattering spectrometer very well suited to such measurements due to its excellent energy resolution: SE (FWHM) < 1 μeV. Measurements were carried out for the bulk solid, partially filled pores, and completely filled pores. Spectra taken at 5 K are shown in Fig. 2.

[graphic][merged small][subsumed][subsumed][subsumed][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][ocr errors][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]
[blocks in formation]

The bulk CH,I spectrum shows two sharp side peaks whose positions are related to the frequency at which the methyl group (CH1) tunnels between three equivalent orientations. The location of these peaks is directly related to the potential barrier height hindering reorientation. These peaks at ± 2.5 μeV correspond to a potential barrier of 42 meV. When the pores are partially filled (50%) a broad set of peaks appear at ± 4 μeV. In addition to these broad peaks, a broad featureless scattering intensity appears underneath the welldefined peaks over the entire dynamic range. When the pores are completely filled the broad peaks at ± 4 μeV increase in intensity and a new set of peaks at ± 2.5 μeV appear. We interpret these different peaks as due to the presence of varying amounts of order in the molecular structure of the confined CH,I. The narrow peaks that occur in the bulk and the full-pore spectra point to the similarity of the potentials experienced by the molecules as expected when there is structural order. The broad peaks at ± 4 μeV correspond to methyl groups under the influence of a distribution of potential barriers. The very broad scattering feature underlying both the full-pore and partially-filled samples is attributed to very strongly disordered methyl groups. Based on the filling dependence of the two broad scattering components, these are attributed to molecules near the

glass surface while the narrow peaks at ± 2.5 μeV are due to molecules located near the center of the pore.

To quantify the disorder, we performed numerical calculations of the effects of a distribution of potential barriers on the tunneling lineshape. In Fig. 3 we plot the variation of the potential barrier as a function of tunneling energy. For a broad but symmetric distribution of barrier heights, the tunneling lineshape is clearly asymmetric. Finally, using a relationship between the tunneling energy and barrier height we may extract the probability density for a particular barrier height, P(V3). The result for the full pore spectra corrected for instrumental resolution is shown in Fig. 4.

Thus neutron inelastic scattering measurements of the rotational tunneling spectrum offer a means of quantifying the disorder of the energy landscape in this system of molecules in a confined geometry. A further challenge is to correlate the energy and structural descriptions of disorder.

REFERENCES

[1] W. Press, Single-Particle Rotations in Molecular Crystals, Springer-Verlag, Berlin, Heidelberg, New York, (1981).

[merged small][ocr errors]

arbon nanotubes, originally discovered as byproducts of fuller

ene synthesis, are now considered to be the building blocks

of future nanoscale electronic and mechanical devices. It is therefore desirable to have a good understanding of their electronic and mechanical properties and the interrelations between them. In particular, single wall carbon nanotubes (SWNT) provide a system where the electronic properties can be controlled by the structure of the nanotubes and by various deformations of their geometries [1,2]. The physical properties can also be altered by intertube interactions between nanotubes packed in hexagonal lattices, as so-called "nanoropes."

60°

The intertube interactions in nanoropes can be probed by applying external pressure to vary the intertube distance. For fullerenes, such high pressure studies have yielded many interesting results including new compounds such as the pressure-induced polymeric phases of C60. It is, therefore, of interest to inquire if similar covalent-bonding can occur between the nanotubes in a rope. This could have important consequences for nanoscale device applications and composite materials that require strong mechanical properties since nanoropes consisting of inter-linked SWNT will be significantly stronger than nanoropes composed of van der Waals (vdW) packed nanotubes.

We investigated possible new pressure-induced ground state structures for (n,0) nanotube ropes from first-principles total energy

calculations using the pseudopotential method within the generalized gradient approximation (GGA) [1]. For simplicity, we model the nanoropes as a hexagonal lattice of nanotubes with one nanotube per unit cell. The pressure dependence of the lattices of nanotubes was determined by calculating the total energy as a function of nanotube separation (i.e., a and b) while the other parameters, including atom positions, c, and y are optimized. We observe that (7,0) nanotubes become elliptically distorted with applied pressure (i.e., decreasing nanotube-nanotube distance). At a critical pressure, we observe a structural phase transformation from the vdW nanotube lattice (as shown in Fig. 1a) to a new lattice in which the nanotubes are interlinked along the [110] direction, where the strain of the nanotube is largest (Fig. 1b). The covalent bonding between nanotubes is therefore the result of curvature-induced re-hybridization of the carbon orbitals. The same structural transformation was observed for the other (n,0) nanoropes.

To quantitatively study the bonding mechanism, we calculated the total energies of the different phases as a function of the lattice constant (i.e., applied pressure). The result for (7,0) nanotubes is summarized in Fig. 2. The energies of the vdW and the one-dimensional interlinked phases cross each other at about a = 9.0 Å with an energy barrier of only 46 meV/unitcell (552 K). The pressure required to attain this lattice constant is only about 0.3 GPa for the vdW phase, indicating that polymerization of vdW (7,0) nano

C

[graphic][subsumed][subsumed][subsumed][ocr errors][subsumed][subsumed][subsumed][subsumed]

FIGURE 1. Optimized structures of the vdW (7,0) (a), and one-dimensional interlinked (7,0) (b) nanotube lattices. The interlinked structure shown in (b) has lower energy than vdW packed (7,0) nanotubes shown in (a).

« PreviousContinue »