Chemical and structural properties of TAG-MNT at ambient conditions.9 (a) Triaminoguanidinium and 1-methyl-5-nitriminotetrazole molecular ions; the anion is deprotonated (leaving the lowermost ring nitrogen with negative charge) in salt form. (b) Molecular conformities, with hydrogen bonds shown as dotted lines and atoms in adjacent molecules labeled in parenthesis. (c) The triclinic unit cell of TAG-MNT with dotted lines showing hydrogen bonds.
Diamond-anvil cell samples of TAG-MNT in transmitted light, showing a powder loaded with a neon medium (left) and TAG-MNT loaded without a medium (right). Red circles indicate location of rubies used to measure pressure.
X-ray diffraction data at ambient pressure and temperature (black), and pattern predictions based upon the previous crystallographic determination at 200 K (grey) (Ref. 9) and refined (blue) structures for room temperature (293 K). Prominent peaks are labeled with reciprocal lattice planes. Small contributions from the Rhenium gasket are identified with a star. Le Bail fitting improvements not included.
Raman spectra of TAG-MNT at ambient pressure, for pristine material and previously compressed material. The black spectrum was obtained using an uncompressed single crystal; the grey (blue) spectra were obtained from a polycrystalline sample nonhydrostatically compressed to 15.6 ± 0.1 GPa (26 ± 1 GPa) and then decompressed; 488 nm (black) and 457 nm (grey, blue) excitation was used. Previously reported Raman lines9 are black vertical lines near the bottom axis. Spectral features appearing as a result of pressure-induced chemistry are marked with a star.
Raman spectra (top), IR spectra (middle), and lineshifts with pressure (bottom). In the bottom panel, vertical red (black) dashes below 0 GPa (horizontal black line) are ambient Raman (IR) vibrational frequencies given by Klapötke et al.;9 colored open (filled) circles represent Raman data from nonhydrostatic (hydrostatic) sample loadings; colored solid lines are second- to fifth-order polynomial fits to Raman lineshift data, pinned (when possible) at the ambient Raman frequency; grey hatched region represents the Raman signal of the diamond anvils; black triangles connected by dashed black lines are IR lineshift data; horizontal dashed grey lines represent the stiffening regime; horizontal dash-dotted grey line indicates the end of the high pressure structural transition. Raman peaks could be tracked with pressure reliably above 300 cm−1; data shown here were collected using 488 or 457 nm excitation.
X-ray diffraction data under pressure. The curves in black (blue) are diffraction patterns on compression (decompression from 13.9 GPa). The grey curves are the predicted patterns based on a unit cell refinement to observed peak locations; le Bail fit improvements not included. Black dashed line is the location of the first nonsample peak (Rhenium from gasket); to its right are additional non-sample peaks from Rhenium, and Neon, which appears as a broad hump below 5 GPa (liquid) and a sharp line above 5 GPa (solid). The x-ray wavelength was 0.3344 A.
Triclinic unit cell parameters for TAG-MNT under pressure. Lines are fits to the data for the low pressure (solid) and high pressure (dashed) phases. Black (grey) circles are data on compression (decompression).
Compressibility data on TAG-MNT, based on unit cell refinement using x-ray diffraction data. The black (grey) filled points were obtained upon compression (decompression). The starred points indicate data collected for phase II. The lines indicate third- and fourth-order Birch-Murnaghan fits to phase I data. Third- and fourth-order fits to 6 GPa are indistinguishable below 6 GPa, such that a third-order fit (K 0 = 14.6, K / = 4.83) is a sufficient model; on extrapolation to higher pressure, the fourth-order fit (K 0 = 16.2, K / = 2.33, K // = 0.98) provides a reasonable prediction of the high-pressure volumes in phase II, however the highest pressure datum for phase I is anomalously stiff and neither low pressure fit can predict it. For fits including the 9 GPa datum, a third-order fit shows systematic deviations from the data near 1 and 5 GPa; a fourth-order fit is necessary to represent all the data in this range. The arrows indicate the path taken by TAG-MNT upon compression (dashed black curve, based on the fourth-order fit to 9 GPa and an assumed volume collapse near 13 GPa), and upon decompression from 14 GPa (solid black curve, based on the fourth-order fit to 6 GPa).
Detail of interacting vibrational modes near 1500 cm−1 and description based on Fermi resonance theory. Both Raman and IR data are presented, shown as circles and triangles, respectively; open (filled) symbols indicate nonhydrostatic (hydrostatic) loading. (a) Raman Spectra (solid black lines) shown with fitted peaks (solid grey lines). The most intense mode (ν NO2, 1507 cm−1 initial) has a frequency ν− and intensity A−, and the weaker, soft mode (ν C=N, 1553 cm−1 initial) has frequency ν+ and intensity A+. (b) Raman and IR frequencies, with solid (dashed) lines showing the observed (bare) frequencies in the Fermi resonance analysis. (c) Peak amplitude ratio for Raman and IR data with solid line showing the model for the Raman data. (d) Observed peak separations for Raman and IR data, with modeled actual and bare separations (solid and dashed lines, respectively). Hydrostatic loading results at high pressure were used in modeling since shear stress had a significant effect on line frequencies.
Detail of selected lines in the Raman spectrum that show decreased pressure sensitivity in the stiffening regime followed by increased sensitivity at higher pressure associated with the transformation to the high-pressure phase.
Comparison of Raman measurements taken on compression (black) with those obtained on decompression. Data from samples decompressed from 25 GPa and 60 GPa are blue and red, respectively; blue curves correspond to samples loaded hydrostatically (upper) and nonhydrostatically (lower). Excitation wavelengths from top to bottom are 488, 488, 632, 457, and 488 nm.
Raman measurements of photochemical transformation of TAG-MNT compressed to 12.2 GPa, with excitation from a 457 nm laser. Spectral intensity is normalized to laser power and shifted arbitrarily in the y-axis. Data correspond to 5 min of accumulation, beginning either with laser shut-on (solid) or ∼15 min after laser shut-on (dashed). For each power level a fresh area of sample was studied. The transformation was accompanied by the appearance of discolored spots in the sample, which grew in size with increasing power and time. Small variations (several cm−1) in Raman line positions (i.e., of the strong line at 1565 cm−1) were not systematically related to laser power, and are attributed to pressure gradients in the nonhydrostatic sample. Most Raman measurements reported in this study were made using excitation of 488 nm, for which photochemical effects were less pronounced.
Structural and thermodynamic properties of TAG-MNT at ambient pressure; ρ is the density; K T is the isothermal bulk modulus (obtained from fitting equation of state data to 6 GPa with a third-order Birch-Murnaghan equation); T is the temperature; α V is the volumetric thermal expansion estimated from the volume difference between the 200 and 293 K data. Uncertainties in the present structural parameters are based on refinement precision reported by UnitCell and were confirmed by le Bail fitting.
Frequencies (in cm−1) of observed Raman and IR lines in the present study at ambient pressure and 0.6 GPa, respectively, compared with previously observed spectral lines at zero pressure.9 Predicted vibrational mode frequencies, and corresponding mode assignments, are based on DFT calculations as described in the text. The assigned modes in most cases consist of vibrations in a number of bonds; these are listed with the strongest contribution to the mode's properties appearing first.
Article metrics loading...
Full text loading...