Comprehensive Analysis of Methyl-β-D-ribofuranoside: A Multifaceted Spectroscopic and Theoretical Approach

This study presents a comprehensive analysis of the vibrational spectra of methyl-β-D-ribofuranoside. Employing a combination of inelastic neutron scattering, Raman, and infrared spectroscopy allows for the observation of all modes regardless of the selection rules. The experimental techniques were complemented by density functional theory computational methods using both gas-phase (Gaussian) and solid-state (CRYSTAL, CASTEP) approaches to provide an unambiguous assignment of the defining vibrational features. Two distinct structures of the molecule were identified in the unit cell, differentiated mainly by the orientation of the furanose ring O–H bonds. The low-energy region of the spectrum (<400 cm–1) is dominated by lattice vibrations and functional group rotation, while the midenergy region is dominated by out-of-plane bending motions of the furanose ring (400–900 cm–1) and by C–H bending in the methyl and methylene groups (1400–1600 cm–1). The high-energy region (>2800 cm–1) encompasses the C–H and O–H stretching modes and offers convincing evidence of at least one H-bonding interaction between the two structures of methyl-β-D-ribofuranoside.


Contents
Additional data for input files (Tables S1 and S2) S7 Optimised structures (Figures S4 to S10) S10

Custom Renishaw ® inVia Raman spectrometer setup
The Raman spectra were measured with the help of a customised setup which has been previously described [1].Over the last year, the setup has been improved and additional functionalities (such as temperature control) were added to it; therefore, an up-to-date description is given here.The modified setup consists of a specially designed centre stick (Figure S1) to which a laser probe head fibre-optically coupled to a Renishaw® inVia Raman spectrometer can be attached.This offers a choice of two lasers: a 532 nm, 200 mW, Class 3B, continuous wave, diode-pumped solid-state laser, or a 785 nm, 300 mW, Class 3B, continuous wave, Toptica ® diode-pumped solid-state laser.
The Raman centre-stick is comprised of two concentric thin-walled stainless-steel tubes which pass through a vacuum port and a number of thermal radiation/convection shields.With the laser head on top, the stick is approximately 1.65 m in length and is suitable for insertion into a 100 mm-bore cryostat, allowing for measurements at temperatures between 6 K and 350 K.The laser beam has a diameter at the aperture of 700 μm, and it traverses the inside of the inner stainless-steel tube under vacuum.This configuration avoids background scattering from optical fibres, which would have otherwise been in front of the filters in the probe head, and prevents their exposure to cryogenic conditions and large changes in temperature.Using a 20× magnification achromatic lens with a long working distance of 14 mm, the laser beam is focused down to a spot size of approximately 50 μm diameter through a 1 mm thick sapphire window in the lid of the sample cell, which can be attached to the bottom end of the stick, as shown in Figure S1(b).The focusing is adjustable as the lens can be moved vertically from outside of the cryostat through a motorized motion controlled using the Z-drive of an XYZ translation stage (either manually or from within the WiRE TM 4.1 software supplied with the spectrometer).After the light scatters in the sample, the Raman signal returns along the same path.
The laser power is adjustable either in steps using a motorized neutral density filter wheel down to 1% of the maximum power available (∼ 3 mW at the sample position) for the 785 nm laser, or using the CoboltMonitor TM 6.0 software designed for the 532 nm laser allowing for fine adjustments to the power in 1 mW increments.Due to low specific heats and poor thermal conductivities, beam-heating effects can be significant at cryogenic temperatures; thus, being able to adjust the power level at the sample position and read precise temperatures is important.For the latter aspect, temperature is recorded both inside the cryostat and on the sample can itself.A platinum resistance temperature detector (RTD) sensor and two wire heaters are attached to the sample as shown in Figure S2(c).The temperature is monitored and controlled through a Lake Shore Model 224 temperature monitor which provides accurate measurements from 1.4 K to 350 K.The time taken to cool from room temperature to the base temperature of the cryostat (3 K for the liquid helium cryostat) is approximately 2.5 hours, although the sample cell never reaches this temperature, settling around 4.5 K, or 6.5 K, with the laser beam on.
Overall, the setup allows for Raman scattering measurements at 532 nm and 785 nm with a resolution S4 of 1 to 4 cm -1 over a wide range (0-4000 cm -1 ) and at cryogenic temperatures matching the INS conditions on TOSCA.Raman spectra were also recorded using a commercial benchtop Bruker Senterra confocal Raman microscope setup, equipped with lasers of three different wavelengths: 532 nm, 633 nm, and 785 nm; and capable of measurements of similar resolution (3 to 5 cm -1 ) and range (40-4000 cm -1 ).The Raman spectra of methyl-β-D-ribofuranoside were recorded using the 532 nm and 785 nm wavelength lasers at a power of 5.0 mW.Data using the 532 nm laser wavelength was of lower quality, while the 785 nm laser worked better with the methyl-β-D-ribofuranoside sample under the different settings that were tried.Figure S3 shows the comparison between Raman spectra measured with a laser wavelength of 785 nm on both setups at room temperature and a Raman measurement at 6.5 K using the custom setup.
It can be seen that the measurements on the modified Raman setup are of the same quality as the spectra

Figure S1 .
Figure S1.Picture (a) and schematic representation (b) of the custom Raman centre stick used for the cold Raman measurements.

Figure S2 .
Figure S2.Pictures showing the open cell loaded with sample (a), the closed cell with the sapphire window highlighted (b) and the wired (heaters and temperature sensor) copper (Cu) plate secured to the sample cell, ready to be attached to the centre-stick (c).

Figure S9 .
Figure S9.The two structures of methyl-β-D-ribofuranoside optimised in CRYSTAL 17 using the B3LYP hybrid exchange-correlation functional (molecules have been manually selected from the optimised unit cell using the same density functional).

Figure S10 .
Figure S10.The two structures of methyl-β-D-ribofuranoside optimised in CRYSTAL 17 using the PBESOL0 hybrid exchange-correlation functional (molecules have been manually selected from the optimised unit cell using the same density functional).

Figure S11 .Figure S12 .Figure S13 .
Figure S11.Comparison in the low-energy region of experimental INS spectrum of methyl-β-Dribofuranoside with theoretical spectra of structure (I) of the molecule simulated with Gaussian (using the hybrid B3LYP functionals) both with no solvent and with water as a solvent using the CPCM variant of the COSMO solvation model.

Table S1 .
Exponents and contraction coefficients used to define the basis sets for the DFT simulations of methyl-β-D-ribofuranoside with CRYSTAL 17.
a Number of primitive gaussian-type functions (GTF) in the contraction for the atomic orbitals in the shell; b Exponent of the normalised primitive GTF; c Contraction coefficients of the normalised primitive GTF.

Table S2 .
Cell parameters and fractional atomic coordinates of the asymmetric unit cell of methyl-β-D-ribofuranoside selected from the crystallographic information file (CIF) deposited by C. A. Podlasek et al. [2].