Comment on “Exploring Dynamics and Cage–Guest Interactions in Clathrate Hydrates Using Solid-State NMR”

Structure II tetrahydrofuran (THF) hydrate is one of the most frequently studied clathrate hydrates as it is easily made by freezing a solution of composition THF·17H2O. The resulting hydrate is stable at atmospheric pressure up to a congruent melting point of 277.5 K (4.4 °C).(1, 2) This hydrate is often used as a proxy for studies of materials that are more difficult to prepare, such as natural gas hydrates.(3-8) As a system for learning about guest–host interactions, the dynamics of both guest (THF/THF-d8) and host (H2O/D2O) have been of wide interest as shown by the large number of solid-state NMR and dielectric measurements performed over a temperature range from ∼4 K up to near the melting point at 277.5 K.(9-23) Both guest and water molecular reorientation and water diffusion have been observed, and various mechanistic models, including the effects of various defects, have been proposed to account for the experimental results. Transient guest–host hydrogen bond formation in THF and other ether and ketone hydrates was first proposed by Davidson(11) and confirmed from spectroscopic measurements,(15, 24) molecular dynamics simulation,(25, 26) and single crystal X-ray diffraction.(27) The paper featured in the title of this Comment(23) recently reported yet another set of measurements, six longitudinal 1H relaxation times (T1) on THF·17D2O, this time under magic angle spinning (MAS) conditions, purportedly to obtain high resolution information as there are two sets of resolvable protons on the inequivalent methylene carbons. Since the results of these measurements were very different from T1 measurements taken on static samples, an explanation was proposed that involved changes in the THF reorientational activation energy from ∼4 kJ/m at T < 200 K to nearly 20 kJ/m at T > 200 K. This radical change was attributed to increased hydrogen bonding between THF and D2O. We think this is rather unlikely for ether–water interactions. If we take the difference in activation energies for guest reorientation for cyclopentane hydrate and THF hydrate, ∼16 kJ/mol, attribute this to hydrogen bond formation in THF hydrate, and compare this to the estimated hydrogen bond energy in ice at 16–32 kJ/mol,(28) it would predict that THF should be a good hydrate inhibitor rather than a weakly interacting hydrate guest as it will compete with the hydrogen bond formation between water molecules. We have taken literature T1 data obtained for static samples of sII deuteriohydrates of THF (ethylene oxide (EO)-d4), cyclopentane, and dioxolane (Figure 1) published by Jacobs, Zeidler, and Kanert,(18) (Figure S1) and compared these with the MAS T1 measurements (gray rectangle marked a, Figure 1).(23) The literature T1 data for the three hydrates show that the relaxation rates decrease with increasing temperature, with some flattening out of the curves at the highest temperature of measurements. In agreement with other measurements, this trend in T1 is associated with 1H dipolar relaxation due to the reorientational motion of guest molecules in the hydrate cavities in the fast motion limit. The flattening out of the T1 curves leads to a T1 maximum,(10, 12) and then to a decrease in T1 with increasing temperature as slow water motion starts to contribute to the relaxation. This gives a frequency-dependent T1 decrease, evident from measurements at lower frequencies (10 and 30 MHz(10, 12)) to a new T1 minimum, not observed as it would occur above the sample melting point. This new motion is a reorientation of the water molecules, which is also observable by T1ρ measurements and usually assigned to be a process driven by the diffusion of Bjerrum defects.(29, 30) The guest T1 relaxation responds to this motion by (a) cross relaxation of 1H to 2H in water, (b) spin diffusion to residual 1H in the D2O, and (c) changes in guest motion as the symmetry of the hydrate cages changes when the disorder in the water proton positions becomes dynamic. This picture is generally consistent with all static line width, T1, and T1ρ experimental results.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.