128
5 Nucleation of Gas Hydrates
before nucleation of ice can take place. Pure water has been reported to become glassy
around the subcoolings of 40 K [6], so it is reasonable to expect that comparable
viscous slowdowns of water may occur at deep subcoolings routinely encountered
in a quiescent system under a pressurized sparingly soluble hydrocarbon guest gas
[56].
A surprising and potentially relevant phenomenon is that a small volume of
clathrate hydrate can form during heating from a deeply subcooled temperature
of about 240 K [60]. Such a formation of clathrate hydrate during heating was often
observed at a temperature far above the ice point of about 273 K but still below
the thermodynamic dissociation temperature, T eq , of the clathrate hydrate [60]. The
solubility of a guest gas in water typically falls with heating between the ice point
and T eq and hence is expected to contribute to the build-up of the supersaturation
of the guest gas. However, it is unlikely that this effect alone is sufficient to offset
the concomitant reduction in the driving force (system subcooling). It was thus very
surprising that clathrate hydrates formed more frequently at a higher temperature
(just below T eq ) than at a lower temperature (just above the ice point) [60].
The mystery does not end there and goes even deeper; given that (1) ice has been
considered an excellent nucleation promotor of clathrate hydrates and (2) melting of
ice is endothermic and would cool the sample, at least locally, one would expect that
it would have been easier for clathrate hydrate to form just above the melting point
of ice (and while the ice was still melting) than just below the T eq of the clathrate
hydrate long after the last trace of ice has disappeared. That clathrate hydrate did not
form while the ice was still melting but did form at a higher temperature long after
the last trace of ice has disappeared suggests that formation somehow became easier
at a higher temperature despite its smaller driving force [60]. However, we note that
ice does serve as a good nucleation agent for Structure II cyclopentane hydrate for
which T eq is only 7.7 K above the ice point under atmospheric pressure [61]. It thus
appears likely that ice indeed facilitates the heterogeneous nucleation of Structure II
clathrate hydrate to some extent.
Both of the two factors of the solubility limitation and the associated mass transfer
limitation of guests and the viscous slowdown of the aqueous phase would lower the
effective attachment frequency during clathrate hydrate nucleation. Still, the effective
attachment frequency would need to be lowered by more than 20 orders of magnitude for the theoretically determined nucleation rate to agree with the experimentally determined nucleation rate. It thus appears clear that a new and comprehensive
theory is required that accounts for all the clathrate hydrate-specific issues to bring
theoretically and computationally predicted nucleation rates closer to experimental
observations.
It is not clear at this stage how said new and comprehensive theory might look
like. Heterogeneous nucleation of gas hydrates requires not only a solid substrate that
could lower the nucleation work and the activation barrier but also ample supplies of
guest gases in its vicinity. This unique attribute of gas hydrate systems renders interfacial gaseous states the key concept in the heterogeneous nucleation of gas hydrates. It
thus appears likely that such a theory must at least incorporate the disjoining pressure
of interfacial gaseous states.
5 Nucleation of Gas Hydrates
before nucleation of ice can take place. Pure water has been reported to become glassy
around the subcoolings of 40 K [6], so it is reasonable to expect that comparable
viscous slowdowns of water may occur at deep subcoolings routinely encountered
in a quiescent system under a pressurized sparingly soluble hydrocarbon guest gas
[56].
A surprising and potentially relevant phenomenon is that a small volume of
clathrate hydrate can form during heating from a deeply subcooled temperature
of about 240 K [60]. Such a formation of clathrate hydrate during heating was often
observed at a temperature far above the ice point of about 273 K but still below
the thermodynamic dissociation temperature, T eq , of the clathrate hydrate [60]. The
solubility of a guest gas in water typically falls with heating between the ice point
and T eq and hence is expected to contribute to the build-up of the supersaturation
of the guest gas. However, it is unlikely that this effect alone is sufficient to offset
the concomitant reduction in the driving force (system subcooling). It was thus very
surprising that clathrate hydrates formed more frequently at a higher temperature
(just below T eq ) than at a lower temperature (just above the ice point) [60].
The mystery does not end there and goes even deeper; given that (1) ice has been
considered an excellent nucleation promotor of clathrate hydrates and (2) melting of
ice is endothermic and would cool the sample, at least locally, one would expect that
it would have been easier for clathrate hydrate to form just above the melting point
of ice (and while the ice was still melting) than just below the T eq of the clathrate
hydrate long after the last trace of ice has disappeared. That clathrate hydrate did not
form while the ice was still melting but did form at a higher temperature long after
the last trace of ice has disappeared suggests that formation somehow became easier
at a higher temperature despite its smaller driving force [60]. However, we note that
ice does serve as a good nucleation agent for Structure II cyclopentane hydrate for
which T eq is only 7.7 K above the ice point under atmospheric pressure [61]. It thus
appears likely that ice indeed facilitates the heterogeneous nucleation of Structure II
clathrate hydrate to some extent.
Both of the two factors of the solubility limitation and the associated mass transfer
limitation of guests and the viscous slowdown of the aqueous phase would lower the
effective attachment frequency during clathrate hydrate nucleation. Still, the effective
attachment frequency would need to be lowered by more than 20 orders of magnitude for the theoretically determined nucleation rate to agree with the experimentally determined nucleation rate. It thus appears clear that a new and comprehensive
theory is required that accounts for all the clathrate hydrate-specific issues to bring
theoretically and computationally predicted nucleation rates closer to experimental
observations.
It is not clear at this stage how said new and comprehensive theory might look
like. Heterogeneous nucleation of gas hydrates requires not only a solid substrate that
could lower the nucleation work and the activation barrier but also ample supplies of
guest gases in its vicinity. This unique attribute of gas hydrate systems renders interfacial gaseous states the key concept in the heterogeneous nucleation of gas hydrates. It
thus appears likely that such a theory must at least incorporate the disjoining pressure
of interfacial gaseous states.