5.3 The Memory Effect
131
5.3.2 Major Attributes of the Memory Effect and Proposed
Hypotheses
Several attributes of the memory effect stand out from the historical findings: (1)
the memory effect is about nucleation of clathrate hydrates and does not influence
the growth rate of clathrate hydrates that comes after nucleation [31], (2) the reproducibility of the memory effect is poor and the variation shows stochastic nature of
the memory effect [71, 72], much like the stochastic nature of the nucleation process
itself, (3) the level of memory preservation depends on specific samples even when
they have the same thermal history [47, 64, 65], (4) no molecular structures that differ
from the ordinary clathrate hydrates have been detected in the dissociated water, (5)
melted ice that has no history of clathrate hydrate formation can also induce the
memory effect that is comparable to dissociated water made from clathrate hydrate
[62], (6) the sample memory can be erased after the dissociated water is heated far
above the thermodynamic phase boundary and/or for prolong period (after a sample
has attained its thermodynamic equilibrium) [27], and (7) transfer of a small amount
of dissociated water to fresh water still can induce the memory effect that is comparable to that in 100% dissociated water [70]. Clearly, some aspect(s) of dissociated
water has not attained thermodynamic equilibrium while the memory effect is present
but it has been hard to pin down what that factor may be.
As for the possible mechanism of the memory effect, five major hypotheses have
been put forward over the years in the literature (1) structural memory hypothesis [62,
68–70, 74, 75], (2) guest supersaturation hypothesis [63], (3) impurity imprinting
hypothesis [47], (4) nanobubbles hypothesis [76, 77], and most recently, (5) interfacial gaseous states we detailed in Chap. 4 [78, 79], which we may refer to as
“interfacial gaseous states hypothesis” in this book.
The structural memory hypothesis invokes molecular level structuring of H 2 O
molecules in the dissociated liquid water that is different from the structure of the
ordinary liquid water [62, 68–70, 74, 75]. As we saw in Sect. 5.1, ordinary liquid
water is already highly structured due to the hydrogen bonding compared to other
ordinary liquid. The structural memory hypothesis assumes an even greater level of
molecular level structuring in the dissociated liquid water.
The guest supersaturation hypothesis postulates that the memory effect is due to
the retarded diffusion of the guest gas in the dissociated water, which would leave the
dissociated water supersaturated with the guest gas. Such persisting supersaturation
of the guest gas would render the nucleation of clathrate hydrate easier when the
dissociated water is brought back inside the thermodynamically stable zone because
the real driving force for the nucleation of the clathrate hydrate is greater.
The impurity imprinting hypothesis postulates that the memory is somehow
imprinted to solid impurities in a sample [47]. A solid wall of the container that
contains the sample can also serve as an impurity in that the solid is made of a
foreign material. The hypothesis is consistent with the general observations that the
level of memory preservation differed from sample to sample (that were usually
contained in different sample cells or after non-identical cleaning procedures of the
131
5.3.2 Major Attributes of the Memory Effect and Proposed
Hypotheses
Several attributes of the memory effect stand out from the historical findings: (1)
the memory effect is about nucleation of clathrate hydrates and does not influence
the growth rate of clathrate hydrates that comes after nucleation [31], (2) the reproducibility of the memory effect is poor and the variation shows stochastic nature of
the memory effect [71, 72], much like the stochastic nature of the nucleation process
itself, (3) the level of memory preservation depends on specific samples even when
they have the same thermal history [47, 64, 65], (4) no molecular structures that differ
from the ordinary clathrate hydrates have been detected in the dissociated water, (5)
melted ice that has no history of clathrate hydrate formation can also induce the
memory effect that is comparable to dissociated water made from clathrate hydrate
[62], (6) the sample memory can be erased after the dissociated water is heated far
above the thermodynamic phase boundary and/or for prolong period (after a sample
has attained its thermodynamic equilibrium) [27], and (7) transfer of a small amount
of dissociated water to fresh water still can induce the memory effect that is comparable to that in 100% dissociated water [70]. Clearly, some aspect(s) of dissociated
water has not attained thermodynamic equilibrium while the memory effect is present
but it has been hard to pin down what that factor may be.
As for the possible mechanism of the memory effect, five major hypotheses have
been put forward over the years in the literature (1) structural memory hypothesis [62,
68–70, 74, 75], (2) guest supersaturation hypothesis [63], (3) impurity imprinting
hypothesis [47], (4) nanobubbles hypothesis [76, 77], and most recently, (5) interfacial gaseous states we detailed in Chap. 4 [78, 79], which we may refer to as
“interfacial gaseous states hypothesis” in this book.
The structural memory hypothesis invokes molecular level structuring of H 2 O
molecules in the dissociated liquid water that is different from the structure of the
ordinary liquid water [62, 68–70, 74, 75]. As we saw in Sect. 5.1, ordinary liquid
water is already highly structured due to the hydrogen bonding compared to other
ordinary liquid. The structural memory hypothesis assumes an even greater level of
molecular level structuring in the dissociated liquid water.
The guest supersaturation hypothesis postulates that the memory effect is due to
the retarded diffusion of the guest gas in the dissociated water, which would leave the
dissociated water supersaturated with the guest gas. Such persisting supersaturation
of the guest gas would render the nucleation of clathrate hydrate easier when the
dissociated water is brought back inside the thermodynamically stable zone because
the real driving force for the nucleation of the clathrate hydrate is greater.
The impurity imprinting hypothesis postulates that the memory is somehow
imprinted to solid impurities in a sample [47]. A solid wall of the container that
contains the sample can also serve as an impurity in that the solid is made of a
foreign material. The hypothesis is consistent with the general observations that the
level of memory preservation differed from sample to sample (that were usually
contained in different sample cells or after non-identical cleaning procedures of the