# Electric buzz in a glass of pure water

Charge transfer vs Proton transfer

What is the difference between the two? How to observe these processes?

What is fully charged molecules?

At any instant, pure water contains 0.3 M of fully charged water molecules. H-bond charge transfer generates a fleeting positive (+) or nagative (-) charge on water molecules. This may give rise to longer-lived partially charged water molecules (H2O+δ\text{H}_2\text{O}^{+\delta}, H2Oδ\text{H}_2\text{O}^{-\delta})

Hydrogen-bonding defects in liquid water lead to the accumalation of negative or positive charge on water molecules with an odd number of hydrogen bonds

why odd number?

Howevery, a prtial breakdown of this geometry is what makes ice float on water, whose lightly denser structure includees water molecules that form an odd number of hydrogen bonds.

Like Schrödinger’s cat, which is both dead and alive, a pair of hydrogen-bonded water molecules can be either charged (i.e., as H2O+^+ \cdots H2O^-) or uncharged (i.e., as H2O\cdots H2O).

This cost further increses dramatically if one also wants to treat nuclear quantum effects (NQEs).[1]

Can you describe nuclear quantum effects? Does CP2K neglects nuclear quantum effects?

Atomistic simulations of chemical, biological and materials systems have become increasingly precise and predictive owing to the development of accurate and efficient techniques that describe the quantum mechanical behaviour of electrons. Nevertheless, the overwhelming majority of such simulations still assumes that the nuclei behave as classical particles.However, neglecting nuclear quantum effects has become one of the largest sources of error, especially when systems containing light atoms are treated using current state-of-the-art descriptions of chemical interactions. [2]

# Quantum dynamics and spectroscopy of Ab initio liquid water

This work is good because they show they can accelerate path integral AIMD calculations. They use self-consistent charge density functional tight binding, which allows them to evaluate the full electronic structure. Thet apply their methods, which allows them to obtain almost nanosecond time scale simulation.

# The collective burst mechanism of angular jumps in liquid water

We show that large orientational motions require a highly collective dynamic process involving correlated motion of up to 10% of water molecules in the hydrogen-bond network that form spatially connected clusters.

Why first NPT and NVT?

Energy minimization was first carried out to relax the system, followed by an NPT and subsequently NVT equilibration at 300K and 1 atmosphere for 10ns each.


What is the difference of ensemble and simulation box?

How did they determine the percentage of 10%?

How did they quantify these variable? They counted the number of jumpers within a time window of 200 fs. Jumpers are the water molecules with large angular jump. Very simple idea.

What is ThZ timescale?

How can they said a new word without any further explaination?

What is results or finding?

How they discribe their signaficance?

  • We illustrate a mechanism that unambiguously reveals the collective and correlated nature of water reorientation dynamics.
  • We also show that regions with lower local density serve as hot-spot sites in the network where large reorientations can occur simultaneously. 2

Water molecules are characterized by a rather large dipole moment.

What is dipole moment?

The magnitude of dipole moment is the product of the charge QQ and distance of seperation rr:

μ=Qr\mu = Q * r

the direction is from negative charge to positive charge, and it is measured in Debye units denoted by D: 1D=3.33564×1030Cm1D = 3.33564 \times 10^-30 \; C\cdot m

What is the direction of dipole moment for a water molecule?

  1. https://byjus.com/chemistry/dipole-moment/#:~:text=A%20dipole%20moment%20is%20a,%CE%BC%20%3D%20q%20%C2%B7%20r
  2. https://www.quora.com/What-is-the-dipole-moment-of-H2O-Does-it-include-dipole-moment-from-hydrogen-and-lone-pairs-of-electrons-of-oxygen-atoms

They developed an automatized protocol for detecting all the various angular changes in water reorientation, which we term angular swings. The large-amplitude angular jumps involves at least three water molecules which alters the local H-bond network topologh.

# Ask good questions?

When these jumps occur, do other water molecules in this network remain as spectators ar are they active participants in a more collective process?

# Methods to answer your questions?

Water reorientation dynamics includes various processes happening at different time scales, from very fast vibrational motions causing limited reorientation, to slower reorientation through sudden large-amplitude angular jumps.

n(t)=vF(t)×dvF(t)dt\vec{n}(t)=\vec{v}_{\mathrm{F}}(t) \times \frac{d \vec{v}_{\mathrm{F}}(t)}{d t}

which corresponds to the vector perpendicular to the plan of rotation of the body-fixed vector vR(t)\vec{v}_{\mathrm{R}(t)}. This implies that, over the time of one angular swing, the direction of n(t)\vec{n}(t) does not change.

q(t)=1n(t)n(t+dt)n(t)n(t+dt)q(t)=1-\frac{\vec{n}(t) \cdot \vec{n}(t+d t)}{|\vec{n}(t)||\vec{n}(t+d t)|}

which is equal to 0 during the swing.

The final output of the protocol is the start time (t), duration (∆t), and magnitude (∆Θ), for each angular swing detected.

From the total of 1019 water molecules, we find that around 5% of them undergo a major change in the direction of their dipole moment vector during this short time interval, suggesting that these are concurrent events.

Fig.1D by following how its dipole vector changes in time through the time evolution of the angle it forms with respect to one of the axes of the laboratory coordinate system. While this only serves as a proxy for the angular fluctuations, it gives us a semi-quantitative measure of the size of the angular reorientation.

A proxy for the angular fluctuations

They don't use the absolute change of angles, even they can. They can set defferent orgin for each water molecues. But they don't.

New idea

The real motion of water molecues is the combination of rotation and translation. Can I focus on this aspect and analysis the dynamic properties? For example, what is difference of rotations/translations between surface and bulk? The frequency of surface and bulk?

Moreover, eye-balling the time series shown, indicates that these reorientation events are simultaneous and possibly correlated in space, which, as we will see shortly, requires a collective reorganization of the topology of the hydrogen-bond network.

Good trick

The trick I learned here is draw the relevent lines in a single plot, and guess the relation ship.

Interestingly, we observe that the fluctuations in the number of defective water molecules occur in waves. These oscillations reflect processes in the network which, on a picosecond timescale, for example, lead to the creation or annihilation of up to 10-20 defective water molecules in the network (large defect oscillations over short time scale are shown in Fig.S1)

Moreover, the oscillations seem to be correlated in time: the larger the number of the defected water molecules, the more large-amplitude swings simultaneously happening in the system. In fact, in panels B and C of Fig.3, we show that when the local topology in the H-bond network is more defective, there are less small-amplitude angular changes in the system and more of the large-amplitude ones.

What is reaction coordinate?

Quantifying the angular jumps illustrated in Fig.1D requires the identification of reaction coordinates that are highly non-local, involving several degrees of freedom that are challenging to identify by eye.

In chemistry, a reaction coordinate is an abstract one-dimensional coordinate which represents progress along a reaction pathway. It is usually a geometric parameter that changes during the conversion of one or more molecular entities. In molecular dynamics simulations, a reaction coordinate is called collective variable.


# The motivations

Large angular motions usually create coordination defects which affect the hydrogen bonding patterns. This, in turn, affects local topology of the H-bond networkd, leading to rearrangements of nearby water molecules and possibly further large reorientation events.

For small angular thresholds, when the selected swings are predominantly of small magnitude, the swing duration peaks at roughly 30 fs. This corresponds to a fast hindered rotational mode.

New observation

It's interesting that another peak appears when the angular magnitude threshold increses.

Note that for the duration of the jump we consider just the angular reorientation of the water molecule, and not the time needed for the H-bond breakage and formation.

Maybe this is the reason for the shape of the bimodal. And they said the characteristic time of 100fs corresponds more closely to the tiem it takes to transition from one stable H-bond state to another.

New idea

Maybe I can consider the rotation and H-bond network at the same time.

Then they asked this following question:

If such a large number of water molecules undergo angular fluctuations, do they occur independently from each other or are there some underlying connections in the hydrogen bond network involved in these reorientational motions? fluctuations, do they occur independently from each other or are there some underlying connections in the hydrogen bond network involved in these reorientational motions?

This figure shows the local distribution around the jumpers.

How do you define water shell?

They said:

The nearest jumping molecule is typically situated in the first water shell of another molecule that performs large jump, while the 2nd nearest neighbor has a wide distribution around the second water shell distance.

but how did they define the water shell?


I can not see what they mean: Instead, if the concurrently jumping molecules would be distributed homogeneously throughout the system, as is the case when choosing random molecules, the two nearest neighbors would have been positioned at larger distances with a rather different shape of the PDFs (dashed black lines in Fig.5A and B). This gives a strong indication that case when choosing random molecules, the two nearest neighbors would have been positioned at larger distances with a rather different shape of the PDFs (dashed black lines in Fig.5A and B). This gives a strong indication that the simultaneous large jumps we observe are not happening independently throughout the system, but are instead interconnected and correlated.

At this point, they give a YES for their question. But doesn't give further explainations.

Since our automated protocol detects angular swings without initial imposition of hydrogen bonding interactions, it is interesting to examine the correlation between the collective reorientational dynamics and changes in the number of hydrogen bonds. They next investigate how many of the detected angular swings are actully bond breaking events.


You did what I want to do: combination of rotations and H-bond dynamics.

Our results in bulk water at room temperature showing the highly cooperative character of the reorientational dynamics of the water molecules, open the doors to exploring how this effect changes upon supercooling and near biological systems where one might expect an enhancement of the phenomenon.

# The computer simulation of proton transport in water

[1] C. Schran, O. Marsalek, T. E. Markland, Chem. Phys. Lett. 678, 289 (2017)
[2] Markland, T., Ceriotti, M. Nuclear quantum effects enter the mainstream. Nat Rev Chem 2, 0109 (2018). https://doi.org/10.1038/s41570-017-0109