- Water is less dense in solid form (ice) than in liquid form. Ice cubes float, and ice forms at the top of lakes, protecting the fish and other organisms below from the cold. Water is one of only a handful of substances for which this is true.
- Water can be supercooled, and “instantly” frozen. Very pure water can have its temperature lowered below the freezing point — supercooled — while remaining a liquid. Any significant disruption of the liquid, such as pouring, will make it freeze almost instantaneously, as the video at this link shows.
- In the right circumstances, hot water can be frozen faster than cold water! In 1963, a Tanzanian secondary school student named Erasto Mpemba noticed that hot ice cream mix froze faster than cold ice cream mix. He pointed this out to a visiting physics lecturer, and the two published their experimental observations in 1969. The effect is now known as the “Mpemba effect“.
Equally odd, but even less known, is the curious ability to bend a stream of water with static electricity. I did this experiment at home a couple of weeks ago, and a snapshot of the result is below:
The rod is made of glass, and is what is generally known as a friction rod. The rod begins electrically neutral, but is stripped of electrons when rubbed with a piece of silk, leaving a net positive charge behind. It is this positive electrical charge that attracts the water.
Why is water so strongly attracted to static electricity? It turns out that the answer is somewhat complicated, and the complete explanation was a bit of a minor mystery for nearly eighty years!
To begin our explanation, we need to look at the molecular structure of water. Its chemical formula is familiar to most people: H2O, which indicates that a single molecule of water consists of 2 hydrogen atoms (H) and one oxygen atom (O). These atoms are held together by interactions of the electric charges of the atoms: their positive nuclei and their negative electrons. If we could take a snapshot of a molecule, we would find that the atoms arrange themselves in a somewhat asymmetric manner:
The hydrogen atoms form a 105° angle with respect to the center of the oxygen molecule. Though the molecule is overall electrically neutral (the amount of positive charge is the same as the negative charge), the oxygen atom exerts a stronger pull on the electrons than the hydrogen atoms do, and the negative charge therefore tends to group around it. This leaves the hydrogen atoms on average positively charged and the oxygen atom negatively charged, and the resulting structure is what is known as an electric dipole. The term “dipole” = “two poles”, referring to the fact that the charge distribution consists effectively of two distinct separated charges of opposite sign. It is often drawn in a simple “ball-and-stick” form, as shown below:
Because a dipole has a spatial structure, it also has an orientation; the direction of a dipole is typically indicated by an arrow pointing from the negative charge to the positive charge. A dipole produces an electric field, indicated by arrows in the following picture, which in turn creates a force on other charged particles:
It is natural to ask at this point: if water is a huge collection of electric dipoles, why isn’t water visibly electric under normal circumstances? The answer is that ordinarily the dipoles are all in constant microscopic motion and any small volume contains a large number of randomly-oriented dipoles:
The effects of dipoles pointed in opposite directions cancel out; with a large number of dipoles in all directions, they produce no electrical effect. This changes when a positive friction rod is brought near the water; then, the negative parts of the dipoles tend to be pulled towards the rod, and the positive parts repelled:
It should be noted that the alignment won’t be perfect: random motion of the molecules will prevent them from being perfectly aligned. However, the stronger the applied electric field E, the more the dipoles will be forced into alignment.
The liquid in essence has a net negative charge on the side closest to the rod, and a net positive charge on the side farthest. The dipoles near the rod are attracted to it, and in turn attract dipoles next to them, and so on, forming a “chain” of dipoles such that the entire body of water gets dragged towards the rod:
This basic description of the electrical properties of water is qualitatively accurate. When researchers turned it into a quantitative model, however, they ran into a problem: water is much more electrically active than the model predicts! Loosely speaking, researchers knew the electrical “strength” of an individual water molecule, and they knew how many molecules were in a given volume of water. The whole body of water, however, was electrically stronger than the sum of the individual molecules: the whole is greater than the sum of the parts.
The answer turns out to be quite complicated, but involves an effect mostly absent from the simple model described above: interactions between the water molecules! We have mentioned that, in a water molecule, the electrons tend to be drawn towards the oxygen atom. This means that the oxygen atom is somewhat negatively charged and the hydrogen atoms are somewhat positively charged, and can attract one another. This weak form of molecular bonding is referred to as a hydrogen bond. The result of this bonding is that any given water molecule can be connected to up to four others in a tetrahedral shape, in which two hydrogen atoms can be stuck to a single oxygen atom (taken from Ref. ):
How does this affect the electrical properties of water? Unlike our previous description of water molecules being randomly oriented, this tetrahedral arrangement means that nearby water molecules are oriented somewhat parallel. This in turn means that, when an electric field is applied, groups of molecules can move together, providing an extra “oomph” to their electrical properties and presumably being more resistant to random molecular motion. Furthermore, in liquid phase, these hydrogen bonds can be bent, allowing the dipoles to align even more parallel.
This alignment must be local, because without an external electrical influence, we expect water to be electrically neutral. Our earlier picture can be crudely revised as shown below, in which molecules are locally parallel in small regions, but different regions have different orientations. The net result is the same, in that every dipole has, on average, another nearby dipole that cancels it out!
With water molecules locally aligned, an additional effect is possible: the hydrogen bonds can then “stretch” the individual water molecules in a sort of “tug-of-war”! This stretching can further separate the positive and negative charges of a water molecule, increasing its electrical strength.
These effects were already known as far back as the 1930s, but turning this qualitative understanding into an actual theoretical model that agreed with experiment turned out to be challenging. Since 1933, a number of authors [2-4] developed increasingly accurate but more complicated models that came closer to experiment. In 2007, a collaboration between workers in Princeton and Trieste  finally introduced a model that agreed well with the electrical properties of water at a range of temperatures, and could even account for the electrical properties of ice.
The strong electrical properties of water also account for its power as a solvent, able to dissolve a wide variety of other liquids and solids. Of course, these strong electrical properties also account for the very dramatic “water-bending”, shown below in video!
It is somewhat eye-opening to realize that even a seemingly ordinary liquid such as water can hold many surprises for science, and very difficult challenges in understanding!
1. J.D. Bernal and R.H. Fowler, “A Theory of Water and Ionic Solution, with Particular Reference to Hydrogen and Hydroxyl Ions,” J. Chem. Phys. 1 (1933), 515.
2. L. Onsager, “Electric Moments of Molecules in Liquids,” J. Am. Chem. Soc. 58 (1936), 1486.
3. J.G. Kirkwood, “The Dielectric Polarization of Polar Liquids,” J. Chem. Phys. 7 (1939), 911.
4. J.A. Pople, “Molecular association in liquids. II. A theory of the structure of water,” Proc. Roy. Soc. Lond. A 205 (1951), 163.
5. Sharma M, Resta R, & Car R (2007). Dipolar correlations and the dielectric permittivity of water. Physical review letters, 98 (24) PMID: 17677991