Sixteen Shades of Ice

By John McCormick

ICE IS A MARVELOUS MATERIAL, probably second only to fire in mankind’s history as far as usefulness goes—mankind discovered that fire could cook his food and that ice could preserve it.

Ice—it goes in your Scotch (unless it is a decent single malt, or if you are a nebbish); more properly, it is a critical part of both regular iced tea and Long Island iced tea, along with any martini. Far too often winter ice involves the insurance deductible on your car, turns your steps into a lawsuit in waiting, and is indispensable to Canada’s and Scotland’s most exciting indoor sport—curling, the greatest contribution to sports, second only to that created at St. Andrews. In fact, if regular everyday water ice had a density of 1.001 grams per cubic centimeter instead of the actual 0.93 g/cm3 there would be no carbon-based life on Earth.

Many chemicals that have crystalline structures when solid can have more than one form of crystal, sometimes with radically different features.

Carbon, for example, comes in both cubic and hexagonal crystalline structures known as allotropes. Hence, carbon is both the hardest and virtually the softest naturally occurring material. The first is clear and known as a diamond, while the other is opaque, black, and known as graphite. In most respects, the two allotropes are polar opposites. Diamond is an excellent heat conductor and an excellent electrical insulator. Graphite is a good thermal conductor and a good electrical conductor.

Di-hydrogen oxide or H2O also has allotropes but water isn't limited to just two or three crystalline structures. Water can turn into a solid in about sixteen different ways—“about” because some forms such as Ice II have been reported as having multiple structures. If shown to be non-identical, they will eventually be given different numbers. Others already given different designations may be identical—hence my use of the word “about.”

But, while the exact number of different forms of solid water there are is somewhat up for debate, it is obvious that there are many.

Fans of Kurt Vonnegut should be familiar with “Cat’s Cradle,” not just “Breakfast of Champions,” and, if so, you know about the concept of Ice Nine, a denser-than-water form of ice that melts above 100 degrees F instead of 32 F. A sliver of Ice Nine acts as a seed crystal that triggers a change in what, to Ice Nine, is a supercooled ocean.

A supercooled liquid is one that, by careful cooling, is actually below the normal freezing point but which is still not a solid. This is similar to the superheated condition that we often encounter in heating water in a microwave.

I’ll describe the superheated condition first because it is easy to recreate for yourself with a simple experiment that is safe enough so long as you anticipate what will happen. When heating pure water for a cup of instant coffee or to brew tea, people sometimes find that especially pure water isn’t yet bubbling, but if they put a tea bag or a bit of instant coffee in the cup it instantly boils over. That occurs because the water is actually more than 212 degrees F (100 degrees C) but hasn’t been disturbed. Touching it will usually cause it to boil but, even curiouser, when the water is almost at boiling and something such as powdered coffee is added, the mixture boils at a considerably lower temperature.

Less commonly seen but just as dramatic is supercooled water where the liquid is barely below the freezing point and requires a trigger of some sort, such as a single ice crystal, that instantly causes the entire volume to solidify. (Because this video was produced for high schoolers, they felt it needed a hip-hop score.) Here's a less Gangsta-targeted experiment.

You can also do this at home in a couple of hours. In fact, it may already have unintentionally happened when you were trying to cool a beer or soda rapidly by placing it in the freezer—then promptly forgot about it (possibly from finishing off the other five cans).

When you open these very cold containers, the pressure drops and the entire bottle flash-freezes to slush so thick it usually won’t pour. Try to avoid doing this with champagne or any glass bottle because it will likely shatter.

To experiment with supercooling, take a very clean plastic bottle and add distilled or reverse osmosis purified and boiled water (to remove the air). Place it in an ordinary home freezer beside a similar-size bottle of plain tap water—both the same amounts. Ordinary tap water has between 100 and 400 parts per million of dissolved solids according to the test meter supplied with PUR Water Filters, so it will freeze while the pure water won’t because there is no nucleation particle to trigger the freezing.

Carefully take the bottle with liquid still in it and, holding it over a sink, open.

Making different kinds of ice is accomplished by getting water to solidify using unusual pressures or colder than usual temperatures. When water is under great pressure this alters the freezing point, resulting in a different crystalline structure that also has different physical properties.

The Ice Nine of Vonnegut's story freezes at just above 100 degrees F, and therefore most of the water on Earth is actually in a supercooled condition with that crystalline structure. Dropping a single Ice Nine crystal into water triggers the formation of a solid.

The danger of Ice Nine is that it is denser than water. Most solids are denser than their liquid form—ice is an exception to the general rule—fortunately for us.

Why fortunately? Consider if water were less dense than the usual forms of ice. Instead of ice forming on the surface of ponds, lakes, oceans, etc., leaving liquid water below for fish and other life to wait out the winter, it would sink as soon as it froze, leaving liquid water on the surface to cool quickly and freeze—in turn sinking to the bottom. Even in a mild winter, rivers, ponds and lakes wouldn’t just have a relatively thin coating of ice on top. They would be frozen solid to a great depth where, in many regions, they would remain frozen to most of their depths even in the hottest summer.

Although this would have made life easier for Captain Smith and his new ship (the Titanic), it would in reality be catastrophic for all life.

Ice Formations

Few people realize there actually are many different forms of ice, and I’m referring to plain old water ice, not the blue ice of frozen oxygen, or toilet bowl cleaner, or other exotic ices.

In fact, there were, until recently, fifteen recognized forms of ice, from amorphous frozen water (frost) to other more densely packed crystalline forms.

This fall scientists announced the discovery/creation of Ice XVI, yet another crystalline form of solid water with some very interesting properties.

With a nod to Vonnegut, there is actually an Ice IX that is also denser than water. Fortunately, it melts at a considerably lower temperature than 100 degrees F; in fact, it only exists at very low temperatures, being created when water is quickly turned into one form of Ice I or when Ice II is cooled in liquid nitrogen, changing its crystalline structure.

But before we get to the latest kinds of water ice we should get more familiar with that stuff we put in iced tea.

The ordinary ice you get when you put tap water in freezer trays looks white because of air trapped in the crystal structure or because it contains some amorphous (non-crystalline) ice, which is naturally white.

To get the beautiful clear ice carvings you see melting at fancy weddings and dinners you need to start with distilled water that has been boiled—not just run through a reverse osmosis system; that is the simplest way to get all the impurities and all the air out of water before freezing.

That is one form of ice, Ice Ih for hexagonal, the most commonly occurring form. Although a close second on the surface is the frost ice your spouse probably is complaining about when telling you it’s time to defrost the freezer or scrape the windshield.

Snow is, of course, only frozen water so it too is ice—Ice Ih in fact.

When I was young and knew everything (as opposed to being old and knowing almost nothing), one thing I always questioned in science was the comment that no two snowflakes were identical. First, I wanted to know who had compared all of them and, second, I wondered how there could be billions upon billions of different crystalline structures, all of them six-sided. In my limited experience, I could believe a couple hundred, perhaps a few thousand, different designs, but not ten to the tenth or so.

[Cubic ice on a leaf, right.]

As I studied physics, I still had questions and even after a stint in mineralogy class at Harvard learning such useless things as which minerals had body-centered cubic crystal structures and which were face-centered, or a dozen more crystalline structures we had to diagram from memory, I simply couldn’t picture an infinite number of snowflake designs.

However, in the dim old halls of the Mallinckrodt Building, I did learn that there were numerous phases in which water could solidify, depending on the temperature and especially pressure of the environment.

Ice in the form of snow forms in nature or from giant snow guns at Killington, Sugarbush, and Mt. Snow. (OK, I’m a New Englander who couldn’t afford a plane ticket to Vail but could fuel up a VW bug to get from Boston to Vermont.)

It can also form as hail or sleet depending on conditions, but there are fifteen more forms of ice.

Being an inorganic solid, ice is classified as a mineral, which is why I studied it in a mineralogy class.

Because of its particular crystalline structure, water ice expands a bit when the water freezes and is therefore about eight percent less dense than pure water—a critical eight percent because if it were denser than water, ponds, lakes, and the oceans would freeze over, the ice would sink to the bottom, and this would repeat until all the fish were dead. The bodies of water would turn into solid blocks of ice that would never melt. I previously pointed this out.

Tough break for life on Earth.

Fortunately, every common phase of frozen water is less dense than water.

The most common form of ice is designated Ih for hexagonal. It often has a small percentage component of Ic for cubic. Both names are obviously derived from their crystal structure.

There is also amorphous ice, which lacks any crystal structure.

To discuss ice structure it is easiest to look at formation temperatures as measured in kelvins (named for, wait for it, Lord Kelvin).

Zero C and 32 F are about 273 K.

Ice Ic is formed at normal atmosphere pressure and temperatures between 130 and 220 K. If it is warmed beyond 240 K it turns into Ih.

Ice II is rhombohedral in structure and formed by compressing Ih at about 200 K.

If you warm Ice II it turns into Ice III, a tetragonal crystal structure that is actually denser than water.

So, how do we get all these different forms of ice? In a word, pressure.

Water freezes at different temperatures and various pressures that affect the kind of crystal the solid forms.

In the SI (Système International d’Unités) system pressure is measured in pascals or Pa.

Other SI units are more familiar: meters, grams/kilograms, seconds, ampere, and kelvin.

Pascal is a derived unit; that is, it isn’t measured but is instead defined in terms of other SI units. One Pa equals one newton per square meter, which means little to most people, but atmosphere pressure at sea level is approximately 101 Pa (obviously this varies minute-to-minute as barometric pressure varies).

These different forms of ice can be very important.

Clathrates (solids with box-like openings in their structure) are now known to store enormous quantities of methane and other gases in the permafrost as well as in vast sediment layers hundreds of meters deep at the bottom of the ocean floor. Their potential decomposition could therefore have significant consequences for our planet, making an improved understanding of their properties a key priority.

In a paper published in “Nature,” (Nature 516, 231–233, 11 December 2014, doi:10.1038/nature14014) scientists from the “University of Göttingen and the Institut Laue Langevin (ILL) report on the first empty clathrate, consisting of a framework of water molecules with all guest molecules removed. Long thought to be purely hypothetical, this empty clathrate plays an important role in our understanding of the physical chemistry of gas hydrates. Such research could help ease the flow of gas and oil through pipelines in low temperature environments, and open up untapped reservoirs of natural gas on the ocean floor.

“In order to create the sample of Ice XVI, the researchers constructed a clathrate filled with molecules of neon gas, which they then removed by careful pumping at low temperatures. Using small atoms such as those of neon gas allowed the clathrate to be emptied without compromising its fragile structure.”

A stable solid form of H2O, the empty clathrate is the newest phase of ice. Ice XVI is the 16th known form and also the least dense.

According to the 2007 World Energy Outlook, “the total amount of methane stored within clathrates on the ocean floor far exceeds the economically exploitable reserves of conventional carbon in the form of coal, petrol or natural gas left on Earth. These reservoirs are difficult to exploit at present but form a domain of intense ongoing research.”

Thomas Hansen, one of the study authors and instrument scientist on the D20 medium to high resolution two-axis diffractometer, says: “... there is a possibility we could extract methane and convert it to useful energy, and replace it with the CO2. In other words, we could pump CO2 down to the ocean floor as a replacement for the methane in the gas hydrates ... extremely intriguing possibility worth exploring.”

Clathrate research is critical for the maintenance of pipelines where gas is transported at high pressures and low temperatures, causing gas hydrates (ice) within the pipes to form substantial blockages costing the industry approximately $500 million a year.

Ice research isn’t of interest just to bartenders.

Slippery When Wet (not Esther Williams)

Taking a look at one thing we all know about ice may make things a lot less clear.

When it is very cold ice isn’t slippery, it is almost tacky, and you certainly don’t want to lick a piece of very cold ice or metal as some idiot proves on YouTube each winter.

But near the melting point for ice we can skate or sled on ice very easily because the pressure from the metal blades melts a very thin layer of ice, coating it with water and making it slippery.

Everyone knows that! Right?

Actually, everyone except chemist Gabor Somorjai and physicist Michael Van Hove, both at the University of California at Berkeley, who “ran the numbers” and discovered that the pressure involved isn’t great enough to cause the ice to melt. In other words, ice isn’t slippery because the narrow steel blade causes it to melt.

This contradiction between theory and calculated results intrigued them sufficiently that they used electron beams bounced off the surface (a lot like x-ray crystallography), and determined that the entire surface of the ice, while still untouched by any object, has a thin layer of ice that behaves a lot more like water than solid ice.

This slippery layer accounts for shoes and other objects, which create relatively low pounds per square inch pressure, sliding easily on ice. Measurements also showed that when it gets quite a bit colder, below about -22 degrees C, this water/ice layer begins to quickly thin out and the ice becomes progressively less slippery as the temperature drops further.

The electron reflection studies showed that the surface of ice near freezing is a lot like the surface of boiling water: molecules aren’t tightly bonded together and while some evaporate from the surface, others condense on it, resulting in a constantly changing surface through what is nearly a one-to-one exchange.

That surface behavior seems to hold true for many forms of ice but enough of generalities; just what are these different phases of ice?

First, there are a couple forms of amorphous ice: amorphous meaning unorganized into any recognizable crystalline structure. Amorphous ice comes in high-density and low-density versions.

Room Temp Ice?

Those familiar with the physics of gases and liquids may recall that gases can be changed to liquid by increasing pressure (this is how A/C systems work) and in many instances you can move one step further to turn a liquid into a solid by increasing the pressure further, while keeping the temperature the same. (The process generates considerable heat energy that must be removed, but no actual cooling below ambient temperature is usually required.)

This occurs for virtually every known substance because the solid form takes up less space (has higher density) than the liquid form. That means that if you have a pot of liquid iron and toss in an iron bar it will sink.

Of course if you drop an ice cube into a glass of water you are not only much more likely to survive the experiment without a fast trip to the emergency room, but you will also prove that water is unlike virtually every other liquid—it is less dense in solid form—i.e., the ice cube will float.

That also means that when putting water under any reasonable pressure—say no more than ten times the pressure at the bottom of the ocean—water won’t turn to ice at room temperature. In fact, if you put sufficient pressure on ice instead of water, you will cause it to melt.

But ice can also form at temperatures as high as room temperature when subjected to strong electrical fields, and, according to a report in the August 26, 2005 issue of “Science” magazine (AAAS), this is much easier than previously thought.

Room temperature ice was long believed to be possible in electric fields with an intensity approximately 100 times that required to create a lightning strike. However, just as with the ice skate liquefaction calculations cited above, the voltage calculations also turned out to be wrong, this time by a factor of about 1,000.

So, room temp ice is possible, as are entirely new forms with potential real-world applications such as the Ice XVI described above. For example, back in 2006 a team of researchers subjected a drop of water to seventeen GPa on an anvil made of diamond, then bombarded the water with high-energy x-rays to break apart the atoms and create a new form of H2 and O2.

Why would anyone care about such esoteric discoveries such as the potential for a seventeenth or eighteenth form of frozen water?

Well, for one thing, water is in great demand but in short supply on Earth, so learning more about something that critical to life as we know it is always important. More specifically, some of these exotic forms of ice may lead to new ways of storing hydrogen fuel for everything ranging from cars to space vehicles.

Why Is Ice Essential To Life?

One of the great questions in science is just why water is seemingly so necessary to life. We know from all evidence that it certainly is essential, playing a role in creating enzymes and proteins, some of the most complex molecules in existence.

But why only water? Why not ammonia? Or hydrogen peroxide? Both are liquid at comparable temperatures.

Some recent computer modeling has shown that there are two different forms of regular water and that at normal temperatures and pressure water is continually changing between these forms.

From “The Journal of Physical Chemistry, B” (J. Phys. Chem. B, 2010, 114, 47, pp 15598–1560): “Water molecules bond with one another in a surprisingly complex and dynamic way. Any given volume of water contains two types of molecular structures—one a blobby, loosely packed agglomeration and the other a tight, regular arrangement resembling a crystal lattice. But both structures tend to break apart and recombine frequently, on the order of extremely tiny fractions of a second. The result is a chaotic mix of water molecules. Within that mix, the hydrogen atoms form connections that function like hooks, onto which carbon or nitrogen atoms can presumably grab to form the beginnings of complex organic molecules. And the process can dramatically influence the motion of even more complex biological systems, such as proteins, by helping their assembly. As far as anyone knows, no other liquid demonstrates this property.”

[Above left, ice crystals collected by Dr. Pablo Clemente-Colon, Chief Scientist National Ice Center, from the Beaufort Sea. Photo courtesy NOAA At the Ends of the Earth Collection.]

This may provide the missing evidence explaining why water is the basis of all life we know.

Amorphous Ice

Ice that isn’t formed into a crystal structure is called amorphous, indicating that the water molecules aren’t organized but are randomly distributed.

However, not all amorphous ice is the same.

Three basic kinds are formed at normal pressures but are cooled at different rates.

Normal cooling results in crystal hexagonal ice; an ice cube is a good example. Snowflakes are another good example, being six-sided crystals. This Ice Ih (Ice One “H”) is the most common form of ice on Earth.

Cooling extremely pure water very carefully and slowly to as low as -40 C without disturbance creates supercooled water that instantly turns to ice throughout if you simply touch it. At higher temperatures the heat generated by energy released by ice formation only permits about 20 percent of the liquid water to turn solid when the beginning temperature is -20 C.

Hyperquenched glassy water (HGW), is formed when liquid water is sprayed into or with a very cold liquid such as liquid helium. Ordinary clear window glass, as well as inexpensive glassware, is also amorphous.

Amorphous solid water (ASW), is similar to HGW but is formed by spraying a fine mist on a cold object such as metal.

Very high density glassy water (VHGW), is yet another amorphous form.

Ice Ic and XIc are both cubic structures that occur when moisture coalesces to form ice crystals at high altitudes and temperatures between -38 C and -80 C. It also forms as a transitional form when ice formed at high pressures and below -77 C experiences reduced pressure.

Ice II is a tetrahedral crystal formed by processing hexagonal ice or by reducing the pressure on Ice V at a temperature of 238 K (273 K is 32 F). Ice II is very important because it is thought to be the primary form of ice found in various satellites, including Ganymede.

Ice III is a high pressure ice that is only stable over a small range of pressures and temperatures. It is essentially a tetragonal structure but the hydrogen bonds are constantly shifting around. One form of Ice III is thought to possibly be identical with Ice IX.

Ice IV is a rhombohedral crystal formed by slowly heating high-density amorphous ice.

Monoclinic Ice V is formed when water is exposed to 500+ million pascals at 253 K (-4.27 F). Sea-level standard atmospheric pressure is about 101 kilo-pascals or kPa. One Pa equals 0.000145 p.s.i.

Ice VI is formed from liquid water at 1.1 thousand million pascals (1.1GPa) at 26 F. That would be the equivalent of an ocean depth of about 375,000 feet. However, the Challenger Deep, the deepest part of the Pacific Ocean just off Guam, is only one-tenth that deep at about 36,000 feet; therefore, you are unlikely to run across Ice VI at the local rink and won’t need to sharpen your blades differently.

The pressure at the Earth’s core is about 330 GPa and you would find enough pressure to generate Ice VI at a depth of about 600 Km, so ice VI and other high-pressure ice forms are not something you can recreate in a pressurized soda bottle in your home freezer.

Ice VII and the rest are rare, difficult to create, and, to tell the truth, not particularly interesting. 

Further Reading

Q&A Supercooled Water
Water Phase Diagram

John McCormick is a trained physicist, science/technology journalist, and widely-published author with more than 17,000 bylines to his credit. He is a member of The National Press Club and the AAAS. His full bibliography can be accessed online.