Tiny Crystal Provides a Big Clue

Lab-grown crystal of blue ringwoodite, about 1.5 times the size of a salt grain (~150 micrometers) Photomicrograph by Jasperox, Wikimedia Commons.

Lab-grown crystal of blue ringwoodite, about 1.5 times the size of a salt grain (~150 micrometers) Photomicrograph by Jasperox, Wikimedia Commons.

Is there an immense ocean, far beneath the surface of the Earth, that replenishes the oceans above? Recent news items describe a deep reservoir containing as much as three-quarters of the Earth’s water supply. Most of these news stories are careful to note that this isn’t some great sloshing underground pool, and you won’t find any fish living there. Rather, the water is “bound up” in mineral deposits and released when these minerals are put under immense pressure. Some news stories compare the minerals to sponges, which is not something you usually associate with rocks. (Here are a couple of examples of the news items: Daily Digest News, The Guardian.)

What’s really going on here? Last March, a research paper in the journal Nature reported the discovery of a tiny crystal of a mineral called ringwoodite, encased in a diamond that was plucked out of shallow river gravel by artisan miners near Juína, Brazil. A research team led by Graham Pearson (University of Alberta, Canada) found the ringwoodite crystal (60 microns, less than the thickness of a sheet of paper) inside a dirty-looking brown diamond that they had bought for about $20 (US), according to an article in Sci-News. (Here’s the article, with a photo of the diamond and ringwoodite crystal.).

This month, a paper in Science used this discovery, along with geological measurements, lab studies, and models, as evidence to support an idea that geologists had been looking at for many years. (Here are a couple of press releases, from the the University of New Mexico and Northwestern University, where the two lead authors are located.)

Scientists suspect that ringwoodite may be common in a “transition zone” that lies 410 to 660 km (250–410 miles) below the Earth’s surface, based on seismic measurements, which track changes in the speed and direction of earthquake vibrations. Ringwoodite has been made in the lab and found in meteorites, but this was the first time that anyone had found a naturally occurring crystal from inside the Earth.

green olivine sand

Green olivine sand on black volcanic rock. Photo by Brocken Inaglory, Wikimedia Commons.

Why do we care about this elusive mineral? Because it’s a high-pressure form of another mineral called olivine. As the name suggests, many forms of olivine are dark green (it can range from yellow to black). The dark-green sands on Papakolea Beach, Hawaii are mostly olivine, and this mineral is common all over the world. At very high pressures, like those you would find more than 660 km beneath the Earth’s surface, olivine transforms into another mineral called perovskite. The presence of one form or another affects the way that earthquake vibrations travel through the Earth. These two minerals have been studied intensively, and geologists use seismic signals as clues to help them map mineral forms and geological activity far beneath the surface, where it’s hard to get the information any other way.

Water and hydroxyl radical

Water molecule (H2O, left) and hydroxyl radical (OH, right). Courtesy of NIST.

Ringwoodite is intermediate between olivine and perovskite. Because of the way its crystals form, they can contain as much as 3% by weight of something called hydroxyl radicals (OH•). (The crystal reported in the Nature paper had 1.5%.) This “radical” has nothing to do with political activism or unorthodox beliefs. Used in a chemistry context, the term refers to a water molecule (HOH, more commonly written H2O) that has one hydrogen atom stripped away, leaving behind a spare electron that it can share with something else, in this case, the elements in ringwoodite’s crystal structure.

Olivine doesn’t contain hydroxyl radicals. When olivine is put under pressure, water from the surrounding environment can be converted into OH•, forced inside, and incorporated into the framework of atoms, forming ringwoodite. The more OH• trapped in the framework, the faster the sound waves travel through it, which is why geologists had suspected that there was a water-containing mineral intermediate between olivine and perovskite.


Perovskite collected by A.E. Foote from Magnet Cove, Arkansas. Mineral collection of Bringham Young University Department of Geology, Provo, Utah. Photo by Andrew Silver, USGS.

Put ringwoodite under even more pressure, and it converts into perovskite. The hydroxyl radicals are forced back outside again, where they recombine with hydrogen (which is just about everywhere) to form liquid water. This water causes the perovskite to melt a little, in much the same way that sprinkling salt on ice causes the ice to melt. The scientists who published the paper in Science had actually seen this melting behavior in the lab, when they used a device called a diamond anvil cell to put immense pressure on a ringwoodite crystal and convert it to perovskite.

Geologists had seen seismic waves suddenly slowing down in regions where they had other clues that one rock layer was sinking down below another. A layer of partially melted perovskite that was taking a dive would slow down a seismic wave like this.

Previous examination of ancient minerals called basalts, taken from mid-ocean ridges, suggested that Earth’s upper mantle has a water content of 0.005 to 0.02% water by weight. Lab and modeling studies show that ringwoodite and a related mineral, wadsleyite, can hold between 1 and 3% water by weight. Seismic evidence suggests that the water content below 660 km is much less. Thus, if these studies are right, the transition zone is where most of the water is. If this zone extends all over the world, and its average water content is 1% (a conservative estimate that needs to be verified), this translates into nearly three times as much water as the oceans contain, which is where the news reports got their three-quarters number (three-quarters in the transition zone, one-quarter in the oceans).

Geologists have been working for decades to create a model that “balances the books” on where water comes from and where it goes — a “whole-Earth water cycle”. They had long suspected that there was a subterranean source that acts as a buffer zone to keep the amount of water in the oceans fairly constant. This latest evidence provides more clues to help them fill in the missing pieces.

These iron curtains are very, very sheer.


Model of the atomic structure of graphene. Source: Wikipedia, licensed under the Creative Commons Attribution-Share Alike 3.0 Unported license.

Recently, a group of researchers in Dresden, Germany found a way to make one-atom-thick sheets of iron. It wasn’t what they had set out to do, but they were alert enough to see this as the intriguing discovery it was rather than an annoying byproduct to be cleaned up.

Thin metal films are used as coatings or wrappings (aluminum foil, for example), components in optical and laboratory instruments, and as chemical sensors and catalysts. For a lot of high-tech applications, including high-density recording media and electronic devices, the thinner the film, the better. The ultimate thin film would be a sheet one atom thick — a monolayer.

Carbon monolayers, known as graphene, have been around for a few years, and researchers have been busy exploring the unusual properties of these ultra-thin carbon sheets. It turns out that structures in which the atoms that are linked side-to-side, but not up and down, conduct electricity very differently than their multi-layer counterparts.

Because of the way that carbon atoms connect with one another, it forms thin sheets very easily. Graphene is just an extreme form of graphite, which is made of carbon sheets that are a few hundreds to several thousands of atoms thick. Graphene was discovered when scientists applied strips of adhesive tape to pieces of graphite and carefully peeled up the top atomic layer.

Making thin layers of metal is a more difficult proposition. Metal atoms tend to form chemical bonds in all directions, forming solid chunks. Thin sheets are formed by pressing metal slabs between rollers or vaporizing the metal and spraying it onto a surface. Applying sticky tape to a piece of metal might pull a few individual atoms loose, but it’s not very effective for peeling up an intact layer.

Also, very thin metal films require some sort of support structure, called a substrate, to hold them intact. That means that only the top of the film is available for use. Until now, a free-standing metal monolayer seemed like an interesting theoretical exercise that would never be realized in the real world.

The Dresden group made their discovery while they were making graphene. They grew graphene layers from a carbon-containing vapor deposited onto a nickel–molybenum substrate. To peel the graphene up from the substrate, they used an etching solution made of iron chloride.

When the researchers examined their graphene under an electron microscope, they found that the etching solution had left behind a residue. Under the microscope’s electron beam, this residue was converted to metallic iron, which formed very small crystals and atom clusters on the surface. Some individual iron atoms were stuck to the edges of holes in the graphene film.

hexagonal iron

Three layers of the type of iron crystal found in this study. Source: Wikipedia, public domain.

What was truly surprising, however, was the appearance of pure iron crystalline layers, one atom thick, suspended across some of the holes in the graphene sheets. Although the widest of these layers was only about ten atoms across, the layers were arranged in an orderly fashion that were definitely monolayer versions of an iron crystal.

The researchers repeated their procedure and examined several iron-contaminated graphene sheets using their electron microscope. They analyzed the structure and characteristics of the iron films, to make sure that these truly were pure iron monolayers. Their observations agreed well with the theoretical properties derived from calculations.

They watched the samples for several minutes each to see how the microscope’s electron beam affected the formation of iron monolayers. They found that, under the beam, iron atoms on the graphene surface moved around and collected in small holes, where they formed crystalline monolayers within a few seconds. The monolayers were stable for several minutes under the electron beam. However, if they left the samples under the beam for longer than that, the monolayers began to collapse to form non-crystalline three-dimensional particles.

Now that they know how to make iron monolayers, the Dresden researchers and their colleagues can try this technique with other metals and see just how large they can make these metallic sheets.

The original research results were published here: Science 2014, 343, 1228–1232
I posted a more technical summary on the American Chemical Society’s Noteworthy Chemistry blog.

The Art of the Possible

I’ve been intending to start a lay-person’s version of my weekly postings on the American Chemical Society website. Several of my Twitter followers have mentioned that they are impressed, but confused, by these tech-heavy synopses, written for an audience of professional chemists. I thought I might wait for a week with an “accessible” topic, like medieval bones or counterfeit currency, but I decided that I might as well just jump in and start today.

I really picked a doozy of a week to start. This week’s post deals with an energy minimization study of silica- and germania-based zeolites. Say what?


Sodalite, a naturally occurring zeolite. Source: Wikipedia, used under Creative Commons Attribution Share-Alike License (posted by Ra’ike)

OK, let’s start with “zeolites”. These are inorganic materials, some found in nature, some made in a lab or a factory. If you’re going to get really picky about it, you can call them “microporous solids”. At the atomic scale, these materials form frameworks with a lot of open space, so you can use them as “molecular sieves”, or you can make them into containers for other molecules. Zeolites are great for catalytic converters in cars, for making all kinds of chemicals from petroleum, or for absorbing huge amounts of water in disposable diapers, among other uses.
Carbon dioxide molecules in zeolite cages. Source: Berend Smit laboratory, UC Berkeley, via Department of Energy’s National Energy Research Scientific Computing Center

Carbon dioxide molecules in zeolite cages. Source: Berend Smit laboratory, UC Berkeley, via Department of Energy’s National Energy Research Scientific Computing Center

Scientists have spent decades assembling databases of zeolite framework structures. Some of these structures already exist in the real world. Others “ought to” exist, from geometrical considerations, but they’ve never been made before. Every now and then, someone runs across one of these hypothetical structures, and they want to know whether it’s worth their time and effort to try and make the actual material. Maybe it has cavities of just the right shape, or it has channels of the right size, for some application that this person has in mind.

In addition to structural databases, scientists have access to databases that contain information on how stable certain chemical compositions and arrangements are. These databases are the results of years of research gained from chemical reactions, melting, compressing, and otherwise poking and prodding various materials. As a result, scientists can examine a series of framework structures to see which ones are the most stable.

A research group from Arizona State University recently published the results of their structural stability calculations, using hypothetical zeolite structures made from silica (silicon dioxide, the same stuff that beach sand is made of) and germania (germanium dioxide, which is chemically very similar to silica). (Chemistry of Materials 2014, 26, 1523–1527). They looked at framework structures based on building blocks containing one silicon (or germanium) atom surrounded by four oxygen atoms, each oxygen atom acting as a bridge between one building block and the next. These frameworks are referred to as “tetrahedral networks”. The group was especially interested in the angles formed by two silicon (or germanium) atoms and the oxygen atom bridging them, or T–O–T angles.

Schematic of tetrahedral building blocks. A silicon or germanium atom is at the center of each tetrahedron, and oxygen atoms are at each point. Source: Wikipedia, used under Creative Commons Attribution Share-Alike License (posted by Trikly)

Schematic of tetrahedral building blocks. A silicon or germanium atom is at the center of each tetrahedron, and oxygen atoms are at each point. Source: Wikipedia, used under Creative Commons Attribution Share-Alike License (posted by Trikly)

The research group did calculations for a series of known structures, distorting the structures over a wide range of T–O–T angles and watching to see whether the stability increased or decreased. For all of the silica structures, the structures became more stable as the angles increased from 120º to 140º. From 140º to 180º, the stability stayed about the same. Germania structures, however, were most stable in a narrow angular range between 128º and 130º.

Several commercially important zeolites have structures with T–O–T angles in a range that would make silica very happy, but would be an uncomfortable contortion for germania. In fact, of the 5824 stable frameworks described in the Atlas of Prospective Zeolite Structures, 994 have all of their T–O–T angles in the optimal range for silica (135º–180º), but only 48 are in the most favorable range for germania (117º–145º). Mixing germanium and silicon in the framework might extend the range of stable bond angles and increase the number of stable structures available for synthesis.