A matter of scale

Nanotechnology is big news, and increasingly hard to escape. But for process engineers, the science fiction-inspired worries about microscopic robots reducing all matter to 'grey goo' which so alarmed Prince Charles recently, and the serious work on producing minute mechanical components such as pumps and sensors (which would be the components of the fabled 'nano-robots'), are all beside the point. Although it's very different from the public perception of the subject, nanotech is part of the real world. And it presents some real processing problems.

Nanotechnology is a new name for a long-established field. Colloid science, used in the development of polymer suspensions, has been dealing with particles of this size for many years, while the chemical industry has long used tiny devices to manage the atom-by-atom construction of target molecules - in this case, of course, the devices are known as catalysts.

The current interest in nanotechnology is fuelling research into applications based on small-scale science, however, and the demand for nanoscale materials is set to increase.

Current industrial nanotechnology is mainly an offshoot of surface science. It's concerned with the behaviour of extremely small particles of matter, less than a micron in diameter, and sometimes so small that they only contain a handful of molecules.

The scale of such particles can confer surprising and useful properties. For example, some can self-organise into ordered two-dimensional structures - useful in coatings, and in toothpastes. Others, such as the pigment titanium dioxide, can be transparent, where larger-sized particles would be opaque. As the mineral absorbs ultraviolet light as a result of its sub-atomic structure, this makes it a useful - although currently controversial -ingredient in sunscreens. The challenge facing scientists and engineers alike is how to produce such small particles, and be sure that their proportions fall reliably within the correct size range.

There are two approaches to the problem of generating nanoscale particles. You can generate particles of the correct size from scratch, or you can make larger particles and reduce their size to the nanoscale by milling. These are known as 'bottom up' and 'top down' approaches, respectively. The former approach takes in techniques such as crystallisation, where particles are grown from a solution or melt; direct generation, such as the manufacture of 'buckyball' fullerenes and nanotubes from solid graphite; and techniques involving plasmas and atomic vapours.

'Bottom up' techniques aren't the sole preserve of synthesis labs, however. Grinding specialist Hosokawa, for example, has opened a Nano Particle Technology Centre in Minneapolis, which is investigating a technique which it calls Mechano Chemical Bonding. This, it says, is 'a novel technique by which materials can be chemically bonded together by using mechanical energy without any binders.' This, it claims, is best suited to producing nanocomposite materials such as ceramics and high-temperature superconductors.

Most of the engineering problems are to be found in the 'top down' technique of milling solids to nanoscale sizes, and the main problem is purely mechanical.

Most types of grinding and milling machinery work by squashing the particles between a fixed plate and a rotating element, but this cannot produce nanoparticles, as the machinery is not made with mechanical tolerances small enough to produce reliably particles of this size. This counts out devices such as hammer mills and knife mills, which are the usual choices for producing small particles.

Some technologies are still suitable, however. The key is that the grinding elements - the moving part of the device, as opposed to the static part - is not fixed to anything, but is free to move.

One such technology is the horizontal bead mill, such as the Dyno-Mill produced by New Jersey-based Glen Mills. This works on wet materials - that is, the solids to be ground have to be suspended in a liquid medium, although this need not be water or water-based.

The crucial section of the mill is the cylindrical grinding chamber. A shaft running along its central axis carries a series of equally-spaced agitator discs. The chamber is filled to around 80 per cent of its capacity with hard beads, which can be made from metals, alloys, ceramics, glass or tungsten carbide, among other materials. The slurry of the material to be milled is pumped in to occupy the rest of the capacity, and the rotor started up. The agitator discs force the beads to impact the solid component of the slurry over and over again, producing an extremely small particle size.

The Dyno-Mill can produce particles of a mean size of less than 500nm, which have widespread applications. Among these are pigment particles for paints and coatings, and for drug active agents, which are more likely to be absorbed the smaller they are.

Working along similar lines to the bead mill is the planetary ball mill, whose manufacturers include Verder subsidiary Retsch. This is a more complex piece of equipment - in place of the horizontal grinding chamber are four 'stations' - cup-like chambers known as planets - mounted on a circular platform called the sun disc. This disc rotates in one direction, while the cups rotate in the opposite direction. The cups hold one large or several small grinding jars, which in turn contain hard beads similar to those used in the bead mill.

As anyone who's been on a cup-and-saucer ride at a theme park will know, the forces imparted by a spinning vessel mounted on a counter-rotating plate are extremely large. Centrifugal forces fling the contents of the jars against their walls, and the combination of the two different spins changes the direction of the force - and therefore the trajectory of the grinding balls - constantly. This allows the mills to grind more intensively than most normal grinders.

Several models can grind wet and dry materials. Retsch's PM200 machine, launched at Achema in May, can grind colloidal materials to sizes below a micron.

Typically, mills for nanoscale production are small, generally desktop-sized, and handling capacities of less than a litre. But although they are small in capacity, they are being used for projects with extremely large scope. For example, NASA is using them to produce materials which could find uses in spacecraft.

The mineral gamma-titanium aluminide (g-TiAl) is hard and stable at extremely high temperatures, making it a possible ingredient for lightweight materials in high speed aircraft and reusable spacecraft, for example as the core of honeycomb sandwich structures. However, it is very difficult to work, because of its high ductility and low toughness at room temperature. However, researchers have found that the room temperature ductility of g-TiAl increases between 100µm and 6µm, but decreases between 200nm and 10nm.

Stephen Hales and Peter Vasquez of NASA's Langley Research Centre in Virginia are using nanomilling techniques to find out whether there is an ideal size between these two ranges which maximises the mineral's ductility. Their process uses a Fritsche Pulverisette vario-planetary ball mill to produce nanocrystals of g-TiAl with a diameter of around 15nm.

Another technology using the impact technique is the jet mill, where the particles themselves are both the grinding agent and the grinding element.

Originating as far back as the 1930s, jet mills are also known as micronisers. They work by 'firing' particles of crystalline or friable (crumbly) materials into a grinding chamber in a high-pressure jet of gas. Because the gas expands through a nozzle, it also cools down rapidly, removing the heat generated by the attrition of the particles.

A US-based jet mill manufacturer, the Jet Pulveriser Company, claims that its Micron-Master jet mill can grind materials such as molybdenum and paint pigments down to sizes around 200nm. The products tend to have a very large surface area, making them extremely reactive - ideal for the production of intermediates in solid form. In this case as many others, small is beautiful.