Today, chemists are spoilt for choice when they want to introduce energy into a system. In the past two centuries, we have learnt how to bring about chemical change using electricity, lasers, ultraviolet (UV) light, pressure, detonation waves, ultrasound, x-rays, and radio-frequency radiation, to name just a few. In the past 15 years, microwave energy has carved out its niche in the laboratory. A paper by Richard Gedye in 1984, in which he describes the rate enhancement of certain chemical reactions in Teflon pressure vessels placed in a domestic microwave oven, led to the explosion of interest in the technique, which continues today.

Chemists are often told that microwave ovens are 'tuned' so that water molecules absorb microwaves into rotational energy levels, and this causes molecular motion, and thus heating. This common misunderstanding comes from a failure to realise that while gaseous water has quantised rotational energy levels in the microwave region, in the liquid phase, the quantisation of rotational levels is, for all practical purposes, non-existent.

The easiest way to visualise the mechanism of microwave heating is to picture a microwave for what it is - ie a high frequency oscillating electric and magnetic field. Anything that may be electrically or magnetically polarised by this 'oscillating' field will be affected to some extent. Two principal heating methods exist - dipolar polarisation, and conductive heating.

Dipolar polarisation
If we consider a polar molecule - eg, water, methanol, THF etc - in a liquid, we know that intermolecular forces give rise to some degree of inertia in the liquid's motion. Microwaves heat polar liquids well because their electric field oscillates at such a frequency that the molecules cannot quite keep in phase with it. As molecules vainly attempt to follow the field, they collide with one another, and we observe heating in the sample. Radiation whose frequency is much higher or lower than this will not heat the sample. With higher frequency radiation, the field oscillates too quickly for the molecules to respond, and with lower frequency radiation, the molecules follow the field so well that no random motion is generated.

Even though there is a specific frequency at which this heating is most efficient, the range over which heating may take place is very broad (ca 1-100GHz for water, see Fig 1). The frequency of a domestic microwave oven (2.45GHz) is not selected so that it is at the maximum absorbency for water (something like 10GHz). If it were, you would find that most of the microwave energy was absorbed by the outer layers of your food, while the inside stayed unheated and so uncooked.

Fig 1. A plot of the dielectric loss tangent, tan δ (effectively a measure of microwave absorbance) for water versus wavelength.

Conductive heating
If we irradiate an electrical conductor or semiconductor with microwave energy, any mobile charge carriers (electrons, ions etc) move relatively easily through the material under the influence of the electric field. These induced currents heat the sample, owing to electrical resistance. If the sample is a metallic conductor, most of the microwave energy is reflected with relatively little energy penetrating beyond a few microns into the surface. However, colossal surface voltages may still be induced, and these are responsible for the dramatic electrical discharges that are observed when a metal is placed in a microwave oven.

Conductive heating can be demonstrated in a domestic microwave oven by using materials such as copper oxide or carbon. One should be aware, however, that these materials become very hot, very quickly, and that the electrical potentials induced in the materials can sometimes lead to dramatic (but otherwise harmless) electrical discharges. Alternatively, if pure water is heated in a microwave oven, where the polarisation mechanism dominates, the heating rate is significantly lower than that observed when the same volume of a dilute salt solution is heated. In the latter case, both dipolar polarisation and conductive mechanisms contribute to the heating effect.

Unfortunately, microwaves cannot be treated in quite the same way as a heating mantle, because of their long wavelength (12.2cm for a domestic oven). In any microwave oven, the microwaves are retained by the metal walls and there is interference of the waves as they are reflected off the sides of the oven. At some points the waves add together to give high intensity standing waves (antinodes) and at other points they cancel out (nodes). You can demonstrate this by using a microwave oven that has had its turntable removed. Place a large plate of evenly spaced marshmallows in the microwave and heat for ca 30s. Several of the marshmallows triple in size and are too hot to touch, while some remain at room temperature and are unaltered in size. Alternatively, use a piece of filter paper wetted with cobalt chloride solution. This compound is pink when surrounded by water molecules and blue when they are removed, eg by heating, and the cobalt chloride paper is dried out much more rapidly at the antinodes than at the nodes.

Although it is possible to modify a domestic oven for chemical syntheses, the wave nature of microwaves means that highly reproducible work requires slightly more sophisticated microwave applicators. In a well-designed system, energy can be imparted directly and efficiently into the reaction components, with little energy lost through reflections or through heating the reaction vessel.

For microwaves to be used as a practical heating method in the laboratory, or in industry, there have to be good reasons for choosing them over existing technology. Studies over the past decade have uncovered several reasons why microwave heating can be advantageous.

Superheating
Consider heating water in a round-bottomed flask by a heating mantle. Heat is slowly transferred from the glass to the core by convection, and boiling occurs when bubbles of vapour form at a nucleation site, a particle or a surface. Because we are heating from the outside in, the core of the water may be as much as 5°C cooler than the edge, even at the boiling point.

Microwaves, on the other hand, heat the water directly and almost uniformly. Under these conditions, the core is hotter than the outside because of surface cooling (often incorrectly expressed as heating from the inside-out), so that when the nucleation sites in the glass are hot enough to allow boiling, the core is some 5°C hotter. Thus by using microwaves, we can raise the effective boiling point of water by as much as 5°C, an effect known as superheating. (The reason why people are scalded as they add coffee to a cup of microwave-heated milk is that the grains of coffee provide nucleation sites on which bubbles form explosively.)

Solvents such as tetrahydrofuran or acetonitrile (ethanenitrile) exhibit superheating levels of up to 40°C. From a chemical perspective this is important. If we consider that, for an average reaction, a 10°C rise doubles the reaction rate, then simply using microwaves to heat a reaction can speed it up appreciably.

Selective heating
An important attribute of microwave heating is the ability to put energy directly into the reaction components, or to heat selectively one reaction component. Consider, for example, the direct synthesis of metal sulphides and selenides. To synthesise these materials, which are used as energy storage devices and as semiconductors, takes several days by conventional methods. The former involves mixing sulphur and the metal, both in powder form, and heating them in a sealed tube. The problem is that sulphur vaporises as it warms up, and if the temperature gets too high or rises too quickly, the pressure of the sulphur vapour will blow the tube apart. To avoid this, the mixture is heated slowly and cautiously, even though this means it may take a week or more for the ingredients to combine and form the metal sulphide. Microwaves, on the other hand, may be used to heat the mixture rapidly, and without fear of an explosion, because microwaves heat only the metal and not the sulphur. Sulphur vapour recondenses in the cool parts of the tube before flowing back to the hot metal. Instead of taking days, the reaction is complete in 15 minutes. This is also a visually stunning reaction because the microwaves stimulate a plasma glow in the sulphur vapour (Fig 2).

Fig 2. Preparation of chromium sulphide using microwave radiation

Ajay Bose at the Stevens Institute of Technology in New Jersey uses the preparation of aspirin in a microwave oven to demonstrate this effect. By the conventional method this synthesis takes 30 minutes, but in a microwave oven this is reduced to 90s (see Box 1). This improved reaction time is not a result of any 'non-thermal' effects, but is an example of how reaction times may be reduced by increasing the specificity and rate of heating.

Green advantages
While microwaves are both financially and energetically expensive to produce, the efficiency with which they can be used make them an attractive 'green' alternative to other forms of heating. Moreover, in recent years there has been a drive within the chemical industry to reduce both the production of waste products and the use of solvents. Waste products equate with wasted resources, and solvents can be toxic, flammable, and expensive to dispose of. Microwave chemistry provides a cleaner alternative, this time by exploiting the ability of microwaves to heat the reactants directly. Using only a minimal amount of solvent, the reactants are absorbed into a sponge-like support material (clays, aluminas, zeolites etc). The reactants are then heated directly with microwaves to generate the products, which are then extracted, again with a minimal amount of solvent. Because microwave heating is essentially uniform throughout the material, there is no time lost waiting for thermal conduction to heat the sample and, consequently, reaction times are often measured in minutes or even seconds.

A 'green' approach has been adopted by Chris Strauss, at the Commonwealth Scientific and Industrial Research Organisation (CSIRO) in Australia. Strauss and his team carry out organic reactions in supercritical water - water at high pressures and elevated temperatures - instead of organic solvents. Under these conditions, the properties of water change markedly from those that we encounter under ambient conditions, and it acts as an excellent organic solvent. The advantage is that the solvent is non-flammable, and when the reaction is completed, the waste solvent may be disposed of down the laboratory drains. Although this may be - indeed has been - performed using conventional heating, the use of microwaves is particularly convenient for generating these conditions because of the efficiency with which water is heated in the microwave oven.

That microwave chemistry is more than an academic interest has been demonstrated recently by the Dow Chemical Company in the US. Faced with tighter regulation of emissions from an existing agrochemical plant, Dow Chemical had the option of closing the plant down or cleaning it up. By switching to a 60kW microwave-based process, the plant has reduced its production of waste and unwanted byproducts, while increasing productivity and reducing energy costs.

Sometimes, microwave heating produces materials that cannot be easily produced by conventional means. One example from our laboratory is the preparation of needle-shaped iron oxide particles. Such particles are used in the manufacture of magnetic tapes and discs, and for audio cassettes and floppy discs. For information storage and retrieval reliability, particle size and shape must be controlled. Unfortunately, the industrial preparation involves several steps. Largely through the efforts of Gui-Hua Wang, a colleague from China, we found that microwave-heated iron salt solutions yield needle-shaped iron oxide particles. By contrast, conventional heating does not produce needle-shaped particles at any temperature under equivalent conditions.

Another example is that of a platinum carborane cluster compound that was prepared by a former colleague, Dave Baghurst, in Oxford. As with the iron oxide particles, extensive attempts to reproduce the results of the microwave-driven reaction using conventional heating failed. The question is why these products form when they do not form by conventional means. At the moment, it seems unlikely (though not impossible) that, in these cases, there is anything special about microwave heating. The direct and rapid nature of microwave heating means that reactants can cross large kinetic barriers at an earlier stage of the synthetic process than they do with conventional heating, making alternative reaction pathways more accessible. In other words, because so much energy is introduced to the reactants within a short time, microwave heating may lead to the formation of kinetic products over thermodynamic products.

All this leads us to the most contentious point about microwave heating - ie whether there is a specific microwave effect. By this, we mean an effect that does not arise from the heat from microwaves but from a direct interaction of the microwave field with specific modes of molecular or ionic motion. As anyone who has followed the mobile phone debate will be aware, this question is at the heart of low-frequency radiation safety, and the question of whether mobile phones can affect our health.

Specific effects, owing to the microwave electric field, have been documented in solids by several groups, most notably John Booske's group in the US, Monika Willert-Porada's group in Germany, and by our group in Edinburgh. All of these effects appear to be different manifestations of a so-called 'ponderomotive effect' - ie an effective increase in ion motion resulting from the influence of concentrated electric fields at grain boundaries in solids. Because ion diffusion is important in solid state reactions, microwave heating can be particularly advantageous here. There is evidence that, for a given reaction rate, the temperature can be up to 200°C lower when microwaves are used than when conventionally heated. This is useful because it lowers the processing temperatures and increases processing efficiency. The lower temperatures may be particularly useful for rapid preparation of thermally fragile materials, for example.

The microwave power used in mobile phones is several orders of magnitude smaller than that used in solid state reactions, and the electric field is correspondingly much smaller. Significant heating of biological tissue by mobile phone emissions is therefore not possible, and the most important question is whether the microwave energy is always equally partitioned through a molecules' vibrational and rotational modes. Normally, molecular vibrations redistribute energy at a rate several orders of magnitude higher than that at which microwave energy is injected. The question that we are currently investigating is whether, in large molecules such as proteins, the microwaves can pump energy into low-frequency vibrational modes at a rate that exceeds the uniform distribution of that energy. This is something that the protein is not designed to deal with, and could be compared with the destruction of the Tacoma Narrows Bridge, where a relatively light wind caused oscillations that destroyed an otherwise robust structure.

Practical microwave sources have been with us for less than a century, and microwave heating has delivered a lot to the chemist in just a few years. With an ever-increasing understanding of fundamental microwave-induced processes, we can look forward to more, better-targeted, applications of low-frequency radiation.

Dr Gavin Whittaker is a lecturer in the department of chemistry at the University of Edinburgh, King's Buildings, West Mains Road, Edinburgh EH9 3JJ.

Go to Education in Chemistry September 2002 issue

Go to Education in Chemistry back issues