This article is reproduced with the permission of New Scientist for exclusive use by Nova users.

The great nanotech gamble
14 July 2007
From New Scientist Print Edition.
Karen Schmidt

"A revolutionary and superior bat with the widest sweet spot ever." So says the marketing blurb on the Easton Stealth Comp CNT baseball bat. I'm shopping for a lightweight bat for my 9-year-old son, and the Easton seems to hit the mark. CNT, you see, stands for carbon nanotubes, and it is this artificial material embedded in the bat that makes it super-stiff yet ultra-light, perfect for my budding Babe Ruth.

The "nano" raises a nagging doubt, though. I've read news reports that suggest carbon nanotubes may be harmful. But if they are in consumer products, surely they've been tested and verified as safe - haven't they?

Carbon nanotubes are just one example of a whole range of new manufactured materials, collectively known as "nanomaterials", that are starting to be used in everyday products. The US National Nanotechnology Initiative defines a nanomaterial as a substance that has at least one critical dimension less than 100 nanometres and possesses unique optical, magnetic or electrical properties. Nanomaterials can have properties that are quite different from those of otherwise similar materials made up of larger particles. Silver nanoparticles, for example, react with hydrochloric acid, something that bulk silver doesn't do.

Nanomaterials are creating a boom industry, often touted as the "nanotech revolution". From carbon nanotubes used to make strong, light materials to silver nanoparticles that function as antibacterial coatings, and titanium oxide nanoparticles to make transparent UV-filtering sunscreens, you are likely to encounter nanomaterials sometime soon - if you haven't already.

Nanotech on the brain

There's a catch. If these particles escape into the environment, their very smallness means they could have as yet unknown and possibly damaging effects. You might inhale or swallow them, or they could collect on the skin. They could then be carried to major organs such as the heart, liver and even the brain. The consequences of all this are still not clear, but following past health disasters caused by substances such as PCBs and asbestos, the prospect has stirred concern among governments and scientists alike.

So in recent years researchers have begun investigating the potential effects of nanomaterials. Proponents, eager to allay safety fears, say that this time we can "do it right" with no repeat of past fiascos. It seems as if barely a month goes by without another report laying out a strategic plan for the development and safe deployment of nanomaterials, and you might be forgiven for thinking that the nanomaterials now finding their way into consumer products are safe for human use.

Yet a little digging reveals things aren't so straightforward. There are significant gaps in the knowledge needed to know what the real effects of nanomaterials are, and they could take decades to fill.

In a review of the UK's policy on nanotechnology published in March 2007, the Council for Science and Technology, which advises the British government, concluded that the government has "placed insufficient emphasis on the need to investigate health, toxicology and environmental effects of nanomaterials". The problem for all governments and regulators is that working out the health risks of nanomaterials is devilishly difficult. The diversity of chemical compounds used to make nanomaterials, coupled with the huge variety of properties that nanomaterials have, means that no one even knows how to classify them in a way that allows general conclusions to be drawn from studies on particular nanomaterials.

This has led to suggestions that we may have to develop a new approach to chemical toxicity that will account for nanomaterials. The first step is to figure out whether a particular nanomaterial can ever be harmful. "Without that piece of knowledge, all the rest is guesswork," says Andre Nel, a toxicologist at the University of California, Los Angeles. Researchers usually answer that question for particular chemicals by exposing animals to increasing doses and looking for signs that this is doing some harm. Such studies are open to criticism for being too artificial to be relevant to the real world, but when multiple studies turn up similar results - showing, say, that a pesticide kills cells - it is taken as pretty good evidence that the chemical in question is worth worrying about.

This approach falls short, however, when testing nanomaterials, because here it is not only chemical composition that matters, but also a particle's size and physical properties. This makes it hard for toxicologists to know precisely which material they are studying, as nanoparticles can exist in myriad forms. One moment they are individually suspended in solution, the next they are clumping together or picking up contaminants. Not only do they change size and shape, their surfaces can differ and their crystalline structures vary. Each of these characteristics can affect their reactivity and so their ability to interact with living things. "A small change in experimental conditions can lead to huge differences in outcome," Nel says.

To take one instance, in January this year there was a report from Swiss researchers which said that rope-like agglomerates of carbon nanotubes were more toxic to cells than the dispersed particles are (Toxicology Letters, vol 168, p 121). Then in February this year, researchers at Rensselaer Polytechnic Institute in New Jersey reported experiments that suggest the opposite: finely dispersed carbon nanotubes, even at low concentrations, were more toxic to cells than were larger clusters (Toxicology Letters, vol 169, p 51).

Such apparently contradictory results call into question whether these two - or any two - research groups are actually studying the same materials under similar conditions, especially as there are no standard ways to describe or identify different nanomaterials. "The nomenclature issue is huge," says Nigel Walker, who coordinates nanomaterials research for the US National Toxicology Program. Even if a regulatory body draws up what seems to be a complete description of a nanomaterial it wants to control, there are so many possible variables that the door will remain open for manufacturers to split hairs and say they are making something slightly different. "You could get into some really hot water legally," Walker says.

Establishing a system for characterising and naming nanomaterials remains a daunting challenge, he says. Consider single-walled carbon nanotubes. These structures resemble graphite - sheets of carbon atoms in a chicken-wire arrangement - that has been rolled up into a tube a few nanometres wide. Tweak the manufacturing process and you can create around 50,000 different versions of the material. Multi-walled carbon nanotubes - tubes stacked inside one another like a set of Russian dolls - also exist in myriad versions. If one is found to be toxic, that doesn't necessarily mean that the others will be.

No one seriously suggests embarking on the mammoth task of testing each and every kind of carbon nanotube for toxicity using the classic animal testing regime. Instead, the aim is to develop a way of predicting the hazards that nanomaterials pose. The National Toxicology Program is examining classes of nanomaterials and trying to figure out which physical and chemical properties distinguish the toxic ones from the benign. The aim is to apply this information to other members of the class to predict which of them will be toxic. Even better, it might be possible to use this information to engineer nanoparticles to be non-toxic from the word go.

So far, researchers have focused on three types of nanomaterials that they think may be toxic: carbon nanotubes, the spheres of 60 or more carbon atoms known as buckyballs, and metal oxide nanoparticles. They already have clues as to what features are likely to make them hazardous. For instance, the toxicity of carbon nanotubes seems to be related most closely to their length and surface characteristics. For buckyballs - which often have chemical groups attached to them - particle size and surface chemistry seem most predictive. For metal oxides, such as titanium dioxide, the key feature appears to be crystal structure. "We have some understanding of mechanisms for certain nanoparticles," Nel says. "We know what is a dangerous particle, the principles by which they function, and some of the tissue responses."

Alarm bells

Tests on carbon nanotubes invariably ring alarm bells. No matter which form is examined, the results suggests that many materials in this group have the potential to be toxic.

The obvious comparison is to asbestos, which is also a fibrous material and is known to cause a kind of lung cancer called mesothelioma. Asbestos is carcinogenic because the fibres are long, thin and can't be broken down in the lungs. They cause inflammation that damages lung tissue over many years. In a major review, published last year, of both animal and lab studies that investigated whether carbon nanotubes are toxic, Chiu-wing Lam, a toxicologist at NASA's Toxicology Group at the Johnson Space Center in Houston, Texas, concluded that carbon nanotubes could produce inflammation in the lung leading to granulomas, a kind of scar tissue that damages lung function.

However, in a debate comparing the potential effects of carbon nanotubes and asbestos on health, a panel of researchers at the Society of Toxicology meeting held in March this year in Charlotte, North Carolina, concluded that they are probably more different than alike - largely because asbestos is stiff and forms splinters, while carbon nanotubes tend to be flexible and scrunch up into a ball.

Don't breathe easy yet, though. The panel said they are more similar to ultra-fine particulates of like size, such as those emitted in diesel engine exhaust. Breathing the very small particles contained in exhaust fumes and smoke is well known to cause health problems, such as damage to the cardiovascular system. Now a study by Petia Simeonova and colleagues at the US National Institute for Occupational Safety and Health (NIOSH) in Morgantown, West Virginia, suggests that inhaling carbon nanotubes might have a similar effect. The researchers injected a dose of single-walled carbon nanotubes - stripped of metal impurities - into the lungs of mice. When they looked at the lining of their aortas, they detected signs of free radicals, which are capable of damaging cells and tissue. Animals genetically susceptible to atherosclerosis also developed more arterial plaques, which are known to cause heart attacks and strokes. The ultra-fine particulates in air pollution may cross into the bloodstream and directly inflame and damage blood vessels, and the NIOSH group is now studying whether carbon nanotubes might do the same.

The similarities that are emerging between carbon nanotubes and ultra-fine particulates have raised the intriguing idea that they are not merely similar, but are actually the same thing: that carbon nanotubes are in fact the toxic component in diesel exhaust and other pollution. Lawrence Murr, an environmental and materials scientist at the University of Texas, El Paso, has already found these nanomaterials in urban air samples and in emissions from gas stoves and wood-burning fires. "We began to see multi-walled carbon nanotubes essentially everywhere we looked," he says. "They're part of the combustion regime." The same goes for buckyballs.

Whether the nanomaterials that Murr finds in the environment are the same as those made in the laboratory, however, with similar health effects, remains to be seen. Engineered carbon nanotubes are certainly more homogeneous and pure; that's what makes them more useful than plain old soot. Lam now plans to do animal studies to test the health effects of carbon nanotubes collected from urban air, and will compare the results with those of earlier studies testing the engineered versions.

If carbon nanotubes do turn out to be highly toxic, the urgent question will be to assess how likely it is that people will come into contact with them, now and in the future. So far, consumers only encounter carbon nanotubes in products like that baseball bat, where they are embedded in some hard and durable material. Even so, a white paper on nanotechnology published in February this year by the US Environmental Protection Agency warns that when such objects are eventually discarded, nanotubes could start to disperse into the environment as the material containing them breaks down.

And what of the people working in factories where mixtures containing carbon nanotubes are handled and mixed? Vincent Castranova, coordinator of NIOSH's nanotoxicology programme, warns that no one knows what industrial users are doing with nanomaterials and whether workers in these plants are handling them safely. "It's very difficult to find out how products are being made, what the processes are, and what the hotspots of exposure are," he says.

Likewise, it is unclear how consumers might be exposed to nanomaterials. Windows, fabrics and even railings in subway stations are now being coated with various nanomaterials to make them antibacterial, self-cleaning and more durable. Could they be shedding nanoparticles as a result of normal wear and tear? In tests at the Industrial Technology Research Institute in Taiwan, Li-Yeh Hsu and Hung-Min Chein found that when they mimicked the action of sunlight, wind and human contact on coatings containing titanium dioxide nanoparticles, some particles did escape, particularly from coated tiles (Journal of Nanoparticle Research, vol 9, p 157).

Understanding the routes of public exposure to engineered nanomaterials is going to be tricky, yet some experts argue it is the most important knowledge gap to fill. "One can be led down all sorts of blind alleys," says Martin Philbert, a neurotoxicologist at the University of Michigan, Ann Arbor. "You could overreact to extremely toxic materials that will never reach high enough concentrations in the environment, and ignore seemingly benign materials that will be released in very large quantities."

Carbon nanotubes appear so far to fall into the category of materials with potentially high toxicity but low levels of exposure, but more exposure might be coming as they appear in an increasing number of products. That's why earlier this year both the US Environmental Protection Agency and the European Commission called for a life-cycle approach to assess the long-term safety of products containing nanomaterials.

Nanomaterials hold huge promise in a wide range of applications - from solar cells to drug delivery - so let's hope we can pin down any potential risk and make sure that they can be used safely. For that to happen, a lot of questions over their potential health effects have still to be answered. So for now, my son will be developing his baseball talents with a good old-fashioned aluminium bat.

Karen Schmidt is a California-based writer and host of the podcast Trips to the Nanofrontier

From issue 2612 of New Scientist magazine, 14 July 2007, page 38-41

For the latest from New Scientiist visit www.newscientist.com