91av

“I FELT fine until I passed out,” says Edward Cussler of the University of Minnesota in Minneapolis. Back in 1980 he had built an artificial gill that extracted oxygen from water. His dream was that the device would one day allow him to breathe just like a fish, giving him an unlimited supply of oxygen while diving.

“I wanted to swim around like a fish,” he says. But when he tried out the gill in his lab, he found it didn’t provide enough oxygen to support him. “We didn’t have quite the capacity to keep me alive,” he recalls.

Cussler is not the first to dream of breathing like a fish. In 1962, underwater exploration pioneer Jacques Cousteau predicted the arrival of Homo aquaticus: people surgically equipped with gills. “The lungs will be by-passed and he will be able to live and breathe in any depth for any amount of time without harm,” said Cousteau.

Of course, most of us would prefer not to have to resort to surgery. So could new technological developments solve the problems that confounded Cussler and allow us to create a practical artificial gill? It’s not just divers who would benefit. Artificial gills could have many other uses, from supplying oxygen for submarines, to powering fuel cells.

Surprisingly easy

Making a crude artificial gill is surprisingly easy. All you need is a watertight box made out of a membrane that is highly permeable to gas. Fill it with air and put it underwater and you’ve got a gill. The levels of oxygen and CO₂ dissolved in water are in equilibrium with the atmosphere above it, so diffusion through the membrane will result in concentrations close to those of the atmosphere inside the box. If the oxygen level in the box falls, more oxygen will diffuse in from the water, and any excess CO₂ will diffuse out.

In 1961, just months after creating the first highly permeable silicone membrane, Walter Robb of General Electric built a gill that could support a hamster. “You just had to keep the water moving,” he recalls, so that water high in oxygen and low in CO₂ was always in contact with the membrane.

Humans, needless to say, need a lot more oxygen than hamsters. There is not nearly as much oxygen in water as there is in air – just 4 to 6 millilitres per litre of seawater, depending on the temperature. So to get a good flow of oxygen you need not only a good flow of water over the membrane, but also a large surface area to extract a greater volume of oxygen at once. The minimum needed is 80 square metres, according to physiologist Charles Paganelli, who also experimented with artificial gills in the 1960s.

To make gills practical, this surface area must be squeezed into a small space, in the same way that the alveoli give our lungs a big surface area. Artificial lungs for oxygenating blood during heart bypass operations or after severe lung damage have long been used in hospitals. Blood is pumped through a network of tiny gas-permeable tubes – which pack a large surface area into a small space – while air is blown around them. Oxygen diffuses through the walls of the tubes and into the blood, while CO₂ flows from the blood into the air.

Coffin-shaped

Cussler realised he could create an artificial gill by cobbling together several of these artificial lungs. Pass breathing air through the gas-permeable tubes while pumping lots of water over them, and oxygen will diffuse into the breathing air. Though his gill could not supply enough oxygen for a human, Cussler proved it worked when he lowered his wife’s beloved terrier, Muggins, into the Mississippi river in a sealed metre-cubed box hooked up to his gill. “I’ve never been so worried about an experiment in my life,” he says. Luckily for his marriage, it worked. Oxygen levels stabilised at around 16 per cent, and Muggins could have survived indefinitely, though Cussler let him out after 3 hours.

“He lowered his wife’s beloved terrier, Muggins, into the Mississippi river in a sealed box”

A Japanese company selling state-of-the-art silicone membranes took over where Cussler left off. In the 1980s, Fuji Systems of Tokyo developed a series of prototype gills for divers as a way of demonstrating just how good its membranes are. The early versions resembled a small fridge strapped to a diver’s back, while the most advanced prototype, called the Donkey III, consists of a coffin-shaped box that has to be pushed in front of the diver. It’s huge, but it does work. In a televised demonstration in 2002, it supported a diver in a swimming pool for 30 minutes.

Size is not its only problem. Normal air is 21 per cent oxygen. Like Cussler’s gill, Donkey III maintains oxygen levels in the breathing air around 16 per cent. Such low levels of oxygen can impair people’s ability to think clearly, which is hardly desirable when diving, where poor decisions can be fatal.

Artificial fish blood

So gills that rely on diffusion alone are not good enough. Some way is needed of boosting the concentration of oxygen. We know it is possible: fish inflate their swim bladders – which keep their buoyancy neutral – with pure oxygen they extract from water.

In the 1980s, Joseph and Celia Bonaventura at Duke University in North Carolina showed that fish do this with the help of a pH-sensitive form of haemoglobin, the oxygen-carrying blood protein. When cells around the swim bladder release lactic acid into the bloodstream, the drop in pH triggers a release of oxygen into the bladder, maintaining its volume when a fish dives deeper.

The couple realised they could create artificial gills for a variety of purposes by mimicking this process. Instead of fish haemoglobin, they planned to use synthetic chemicals that bind strongly to oxygen but release it when they pass over an electrode. The gill the Bonaventuras designed for divers consisted of two loops. In the first loop, the artificial haemoglobin would extract oxygen from the water. At the other side of the loop it would flow over an electrode and release this oxygen which would pass through a membrane into a second loop carrying the breathing air. “All our calculations showed it would work,” Celia Bonaventura says.

But the plan never made it off the drawing board. The technology was eventually sold to a company interested in making blood substitutes rather than gills, and the two-loop gill died. Well, almost. A team at Waseda University in Japan has been experimenting with simpler systems for several years (see Diagram), using either haemoglobin or perfluorocarbon, an inert fluid in which oxygen is very soluble, to extract oxygen from water and transport it to the breathing-air loop (Journal of Membrane Science, DOI: 10.1016/ j.memsci.2005.01.008). “We think it would be possible to develop a reasonably sized gill,” says team member Kenichi Nagase.

Fundamental problem

But Cussler and Paganelli are not convinced that two-loop systems are the way to go. They say there is a fundamental problem with membrane-based gills, one first identified nearly a century ago by Richard Ege, a Danish physiologist who studied diving beetles.

Diving beetles regularly come up from the bottom of the pond to grab a bubble of air which they place under their wing covers, or on the tip of their abdomens, over the holes through which they breathe. These bubbles are more than just a fixed air supply like a scuba diver’s tank. The surface of the bubbles acts like gas-exchange gills: oxygen diffuses into the bubble from the water, while CO₂ diffuses out. But if the bubble acts like a gill, why do most diving beetles have to keep making risky trips to the surface?

The answer has to do with the effect of the increasing pressure on the air bubble as the beetle descends. This compresses the air in the bubble and also makes the gases more soluble in water. The combined result is that the bubble shrinks until it is either too small to work as a gill, or collapses entirely. That’s when the beetle needs to replenish the bubble.

The same problem applies to artificial gills. As the diver descends, the higher pressure will both compress the breathing air and make more of it dissolve in water. Most of that lost gas is nitrogen, since it constitutes nearly 80 per cent of air. To prevent the lungs collapsing, the gill will have to pump more oxygen into the breathing air. The proportion of oxygen will therefore increase, which poses a problem because pure oxygen becomes toxic at depths of just 9 metres. So in an ironic twist, it appears that if you get your oxygen from water, you have to carry a supply of inert nitrogen to top up your air supply and avoid oxygen poisoning as you dive deeper.

Air extractor

The problem is hard to avoid. Two-loop gills reduce nitrogen loss but do not eliminate it. Even if you could completely prevent loss through the gill, divers would still lose nitrogen through their skin.

Enter Israeli inventor Alon Bodner, who unveiled a novel approach last year. Instead of a membrane gill, he plans to use an industrial process for separating gases from a liquid, based on the principle that if you lower the pressure of a liquid, for example with a centrifugal pump, the gas dissolved in it bubbles out (see Diagram). Bodner claims his battery-powered device will be able to extract virtually all the air dissolved in water. With seawater, this would typically yield a gas containing 34 per cent oxygen. Crucially, because Bodner’s device extracts nitrogen as well as oxygen from the water, nitrogen loss is not an issue.

But there is a problem with this approach. The system would have to process more than 1000 litres of water per minute just to provide enough air for a diver to breathe at the surface. Descend 10 metres and the pressure doubles, so you need to extract twice as much air to provide the same volume. Go deeper and you need to extract even more. The only way to make the system practical is to make it part of a rebreather.

The vast majority of the compressed air carried by scuba divers is wasted, bubbling back to the surface with most of the oxygen unused. In rebreathers (and in membrane gills) the air is recycled, with the lost oxygen replenished and CO₂ removed. Rebreathers give divers hours of bottom time with just a small oxygen tank. So Bodner plans to make his system part of a rebreather, which means it would only need to extract all the air from 200 litres of water a minute. Because the system would add oxygen-rich air to the breathing air rather than pure oxygen, nitrogen would have to be vented periodically to prevent its build-up, but this is already done in existing semi-closed-circuit rebreathers.

Scrubbing CO₂

However, the limiting factor with rebreathers is not carrying oxygen but getting rid of CO₂. The canisters of soda lime that “scrub” out the gas last only a few hours. They cannot be reused and are expensive to replace. So while Bodner’s approach does solve some of the big problems with membrane gills, it also sacrifices their great strength: they excel at getting rid of CO₂.

Membrane gills have been considered as CO₂ scrubbers to replace the chemicals used in rebreathers and small submarines, and as emergency systems for larger submarines. In 2003, a feasibility study for the US Office of Naval Research recommended further trials of membrane gills as an emergency scrubber system. One of the researchers, Dan Warkander of the University of Buffalo, New York, thinks that it should also be possible to develop a membrane gill small enough for divers to use as part of a rebreather. This would provide “infinite scrubbing capability”, he points out.

The most likely use of artificial gills in the near future is to supply oxygen to fuel cells for powering underwater machinery. Prototypes have already been developed. At the moment, remotely operated vehicles and submarines powered by fuel cells carry liquid oxygen with them. Taking it from the sea instead would allow these vehicles to carry more fuel.

And in the future, artificial gills might be used to supply submarines or underwater habitats with breathing air. Bodner sees this as the most likely use for his device.

Avoiding the bends

But for divers, without some amazing technological advance such as membranes that can actively pump specific gases in or out, the only likely use for artificial gills is as CO₂ scrubbers. Bodner and others may well produce working prototypes. But they are not going to catch on with divers unless they are smaller and safer than scuba sets and rebreathers, or offer a big advantage.

The main advantage is supposed to be an unlimited supply of air, but all proposed designs rely on limited-duration batteries or fuel cells to power pumps. And even if you could drastically reduce the power requirements, most divers don’t need or want to stay down for more than the 12 or so hours already possible with some rebreathers.

In fact, the main reason for long dives is to avoid getting the bends. The bends are caused by exactly the same phenomenon that causes nitrogen loss in membrane gills: put a gas under pressure and more of it will dissolve in blood or water. Surface too fast and potentially deadly bubbles form in your tissues. In joints, they cause excruciating pain and damage cartilage. In blood, they can block capillaries in the brain. The longer you stay down, the more slowly you have to come up.

So perhaps the inventors of artificial gills are trying to solve the wrong problem. What’s needed is not a way to get air from water so divers can stay down longer, but a way to avoid the need for decompression. And that means eliminating the need to mix oxygen with an inert gas like nitrogen.

Liquid breathing

One way would be to replace the inert gas with an oxygen-carrying liquid. Liquid breathing, as made famous in the movie The Abyss, would transform diving. We could descend to extraordinary depths and then ascend as fast as we liked. But while medical trials suggest it is safe, it is not as easy as the movie suggested. Human lungs cannot move a dense fluid in and out fast enough to provide enough oxygen, so you need tubes down the throat to circulate the liquid.

“Liquid breathing, as made famous in the film The Abyss, would transform diving”

Ronald Hirschl, a surgeon at the University of Michigan, Ann Arbor, is developing the technique to treat various lung conditions. “None of our patients were able to tell us what it was like, as they were unconscious,” he says. But he thinks it would probably be extremely unpleasant. Despite this, Hirschl thinks it might one day be possible to adapt it for divers. “The bottom line is that right now we are using it for sick lungs. Would it ever be used for situations where lungs are normal? Possibly. That’s much more futuristic, but I think it is realistic.” And when liquid breathing is perfected, perhaps artificial gills will supply the oxygen. It’ll happen some day. Just don’t hold your breath.