Welcome to modern nanorobotics, where robots are made of chemicals and magnets and powered by heat and light. Recent advances could revolutionise everything from healthcare to cleaning up oil spills.
Microgrippers are strange little critters. They look a bit like a six-armed star fish and measure no more than 4.5mm tip-to-tip when open. But these devices can grab tiny objects and move them from place to place.
Presented this January by a team from the Johns Hopkins University in Maryland, what makes them really clever is what they're made of: hydrogel – a gel that readily absorbs water – blended with a stiff, segmented polymer, and embedded with iron oxide nanoparticles. They open and close in response to temperature, while the iron oxide makes it possible to move them around very precisely using magnetic coils.
"Have you seen those toys that grow very large when you put them in water?" says David Gracias, one of the study's co-authors at Johns Hopkins University. "Those are hydrogels. They're just polymers with space that can be filled with a lot of water."
These little grippers are a perfect example of where nanorobotics is heading. In the science fiction movie 'Fantastic Voyage' a medical team in a shrunken submersible navigates a patient's body to repair damage to his brain. Real scientists, too, have been working on building robots on a similar scale – and they've realised tremendous progress during the past decade.
The alliance of medicine and 'soft robotics' (the idea that working robotic devices can be made from pliable materials) has been key. Robotics has traditionally been the domain of physicists and engineers, but they've had to learn new skills, transferring their knowledge to biological environments.
Gracias is an engineer at the university's department of chemical and biomolecular engineering, and microgrippers are his baby. The reason why he and his team are relying on chemicals, and why they've become the go-to solution for nanorobotics researchers around the world, is that they solve some of the big problems of traditional robotics. It's easy to conceive of a robot that could navigate inside the human – it's been a staple of science fiction for decades – but actually creating one is somewhat trickier.
"One of the challenges with small devices is: how do you power them?" Gracias says. "Batteries are typically challenging to miniaturise, and wireless energy isn't very effective at that scale."
He explains that hydrogels not only let them activate the grippers without the presence of traditional power, but also removed the problem of getting them back afterwards: they simply safely dissolve in a living body once they've performed their task. "They are biodegradable. Our initial devices were made out of inert metals like gold, but there was a concern that machines this small might get lost. We didn't want them to migrate somewhere that could prove difficult for the patient," says Gracias.
His team is not alone in aiming to get bots into your body. A number of researchers are looking into using tiny chemical-based devices that could be inserted and guided by magnets or heat or light. They could deliver drugs, move individual cells around, guide catheters through the body. After they've done their job, they'll dissolve; there's no point retrieving them when you can 3D-print millions more.
"Microtechnology and nanotechnology have reached a space where people can actually build these things," says Gracias. "The tools for fabrication and design have evolved. The first generation was passive – they don't have any moving parts. But the generation coming online now are devices like ours, which have things that move."
There's still a long way to go, but a team at the University of California San Diego recently performed the first-ever 'in vivo' test, where a swarm of devices were inserted into a living organism. In this case, the organism was a mouse, and the devices in its stomach were no more than 20μm in size. The nanobots were made of zinc, and propelled themselves by reacting with the gastric acid in the mouse's stomach to produce hydrogen bubbles. "All the materials are based on polymers and zinc, which are all biocompatible," says nanoengineer Joseph Wang. "The zinc dissolves, and it's an essential element anyway. We want to expand it for a large-scale animal study, and test it with therapeutic drugs."
At the scale of Wang's zinc devices, the environment is different from what one might expect. "Biological tissue is not like water," says Peer Fischer, a physicist at the Max Planck Institute in Germany. "It's a complicated, mesh-like network of polymeric macromolecules. In order to navigate these, it is actually advantageous to have a very small structure because it can sneak through this network. A much larger structure will get entangled, and you need much larger forces to move it."
Bubble propulsion is one way of navigating this environment. Moving devices around with magnets is another. But Fischer's team has come up with something a little more elegant, and it's a design that demonstrates the new directions that researchers are taking.
The fluids in the human body are non-Newtonian fluids: their viscosity is dependent on the velocity of objects travelling through them (for water, the viscosity stays the same). Their device, which is 200μm in size, is scallop-shaped, with two 'wings'. It exploits the change in viscosity to move through the non-Newtonian fluid by varying its propulsion rate – in other words, varying the speed of the opening and closing movement of its wings. That's something that doesn't work in water, but it's perfect for most of the complex fluids found in the human body."The flapping motion is perhaps best understood in that the scallop is a polymeric structure," says Fischer. "The body likes to be straight – that is, an open scallop shape. At the tips of the two shells are tiny magnets, which are arranged in such a way that if a magnetic field is applied, they can be aligned with the field, which closes the scallop." (Fischer points out that the scallop bots aren't designed to be used in the body; they were simply built to show that simple actuators can be used to get propulsion in the fluids of the body and tissue.)
The scallop also shows how technologies like 3D printing are being put to good use in nanorobotics, reducing the costs of production. Fischer says that producing the devices this way is extremely convenient. They've even been taking it one step further, creating tiny corkscrew nanobots with a fabrication method that can produce a hundred billion of them in an hour. Fischer's team places a substrate in a vacuum chamber at an angle, and seeds it with nanoparticles of 5nm to 10nm in size. The substrate is then blanketed with growth material. "It's like snow on a tree," Fischer says.
The big goal in the next few years is to actually get these devices to do useful things inside a living body, on a regular basis. "We need to add some kind of intelligence to these devices," says roboticist Brad Nelson at the Institute of Robotics and Intelligent Systems at ETH Zurich. "We need to make them respond to their environments in some way. Bacteria do that with chemo-receptors and little flagellar motors (molecular engines). Maybe ours could be guided by temperature changes or a change in acidity. That's the challenge: how do you get these things to somehow change their motion?"
Nelson is another example of how engineers are colonising the field: he's a mechanical engineer with a PhD in robotics, and he's become one of the foremost experts in the field. "I am an engineer and I don't come at this from a medical side," he says. "Our specialty is how to make small things move very precisely, in controlled ways. When we start thinking of applications, surgeons and doctors understand the potential and are interested in the technology... We start developing relationships with them. They make the call on the medical side as to whether this makes any sense. I just tell them the possibilities."
Drug delivery is one of the more interesting applications of nanorobotics. It has yet to be fully realised, but it's getting there. The idea of delivering drugs to a very precise point in the body isn't all that exciting – until you start thinking about what it actually means.
Imagine if you had cancer, and required chemotherapy. Currently, you'd need to go in for several treatments to deliver vast quantities of highly toxic drugs into your bloodstream, eliminating healthy cells along with cancerous ones. If you could deliver the drugs directly to the cancer site, you'd not only avoid the deaths of healthy cells, but you'd also need to use a fraction of the drug currently required.
"It's a non-linear law," says Nelson. "If the drug is ten times closer to the location where I want to be – say, instead of a centimetre from the location, I can make it a millimetre – I don't just need ten times less drug, I need a hundred times less. If I go from that centimetre to a micron, I've gone a thousand times closer, so that means only a million times less drug for the same efficacy. The drug could be toxic to the rest of your body, but if it doesn't reach it, that's OK."
The implications are, then, potentially very significant. It might change what patients pay for drugs, or how it might affect the profits of the pharmaceutical multinationals that produce these drugs. Nelson says that this is still a very long way off, and that it's never going to be quite that simple. He also points out more fundamental challenges that researchers will need to overcome. "These things are small: how are you going to get enough drugs on them? How do you get it to diffuse off at the rates you want? These are not unsolvable problems, but they are questions that people are thinking about."
And in-body applications aren't the only goal here. Nanobots could conceivably be used to deal with environmental disasters. Nelson says he's had interest from oil companies, asking if nanobots could be unleashed on an oil spill, converting the oil molecules into harmless substances. Also, he says, they want to send nanobots down into the ground, navigating to places where more traditional machines can't go, in the hope of discovering new sources of oil.
"We started thinking about microswimmers around a decade ago, without any real idea of the physics or the manufacturing," says Nelson. "I guess what surprises me the most is how much progress we've made in just a decade. I'm anxious to see what the next ten years will bring."