The world’s smallest light-trapping silicon cavity

Nick Petrić Howe

Welcome back to the Nature Podcast, this time: how to craft a photon-trapping gap…

Shamini Bundell

…and is it possible to reduce poverty without increasing emissions? I’m Shamini Bundell...

Nick Petrić Howe

...and I'm Nick Petrić Howe.

Nick Petrić Howe

First up this week, there's a paper in Nature about a tiny bowtie-shaped structure that researchers have made between two pieces of silicon. In addition to being tiny, at the centre of the tie is a cavity, a hole, which is the smallest of its type ever made, just two nanometres across – around the width of a DNA molecule, and it can be used to trap light. Now, trapping lights may sound like a strange way to spend your time, but doing so — even fleetingly — can increase its interaction with matter, which allows the researchers to use it in different ways. They can even use the trapped photons bouncing around inside to produce more photons, which could then be used in things like quantum computing. But none of this is possible without a cavity.

Søren Stobbe

So, a cavity in its simplest form is two mirrors facing each other and light can bounce back and forth. But it is interesting to make them as small as possible, because the smaller they are, the stronger the light-matter interaction.

Nick Petrić Howe

This is Søren Stobbe, one of the authors of the new paper. And when he says small, he means really small. Now, the size of the cavity you need actually depends on the wavelength of light you're trying to capture, but the smaller the cavity the greater its ability to trap light, so generally you want something that's only a few nanometres wide.

Søren Stobbe

And it has been understood for nearly two decades that it's possible to build structures that really confine light to extremely small dimensions. However, it turns out that to reach that regime, you need to make extremely small structures, you really need to push the boundaries of what can be done with nanolithography.

Nick Petrić Howe

Nanolithography is a method that lets you create patterns on very tiny materials. You can think of it as putting a kind of stencil on top of a surface to protect the area you want for your desired pattern, and then etching away everything else. In this case, you'd want a stencil with a bowtie-shaped gap in the middle, which you could then blast away using ionised gas to create your cavity. But you can only get so small with this technique. Previously, the smallest cavity the team produced was eight nanometres across.

Søren Stobbe

But it's really hard to go below that. So, eight nanometres is already very small. And what we thought about here was: how can we go below that?

Nick Petrić Howe

Søren and the team use silicone when making bowtie shapes, as it's a good material for confining light, and lots of technology is available to create tiny cavities on silicon, thanks to decades of innovation in the semiconductor industry. However, the normal practice of carving up cavities using blasts of ionised gas won't work at the smaller scales, because the gas struggles to fit through the hole shape you're trying to make. So, they needed to come up with an alternative. And Søren thought maybe he can turn a common manufacturing problem on its head.

Søren Stobbe

It's a known problem in this type of devices, that whenever you have two surfaces, and they get close, they stick together. And at some point, I realised that this is not a problem, it is a solution. Because the fact that they stick together means that it's possible to design new types of cavities.

Nick Petrić Howe

At these very tiny scales a certain type of intermolecular attraction, the Van de Waals force, pulls molecules together — kind of annoying if you're trying to make something with a tiny bowtie-shaped gap in the middle. But Søren figured this was something that could be exploited. Instead of making a cavity, they could etch edges, which is a bit more straightforward, on two tiny bits of silicon in such a way that they would get pulled together and self-assemble into a tiny light-trapping cavity. So each bit of silicone would have half the bow tie on either side, and then, when they're released, they would snap together to create a cavity, with a tiny two nanometer gap at the centre. To achieve that though, they first needed to understand these forces better.

Ali Nawaz Babar

This was the first experiment was to really map out the surface forces.

Nick Petrić Howe

This is Ali Nawaz Babar, another of the paper’s authors.

Ali Nawaz Babar

We did the experiment with 1000s of devices — to be precise 2688 devices — we characterised those devices, and then we examined that what parameters should we have to either use these surface forces to self-assemble our devices, or what parameters can we use to avoid these forces.

Nick Petrić Howe

Once they understood how the forces would interact with their materials, the team could then use more conventional methods to etch away unwanted parts. The key here was to get very smooth edges with slight offsets so that when they came together, they're the perfect cavity. And by doing so, they were able to create their tiny two nanometer gap — around the width of DNA — and a great size for some light trapping. In fact, the team showed that their cavity does trap light well. And the next step would be to integrate a material that produces light into the bowtie, to create a device that emits photons, which they believe is the first step towards the creation of new kinds of devices that could be useful in the world of quantum computing, telecommunications and beyond. But Søren also thinks that this method could be applied to all sorts of other fields where having a tiny cavity, which you can pass molecules through, is a big advantage.

Søren Stobbe

We hope that many researchers across a lot of different disciplines will pick up on this because this could be for DNA sequencing and a lot of different applications where you need to make so small devices that it's impossible today and seems also beyond reach for the many next decades. Here is a relatively simple and direct path to realising such devices.

Nick Petrić Howe

That was Søren Stobbe from the Technical University of Denmark. You also heard from Ali Nawaz Babar, from the same institution. For more on this story, check out the show notes for a link to the paper and an accompanying News and Views article.

Shamini Bundell

Coming up, an analysis of how tackling poverty could impact climate change. Right now though, it’s the Research Highlights, with Dan Fox.

<music>

Dan Fox

Dark matter is tricky to pin down. And so, scientists have headed to the remote wilderness to see if it can be detected far from human influence. The invisible material called dark matter is thought to fill the universe, but its make-up is mysterious. Some scientists think it could be composed of ultralight particles, which interact weakly with themselves and other matter. And candidates include “hidden photons” and axions. Both might interact with Earth's electromagnetic radiation as they pass through, producing a detectable oscillating magnetic field. To look for this interaction, researchers deployed three sensitive magnetometers in rural locations across the United States that were remote enough to reduce magnetic interference from human-related sources. The experiment did not find evidence of the ultralight dark matter particles, but the team were able to gather useful data and future upgrades could allow them to target heavier particles. You don't have to travel to the rural U.S. to find that research, it's in Physical Review D.

<music>

Dan Fox

Bottlenose dolphins can detect weak electric fields, which helps them to find prey buried in the sand and avoid predators. Researchers worked with two dolphins training the animals to place their snouts near two submerged electrodes that generate an electric field. The dolphins learned to leave the experimental apparatus within five seconds if they felt an electric field and to remain in place for more than 12 seconds if they did not. Through trials the pair showed that they could sense direct current electric fields as faint as 2.4 and 5.5 microvolts per centimetre. The team suggests that with this sensory skill, bottlenose dolphins can detect fish at a distance of three to seven centimetres – useful for finding prey buried in the sea floor. They also predict that the animals’ ability to detect electric fields could enable them to use cues from the Earth's magnetic field to navigate the ocean. Navigate to the Journal of Experimental Biology to read that paper in full.

<music>

Shamini Bundell

Historically the ways that nations have often reduced poverty has meant more emissions. Bad news for the climate, but ending poverty is an important goal. So that begs the question: is it possible to end poverty without making climate change worse? Well, a new analysis in Nature tries to model the relationship between emissions and wealth. Reporter Alex Lathbridge is here with more.

Alex Lathbridge

With COP28 underway, climate change is on everybody's lips. And rightly so. It's one of the biggest problems facing humanity, which is why taking action to combat climate change is one of the UN's key objectives – part of their sustainable development goals. To tackle this, we need to reduce our emissions. But at the same time, the UN — and humanity more broadly — have a lot of other goals. For example, more than 10% of the global population is still living in extreme poverty, defined as surviving on less than $2.15 a day. This too, is an urgent problem, and one that could be tackled by economic growth.

Daniel Mahler

Historically speaking, if we were to paint a really wide stroke, there has been a tight relationship between economic growth and greenhouse gas emissions. So that is when economies have grown, in general, so have their emissions.

Alex Lathbridge

This is Daniel Mahler, an economist at the World Bank. And with such a relationship between emissions and economic growth, does this mean that tackling both extreme poverty and climate change is impossible?

Daniel Mahler

But there are important country differences beyond this general pattern. There are countries that have managed to essentially delink economic growth and greenhouse gas emissions. But they're also countries where greenhouse gas emissions have grown even faster than economic growth.

Alex Lathbridge

All right, so there's some complexity here – economic growth doesn't have to mean increased emissions. So, to untangle this, Daniel and the team from the World Bank have developed a model to simulate potential futures, where countries across the globe have alleviated extreme poverty, meaning that only 3% of people are living on less than $2.15 by 2050. But to know how the future could play out, we must look to how we've done it so far.

Daniel Mahler

So, we– we first tried to calculate how much all countries that currently have poverty as defined by this international poverty line, how much they would need to grow their economies to alleviate poverty.

Alex Lathbridge

In essence, if countries didn't do anything different, how much economic growth — and attached emissions — would it require for them to reduce the number of people living in extreme poverty below that 3%? Which is a very different challenge, depending on where you are.

Daniel Mahler

Most of the global extreme poor today live in Sub Saharan Africa, more than half perhaps two thirds, and the remaining are primarily in in South Asia, but also in in countries around the world that often are in conflict or in other sort of dire situations. For the typical country in Sub-Saharan Africa, the economy, we need to triple we estimate for them to end extreme poverty.

Alex Lathbridge

Now, a tripling of the economy might sound like a huge step up, and therefore a big increase in terms of emissions. But Daniel, and the team found that this wasn't really the case.

Daniel Mahler

Yet given that these countries tend to emit so little at a global scale, even if their emissions were to triple for them to end extreme poverty, it would not make a very large dent to global emissions. And that's why we find the total impacts of ending extreme poverty globally do not add up to more than 5% of global emissions.

Alex Lathbridge

5% of global emissions would be less than half of the emissions produced by road transport, you know, things like cars, trucks and whatnot. And it's a number that didn't really surprise Kate Ricke, a researcher of climate sustainability.

Kate Ricke

The baseline result doesn't surprise me in that the 5% is low in terms of annual emissions, because even if it's a lot of people that are in extreme poverty, and we're developing these economies in order to lift them out of extreme poverty, they're still emitting very little because they're consuming very little compared to most people in the world.

Alex Lathbridge

As Daniel said, this model used past relationships between economic growth and emissions to simulate futures. Helpful, yes, but it's hard to accurately predict the future. But one thing that we do know, unfortunately, is that climate change will continue to affect everyone. But what could that mean for alleviating extreme poverty?

Kate Ricke

Difficulties producing food, difficulties with heat waves and tropical storms are always going to be cutting into that progress, that development progress, right? If we don't fix the climate risk problem, it's going to become harder and harder and harder over time to see substantial development gains. And they couldn't account for that in this model, because it just, there's so much uncertainty around those effects. But there's an abundance of evidence that those effects exist.

Alex Lathbridge

But in the future, our technology may also get better. So, one optimistic yet plausible alternative scenario that Daniel and the team modelled, asked, “what if these countries managed to replicate the best performers in terms of energy efficiency, and decarbonisation?” Or in other words, “what if these countries can grow their economies while matching the rate of technological advancement seen in the best performers?” and they found that instead of a 5% increase in global emissions, it would go up by just over 0.5%.

Kate Ricke

I was actually surprised by the results of the alternative scenarios; you don't need some magic solution that we haven't thought of yet. And it has very cool implications in terms of how can we make development policy and climate policy work together in these situations?

Alex Lathbridge

The fact that alleviating global extreme poverty by 2050 only increases emissions by 5% or less, highlights how 95% or more of emissions are likely going to come from higher income, more developed countries. And for a long time, there has been a push to get these high-income countries to fund development in lower income countries that are less responsible for climate change, which could be funding in a way to avoid many of the possible emissions and get that 5% even lower. But more needs to be done.

Kate Ricke

The high-income countries already aren't meeting their pledges. So, there's this commitment that high income countries made to provide 2.7% of national income to development in low-income countries and they’re way below the bar there. And also, there's this commitment for high income countries to donate $100 billion per year to climate finance. We're also way below the bar there. You could think of it in terms of how can resources be increased so that high income countries are meeting their obligations? And then how can they be best deployed? So that we can meet these objectives, both important objectives to reduce climate related risks and to lift other humans out of poverty. How can we reach these goals in a synergistic way?

Shamini Bundell

That was Katharine Ricke, from the University of California, San Diego, in the US. You also heard from Daniel Mahler, from the World Bank, based in Washington, DC, also in the US. For more on that story, and Nature’s COP28 coverage, check out the show notes for some links.

Nick Petrić Howe

Finally, on the show, it's time for the Briefing Chat, where we discuss a couple of articles that have been highlighted in the Nature Briefing. Shamini, what's caught your attention this week?

Shamini Bundell

I have a very cool story about octo-arms. That's what I'm calling it. It's a um– it's a Science Robotics paper that I've been reading about in a Nature article. And we talk a lot about soft robots and octopuses — great model for soft robotics — really cool animals, can do all sorts of things that we might try and copy. And this one is trying to make a little octopus tentacle. But they've also made it so that you can control it with your finger.

Nick Petrić Howe

Oh, that sounds very cool. And maybe a nice treat for a kid at Christmas but I'm not quite sure how it relates to sort of science and you know what we talk about on the podcast. So how would making your finger more octopus-like be useful for science?

Shamini Bundell

Okay, the intention is not to give the small children octopus fingers. This is you know, part of research into how we can make more flexible robots, different kinds of robots, and also how we interact with robots? What different ways are there to control robots? But let's start off with octopuses. One of the reasons that octopus is particularly interesting and roboticists want to kind of try and emulate that is they have so much flexibility in their tentacles. As you can imagine, if you compare it to your arm, which bends in a few places, octopus tentacles have hundreds of muscles, they can control them all, they have all these degrees of freedom, they have a lot of range and flexibility. So, people are thinking that would be great to be able to recreate that in a soft robot. But the other thing that octopuses do is that they are not in the same way that we are controlling every one of those muscles from the brain—

Nick Petrić Howe

—mmm—

Shamini Bundell

—so, their nervous system is set up a bit differently. One of the researchers described it as “embodied intelligence”—

Nick Petrić Howe

—oh—

Shamini Bundell

‑—the arm itself is somewhat independent, there are the signals that go from the brain to the arm. But then within the arm, the nervous system is arranged in this way that there's this sort of sequential activation, so that you don't need all this sort of processing and computing going on in your brain to– to make the arm do the basic things that it does, which in this case, the thing that they're looking at is grabbing. So if you imagine a tentacle kind of coming out, in a sort of like, wave-like bend starting from the base, then it kinda– I’m not sure I'm describing this very well, I'm trying to mime but this is a podcast. Um, it’s sort of yeah, the wave propagates out from the base, and then kind of ends with it curling at the very end, and like grabbing, you know, its prey, say, with its suckers, and then drawing it back in. And that's the sort of movement that they're trying to recreate.

Nick Petrić Howe

I mean, it sounds like that'll be very cool if you want a robot that can sort of like grab stuff. And it's not made from so many like hard components and stuff. But it must be quite challenging to make something like this, it sounds incredibly complicated.

Shamini Bundell

Mm, I think grabbing things, in general is one of these, like big challenges of soft robotics. Because it's really sensitive, you need to be able to grip something tight enough that you can hold it, but not crush it, which kind of requires feedback. So I'm not sure how much they focused on that element in this, but they have got in the end of the tentacle, suckers and temperature sensors. And then they've got the sort of recreation of an octopus nervous system using these sorts of liquid metal wires that are embedded—

Nick Petrić Howe

—oh, wow—

Shamini Bundell

—in soft silicone. And they've got these sort of different elements of the silicone arm. And you can actually see, there's a cute little gif of it in the News article, if you check it out of this thing, of yeah, extending out and then as it extends, the actual arm sort of expands slightly as well, like it stretches slightly and then kind of contract as it grabs some sort of small toy looks, I think, oh, it's a little robot toy. Possibly, possibly a Furby? Not quite sure what's going on there? But um, yeah, it's quite cool. And then the other element, of course, is this whole, okay, so if you've got a robotic octopus arm, what's the best way to interact with that? And hence them sort of designing this like glove mechanism where it's just one finger. And again, it's that whole thing about like, you don't have to control every muscle in that robot arm, right? Consciously, just like an octopus isn't controlling every individual muscle, it's going to translate your movements into that extend, and grab. But there's also feedback, so they've got like, little suction cups inside the glove, so that you can actually sense when the robot suckers at the other end of the arm have attached on to something.

Nick Petrić Howe

That's super cool. So like, in a way, it makes you able to almost feel like an octopus.

Shamini Bundell

Yeah, I mean, I think it's interesting because we do talk about soft robotics a lot. But we haven't often talked about what is the best way to control a robot. So one of the quotes from a researcher here is like, “it's a big unanswered question, what is the best way for humans to interact with a robot?” And he says, “this is interesting, because it's like a one-to-one mapping almost, between the human and the octopus movement.” And it's certainly something that, you know, I haven't thought you know, you think about robots are sort of wandering around being independent. Or you might think of like big mecha suits, like in sci-fi, where you climb into a big suit and have to do things. But yeah, I've certainly never– never thought about having my fingers be controlling robotic tentacles, which you could even do remotely as well. So yeah, just sort of lots of interesting elements to that one.

Nick Petrić Howe

That is very interesting. And I suggest listeners go and have a look at the gif — it's quite fun. But for my story, this week, I was reading an article in Nature about how breakthroughs happen in science and how the distance between the people who make up the teams of science, how that impacts how many breakthroughs that are.

Shamini Bundell

Oh, right, so I assume you mean sort of geographical distance, which is more of a thing these days, you know, so science started off with the idea of the lone genius by themselves and then I suppose you'd have teams that are likely to be in the same place and now, international collaborations spread across the world. So is that the kind of thing that they were sort of trying to compare?

Nick Petrić Howe

Yeah exactly that, they didn't look at emotional distance, it was geographical distance—

Shamini Bundell

—I did wonder, yeah I just wanted to clarify—

Nick Petrić Howe

—maybe that has an impact on scientific teams, but it wasn't assessed in this paper. So basically, in this paper, they had a measure of kind of what is disruptive science. And they measured it in this way called a D-score, which is like if a paper is cited a lot and the papers that came before it, that that paper built on, aren't cited than that paper is more disruptive, it's more foundational, because you know, everyone's just citing that and not citing the stuff again before, so that's a new like paradigm. And they've basically compared this metric to the sort of distance between authors in different teams. And so they look to millions and millions of research articles from 1960 to 2020. And they found that in general, when there is more geographical distance between authors, papers, and even some patents, seem to be less disruptive.

Shamini Bundell

And I can think of a lot of, sort of, other variables that might be impacting this, like, what kind of things were they controlling for?

Nick Petrić Howe

So they controlled for like time periods and things like that, because you might think that, you know, now we have more technology, so people—

Shamini Bundell

—that’s what I was wondering, yeah—

Nick Petrić Howe

—tend to be further apart. So they controlled for that and they didn't find an effect of that. They also controlled for the size of teams, and also the field as well. And generally, that didn't seem to have a great effect. The biggest thing was the actual geographical distance between the teams.

Shamini Bundell

Yeah, so the sort of correlation is that, you know, UK to Australia has a proportional impact on this disruptiveness compared to UK to Berlin say.

Nick Petrić Howe

Yeah exactly that. So to give you an exact metric, if the average collaboration distance was 600 kilometres compared to zero kilometres, then the probability that this is going to be a disruptive study was 6% lower, and about 12% lower for patents. So, there is a like a measurable effect as you increase this distance as well.

Shamini Bundell

And did they actually look into any of the reasons or just sort of hypothesise about some potential explanations?

Nick Petrić Howe

The why question hasn't quite been answered yet, it’s more an observation that this is a thing; like when teams are further apart, then there seems to be less disruption. There are some suggestions. So for example, the papers where authors were more distant, they tended to be papers where you were sort of building on an existing idea, sort of more incremental research. And that might take advantage of where you have many experts across the world. And you want to sort of say, okay, this is the next incremental step. So, say, a field has been established for a long time, and you want to be like, this is what we know about this field. These are where we need to go next, maybe you would look around the world and try and get all the sort of top minds in that one. So perhaps, that could be effect. Whereas when you're doing more disruptive science, this may be a moment where ideas are harder to sort of actually put your finger on. So you're sort of brainstorming together, and you're sort of coming up with things together. And so there could be an advantage of maybe being closer and maybe more face-to-face with that. For example, you know, last year, you did a video about how—

Shamini Bundell

—I was just gonna say about that, yeah—

Nick Petrić Howe

—you did a video about how people tend to brainstorm better when they're in person. There were some caveats to that but in general, it seemed that being face-to-face led to sort of more ideas, and maybe this has something to do with that.

Shamini Bundell

That's so interesting. I feel like it's quite a sort of politically relevant finding, in terms of I'm sure scientists have a lot of opinions about whether they sort of prefer to work with people close by or far away, or sort of, there might be potential benefits of working more globally that maybe aren't captured by this.

Nick Petrić Howe

Definitely. And I think answering that why question will be really important to this, because obviously, international collaboration is really valuable part of science that a lot of scientists put a lot of stock in, but obviously, so it's disruption, being disruptive, coming up with new ideas, shifting the paradigms, as it were. So understanding why this is will be really important. And obviously, maybe many more people are doing more remote working now — so that's another aspect of this. And there has been an ongoing transition in science from more local teams to more global collaboration. So what does that mean for the future of science? We don't know. There's been a lot of work to trying to understand this sort of disruptive nature of science. There's been several studies that have sort of like said that disruption in science is declining and tried to figure out why. We looked at another study when I first started at Nature four years ago that was saying like small teams were more disruptive. So there's a lot of different things that we don't quite understand yet. But nonetheless, it's an interesting sort of aspect to this sort of changing nature of science.

Shamini Bundell

Well I always like it when we cover the science of science, more more science about the doing of science. Really interesting, Thanks, Nick. Listeners, if you want to read more about those stories you can check out we'll put the links in the show notes. And they both came from the Nature Briefing, and we'll put a link to where you can sign up for it and get more stories like this in your emails.

Nick Petrić Howe

That's all for this week, as always you can keep in touch with us on X, we’re @NaturePodcast, or you can send an email to podcast@nature.com. I'm Nick Petrić Howe…

Shamini Bundell

… and I'm Shamini Bundell. Thanks for listening.