Dr. Andrei Kolmakov and a team of researchers at the National Institute of Standards and Technology (NIST) have developed a method for 3D printing of tiny gel structures in liquids with electron beams — a method that has previously been limited to solids.

Dr. Andrei Kolmakov from NIST
Dr. Andrei Kolmakov

Dr. Andrei Kolmakov: One of the projects we are running, is using electron microscopy in unusual environments. Electron microscopes usually work in a vacuum. There are many processes, for example, in batteries, catalysis, and in the semiconductor industry, where you want to look at objects that are in high pressure gas or liquid environments. That’s hard to do with an electron microscope because it doesn’t penetrate too deeply into dense materials.

Tech Briefs: What got you interested in this project?

As part of our research, we were working on techniques to develop electron imaging capabilities for different applications. Once, at a Materials Research Society (MRS) meeting I noticed a bio-related exhibit, where an extrusion printer was printing hydrogels by extruding a small amount of liquid gel, which gets gelated — solidified — by UV light. I immediately felt that our work on electron microscopy in liquids could contribute to printing gels. For us it doesn’t matter if we are imaging or doing something in a liquid or in a liquid gel precursor.

The next week, my post-doc and I did a test to see if it was feasible, and to our amazement, it was easy. So, we decided we were onto something big. We spent a year or more on different kinds of tests, developing these techniques and that’s how it happened.

It was challenging because not that much is known about our process. The study of cross-linking, the formation of larger molecules from liquids, from smaller precursors, using electrons or gamma rays or x-rays basically comes from 1960s radiation physics. But before us, no one was using highly focused low energy electron beams for these kinds of processes. We decided this could open a new door in synthesis, lithography, or even 3D printing.

Tech Briefs: How are gels usually created?

Dr. Kolmakov: For commercial gel printers, it’s usually done with UV light. However, these printers have very low resolution compared to us. They typically have a millimeter feature size, whereas we can go to the nanometer level — a million times smaller.

The way a standard 3D printer works, is you have a liquid: molten plastic, or liquid gel precursor solution in the case of bio-printers, and because it’s viscous, the liquid can be slowly extruded through a nozzle. You can paste the extruded liquid in a controllable way by moving the nozzle over a surface. Then, you can use UV light to cure — solidify — the layer you made. Special chemicals, called initiators, have to be embedded in the solution to enable this kind of solidification upon being irradiated with UV light. You have to use these chemicals because the usual gel doesn’t do anything with light, it’s transparent.

In our case we don’t use either a nozzle or initiators. We can work just with the liquid precursor as it is, because the electron beam itself does the initiation in the water.

The gel precursor solution is a water solution of an ensemble of molecules that are cross-linked — very long molecules connected together chemically. You can fill it with water, and it swells because water fills the spaces between the molecules, or you can dry it and it shrinks.

One example of a typical gel application is contact lenses. However, the need exists, especially in biology, to make more complex structures. For example, if you wanted to create artificial organs, say an ear, you could make a scaffold from the gel and populate it with biological cells that will stick to the gel and grow there. That’s why the bio industry is interested in these kinds of techniques.

Imagine now that you want to make something really small, basically on the level of the individual biological cell itself. Or say you want to make an electrical contact to the cell, to send a signal back and forth. You have to do it gently, without disturbing the cell too much because a biological cell is a very fragile organism. You could try to connect to it with a wire, but that could be destructive even if you’ve done it gently. In our case, we are able to produce gels so small that we can make a very small contact with extremely high precision. We can do this because of the ability of the electron beams to be focused on very, very small areas.

And, another thing, the initiator chemicals I spoke of before are often toxic. If you want to print something really small using state-of-the-art two-photon 3D printing techniques, you have to increase the concentration of the initiators, so the gel becomes even more toxic for cellular material. In our case, we can make extremely small features without using any toxic initiators.

Tech Briefs: Let me make sure I follow the basic process. The way I see it, you 3D print with the gel — you deposit the gel on a substrate. Is that right?

Dr. Kolmakov: Let me describe the process in greater detail. Imagine that you have a standard scanning electron microscope. It is a vacuum chamber with a very, very fine electron beam inside. The beam can be as small as three nanometers. If your sample is inside the vacuum chamber, you can scan the beam over the surface and get a signal from it, and from that, you get an image. Or, if you want to fabricate something, you could put, say, a layer of material — people use this for the semiconductor industry — you lay down a film of photoresist. You can then draw something on this resist, chemically modify it with an electron beam on the solid film and treat it afterwards with special chemicals for removal. Then you’ll get a pattern on the surface of the sample. Those are standard electron microscopy and electron lithography procedures.

That’s fine with solid films or objects, but we want to do something like that in liquids. The problem is that liquids don’t last in a vacuum, they evaporate. The microscope would get contaminated — and that’s very expensive.

An illustration of a prospective biocompatible interface shows that hydrogels (green tubing), which can be generated by an electron or X-ray beam 3D printing process, act as artificial synapses or junctions, connecting neurons (brown) to electrodes (yellow).
An illustration of a prospective biocompatible interface shows that hydrogels (green tubing), which can be generated by an electron or X-ray beam 3D printing process, act as artificial synapses or junctions, connecting neurons (brown) to electrodes (yellow). (Image Credit: NIST)

To handle this challenge, we use a very thin membrane, in the 10-nanometer range. It’s made from silicon nitride, which is a standard semiconductor material. The membrane is so thin that electrons can penetrate it with only a small amount of scattering or attenuation, but gases and liquids cannot. We use this trick to deliver the beam into the liquid. We did it by creating a small secondary chamber with a silicon nitride window and filling it with a liquid precursor for the gel formation. We then irradiated the liquid very precisely with electrons, creating certain patterns. In the areas where the electron beam hits it, the liquid is chemically modified, and a gel is formed.

That was our major idea: to create the soft layer in this manner. Then you can delaminate it because it’s formed very, very close to the membrane. Following that, you start growing the second layer, delaminate that, start growing the third layer, and so on. This was our aim — to use an electron beam to create a layer by layer gel structure inside the liquid.

Tech Briefs: So, the gels are laid down in certain patterns?

Dr. Kolmakov: Yes, we haven’t created extremely complex structures so far. But we have demonstrated the kinds of simple structures that are possible. Importantly, we have also demonstrated the way the delamination could be done. When you do 3D printing, the delamination of the first layer from the membrane becomes an issue because it sticks. So, you have to create a procedure for delaminating it, to be able to write a second layer on top of the first one.

Tech Briefs: Is the gel a basic structure on which you could put a biological cell or some kinds of sensors?

Dr. Kolmakov: Yes, with the gels you can do a lot of things. For example, conductive gels can be used as electrical contacts. Or since they are transparent, they can be used to make optical fibers. Also, some gels could be made to be reactive to certain stimuli. For example, they can be made sensitive to temperature or pH. You can create a lot of functionalities by modifying the molecules of the gels. In this way, you can build functional objects like nanoswimmers or soft micro-robots.

Tech Briefs: Does the electron beam do all of these modifications?

Dr. Kolmakov: No, so far, the electron beam itself just makes a shape.

Tech Briefs: So, how do you do all of the other things?

Dr. Kolmakov: You introduce the functionalities to the gel itself. For example, we wanted to sense humidity and we wanted to make the sensor very, very small. We added gold nanoparticles into the solution and during the writing process, we encapsulated the particles inside the gel structure.

Tech Briefs: So, you’re saying you put the particles into the mixture and then you used the electron beam to make the structure.

Dr. Kolmakov: Yes, the particles now become encapsulated inside the gel. The size of the gel material is very sensitive to humidity. Let’s say it shrinks if it’s dry outside and swells when it’s wet or humid. Then, the distance between the embedded particles changes because of the humidity variations. You can then determine the humidity by monitoring the color of the composite gel. The technique we use is called plasmonic excitation. You can look at the optical spectrum of the material and determine the distance between the particles. So, this is a simple way to monitor humidity. But there are many other things you can do. For example, you can change the gel molecule itself so it will be responsive to pH. You can then make something like a robot that moves when the acidity changes. A nanoswimmer robot inserted into certain areas of the body could move when the pH of the solution changes. The advantage is, that different from other technologies currently used for these purposes, we can make the structure extremely small — we can actually make it smaller than the cell itself.

Tech Briefs: Can you use x-rays instead of electron beams?

Dr. Kolmakov: To a large extent, it doesn’t matter what kind of ionizing radiation we’re using. The benefit of both electron beams and x-rays is that you can focus them into a very, very small spot — you can use either of them to write very small structures. However, x-rays have their own advantages. You can change the energy of the beam by changing its wavelength. Since each chemical element absorbs X-rays at very specific wavelengths, you can add chemical specificity into the writing process. For example, you can write oxygen-containing gels shallower or deeper if you tune the wavelength of the x-rays closer to, or away from, the point of maximum oxygen absorption.

Tech Briefs: But aren’t they more dangerous?

Dr. Kolmakov: Well, this is ionizing radiation, so appropriate safety measures must be taken, such as sufficiently shielding the beam from the user. But it’s a question of the dose needed to modify the media. The ability of the electron beam to ionize water in solution is very effective and does not require large doses — this is exactly what we are using as a cross-linking agent.

Tech Briefs: Do you see this being used commercially anytime soon?

Dr. Kolmakov: The interest of industry in this technology will depend on the capabilities we can demonstrate. I think the major challenge now, for example in 3D printing, is that we still need to improve the reliability of layer-by-layer delamination from the substrate. So, as soon as we show 3D complex submicron structures, industry should be interested in this technology for printing very small stuff. We are continuing to work on these.

Tech Briefs: Could this be done with commercially available energy sources?

Dr. Kolmakov: Exactly! That was our goal, we wanted to demonstrate this to the community of people who work with standard scanning or transmission electron microscopes, and there are thousands of them all over the world. Also, there are people who possess x-ray microscopes (which is a new industry) — they’ve become available for labs recently. All the machines we use in our lab are commercial. We just added very simple custom-made setups. So, it’s definitely possible to do this on a large scale. Even more, there are new developments in the microscopy itself. Some companies, have become interested in producing electron microscopes that are capable of operating in an ambient environment, like in air. That would be even easier, then, because you would just put your sample in air under the microscope.

Tech Briefs: What excites you most about this project?

Dr. Kolmakov: What excites me most is that this is a brand-new technology and we are in on the very beginning of it. My wish is to find enthusiastic partners and enough resources and manpower to move it forward.

Tech Briefs: Would you be able to work with a commercial company to implement your techniques?

Dr. Kolmakov: Definitely, I would be delighted to. Our mission at NIST is to help industry to develop new technology or metrology.

Tech Briefs: It would seem to me that plenty of people should be interested in this.

Dr. Kolmakov: Right, biologists working with the printing community would be interested. For example, using current 3D bioprinting technology, people are building centimeter-size gel structures for tissue engineering. However, there is also a need to print very small structures, let’s say on the level of the cell, or inside the cell, but that is still science fiction. There’s no market yet — it’s pretty much an open space. If someone comes up with a marketable idea for artificial subcellular gel structures, industry will become more interested. It may be that we are too early.

Tech Briefs: Can you predict any markets?

Dr. Kolmakov: Well, one of the things I think will be interesting, is to link our technology to computer-brain interfacing. There are two major challenges there. One is developing soft electrodes that will not damage brain tissue and the other is to deliver these electrodes into the brain.

Tech Briefs: I heard Elon Musk talking about that.

Dr. Kolmakov: Yes. The problem is that he is using an older technology. Their electrodes are solid — not very flexible — and are not extremely friendly to the tissue. The second thing is they have to do an operation on the skull to implant the electrodes. What I see with our kind of method, is that we can make the electrodes much thinner, much more flexible, and much more bio-friendly. Also, our electrodes can transmit electronic and ionic signals, and are optically transparent, so they can transmit optical signals back and forth. So, in my opinion, this is a much better prospect for brain activity imaging than anything else. That’s probably the hottest application I can imagine. Practically everyone working in soft electronics now is keeping in mind brain-computer interfacing. Initially, it will be for people who have lost some functioning, for example mobility, because they are desperate. But eventually — imagine that you have a second brain in your possession.

There is a very small gap, I think, between science fiction and reality now… This is a huge field and what we are doing is just a very small contribution. People have learned a lot by starting to read the signals that brains generate. Understanding the brain has already changed the way we do computing and led to the beginning of a new technology: neuromorphic computing. People are trying to create computers with a completely different architecture and language, and even logic, to work with, while still using the normal elements, the usual semiconductors. It would be less digital and more into analog and pattern recognition and may use different, for example, soft materials, instead of inorganic transistors or other devices such as memristors.

An edited version of this interview appeared in the December 2020 issue of Tech Briefs.