With Lasers, 3D Printing on a Miniature Scale

HANK HOGAN, CONTRIBUTING EDITOR, hank.hogan@photonics.com

Copies of buildings and the Statue of Liberty that are only microns tall — these are examples of what’s been done with three-dimensional nanolithography, known as 3D nanoprinting. Also possible today is the manufacturing of optical, photonic, microfluidic, and other components on the nanoscale. Compared to mask lithography and large-scale manufacturing methods, 3D nanoprinted components offer solutions that can be less expensive, better performing, quicker to create, and more compact.

For instance, in February 2018 researchers at New York Genome Center and New York University reported on a 3D-printed, low-cost, microfluidic system for single-cell analysis, demonstrating the usefulness of the approach in a clinical environment. Thanks to custom printing of the central device, the total cost of the system was a fraction of a comparable traditional solution, and it had a smaller footprint.

Metamaterial crystal built using Nanoscribe 3D laser nanolithography techniques. It responds to compression with a twisting motion. Courtesy of T. Frenzel, Karlsruhe Institute of Technology.

Microscopic 3D manufacturing depends upon lasers, positioning systems, and photosensitive materials with the right properties. Hence, advances in these areas are required if 3D nanoprinting cannot currently satisfy requirements, such as hitting cost targets.

But, in some applications, 3D nanolithography already offers the necessary performance in terms of quality and cost. A case in point is diffractive-refractive optics, which can increasingly be found in the time-of-flight and lidar technology in products such as self-driving cars. For such optics, 3D nanoprinting can do what conventional manufacturing cannot.

Gold-coated hollow polymer pyramid with a submicron aperture for subwavelength plasmonic light focusing. Structure created with 3D laser nanolithography. Courtesy of Changzhi Gu, IOP/CAS.

“Conventional lenses with a size down to a micron diameter can be produced by diamond turning, but other DOEs (diffractive-refractive optical elements) have a structure size in the nanometer area,” said Peter Pagnin, the European business development manager for Singapore-based Moveon Technologies Pte. Ltd., in explaining why diamond turning is not suited for the task.

In the past, manufacturers relied on a mask-based lithographic process to make these fine-feature optics, which meant masks had to be designed and manufactured first, and then components were made using those masks. This mask-and-component process had to be repeated with each development turn. During prototyping, several cycles were often needed.

3D-printed tubular structure mimicking a brain capillary of the blood-brain barrier. Courtesy of A. Marino, Italian Institute of Technology.

That’s why Moveon turned to a two-photon, 3D nanoprinting process for a rapid and inexpensive prototype turnaround of microlens arrays, mirrors, and other DOEs. In it, an NIR laser fires femtosecond pulses, which are focused within a UV photocurable resin that sits on a substrate. The intense light of the pulse at the focal point exceeds a threshold, leading to two-photon absorption and polymerization of the resin. This spot can be moved around as desired by shifting the stage, the beam, or both.

Freeform 3D micro-optical elements fabricated by laser lithography. There are hybrid optical elements — spheric and axicon lenses as individual elements and monolith components mounted on tips of single-mode optical fibers, thereby creating integrated micro-optical devices. Courtesy of Mangirdas Malinauskas, Vilnius University.

After development and curing of the photoresin, the resulting structure has the required form and dimensions for its particular application. In Moveon’s case, the products measure a few tens of microns in size, with a smooth enough surface for optical applications. Typical turnaround time is a few days, versus the weeks needed with mask lithography. After the design is proven and production scales up, tooling suitable for mass production can be derived from the nanoprinted sample.

Need for speed

Current two-photon printers can handle prototyping, but there are still improvements needed. One advancement, for instance, could make the manufacturing time shorter and therefore lower the cost.

“We are requesting higher writing speed because time influences cost,” Pagnin said.

Moveon, which uses a two-photon 3D printer from Nanoscribe GmbH of Eggenstein-Leopoldshafen, Germany, was the first company to sell parts and products based on processes using these 3D nanoprinting machines, according to Andreas Frölich, head of sales and marketing at Nanoscribe.

Frölich said Nanoscribe’s 3D nanoprinters initially used a fixed laser spot, moving the part being written in such a way that the spot traced the desired trajectory through the resin. However, with the addition of two galvo mirrors, one for X and another for Y movement, it became possible to rapidly shift the spot location, which sped up the manufacturing process a thousandfold, according to Frölich.

In some applications, such as diffractive-refractive optics, 3D nanolithography already offers the necessary performance in terms of quality and cost.

The mirrors move the laser focal point around so that everything is written in a given layer. Then, the part is repositioned and the next layer written, and the sequence is done repeatedly until the entire part is complete.

In the quest for more speed, peak intensity is important because only the volume above the minimum intensity threshold will be polymerized. In general, a more powerful beam will mean a shorter write time because less exposure time is necessary. Alternatively, a larger volume can be exposed. However, too great an intensity will overexpose the photoresin instead of polymerizing it. Thus, the peak intensity must be kept below this maximum intensity threshold.

Performance success also relies on beam quality. Higher-quality beams, as well as beams with shorter wavelengths, can be focused down to a smaller spot size, making it possible to create smaller features. According to Frölich, the company’s current technology can do submicron-size features. It cannot yet produce structures that are smaller than 100 nm.

Bringing the achievable size and spacing down by a factor of two or three would mean that new structures could be built, such as photonic crystals for the visible instead of the IR. Although there may be a desire among users for that, there is a stronger pull toward faster writing, higher throughput, and the capability to build bigger parts, according to Frölich.

There are other, nonlaser aspects of the system that also play a critical role in the 3D nanoprinting performance, which is why Nanoscribe has been developing new photoresins. The speed with which parts can be made depends somewhat on the size of the polymerized volume. A larger volume means fewer writing passes are needed, reducing the time needed to make a part. The company has created more sensitive photoresins, leading to a larger volume that is polymerized.

“You can trace out more solidified material within the same laser sweep,” Frölich said, noting that this results in higher throughput.

Further development is needed and may include tissue engineering applications, in which the polymer must be biocompatible and also suitable for two-photon printing.

A scaffold created by laser 3D nanolithography made out of biocompatible hybrid SZ2080 material. Hexagon pore shapes and their dimensions and the overall size of the microporous scaffold are optimized for cartilage tissue engineering. Mangirdas Malinauskas, Vilnius University.

Another example of 3D laser nanolithography, and an indication of where the field may be headed, comes from research underway at the Laser Research Center of Lithuania’s Vilnius University. In collaboration with industry and other academic centers, its Laser Nanophotonics group produces 3D micro-optical and nanophotonic components using femto-second lasers and nonphotosensitized organic-inorganic polymers.

“Laser 3D nanolithography offers true 3D writing capability, unmatched variety of processable materials, and flexible integration directly onto or into chips,” said Mangirdas Malinauskas, the Laser Nanophotonics group leader, in outlining why this is an area of research interest.

Many of the resins used in laser nanolithography come from other lithography applications, which is not necessarily optimal, Malinauskas said, so research is underway to make improvements. For instance, Malinauskas and colleagues published an article in Materials in January 20171 that reported on an investigation of an organic-inorganic material that is 20× more resistant to high irradiance than standard lithographic photoresin. The ability of finished components to withstand high-intensity light is important for micro-optics in some settings.

Increasing throughput

In addition, the Laser Nanophotonics group is working with industry partners on ways to overcome the biggest bottleneck facing laser nanolithography: throughput. By synchronizing workpiece translation with beam delivery, the goal is to enable continuous, true 3D writing in all directions without layering or stitching. The technique they developed enables the creation of millimeter- or even centimeter-scale 3D objects with nanometer-size individual features, according to Malinauskas.

With lasers, resin chemistries, and other aspects improving, there’s also a need for the mechanical systems to keep pace.

He added that further increases in throughput may demand more laser power, but this can also lead to a loss of resolution, accuracy, or surface quality. One alternative is to use parallel processing, with multiple spots being traced through the resin. That approach, Malinauskas said, “can dramatically increase the throughput, yet limit the flexibility. Thus, it becomes efficient, but case-sensitive.”

As for laser improvements, Malinauskas said the key advances needed are tunable wavelengths that are optimized for specific materials, ultrashort pulses, and power stability. These features are available today, but the cost of systems equipped with them is high. So, they may be suitable for a research setting but not mainstream industrial application.

In addition to additive techniques, such as two-photon printing, other forms of 3D laser nanolithography may be ablative. Microscopic embossing is an example, according to Scott Jordan, head of photonics at Physik Instrumente of Auburn, Mass., which provides motion and positioning solutions.

One thing the various 3D laser nanolithography techniques have in common is the need to precisely and accurately position both the beam and the workpiece, Jordan said. Positioning must be done rapidly enough for high-throughput processing, must be stable over minutes of processing time, and must be on the nanoscale.

“The positioning capability needs to be considerably better than the form factor of the devices or structures being produced,” Jordan said.

With lasers, resin chemistries, and other aspects improving, there’s also a need for the mechanical systems to keep pace. For that, there are new piezo motors that provide high stiffness, extreme stability, and high speed; set-and-forget nanoscale structural compensation; and other technology enablers, Jordan said.

Moving manufacturing techniques into the mainstream requires further improvement in the process and equipment, particularly when it comes to increasing throughput and decreasing cost. Solutions being developed to do that may often be hidden from public view until a product implementing the new approach is launched. But, those with knowledge of the field believe such proprietary solutions will work out well for 3D nanolithography.

“Those (solutions) are what will propel these technologies out of the research lab and into the mainstream,” Jordan said. “Based on what I’m seeing in the field, this rollout is imminent.”

Reference

1. L. Jonušauskas et al. (January 2017). Optically clear and resilient free-form µ-optics 3D-printed via ultrafast laser lithography. Materials, Vol. 10, p. 12.