We chatted with Greg Guymon, a graduate student from the MacKenzie Group, and David Sharp, a graduate student from the Majumdar group, to find out more about the research behind this report. We also had a chance to catch up with Stephen Gibbs, a postdoc from the Gamelin group who has recent taken a position with Bruker Nano Surfaces & Metrology.

Nanophotonic structures are foundational building blocks for the emerging field of light-based devices and quantum networks, enabled by their ability to couple with and manipulate photons.

Colloidal quantum dots are excellent materials for such photonic devices, they can be made on scale, in solution, and their properties can be tuned by adjusting their size and chemistry. The utility of colloidal quantum dots can be significantly expanded by their hetero-integration into a photonic cavity. The cavity is like a series of mirrors in which the photon can be reflected, controlled, and amplified. The Purcell effect offers the opportunity to maximise the emission intensity, narrow the linewidth, and enable novel effects such as on-chip lasing and low-power optical nonlineararity.

For the optical cavity that is fabricated in a silicon wafer, silicon nitride has emerged as the material of choice. Overall, this positively supports a future pathway to integrating quantum optoelectronics with ubiquitous silicon CMOS technology. However, silicon nitride’s low refractive index (~2) limits the optical bandgap size, and thus the quality factor and mode confinement attainable. One strategy that has been proposed to overcome this limitation and achieve higher Q-factors is to use suspended nanobeam cavities, as opposed to designs monolithically built into the substrate.

By suspending the nanocavity, you are isolating it from the substrate, reducing losses due to absorption and scattering typical in monolithic cavities that are in direct contact with the substrate. Minimizing the interactions with the substrate also significantly improves the Q-factor. Suspension of the cavity means that it is surrounded by air or vacuum, which can help increase the optical confinement and enhance the light-matter interaction within the cavity.

Previous approaches for placing colloidal quantum dots into such cavities have relied on methods such as drop casting and spin coating, which typically spread quantum dots over a large area. This is like trying to paint a detailed feature on a wall with a paint roller – you cover much more of the wall with paint than you want to. More accurately it would be like throwing the contents of the tin of paint at the wall. This presents a number of drawbacks, including the inefficient use of quantum dot materials, detrimental effects on the function of other parts of the system you are building, and it can cause unwanted background emission or absorption from the quantum dots that are not in the correction location in the cavity that can dilute and mask the desired action of the device. These approaches can also put significant mechanical stresses on fragile structures in the device, such as the suspended cavity, which can have significant detrimental impacts on the physical properties of these devices. All of these factors mean that these approaches are not suitable for sustainable and large-scale manufacturing.

A new approach was needed. Essentially the team needed a correctly sized paintbrush instead of a paint roller (or tin of paint). This would ideally make the process more accurate, more efficient in terms of how materials are used and placed, produce devices with the desired properties, and be applicable to large-scale sustainable manufacturing processes.

Electrohydrodynamic inkjet (EHDIJ) printing has emerged over the past decade as an advanced manufacturing process that enables the accurate and reliable placement of droplets down to the attoliter scale. This is small, really, really small. An attoliter is 1×10^-18 liters. So, in 1 liter (~33 oz), there are 10 million times the number of stars in the Milky Way attoliters present. Put another way, if you could split one liter into attoliters, you could give every person on the planet 125 million of them.

Conventional printing techniques, like your desktop inkjet printer can only form much larger drops (10,000 x larger) through mechanical pressure processes that are limited by surface tention and intertial forces that limit the productions of smaller ink droplets. EHDIJ achieves its very high precision precision through the use of an electric field to extract the attoliter scale drops and enabling a zero-waste, additive process. By applying a strong electric field across the nozzle and the target, attoliter droplets containing the quantum dots are selectively deposited onto the target – this paint brush never actually touches the wall, the electric charge “pulls” the charged paint (the quantum dots) from the brush onto the surface. While EHDIJ has been explored for a number of applications in the past, including the manufacture of circuitry, flexible electronics, photodetectors, LEDs, and solar cells, this report is the first time it has been used to deposit quantum dots into suspended nanophotonic structures.

Achieving this goal required contributions from teams across the entire research spectrum in IMOD. Robust quantum dots that could be produced in an ink co-solvent were designed and synthesized for this program, with feedback and iteration possible as the process was refined. Specific dielectric-mode silicon nitride suspended cavity structures were designed and built based on the properties of these quantum dots so that they effectively couple with the peak emission wavelength of the quantum dots. These then had to be precisely placed and the properties of these new devices examined and explored through cutting edge spectroscopic methods. Each of these contributions needs expert knowledge and skills. By working together this collaborative team was able to advance much more rapidly than if any single investigator had attempted to explore this challenge.

Greg gave some insight into how the different parts of the team operated. “The folks in the Majumdar group were our cavity fabricators, and they were giving us feedback on what their needs were, what was working, what wasn’t working. The folks in the Gamelin Group were our quantum dots synthesizers, adjusting the ligands and the properties of the quantum dots to improve the printability of the dots”.

As the teams worked together to develop a process to accurately place these quantum dot emitters onto a suspended cavity structure for the first time using the EHDIJ printing approach, there was a great deal of interaction between the groups.

“As the synthesizers in this project, our goal was to provide robust quantum dots with bright illumination in the visible spectrum” explained Stephen. “This led to the choice of CdSe quantum dots with a CdS shell. CdSe is a bright emitter in the visible range whose photoluminescence intensity and resistance to degradation is added by a thin CdS shell – not dissimilar from how steel is galvanized with a think layer of zinc to prevent oxidation.” The stability of the quantum dots in solution was critical to this work. Stephens work had a strong focus on this aspect “If the particles are not well dispersed, they can aggregate and crash out of the solution before being printed. To prevent this we did a good amount of work on the concentration and identity of the ligands coating the quantum dots to get good, long-term colloidal dispersibility”.

“From the hetero-integration side, we provided a lot of feedback to the colloidal synthesis experts and the device fabrication experts” explained Greg. “This included things like iterative performance optimization of the inks to maximize their printing stability, minimization of feature sizes, exploring different chemistries and concentrations to better understand how these impacted the array of process parameters involved in this printing process. The collaborative nature of this work was crucial. Everyone involved is a specialist at what they do, so being able to interact with each other was fantastic. Whenever we needed to make any sort of adjustment, you knew who to speak to, and they knew how to solve a problem that really requires in-depth knowledge to understand”.

“I designed and fabricated all the cavities that we explored in this work” explained David. “I confirmed with the folks who made the quantum dots the wavelengths and properties of them so I could design the cavities appropriately, and then worked with Greg on the printing side, getting the alignment marks for the printhead sorted out.” David reflects on how the team approached the project together: “Something I really enjoyed about this collaboration was how we built a dialogue as the project matured, instead of being by myself just thinking about things, we could ask each other questions and different parts of the team stepped up to the challenge and got us there.”

This collaborative approach paid off. The team showed, for the first time, that EHDIJ could be used to precisely place colloidal quantum dots directly into suspended nanophotonic cavities with exquisite accuracy.  “The level of precision we achieved with this process was amazing” says David. “I was excited to see that we could print accurately onto these suspended structures. We were able to selectively put QDs into cavities that were only 100 nm apart”. “Hitting a sub-micron target is an ever-present challenge” says Greg. “It doesn’t leave much room for error, and whenever you are making necessary alignments or calibrations to the printer it really requires rigorous attention to details and procedures.” Greg continued “This was where strong communication with the device fabrication team was needed, we needed to know exactly where everything was so that we could properly design our prints in an area where we can’t even see what is going on until we put it under a microscope”. Using a technique called Scanning Electron Microscopy, or SEM, the team were able to show the accuracy of this process and confirm that the physical integrity of the cavity structures was not damaged by the process. 

“Seeing this work for the first time was really exciting” explains Greg. “EHDIJ benefits from a phenomenon where electric fields become concentrated around non-planar features, like edges or sharp points, almost like a lightning rod. Given that there are air gaps around the suspended cavities, that decreases the strength of the electric field. When the print head passes over these high-aspect suspended nanobeams, the electric fields should become attracted to them and focus the ejected quantum dots toward the cavities. That was the proposal, and when it worked, it was really something!”

This demonstration should just be the start of interesting investigations in the future. While more broadly this success should prompt other teams to adopt this approach for the fabrication of heterointegrated suspended devices, the IMOD team already has plans, exploring different aspects of this process, probing how much control they have over levels of deposition and structures that can be “painted” onto.