Researchers measure near-perfect performance in low-cost semiconductors

Researchers measure near-perfect performance in low-cost semiconductors
A close-up artist's rendering of quantum dots emitting light they've absorbed. Credit: Ella Marushchenko

Tiny, easy-to-produce particles, called quantum dots, may soon take the place of more expensive single crystal semiconductors in advanced electronics found in solar panels, camera sensors and medical imaging tools. Although quantum dots have begun to break into the consumer market—in the form of quantum dot TVs—they have been hampered by long-standing uncertainties about their quality. Now, a new measurement technique developed by researchers at Stanford University may finally dissolve those doubts.

"Traditional semiconductors are single crystals, grown in vacuum under special conditions. These we can make in large numbers, in flask, in a lab and we've shown they are as good as the best single crystals," said David Hanifi, graduate student in chemistry at Stanford and co-lead author of the paper written about this work, published March 15 in Science.

The researchers focused on how efficiently reemit the light they absorb, one telltale measure of semiconductor quality. While previous attempts to figure out dot efficiency hinted at , this is the first measurement method to confidently show they could compete with single crystals.

This work is the result of a collaboration between the labs of Alberto Salleo, professor of materials science and engineering at Stanford, and Paul Alivisatos, the Samsung Distinguished Professor of Nanoscience and Nanotechnology at the University of California, Berkeley, who is a pioneer in quantum dot research and senior author of the paper. Alivisatos emphasized how the could lead to the development of new technologies and materials that require knowing the efficiency of our semiconductors to a painstaking degree.

"These materials are so efficient that existing measurements were not capable of quantifying just how good they are. This is a giant leap forward," said Alivisatos. "It may someday enable applications that require materials with luminescence efficiency well above 99 percent, most of which haven't been invented yet."

Between 99 and 100

Being able to forego the need for pricey fabrication equipment isn't the only advantage of quantum dots. Even prior to this work, there were signs that quantum dots could approach or surpass the performance of some of the best crystals. They are also highly customizable. Changing their size changes the wavelength of light they emit, a useful feature for color-based applications such as tagging biological samples, TVs or computer monitors.

Despite these positive qualities, the small size of quantum dots means that it may take billions of them to do the work of one large, perfect single crystal. Making so many of these quantum dots means more chances for something to grow incorrectly, more chances for a defect that can hamper performance. Techniques that measure the quality of other semiconductors previously suggested quantum dots emit over 99 percent of the light they absorb but that was not enough to answer questions about their potential for defects. To do this, the researchers needed a measurement technique better suited to precisely evaluating these particles.

"We want to measure emission efficiencies in the realm of 99.9 to 99.999 percent because, if semiconductors are able to reemit as light every photon they absorb, you can do really fun science and make devices that haven't existed before," said Hanifi.

The researchers' technique involved checking for excess heat produced by energized quantum dots, rather than only assessing light emission because excess heat is a signature of inefficient emission. This technique, commonly used for other materials, had never been applied to measure quantum dots in this way and it was 100 times more precise than what others have used in the past. They found that groups of quantum dots reliably emitted about 99.6 percent of the light they absorbed (with a potential error of 0.2 percent in either direction), which is comparable to the best single-crystal emissions.

"It was surprising that a film with many potential defects is as good as the most perfect semiconductor you can make," said Salleo, who is co-author of the paper.

Contrary to concerns, the results suggest that the quantum dots are strikingly defect-tolerant. The measurement technique is also the first to firmly resolve how different quantum dot structures compare to each other—quantum dots with precisely eight atomic layers of a special coating material emitted light the fastest, an indicator of superior quality. The shape of those dots should guide the design for new light-emitting materials, said Alivisatos.

Entirely new technologies

This research is part of a collection of projects within a Department of Energy-funded Energy Frontier Research Center, called Photonics at Thermodynamic Limits. Led by Jennifer Dionne, associate professor of materials science and engineering at Stanford, the center's goal is to create optical materials—materials that affect the flow of light—with the highest possible efficiencies.

A next step in this project is developing even more precise measurements. If the researchers can determine that these materials reach efficiencies at or above 99.999 percent, that opens up the possibility for technologies we've never seen before. These could include new glowing dyes to enhance our ability to look at biology at the atomic scale, luminescent cooling and luminescent solar concentrators, which allow a relatively small set of solar cells to take in energy from a large area of solar radiation. All this being said, the measurements they've already established are a milestone of their own, likely to encourage a more immediate boost in quantum dot research and applications.

"People working on these quantum dot materials have thought for more than a decade that dots could be as efficient as single crystal ," said Hanifi," and now we finally have proof."


Explore further

More stable light comes from intentionally 'squashed' quantum dots

More information: David A. Hanifi et al, Redefining near-unity luminescence in quantum dots with photothermal threshold quantum yield, Science (2019). DOI: 10.1126/science.aat3803
Journal information: Science

Citation: Researchers measure near-perfect performance in low-cost semiconductors (2019, March 15) retrieved 26 May 2019 from https://phys.org/news/2019-03-near-perfect-low-cost-semiconductors.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no part may be reproduced without the written permission. The content is provided for information purposes only.
1114 shares

Feedback to editors

User comments

Mar 16, 2019
Whatever ya call it, ya manipulating charge motion. So juz stop trying to justify that QM nonsense. Those day's died around 1955. Ya even pointed out how nothing quite clicked. Please, you are ruining the future. By giving students so much bull$hit to study that leads them no where but stupid is juz wrong! Logic is required to make sense. 21st century. Stop the Nonsense!

Mar 16, 2019
N × 6.022140857 × 10^23 per gram-Mole! com'on, geometry; they each respond! Set C=1; T=Lambda, follow the field "changes"! Jeez! Excuse my nutty nomenclature. I'm rotten wit names!

Mar 16, 2019
You can buy GaN based LEDs that have wall plug efficiencies north of 80% by the square meter.

Wall plug efficiency is the real work measure. After a photon is created/emitted it has to escape the package and get into the real world. LEDs are pushing up against that same quantum efficiency limits listed here - no real surprise, only researchers who don't do a proper literature search or perhaps university press offices that are desperate for a headline.

Mar 16, 2019
Oh and quantum dots aren't cheap - not compared to bulk semiconductors.

Mar 16, 2019
Not so sure about how expensive quantum dots are any more, @LED. They used to be fabricated, like chips are. Now these guys appear to be making them in a chemistry lab. That's gonna be a lot cheaper than wafer fab methods.

Mar 16, 2019
Wanna bet? Semiconductor fabs crank out semiconductor layers by the acre and they produce the entire layer sequence needed for a device.

Just how do you think quantum dots are incorporated into devices which still need layers above/below? That solvents, conditions and time required to make quantum dots are expensive in comparison.

Don't forget you have to add ligands to the outside of quantum dots to prevent surface states from forming.

Mar 16, 2019
I worked in the semiconductor industry, @LED. I know exactly what's involved, and it makes a chemistry lab look like Romper Room. In fact, there are apparently YouTube videos on how to make quantum dots at home. I haven't reviewed them, so I don't vouch for them, but apparently they're now making quantum dot displays and TVs, arguing the fabrication price isn't very high.

Mar 16, 2019
Let's juz say, this is like spinning ya tires while Tesla's already at the finish line. Ya can model $hit billions of times cheaper, faster, and easier; but, that's juz to begin to bring us into the 21st century. It simply puts all this $hit in the dust. Go ahead and use it and publish it as QM nonsense as ya world burns! LOL!

Mar 16, 2019
Production of quantum dots via solution chemistry has been known at least since the 1980s.

It's also difficult to make quantum dots that don't have a perfect crystalline structure. The energy required for a defect is enormous compared to the number of atoms in the QD. I suspect the real difference may be in the sensitivity of measuring the quantum efficiency. There really isn't much of a difference between 99% and 99.9% or even 99.999%.

You still lose energy (Stokes loss) between absorption and emission.

Mar 16, 2019
I've been making LEDs for 30 years and started working with quantum dots for use with LEDs 20 years ago. I might have a reasonable idea of how they are used and the issues. But feel free to look into it yourself.
There is a difference between making quantum dots and making a core/shell quantum dot with the appropriate surface ligands to stablize them over the long term.

Mar 16, 2019
So where are these cheap quantum dot TVs and displays coming from?

Mar 16, 2019
I've been making LEDs for 30 years and started working with quantum dots for use with LEDs 20 years ago. I might have a reasonable idea of how they are used and the issues. But feel free to look into it yourself.
There is a difference between making quantum dots and making a core/shell quantum dot with the appropriate surface ligands to stablize them over the long term.

Juz ignore the word quantum and add science and logic. It becomes obvious! QM is a Gerry rigged wave equation; we are not stupid.

Mar 17, 2019
Quantum dot TVs/displays sell at a premium compared to standard LED backlight TVs/displays. Taking into account that you don't use a lot of quantum dots in those TVs and that should tell you something.

The phosphors used for LEDs are a few hundred $/kg. The equivalent amount of QDs for the same degree of color conversion are expensive in comparison. I suggest you look at what it takes to make a phosphor. Mix some powders and cook at a high temperature for a few hours. It is very simple chemistry, the complication is the high temperature furnace.

Mar 17, 2019
Ah, but since the announcements of chemical means of making quantum dots have only recently been announced in the scientific literature, I strongly suspect that the current ones are still being fabricated. They'll skim the gravy off the market, then compete for market share. At which point the cool will have worn off anyway, and the emphasis will be on volume, and that means low prices.

That's how it usually works.

Mar 17, 2019
And just think: you won't need a foundry to make the silicon crystals, you won't have to do photolitho, and inspection, and ion implant, and it doesn't need to be in a vacuum. Just a chemistry lab. Much cheaper equipment-wise. That makes it more agile.

I worked in processor fabs, mostly.

Mar 18, 2019
I am guessing that your semiconductor experience is with silicon and not compound semiconductors. While you can fabricate quantum dots using standard fabrication methods and multiple growth steps, it is a completely different process from colloidal synthesis. Here is a quick link (https://pubs.acs....174a003)

You can also search Steigerwald and colloid to find a host of other links.

Seth Coe-Sullivan founded QD Vision in 2004 based on colloid/solution synthesis of QD materials (http://www.rle.mi...005.pdf)

Do you need more to convince you that I know what I'm talking about and that colloidal (solution/beaker) synthesis of QDs has been around for decades?


Mar 18, 2019
But see, the thing is, what these folks appear to be saying is that previously used measurement techniques may not have been effective. So that kinda makes one think about how good your synthesizing techniques could be if you couldn't tune them with quality control.

Mar 18, 2019
By 2000 GaN based LEDs had internal quantum efficiencies north of 80% (note that doesn't include electrical or optical loses, only photons created per electron).

Today I can buy a blue LED that has a wall plug efficiency >80%. When you factor in the electrical losses and the optical extraction efficiencies (chip to package and package to air) you are in the neighborhood of 100% quantum efficiency.

The only thing they are really talking about is being able to measure the difference between 99.9% quantum efficiency and 99.99% or 99.999% quantum efficiency. Sorry, but you aren't going to enable much since you have to account for Stokes loss in the process.

Do you understand the difference between a measurement technique and a synthesis technique?

Also this has a university press office spin. What are you really going to enable with a fluorescence dye with a quantum efficiency of 99.99% that you couldn't do with a fluorescence dye with a quantum efficiency of 99%?

Mar 18, 2019
N × 6.022140857 × 10^23 per gram-Mole! com'on, geometry; they each respond! Set C=1; T=Lambda, follow the field "changes"! Jeez! Excuse my nutty nomenclature. I'm rotten wit names!

Still don't get it? Set q=+/- 1; Neutron = Both at the same center.. Finish!

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more