Researchers build first physical 'metatronic' circuit

Feb 22, 2012
Figure A. When the plane of the electric field is in line with the nanorods the circuit is wired in parallel.

(PhysOrg.com) -- The technological world of the 21st century owes a tremendous amount to advances in electrical engineering, specifically, the ability to finely control the flow of electrical charges using increasingly small and complicated circuits. And while those electrical advances continue to race ahead, researchers at the University of Pennsylvania are pushing circuitry forward in a different way, by replacing electricity with light.

“Looking at the success of electronics over the last century, I have always wondered why we should be limited to electric current in making circuits,” said Nader Engheta, professor in the electrical and systems engineering department of Penn’s School of Engineering and Applied Science. “If we moved to shorter wavelengths in the electromagnetic spectrum — like light — we could make things smaller, faster and more efficient.”

Different arrangements and combinations of electronic circuits have different functions, ranging from simple light switches to complex supercomputers. These circuits are in turn built of different arrangements of circuit elements, like resistors, inductors and capacitors, which manipulate the flow of electrons in a circuit in mathematically precise ways. And because both electric circuits and optics follow Maxwell’s equations — the fundamental formulas that describe the behavior of electromagnetic fields — Engheta’s dream of building circuits with light wasn’t just the stuff of imagination. In 2005, he and his students published a theoretical paper outlining how optical circuit elements could work.

Now, he and his group at Penn have made this dream a reality, creating the first physical demonstration of “lumped” optical circuit elements. This represents a milestone in a nascent field of science and engineering Engheta has dubbed “metatronics.”

Engheta’s research, which was conducted with members of his group in the electrical and systems engineering department, Yong Sun, Brian Edwards and Andrea Alù, was published in the journal Nature Materials.

Figure B. When the plane of the electric field crosses both the nanorods and the gaps the circuit is wired in series.

In electronics, the “lumped” designation refers to elements that can be treated as a black box, something that turns a given input to a perfectly predictable output without an engineer having to worry about what exactly is going on inside the element every time he or she is designing a circuit.

“Optics has always had its own analogs of elements, things like lenses, waveguides and gratings,” Engheta said, “but they were never lumped. Those elements are all much larger than the wavelength of light because that’s all that could be easily built in the old days. For electronics, the lumped circuit elements were always much smaller than the wavelength of operation, which is in the radio or microwave frequency range.”

Nanotechnology has now opened that possibility for lumped optical circuit elements, allowing construction of structures that have dimensions measured in nanometers. In this experiment’s case, the structure was comb-like arrays of rectangular nanorods made of silicon nitrite.

The “meta” in “metatronics” refers to metamaterials, the relatively new field of research where nanoscale patterns and structures embedded in materials allow them to manipulate waves in ways that were previously impossible. Here, the cross-sections of the nanorods and the gaps between them form a pattern that replicates the function of resistors, inductors and capacitors, three of the most basic circuit elements, but in optical wavelengths.

“If we have the optical version of those lumped elements in our repertoire, we can actually make designs similar to what we do in electronics but now for operation with light,” Engheta said. “We can build a circuit with light.”

In their experiment, the researchers illuminated the nanorods with an optical signal, a wave of light in the mid-infrared range. They then used spectroscopy to measure the wave as it passed through the comb. Repeating the experiment using nanorods with nine different combinations of widths and heights, the researchers showed that the optical “current” and optical “voltage” were altered by the optical resistors, inductors and capacitors with parameters corresponding to those differences in size.  

“A section of the nanorod acts as both an inductor and resistor, and the air gap acts as a capacitor,” Engheta said. 

Researchers build first physical 'metatronic' circuit
An illustration of an array of silicon nitrite nanorods. The entire array is about half a millimeter long.

Beyond changing the dimensions and the material the nanorods are made of, the function of these optical circuits can be altered by changing the orientation of the light, giving metatronic circuits access to configurations that would be impossible in traditional electronics.

This is because a light wave has polarizations; the electric field that oscillates in the wave has a definable orientation in space. In metatronics, it is that electric field that interacts and is changed by elements, so changing the field’s orientation can be like rewiring an electric circuit.  

When the plane of the field is in line with the nanorods, as in Figure A, the circuit is wired in parallel and the current passes through the elements simultaneously. When the plane of the electric field crosses both the nanorods and the gaps, as in Figure B, the circuit is wired in series and the current passes through the elements sequentially.

“The orientation gives us two different circuits, which is why we call this ‘stereo-circuitry,’” Engheta said. “We could even have the wave hit the rods obliquely and get something we don’t have in regular electronics: a circuit that’s neither in series or in parallel but a mixture of the two.”

This principle could be taken to an even higher level of complexity by building nanorod arrays in three dimensions.  An optical signal hitting such a structure’s top would encounter a different circuit than a signal hitting its side. Building off their success with basic optical elements, Engheta and his group are laying the foundation for this kind of complex metatronics.

“Another reason for success in electronics has to do with its modularity,” he said. “We can make an infinite number of depending on how we arrange different circuit elements, just like we can arrange the alphabet into different words, sentences and paragraphs.  

“We’re now working on designs for more complicated optical elements,” Engheta said. “We’re on a quest to build these new letters one by one.”

Explore further: A new multi-bit 'spin' for MRAM storage

Related Stories

Running electronics using light

Oct 19, 2009

(PhysOrg.com) -- "If you open up almost any electronic gadget, you will see various elements that operating using electric circuitries," Nader Engheta tells PhysOrg.com. "Many of them have different functi ...

Physicists rotate beams of light

Apr 05, 2011

Controlling the rotation of light – this amazing feat was accomplished at the Vienna University of Technology (TU Vienna), by means of a ultra thin semiconductor. This can be used to create a transistor ...

Recommended for you

IHEP in China has ambitions for Higgs factory

4 hours ago

Who will lay claim to having the world's largest particle smasher?. Could China become the collider capital of the world? Questions tease answers, following a news story in Nature on Tuesday. Proposals for ...

The physics of lead guitar playing

6 hours ago

String bends, tapping, vibrato and whammy bars are all techniques that add to the distinctiveness of a lead guitarist's sound, whether it's Clapton, Hendrix, or BB King.

The birth of topological spintronics

7 hours ago

The discovery of a new material combination that could lead to a more efficient approach to computer memory and logic will be described in the journal Nature on July 24, 2014. The research, led by Penn S ...

The electric slide dance of DNA knots

10 hours ago

DNA has the nasty habit of getting tangled and forming knots. Scientists study these knots to understand their function and learn how to disentangle them (e.g. useful for gene sequencing techniques). Cristian ...

User comments : 12

Adjust slider to filter visible comments by rank

Display comments: newest first

Lurker2358
4.4 / 5 (7) Feb 22, 2012
If they can do this, then they should be able to build an optical phased array at the nano-scale using carbon nanotubes or similar structures, which when taken together with optical circuitry, would allow the ultimate in 3d camera technology, range finding, and I believe microscopy.

One of the problems I had with the idea was a computer to process and store the data.

The type of circuitry described here should be able to accomplish this with no problem.

The orientation gives us two different circuits, which is why we call this stereo-circuitry, Engheta said. We could even have the wave hit the rods obliquely and get something we dont have in regular electronics: a circuit thats neither in series or in parallel but a mixture of the two.


Right. That's why I had proposed a phased array of planes of rods of different sizes, offset by fractions of their center to center distance, and also diagonal arrays.

The concept was from phased array radar.
Lurker2358
5 / 5 (4) Feb 22, 2012
These sorts of devices are the future of computing and sensor technology, IMO.

It should be possible to develop a single nano-scale optical array device capable of not only "faking" certain classes of quantum calculations in a single step, assuming very clever design implementation, but also detecting frequency, polarization, intensity, and point of origin light sources in a single step.

This should, IMO, allow the design of nano-scaled router systems for managing multi-cored systems better, neural nets (both series and parallel simultaneously, and generally analog processing in some applications.

Attempting to build "quantum" computers through expensive spintronics or entanglement schemes may be fruitful, but it may be much more cost-efficient to build optical devices that automatically perform "pseudo-quantum" or analog calculations through clever design tricks. After all, light already has a quantum nature, it's just a matter of circuitry that is able to abuse it for useful work
finitesolutions
5 / 5 (3) Feb 22, 2012
How difficult can it be to use some optical nanodevices to implement simple Boolean gates? If XOR, OR, NOR, AND, NADN, NOT are being implemented using light optical processors will be easy to build. Adding, substracting, multiplying, dividing is easily implementable with gates and microcode.
Lurker2358
4 / 5 (4) Feb 22, 2012
I think one of the problems right now is there is no secondary storage technology capable of storing the amount of data this class of sensors and circuitry can theoretically collect in just one instant, one "frame"...

en.wikipedia.org/wiki/Phased_array_optics

Consider this, if your nanorods are 6nm diameter with 6nm spaces between them then at is 83,333 rods in one millimeter, and this is assuming just one layer of rods.

If you had a second layer behind it, staggered by 3nm, OR perpendicular, or diagonal, etc, then you'd have twice as many rods, or "N" layers with various configurations of perpendicular, diagonal, and staggered arrangements...This should allow different types of instantaneous calculations such as point light origins, distance, frequency, etc, for perfect 3d cameras, etc.

It's on the order of potentially ten-billions of pixels per millimeter of surface of the sensor or circuit.

Each rod should be theoretically capable of doing both digital and analog data.
StarGazer2011
3 / 5 (2) Feb 22, 2012
@finitesolutions: FYI, if you can do NAND gates you can replicate everything else using NAND gates.
kaasinees
0 / 5 (21) Feb 23, 2012
How difficult can it be to use some optical nanodevices to implement simple Boolean gates? If XOR, OR, NOR, AND, NADN, NOT are being implemented using light optical processors will be easy to build. Adding, substracting, multiplying, dividing is easily implementable with gates and microcode.

Optical memory?
infinite_energy
not rated yet Feb 23, 2012
Optical memory?
You mean bluray discs?
Lurker2358
2.3 / 5 (3) Feb 23, 2012
Optical memory?
You mean bluray discs?


No. You need photonic ram for starters.

and then for storage you need something like racetrack memory or some other exotic media that can store completely absurd amounts of data.

Like, for example, an artificial diamond with spin states stored in defect atoms between the crystal lattices.

Or you could just use rooms full of backup drives, but that will become a major bottleneck and kill most of the performance gains of the optical circuitry. Which is a big problem IMO.

If only there was a way to actually capture a photon and save it indefinitely for later use...
infinite_energy
not rated yet Feb 23, 2012
Light can travel around the world 7 times a second. There are fiber optics circling the planet. If you send something to yourself on a planetary loop it will arrive a seventh of a second later thus acting like a memory.
If you make the loop bigger the period of storage extends.

I guess the photons are somehow absorbed or lost by the medium in the end.
kaasinees
0.3 / 5 (22) Feb 23, 2012
Optical memory?
You mean bluray discs?


No, you need memory for registers and cache. stack memory. Then you need heap memory. Transistors are useless if you don't have these.
infinite_energy
not rated yet Feb 24, 2012
No, you need memory for registers and cache. stack memory. Then you need heap memory. Transistors are useless if you don't have these.

Check this: http://en.wikiped...ic_logic

I am sure solutions to this problems will be found. Logical operations can be implemented without transistors.
TabulaMentis
not rated yet Feb 24, 2012
From Physorg.com article:
"The technological world of the 21st century owes a tremendous amount to advances in electrical engineering."

I concur electrical engineers are becoming more relevant than physicists. As a matter of fact, when gravity engines are developed it will be done by electrical engineers, not rocket scientists.

"Researchers at the University of Pennsylvania are pushing circuitry forward in a different way, by replacing electricity with light."

I wonder if metatronics could be used to replace AC wiring so electromagnetic radiation poisoning would be avoided?