Metamaterials and Photonic Crystals

By Ctein

The future of photographic optics and sensors is going to be profoundly affected by advances in materials science that most photographers have never even heard of.

Broadly defined, metamaterials are substances whose properties depend on their microstructure more than their chemical composition and that exhibit novel properties not usually found in nature. Most of the definitions I've seen don't include photonic crystals, although I must confess that I don't understand why. Photonic crystals are also materials that have novel properties due to their structure.

The most notable thing about photonic crystals is that they can be what amounts to an optical semiconductor. Electronic semiconductors are valuable because they only allow electrons to have certain energies; between those allowable energy bands are what are called bandgaps. An electron cannot exist inside the material in that bandgap. The way an electron gets from one conduction band to another is by absorbing or emitting a photon that changes its energy by enough to let it jump across the bandgap. You can see where that leads: photodetectors, LEDs and laser diodes. By adding small amounts of contaminants (a.k.a. dopants) to a semiconductor, one can tailor the energy levels and bandgaps to one's needs, yielding the whole array of solid-state electronics and integrated circuits that exist today.

Photonic crystals can play the same tricks with photons. Within them, certain energy levels (a.k.a. wavelengths, a.k.a. colors) are forbidden while others are not. A natural example of a photonic crystal is opal; it's not chemistry but the geometric packing of silica spheres in bulk opal that cause the brilliant flashes of color that we see. Those packed spheres will only allow certain energy levels to exist. The phenomenon is very different from diffraction; when you tilt an opal, you don't see a continuous change in color as you would with a diffraction grating, instead you see it flash brilliantly when the angles are just right to reflect light at an allowable energy level into your eyes.

Artificial photonic crystals, like semiconductors, can be made to do all sorts of interesting things. You can make photonic crystals that act like diodes: Light can pass through them in one direction but not in the other. True one-way mirrors. You can make converters: if a photon of forbidden energy attempts to pass through the crystal, it's converted to a photon of a different energy.

Applications in sensor design are immediately apparent. Near-perfect anti-reflection coatings are one possibility. Another is much better color filters for the sensor. For example, digital cameras don't do well with blue light for two reasons only one of which is that today's blue filters aren't anywhere near perfect. The other is that silicon is relatively insensitive to blue light. Consequently, digital cameras don't have anywhere near perfect efficiency in this part of the spectrum.

One can imagine a photonic crystal blue filter. It blocks red and green light entirely. It doesn't allow blue light to passthrough, either, but it does convert blue photons to, oh say, infrared photons. Silicon is very responsive to infrared photons; the same sensor would have much more sensitivity with this filter than with a conventional one. A filter  might even convert a blue photon into two infrared photons, thereby doubling the signal in the sensor and improving the signal-to-noise ratio in the final output.

This is relatively normal stuff compared to metamaterials. They just plain break the rules, at least as we thought we understood them. When modern metamaterials were proposed about 10 years ago, the mathematics of them allowed for such outrageous possibilities that many physicists fairly felt that they would turn out to be a theoretical concept that could never be realized in actuality. Well, that turned out to not be the case. We can make metamaterials, and they have proven themselves to be extremely weird.

The most important weird property for photographers is that metamaterials can have refractive indices of less than one. Nothing "normal" does that; a pure vacuum has a refractive index of one and all ordinary materials and gasses have refractive indices higher than one. Not only can metamaterials get below one, they can even go negative. The holy Grail has been a refractive index of less than –1 at optical wavelengths, broadband, and we're closing in on that.

What do such negative indices allow that you didn't have before? Superlenses and hyperlenses: optics that perform better than what conventional optical theory allows in a perfect lens. You think the Rayleigh diffraction limit is the ultimate resolving limit for a perfect lens? Not for a superlens; it can resolve much better than that theoretical boundary. Do you believe that you can't build a lens that can resolve details smaller than a half wavelength of light? Hyperlenses blow right past that.

These technologies are working their way out of the laboratory into scientific (translated: expensive!) instruments right now. They will migrate down to the photographic equipment that we use. Unless, of course, something even better comes along in the meantime.

Timeframe? My guess is during the next decade; I'd be surprised to see much appearance of these materials in our toys before 2020, but I'd be utterly amazed if it was after 2030.

Ctein

Physicist/photographer Ctein's regular column on TOP is published on Wednesdays.

Comments

[David Dyer-Bennet]

I'm amused that, at least in the 2007 papers I've glanced at, silver is still playing a role in the process :-).

[Phil Allen]

Cool thing is that these things are found nature, we just didn't really know what was going on. For example, iridescent beetles and butterfly wings behave like photonic crystals. How all this will benefit photographers is an open ended and exciting question.

[John Woods]

Sir John Pendry at Imperial College London has studied flat lenses with negative refractive index. Hasselblad's magazine Vistor had an interesting article by him a few years ago.Read more at
http://www.cmth.ph.ic.ac.uk/photonics/Newphotonics/PerfectLens.html

[Steve Wake]

Interesting! Can you draw a ray diagram or some other illustration demonstrating the properties of a lens made from material with a refractive index <1; <0 ? This might help me wrap my head around this unnatural phenomenon. Thanks!

[Peter]

Actually you've probably heard of these before, but only in the outrageous realms:
http://www.msnbc.msn.com/id/12961080/ns/technology_and_science-science/

[Nick]

In a session on metamaterials (at Photonics West 2009, if I recall correctly), one of the presenters drew the distinction between photonic crystals and metamaterials: Photonic crystals have structure on the order of the wavelength of the light it is intended to modify, while metamaterials have strucutre much smaller than the wavelength of the light.

[Nick]

Also, I will say that I am not quite as sanguine about the prospects of these materials as Ctein. Much of what has been proposed looks great when you run a simulation of it, but fabricating these materials is... well, even the usual scientists jokey understatement, "nontrivial," is probably not adequate here. I expect that there will be some real applications of these things (optical fibers with photonic crystal structures are already available), but I guess I'll be (pleasantly) surprised if we ever see a true 3-dimensional superlens that works at optical frequencies.

[richard]

Crikey!

Ain't it tough being a photographer when all these technological advances are just screwing with your ability to take a photo

I guess I'll have to bypass that and crack open some more Tri-X.

[Arne Croell]

"A natural example of a photonic crystal is opal; it's not chemistry but the geometric packing of silica spheres in bulk opal that cause the brilliant flashes of color that we see. Those packed spheres will only allow certain energy levels to exist. The phenomenon is very different from diffraction; when you tilt an opal, you don't see a continuous change in color as you would with a diffraction grating, instead you see it flash brilliantly when the angles are just right to reflect light at an allowable energy level into your eyes."

While I do agree that Opal is a photonic crystal, and that the effect is based on the internal structure, that specific effect is still based on diffraction. Compared to a grating the opal is a 3D lattice, giving you high amplitude diffracted beams for a given wavelength just in certain directions, not a continuous spread. It is the same as a Laue pattern for "white" X-rays being diffracted by a crystal, giving a set of discrete spots fulfilling the Bragg condition; in the case of Opal its for visible light, since the silica sphere size is similar to its wavelength.

[James]

Ctein,

fascinating stuff. I'm beyond the limit of my understanding of the science, but I know enough about military R&D and the budgets that support it to suspect that there'd be lots of research grants available if you can tie the use of photonic crystals or metamaterials to improvements in field detection of explosives in near real time.

All current technologies seem to work in theory or even in the lab, but don't in the field (remember, time is a critical factor) because the information from specialist instruments for detecting changes in radiation or UV fluorescence cannot be "pictured" by humans. Audible alerts are better than nothing, but a very quick decision needs to be made of the size of explosives detected to either deal with the explosives, or seal off the area. Either way, the local military commander - usually a young officer with no particular science background - has no "picture" of what has been detected, just a bleeping machine and a direction of threat within a +/-500 millirad subtension. If these metamaterials can aid in converting information from one part of the EMS into a part that a simple soldier can literally see on a screen, then there's a huge step forward in practicability.

An example of the sort of approach now being taken is in the Patent at http://www.wipo.int/pctdb/en/ia.jsp?IA=US2007063789.

[Archer]

I found a reasonably clear explanation of negative index of refraction at http://people.ee.duke.edu/~drsmith/negative_index_about.htm

Nit: its the Rayleigh criterion, not Raleigh.

[Geoff Wittig]

Can't remember whether it was Azimov or Arthur C Clarke who said "A sufficiently advanced technology will be indistinguishable from magic". But that seems appropriate here.

[Nicolas]

"Broadly defined, metamaterials are substances whose properties depend on their microstructure more than their chemical composition and that exhibit novel properties not usually found in nature. Most of the definitions I've seen don't include photonic crystals, although I must confess that I don't understand why...."

Definitions doesn't include photonic cristals probably because they are "broadly" found in nature...think about some insects and butterflies physical colors, etc.
You can find papers from the University of Namur(Belgium), Physics department, research center in Physics of Matter and Radiations(PMR).

This paper could be interesting (friends and colleagues works):
http://iopscience.iop.org/1367-2630/10/1/013032/pdf/njp8_1_013032.pdf

Evolution made it!-)

It's funny to speak about these works on my preferred website.
Thanks for your physics essays...

Nicolas (Belgium)

[yunfat]

Japanese companies manufacture and design all the sensors in todays DSLR's, save Kodak (ala M9). Given the choice between a progressive sensor technology (say Foveon, or Fuji S-R) and a conventional one, the choice of a radical sensor historically guarantees you will sell zero units. Since Japanese companies tend to err on the side of caution, the chance of these technologies coming online in a meaningful way will have to happen by one of the bit players.

The bit players right now are Fuji, Sigma, and Kodak. Fuji has given up on its sensors, Sigma has yet to leverage Foveon in a meaningful way, and Kodak, well, we can count on them to screw up.

Nikon and Canon, by the time any new technologies come around, will have leveraged even more from conventional bayer array sensors, as they have always done in the past, all the while conserving their R&D budgets and extracting greater profits from an established process.

In other words, I hope so, but the odds of someone putting profits into a new sensor in this economy are zero. The odds of a new sensor propelling sales of a new camera, are also zero, since by the time we have extracted the promise of the new sensor via RAW conversion (software takes a little time to catch up with the hardware), Nikon and Sony and Canon have enhanced their traditional sensor to match the performance of the new tech.

So, while the new technology may sound promising, profits will determine the future of sensor development in a far greater way than any new process.

If you want to see the technologies you speak of, it would be faster for you to start a camera company and assemble investors than count on any existing camera company to deliver it. In other words, your prediction of 20 years sounds about right.

Just my 2ç.

[John Ashbourne]

Thank you, Ctein, for your patient courteous approach to this sort of material.
It is as if it never occurs to you that most of us have not the slightest clue about what you take for granted - though I suspect that you must realise this to be able to deal with it with such direct simplicity.
However to do it without condescension even when using phrases like,"you can see where that leads" and "applications..are immediately apparent" is a true definition of intellectual courtesy.
I always read these articles but have to confess that as a mere historian i usually end up none the wiser but having enjoyed the ride .. oh yes and very grateful that someone out there who does understand it all is keeping an eye on it for us mere non-physicists photographers.

[David Dyer-Bennet]

Geoff, it was Clarke; that's "Clarke's Law", and the canonical phrasing is "Any sufficiently advanced technology is indistinguishable from magic." (Or "Clarke's Third Law"; #1 is "When a distinguished but elderly scientist states that something is possible, he is almost certainly right; when he states that something is impossible, he is probably wrong." #2 is "The only way of discovering the limits of the possible is to venture a little way past them into the impossible." #3 is much the most widely quoted, hence the numbering starting to fall away, I guess.

The modern tech joke take-off from it is "Any sufficiently advanced technology is indistinguishable from a rigged demo." :-)

(And the spelling of the other author you were thinking of is "Asimov". Now I've guaranteed at least three spelling errors in my message. :-) )

[MJFerron]

Interesting stuff Mr. Ctein but please forgive me when when I admit that advanced tech talk makes my head hurt. Personally I like the world of B&W film tech talk where cutting edge TMX is still looked at with suspicion by many. I can handle it at that level. :)

[ctein]

Dear DDB,

It's amused me, too. Now the silver's in the lenses, not the sensor.

~~~~~~

Dear Steve,

Archer's and John's references have pictures. (Thanks, guys.)

~~~~~~

Dear Nick,

Oh, that's a very nice distinction. Both physically significant and practical.

For the lay audience, an example of the difference that scale makes can be seen by looking up at the sky. The sky is blue because of Rayleigh scattering, which occurs when particles are much, much smaller than a wavelength of light (air molecules, in this case). Clouds are white because of Mie scattering, which occurs when the particles are of the same order as wavelengths of light (the water droplets).

I don't think progress will be fast. That's why I am putting it in the 10- to 20-year timeframe. But over the past decade we've seen a steady march down in wavelengths from centimeter scale down to near-infrared. I'll actually go so far as to predict mid-wavelength optical in less than five years. I think a bigger problem is designing broad bandwidth materials rather than ones that operate over a very narrow range of wavelengths.

~~~~~~

Dear Arne,

My apologies for the sloppy writing; what I should've written was that “the phenomenon is very different from simple diffraction…” It's probably worth mentioning to the general readership that metamaterials don't involve any novel physics. The way the light interacts with the material is through the traditional mechanisms of absorption, refraction, and diffraction. At the molecular/atomic level, the stuff behaves completely normally. It's the geometric arrangement that gives its novel properties.

pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
======================================
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 
======================================

[ctein]

Dear Archer,

Thanks for the spelling correction. Problem is that I dictate these things, and the transcription software doesn't always get it right. It's really, really hard to proofread proper nouns; the brain just skips over them. One way (the only way) that typing is better.

~~~~~~

Dear Yunfat,

Off the top of my head, I don't see ready ways to apply this to sensor technology. Where it's going to have the most impact is on how stuff gets to the sensor, hence my example of filters. But I don't think metamaterials are an alternate sensor technology.

As another example of the way they could have an impact on digital photography, I can see a ready application of hyperlenses, which can image below half a wavelength of light. Relay hyperlenses for microscopes have already been demonstrated, although I don't believe they're in production. The idea's that the hyperlens images stuff much smaller than a wavelength of light and magnifies it enough so that conventional optics can magnify it further.

Turned the idea around, and you have a conventional lens followed by a relay hyperlens that produces sensor plane detail much smaller than a wavelength of light. Now, why would you ever want such a thing? Well, as embedded cameras get thinner and thinner, the sensor has to get smaller and smaller. We already have cameras with pixels not much bigger than a micron; there aren't any fabrication roadblocks to making them a 10th that size. Except that 0.1 µ pixels wouldn't do you much good with conventional optics. No such problem with a relay hyperlens.

Not saying that anyone will ever follow this route. It's just another example of a direct and practical use of metamaterials, like the blue filter idea. Helps the audience see how this stuff might be relevant to them, someday.

pax \ Ctein

[ Please excuse any word-salad. MacSpeech in training! ]
======================================
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 
======================================

[Ed]

Hmmzzzzz, Chlorophyll A, can turn photon's into power due to a stateshift in the molecule when excited, and it acts as an electron donor.....there is a P680 (nanometer) and a P700 complex nicely tuned to different wavelength. What you are suggesting is that these photonic crystals are tuned to a specific energystate? Isn't that what a frequency is. A photon has a set amount of energy (an individual photon can't be brighter then it's college) if i'm not mistaken, but that energy is based on the frequency with which the photon swings. The higher the frequency the more energy is stored, wasn't it. What I would need is a photodetector (single state) that can detect photons of 3 (or more in order to make turtles happy as well) different energy levels simultaniously and make an addition for these over a set period of time. In addition to that these could be stored in a small array behind the photocell (3 bytes or better words per cell) and then a software could be used to ask the values of all these cells over a given amount of time. Thus.......bayer array adieu, and shutter adieu.

The future looks bright enough, but for now my road is fixed on 4x5.......:-).

Greetings, Ed

[Mel]

As a sharpness fanatic I'm curious about the group's opinion - how much resolution do we need in photography? And mirroring an earlier post, how much ISO? Everyone who needs to photograph individual atoms across the Grand Canyon in the dark please raise your hand....

[calvin amari]

With that doozy of a title - Metamaterials and Photonic Crystals - I had about zero expectation of being brought to ground by reading the piece but, as John Ashbourne notes, Ctein's patient and courteous approach pays off again. Well done.

[David Bennett]

Will it get me thinner spectacles (eyeglasses)?

[ctein]

Dear Mel,

Why does it matter? You know what you need. I know what I need.

There are folks out there having fun making gigapixel images-- they could clearly use a camera with billions of pixels. In response to my high ISO column, several readers suggested entirely reasonable photos they wanted to make (e.g. fireflies on a summer night) that required ISOs in seven figures.

So, pretty evidently, anything that's physically possible already has its potential users.

In which case, who cares what "the group" thinks?

Also consider the downward migration. *IF* I could build an ISO 1,000,000 gigapixel camera (I can't), I could have a cell phone camera that would produce photos of as high quality as the high end Canon and Nikon SLRs. How handy would that be?

The history of photography is littered with the corpses of folks who have pooh-poohed each improvement in quality and technology. They have always been wrong.

There's a good reason for that.

Time to learn from history instead of repeating it.

pax / Ctein

[James]

Ctein says "...Clouds are white because of Mie scattering"..... He is no doubt correct, but neglects to inform the readers that God - for a joke, and to give the British the ability to conquer 1/4 of the world's surface with only a cup of tea and a day old ham sandwich to sustain us - placed an ND8 filter permanently into the cloud above this benighted isle, and made the cloud permanently damp as well.

;)

[Eric Rose]

That's all very nice and gee whiz but what will the bokeh be like?

[David Dyer-Bennet]

My resolution demands are modest -- I only really need entire organic molecules, not individual atoms.

Still in the dark across the Grand Canyon, though. Hand-held, of course.

[Barbu]

So... I´m writting here to express my gratitude for having Ctein; from my position of a simple camera salesman (recently converted to the dreaded marketing domain) I´m scouring the interwebs :) long and hard for any bit of info which could have a connection with the photo technology, but this present article is simply from another world. I could look for this kind of info for years on end, and still couldn´t have find it anywhere else but here; maybe I even read about metamaterials, but surely couldn´t have made the connection with photo optics, supposing that I could even understand what´s the fuss about. Ctein simply explained it in layman´s (ahem... me) terms, and rekindled my curiosity about a domain (optics) that, up until now, I regarded as already stationary, everything worth discovering being already done since 50 years ago.
These being said... Where´s that page with Ctein sponsorship? I don´t need any prints or any kind of acknowledgment; I just want Ctein to be able to do what he does best.

[Luke]

How much resolution do we need? How much sensitivity?

Mu.

[Bruno Masset]

yunfat wrote:
> Since Japanese companies tend to err on the side of caution, the chance
> of these technologies coming online in a meaningful way will have to
> happen by one of the bit players.

If a technology is useful, it will be adopted, and not necessarily by a "bit" player.
Panasonic, for example, has apparently already thought of an application:

http://www.physorg.com/news98373788.html

[ctein]

Dear Barbu,

http://ctein.com/CollectCtein.htm

pax / Ctein

[ctein]

Dear Bruno,

OK, that's very cool!

When I wrote the column, I didn't know about this. I proposed photonic sensor filters for the future, and here they turn out to be four years old.

In fairness to yunfat, I think he was looking at the issue of altering the underlying sensor, not the add-ons.

pax / Ctein

[Steve Smith]

Now I've guaranteed at least three spelling errors in my message

Muphry's law: http://en.wikipedia.org/wiki/Muphry's_law