a brief physics distraction

Why not distract yourself from a doom scroll by thinking about photons?
The story of two-photon absorption.

[gingerrocco]

I’ll need one of these everyday for the next 4 years please

[CaptainDisillusion]

I can't believe the two photons are polling neck and neck.

[Leveret]

Ooh ooh I finally know something relevant! My PhD is in nonlinear optics, but we collaborated closely with people on the nonlinear bioimaging side of things, so that's the perspective I'll be coming from.

Say you want to image deep into live tissue -- for instance, ~1 mm into the cerebral cortex, which is where all the good thinky stuff happens. You're generally going to run into two problems: scattering and absorption. Scattering means the light you shine down isn't all going to end up where you want it to, and the fluorescence you get back isn't all going to come from where you think it does. Both of those effects increase your background. On the flip side, absorption means even the unscattered light that makes up your signal is going to be super weak. Together, that's a recipe for a really terrible signal-to-noise ratio. You can play with these effects somewhat, however, by changing the wavelengths you operate at. Scattering gets worse at shorter wavelengths, while absorption mostly follows the water spectrum, peaking strongly at ~1400 nm and ~2000 nm. If you overlay those spectra, you can actually see "windows" appear at ~1300 nm and ~1700 nm. Those are the two wavelength bands that'll let you see deepest inside dense tissue.

Now, there's a problem with that: you still need a fluorophore that matches your laser wavelength. There are all sorts of wonderful dyes that have been developed for use around the 500-800 nm range, but that's a terrible wavelength band for deep-tissue imaging. Scattering is very strong around that region, as you can see by holding your finger up against your phone light. It just so happens, however, that 1300 nm and 1700 nm are very nicely multiples of 2x or 3x from a lot of those dyes' absorption bands, making them perfect candidates for two- or three-photon absorption.

It's actually even more perfect than that, for another reason your video touched on. Sure, the emitted fluorescence will still be in that highly-scattering 500-800 nm band, but you don't actually care! You already know where every photon you detect is coming from: it's coming from the focus of the laser beam, because that's the only volume where you'll get efficient multi-photon absorption.

The catch, because of course there is one, is that there aren't a lot of great laser sources at 1300 nm or 1700 nm. To get the most signal for the least amount of photodamage to the tissue, you want to use very short pulses of light, which requires very specific kinds of laser crystals that nature ended up not providing us. That's where nonlinear optics comes in, as a means to take pulses of light generated at one wavelength and efficiently convert it to the wavelengths you actually want.

But I digress. For the specific application I worked on, it's hard to do much higher order than three-photon absorption for the reasons I laid out. If your excitation wavelength gets too close to 2000 nm, you'll lose too much due to absorption; if the emission wavelength gets much shorter than 400 nm or so, you won't have any good dyes to work with. That really hems you into two- and three-photon absorption using the wavelength bands I mentioned (I think I've seen one or two papers that go to four-photon absorption, but I don't think they really caught on). The base efficiency of the process also gets worse very quickly the more photons you mix. Nonlinear optics can be understood classically as a perturbation series of Maxwell's equations in a dielectric where instead of the P = chi * E we learn in kindergarten E&M, it's now more like `P ~ chi1 * E + chi2 * E^2 + chi3 * E^3 + ...`. Past the first handful of terms, obtaining any appreciable polarization requires unreasonably high field strengths. Like, in the "can't propagate through air because the beam will start tearing apart the nitrogen molecules" strong.

Anyway, apologies for going long -- I, too, needed a distraction tonight. Thanks for the video, and for all the other amazing science communication content you make!

[trickvro]

My internal monologue: ᵃᵃᵃᵃᵃᵃᵃᵃᵃᵃᵃᵃᵃᵃᵃᵃᵃᵃᵃᵃ
Dr. Angela: Hey, here's a fun story.
Me: Oh, that sounds like a nice distraction.
Dr. Angela: So it's the 1920s, and we're in Germany...
My internal monologue: ᴬᴬᴬᴬᴬᴬᴬᴬᴬᴬᴬᴬᴬᴬᴬᴬᴬᴬᴬᴬ

[IanMcKellar]

This is 100% what I needed today. And tomorrow. And probably for the next week.

[jacklefevre]

thanks for the physics distraction, here's an even more brief engineering distraction:
Two photon absorption is also used in micro-scale 3d printing! They focus femtosecond laser pulses inside a material that polymerizes into a solid when it absorbs two photons, which lets them solidify a single point at the focus. Two-photon absorption lets them cure the material just at the focal point rather than the entire path the laser takes. Move around the focus in the shape of your object, and you get a solid part!

[error.]

11:20 In her paper she called it "Der Simultanemission zweier Lichtquanten" which is roughly "The simultaneous emission of two light quanta" with "Lichtquanten" basically being the German precursor word to "photons." She also called it "zwei Quantensprüngen" or "two quantum leaps" which is lovely. She also talks about the inversion of this process, where "zwei Lichtquanten" (two light quanta) whose frequency sum is equal to the excitation frequency of the atom, act together to excite the atom.

[Lambda_Ovine]

It will never cease to amaze me that, through a bunch of very rigorous math, you can basically stumble upon a conclusion about reality, and years and even decades later it's confirmed through experimentation that, yes, that's indeed reality. That's the most fascinating aspect of physics to me, that there are times you can make discoveries before realizing you've made a discovery

[xBris]

6:03 - Habilitation is a process at German universities to get your "licence to teach" - your venia legendi. It's a thing you do after your PhD and in the vast majority of cases also *after* a Post-Doc. It usually consists of a couple of years of research which you then have to publish like a doctoral thesis (usually as a cumulative thesis though). In the Anglo-saxon sphere you skip this whole step and become a Junior-Professor instead. In the alter years, the Junior-Prof track is also getting more common in Germany, but the "old" Habilitation track is still the more common one.

[markTheWoodlands]

Nothing better than an exceptional distraction based on light - on an otherwise very dark day. Thank you Doctor Collier.

[DTcorn]

I don't understand most of what Angela says, but I cannot state how much I enjoy listening to her say it. The excitement just comes through beyond the words spoken.

[rott]

Awesome video! From a biologist here, 2 photon microscopy helps to image deep tissues because infrared light scatters less in tissue than visible light. Also since the only place that has enough excitation is the center of focus (as the lens focuses the infrared light into a little dot) you can observe any light that comes out of the tissue and assume it is from the center of focus. These two factors help to image deep inside tissues. longer wavelengths for 3x or 26x microscopy may penetrate tissues but you get into resolution problems as the imaging resolution is related to the wavelength of light.

[jpsenna]

Thank goodness for Angela's little island of knowledge, civility and good humor.

[richardv.]

An interview around 2050:
-Angela, where did you get your inspiration for your Nobel prize-winning research?
- I don't know, I just needed a major distraction 24/7 for the last 25 years.

[xelaxander]

German here. The title of Göppert‘s thesis translates to „About Elementary Acts with Two Quantum Leaps“.

[roberttorres]

A lifetime of study in under 23 minutes along with schematas explained, in very basic terms, is the sign of pure genuis. Of course I subscribed, I know a gold mine when I hit gold❤.

[stiofanmacamhalghaidhau]

possible take-aways from this video:
- Angela is not fluent in German
- working together is better than a solo run
- women who are not blocked from opportunity can do great things
- n^z1s are the bad guys, mkay?
- it is easy to mispronounce Göttingen
- if you mispronounce Göttingen there will be people [see comments section] happy to help
- you can try, but you will never really distract yourself from 'things happening today'
- try anyway
- no really, even if it gives only momentary respite
- Maria Goeppert-Mayer was pretty cool
- Paul Dirac was also pretty cool
- Angela Collier is also pretty cool

[thecloofer]

But, but my TV gives me photons! (also, thank you for reminding people that the nazis are the bad guys two times. I don't think there is an upper limit with diminishing returns on that one, so we can, and should, all just keep on saying it)

[neilcarver]

Arguably the greatest video ever posted to Youtube. Thank you.

[SteveMassing]

Thanks for the distraction 😢 And thanks for all the rest of your videos!