Optical Microscopy: the Resolution Revolution
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 340 | |
| Author | ||
| License | CC Attribution - NonCommercial - NoDerivatives 4.0 International: You are free to use, copy, distribute and transmit the work or content in unchanged form for any legal and non-commercial purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
| Identifiers | 10.5446/45022 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
|
00:00
NobeliumChemistryChemical compoundMolecular geometryLecture/Conference
00:35
NobeliumCell (biology)RibosomeAtomMitochondrionElectronMoleculeBase (chemistry)Functional groupChemical structureCell (biology)AtomChain (unit)WursthülleMoleculeComputer animationMeeting/Interview
01:04
Columbia RecordsNobeliumElectronMoleculeMoleculeAtomWursthülleProteinBase (chemistry)Cell (biology)Pharmaceutical drugElectronComputer animation
01:55
FluorescenceNobeliumDeterrence (legal)Cell (biology)ProteinOrganelleMoleculeSetzen <Verfahrenstechnik>ElectronSurface scienceRadioactive decayChemistryExciter (effect)FluorescenceSpeciesSeleniteLecture/Conference
03:18
FluorescenceElectronMoleculeAtomBase (chemistry)AreaIntergranular corrosionChemical elementWave propagationBase (chemistry)Chemical plantMoleculeMeeting/Interview
04:49
Base (chemistry)NobeliumSeparation processMoleculeCell (biology)FluorescenceElectronic cigaretteMemory-EffektMolecular biologyCrystallographic defectCoffeeOctane ratingBlock (periodic table)Meeting/Interview
06:35
Physical chemistryCell (biology)TopicitySequence alignmentTaxisProcess (computing)CancerNanoparticleAusgangsgesteinQuantum chemistryTodesbescheinigungAgeingElectronic cigaretteNobeliumHuman subject researchLibrary (computing)Medical historyHyperpolarisierungReflexionsspektrumGermanic peoplesCrystallographic defectAreaFracture (mineralogy)CheminformaticsOrlistatLecture/ConferenceMeeting/Interview
10:50
Quantum chemistryNobeliumMoleculeEmission spectrumStimulantSample (material)Phase (waves)Breed standardCell (biology)Assembly (demo party)ProteinQuantum chemistryEmission spectrumChemical propertyMoleculePhase (waves)WursthüllePatentElectronic cigaretteNuclear poreBase (chemistry)Essenz <Lebensmittel>SchmerzschwelleMolecular beamÜbergangszustandFunctional groupWine tasting descriptorsFluorescenceCheminformaticsExciter (effect)GesundheitsstörungZunderbeständigkeitWaterfallStuffingNahtoderfahrungSample (material)SimulationSense DistrictTractorSilencer (DNA)Setzen <Verfahrenstechnik>AreaIonenbindungMultiprotein complexCrystallographic defectElectron donorFracture (mineralogy)Ice frontRadioactive decaySeleniteOrlistatSunscreenScreening (medicine)Lecture/ConferenceMeeting/Interview
18:50
ProteinCell (biology)ProteinPatentShear strengthNanoparticleProcess (computing)
19:22
ZunderbeständigkeitMortality rateSekretionsvesikelColumbia RecordsNobeliumMilchsäureSample (material)MoleculeWine tasting descriptorsSchmerzschwelleShear strengthDeformation (mechanics)MolekulardynamikPhysical chemistrySetzen <Verfahrenstechnik>Cell (biology)Molecular beamCrystallographic defect
20:50
Sample (material)NobeliumLebensmittelfarbstoffMoleculeWursthülleElectronic cigaretteCrystallographic defectQuartzComputer animation
21:39
LebensmittelfarbstoffNobeliumSample (material)MeatThylakoidFertigteigKeratineCell (biology)HematiteZunderbeständigkeitColumbia RecordsSchmerzschwelleWursthülleProtonationSea levelCell (biology)Atomic numberÜbergangszustandMoleculeFluorescenceEmission spectrumQuantum chemistryOrders of magnitude (radiation)IsomerStickstoffatomWhitewaterDiamondShear strengthCHARGE syndromePotenz <Homöopathie>ChemistryBearing (mechanical)PressureOceanStatinElectronMortality rateLecture/ConferenceComputer animation
25:26
SolutionMalerfarbeDiatomics-in-molecules-MethodeMoleculeFluorescenceZunderbeständigkeitPig ironSoilNobeliumElectric currentDerivative (chemistry)SolutionMoleculeMolecular beamÜbergangszustandFunctional groupElectronic cigaretteZunderbeständigkeitNobeliumChemistrySunscreenCoordination numberChemical elementSample (material)Separator (milk)BET theorySet (abstract data type)Storage tankBleiglanzOceanCrystallographic defectBody weightFloodSea levelHydrophobic effectEngineering drawing
30:44
NobeliumComputer animationLecture/Conference
Transcript: English(auto-generated)
00:17
Thank you very much, Lorsch. I think everyone of us is familiar with the saying,
00:22
a picture's worth a thousand words or a seeing is believing. That not only applies to our daily lives, it definitely also applies to the sciences. And this is also why microscopy played such a crucial role in the development of the natural sciences. With light microscopy, mankind discovered that every living being consists of cells as basic units of structure and function.
00:43
Bacteria were discovered, and so on. However, we learn in school that resolution of a light microscope is fundamentally limited by diffraction to about half the wavelengths of light. And if you want to see smaller things, of course you have to resort to electron microscopy. As a matter of fact, the development of electron microscopy
01:01
in the second half of the 20th century changed things dramatically. You could see smaller things such as viruses, protein complexes, in a number of cases you can get a spatial resolution that goes down to the size of a molecule or an atom. And therefore, the question comes up, at least for a physicist, why should we care about a light microscope and its resolution
01:21
now that we have the electron microscope? The answer to that question is given in the next slide, where I counted a number of papers in this basic medicine journal that use an electron microscope and those that use the light microscope. And now you see which of the two won. And this is actually quite representative. So the answer is that the light microscope
01:43
still is the most popular microscopy modality in the life sciences, and that for two very strong reasons. Now the first reason is that it's the only way by which you can look into the interior of a living cell. It doesn't work with the electron microscope. Usually you have to dehydrate and electrons get stuck at the surface. But light can penetrate into the cell
02:01
and get information from its interior. But there's a second reason. We know there are different types of organelles already, but we would like to know what, say, specific proteins do. Say, a certain protein species. There are 10,000s of different molecules in the cell. And this can be done in light microscopy relatively easily by attaching a fluorescent molecule
02:23
to the protein of interest, for example. And then, of course, by attaching that molecule you can see that fluorescent molecule, then you know exactly where the protein of interest is located. So here, where, if you will, the chemistry comes in, you attach a fluorescent molecule to the protein of interest.
02:40
And because it's a fluorescent molecule, if you shine, for example, green light on it, it can absorb a green photon. Then it goes to a high, up-lying, electronically excited state. And then there's some vibrational decay. And after about a nanosecond, the molecule relaxes by emitting a fluorescent photon, which is likely red-shifted due to the vibrational energy loss.
03:01
Now this red-shift is very useful because it makes the method very sensitive. You can easily separate out the fluorescent light from the excitation light. That makes the method extremely sensitive. As a matter of fact, you can label a single molecule in a cell and still see that single, say, fluorescent molecule, hence that single protein, just by the fact that the method is so sensitive,
03:22
so background suppressing. However, it's very obvious. If there is a second molecule, a third, a fourth, a fifth, coming very close to the first one, closer than the 200 nanometers, you cannot tell them apart. So I'm saying this in order to remind you of the fact that resolution is about the ability to tell things apart. And therefore it's clear that in a light microscope
03:41
and of course also in a fluorescent microscope, despite its sensitivity, if something is more densely packed than 200 nanometers, you cannot tell it apart. And if you manage to overcome that barrier, you will do something good for the field and therefore also for the life sciences. Now I would like to dwell a bit on the question why there is this diffraction barrier.
04:01
I'm sure all of you know about it and you may have heard about Abbess theory, very elegant explanation of diffraction barrier. It's very elegant, but my view, it's very deceptive. Sometimes elegant theories can be very deceptive. And the perhaps best way of explaining it is the way you would explain it to a biologist. Now it's shown here. You can say that the most basic or important element
04:23
of a light microscope is the objective lens. And the role of the objective lens is nothing but to concentrate the light on a single point, to focus it on a point. But because light propagates as a wave, we will not get the light focused on a geometric focal point. We will get a blob of light that is smeared out here and that is at least about 200 nanometers
04:41
in the focal plane and about 500 nanometers on the optic axis at best. And this has, of course, major consequences. And the consequence that it has is that if you have several features in here, say molecules, they will be flooded more or less at the same time by that light. And hence, they will also scatter back light more or less at the same time.
05:01
And so if you place a detector here to detect the fluorescent light, the detector will not be able to tell the signal from the molecules apart. Now you will say, okay, I wouldn't place a detector here. It's silly. I'm using the lens to project the light back. But okay, you have the same problem here because each of the molecules here that fluorescence light, if you collect the light,
05:21
will also produce a blob of focused light in here that has a minimum spatial extent as a result of diffraction. And if you're having a second one, a third one, a fourth one, a fifth one, and so on, they will also produce blobs of light. And as a result of diffraction, well, you cannot tell them apart. So no matter what detector you placed in here, be it a photomultiplier DI
05:41
or even a pixelated detector such as a camera, you will not be able to tell the signal from these molecules apart. Now several people realize this problem at the end of the 19th century. Ernst Abbe was perhaps the one who had the most profound insight. And he coined this diffraction-bearing equation that is named after him. And you know you find this equation
06:01
basically in any textbook of physics, optics, but also in textbooks of cell and molecular biology because of the enormous relevance of the diffraction-bearing of light microscopy to this field. But you will also find it on a memorial that was erected in Ernst Abbe's honor in Jena, where he lived and worked.
06:20
And there you will find this equation written in stone. And this is what people believed throughout the 20th century. Not only believed it, it also was a fact. So when I was a student at the end of the 80s in Heidelberg and I did some light microscopy, confocal microscopy to pre-precise, this was the resolution that you could get at that time.
06:41
So this is a picture taken from a cytoskeleton, so from a part of a cell. And you see there is some blurred features in here. Now the development of STAT microscopy, about which I will talk in a minute, showed that there is physics, so to speak, in this world that allows you to overcome this diffraction barrier.
07:01
And if you play out that physics in a clever way, you can see features that are much finer and you see details that are beyond the diffraction barrier. Now of course, this discovery has earned me a share of the Nobel Prize. And personally I believe that very often,
07:21
and you will hear it at this meeting, discoveries of certain people are tightly connected with their personal history. And this is why I will dwell a bit on how I got actually the idea to work on this problem and to overcome the diffraction barrier. Now some of you will know it, some won't. I was born in the western part of Romania in 1962.
07:41
I belonged to a German ethnic minority at that time. And I realized at the age of 13 that it was communist Romania and being part of a minority, this is not the best way of spending your life, perhaps, in a communist country. So I convinced at the age of 13 my parents to get out of the country and to emigrate to West Germany,
08:01
taking advantage of our German ethnic background. And this is why I managed after some difficulties. Of course, in 1978, we settled here in Ludwigshafen. I like this place also because it's quite close to Heidelberg. And I wanted to study physics and I wanted to enroll myself to study physics in Heidelberg, which I did after getting my high school living certificate.
08:22
Now like most physics students, I think this still applies today. I was fascinated by the modern physics, of course, by quantum mechanics, relativity, wanted to do particle physics, wanted to become a theoretician, of course. But by the time I had to make a choice for opting for a certain topic for my PhD thesis, this is what I looked like
08:40
as a PhD student back then, believe it or not, I lost courage, I lost courage. And it's quite understandable because my parents still had to struggle to settle in the West. My parents got jobs, but the job of my father was threatened by unemployment. My mother was, at that time, diagnosed with cancer and finally died.
09:01
And so I felt, okay, I have to do something that supports my parents, not theoretical physics and so on. And the older semesters told me, oh, you shouldn't do theoretical physics. You may end up as a taxi driver. And so don't do that, don't do that. And so against my inclination, I must say, I signed up with a professor
09:23
who set up a startup company specializing in the development of light microscopy, of confocal microscopy to be precise for the inspection of computer chips and so on. And so I signed up with, my topic was to inspect computer chips
09:40
and align with the computer chips with light microscopy. I thought, okay, if you do this, you get a job with Siemens or IBM and so on. But, well, as you can imagine, if you have different inclinations, I worked for half a year, even a year, but I was bored. It was terrible. I hated to go to the lab. I was about to drop it. I secretly shopped around for other subjects.
10:02
And I had two options now, either to become miserable or to think about whether I still can do something interesting with light microscopy with this physics of the 19th century where there's nothing, just diffraction and polarization. Nothing you can do. I mean, really, honestly, there's aberrations perhaps, but that's boring. So I thought, okay, is there anything left?
10:22
And then I realized, well, breaking the diffraction barrier, that would be cool, because everyone believes that it is impossible. And so I got fascinated with this idea and realized at some point, yes, that should be possible. And my logic was very simple. I said to myself, now this diffraction barrier was coined in 1873. That was end of the 80s.
10:40
So much physics happened, actually, in the interim. There must be at least one phenomenon that gets me beyond the diffraction barrier for some modality. Could it be reflection, fluorescence, whatever. I don't care. There must be something that gets me beyond the diffraction barrier. At some point, I put this philosophy down in writing. I realized it won't work just by changing the way the light is focused,
11:00
because that is not changeable if you use a lens. But perhaps if you look at, in fluorescence microscopy, at the states of the fluorophore and the spectral properties, or maybe there's a quantum optical effect. You can play perhaps with the quantum nature of light, so to speak, and do it. And so this was the idea. Now I should say, at that time, there was already a concept in place.
11:20
This is why I wrote here far field, which is called a near field optical microscope. And that used a tiny tip. It's like a scanning tunneling microscope to confine the light specimen interaction here to small scales. But I felt, does this look like a light microscope? Of course not. This looks like a scanning force microscope. I wanted to break the diffraction barrier of a light microscope that looks like a microscope and operates like a light microscope.
11:41
So something that no one anticipates. And so I tried to do this in Heidelberg, but I couldn't find a place actually to do it. And I was lucky that I met a person who gave me the chance to work here in Turku in Finland. This was not where I wanted to go, actually, but I had no choice, and I ended up there.
12:02
And in 93, actually, in fall on a Saturday morning, of course I did what I always did. I screened textbooks in order to find a phenomenon that gets me beyond the barrier. I opened this textbook hoping to find something quantum optical, look at the squeezed light and tangled photon pairs and this type of stuff. And I opened this page and I saw stimulated emission.
12:23
And all of a sudden I got struck. And I said, oh wow, this could be a phenomenon gets me beyond the barrier, at least in fluorescence microscopy. And I'm telling you now why I was so fascinated. I went back to the lab to make some computations and now I'm explaining the idea. Now I said, there is no way of changing the fact that all the molecules are flooded with light.
12:42
And so normally they would also produce this blob of light here. But we can't change that, of course. We may change the fact that the molecules that are all flooded with light, with excitation light, are in the end capable of producing light at the detector. So in other words, if you manage to place these molecules here in a state in which they are not capable of producing light
13:02
at the detector, then I can separate those that can from those that can't. And so if I manage to keep those molecules in a silent state in which they are not capable of producing light at the detector, I can separate that molecule from the rest. So the idea is, as I said, not to play with the focusing, but to play with the states of the molecule.
13:21
So that was the decisive idea. Keep the molecules in dark state. And I will say, are there dark states in the fluorescent molecule? Now look at the basic energy diagram. We have the ground state and the excited state. Of course the excited state is a bright state. But the ground state is of course a dark state because if the molecule is in the ground state, it cannot emit light. It's very obvious. And now guess why I was so excited.
13:41
I knew that stimulated emission does just that. It produces molecules that are in the ground state. It sends molecules from the bright state back to the ground state. I'm not interested in the stimulated photon of course, but I'm interested in the fact that a phenomenon of stimulated emission produces dark state molecules. And now the idea is very obvious. You have a lens, you have a sample, you have a detector, you have light for turning molecules on so they excite them,
14:02
make them fluorescent, go to the bright state. And then we have a beam of course for turning molecules off. So send them from the bright state back down to the ground state to make them assume the ground state instantly at will, so to speak. And now the condition for making sure that this transition happens is of course that you have to make sure that there is a photon
14:22
at the molecule if the molecule gets excited. So in other words, you have to have a certain intensity in order to outperform the spontaneous decay, the fluorescence. So you must put a photon within this one nanosecond lifetime. That means you have to have a certain intensity that is shown in here, so this intensity of the stimulating beam. You have to have a certain intensity at the molecule
14:41
to make sure that this transition really happens. And so this is a probability of occupation of the excited state as a function of the brightness of the stimulating beam. And once you have reached this threshold, you can be sure that the molecule will not be able to emit light, the fact that it's exposed to green light.
15:01
Of course there is a stimulated photo going this way, but we don't care about it because the only thing we care about is that the molecule is not seen by the detector. And so we can turn molecules off just by the presence of the stimulating beam. And now you can imagine, we don't want to turn all the molecules off here in the diffraction zone because that would be silly, so what do we do?
15:21
We modify the beam such that it's not focused into an airy disk, into a spot, but into a ring. And this can be done very easily by placing a phase-modifying mask in here that produces a phase shift in the focal region, and so in the wave front. And then we get up a ring in here that would turn the molecules off
15:41
by the presence of the red intensity. But now, of course, let's assume we want to see just the molecules here at the center. So we have to turn off this one, and this one, what do we do? We can't make the ring smaller because it's also limited by diffraction. But we don't have to because we play with the state. That's the point, so we make the beam bright enough so that even in the region where the beam is weak,
16:01
the red beam is weak, in fact, here in the center it is zero, where the red beam is weak in absolute terms is beyond the threshold because I need only this threshold value in order to turn the molecule off. And so I increase the intensity to such an extent that only a small region is left where the intensity is below the threshold, and then I see only those molecules in the center.
16:20
And now it's obvious that I can separate features that are closer than the diffraction barrier because those that are adjacent to it within the diffraction barrier are transiently turned off. Now you can say, okay, now what I'm doing here is I'm using the nonlinearity, so to speak, of this transition, but that's the old-fashioned way of thinking it, the 20th century way of thinking it. And it's not a good picture
16:41
because nonlinearity is about number of photons, that many go in, that many come out, or fewer or less. Here it's about states. In order to make it clear to you, I'm removing the donut. So what we actually do is repair the molecules within this diffraction zone in two different states. You see? Here the molecules are forced to stay all the time in the ground state
17:01
because as soon as they get up, they're instantly pushed back down. So those are constantly in the ground state, whereas those in the center are allowed to assume the on state, excited electronic state. And because they're in two different states, we can separate them. This is the way we separate features here in this concept. So I've removed the donut just for clarity so that you can see that the essence
17:21
is the difference in states, but now I'm putting it back on so that you don't forget we have it. And now you may say, okay, now a clever idea. People were excited at the time I published it. This was not the case. I went to many meetings, partially paid by my own pocket because I was a poor guy, but nobody believed really in this lens-based
17:41
nanoscale resolution. There were too many false prophets, in fact. In the 20th century they promised that, but they didn't deliver. So they said, why should this guy from here be successful? So it was very difficult to survive. At some point I survived with money that I license out the old personal patents to a company and they gave me research money in return.
18:02
But one thing I would like to stress, I was much happier than during my PhD thesis because I did the thing that I wanted. I was fascinated by the problem, was passionate for what I was doing, and though it was difficult in, say, if you really look at the facts, I enjoyed it. And so I became successful.
18:21
In the end, somebody discovered me here at the Max Planck Institute in Gottingen and then we developed it. In the end it was also a team effort, of course, over time. Now, you've seen this slide. It has become classical already. Initially it looked like this and now it looks like this. It's a nuclear pore complexes and you see it quite nicely, this eight-foot symmetry.
18:42
And I still enjoy comparing it with previous resolution, I must admit. And it's very obvious that in the end the method worked. Not only it worked, of course, you can also apply it. And this is the first application to viruses. So if you have an HIV particle, you have here, for example, proteins that you are interested in.
19:01
How are they distributed here on the viral envelope? If you just take the fraction-limited microscopy, you can't see it. But if you take, say, stats or a high-resolution image, you see that they form patterns, all kinds of patterns. And then what has been found out is that those that are capable of infecting the next cell have their proteins or these proteins in a single place.
19:21
So you learn something. You can also see dynamical things at bad resolution if you don't have this super resolution, this, that. But now you see things moving and you can learn something about moving. So this is the strength of the light microscopes. Another strength of the light microscope is, of course, that you can focus into the cell, as indicated,
19:40
and take three-dimensional pictures. And here you see a three-dimensional edition of a stretch of a neuron, just to show you what you can do as a method. Of course, now, the method is being applied in hundreds of different labs, and I'm not going to bore you with the biology. I will come back to a more physics-type question
20:02
and talk about the resolution. You will ask, what is the limit of resolution that we have now? Now, first of all, the resolution that we get, it won't be obvious resolution, because obviously the resolution is determined by the region in which the molecule is allowed to assume the on state, the fluorescent state, and that becomes smaller. And that depends, of course, on the brightness of the beam, so the maximum intensity,
20:21
and on the threshold, which is, of course, a feature of the light molecule interaction. Of course, molecule-dependent. If you calculate it or measure it, you will find that these get inversely, actually, is the ratio of I over IS, so IS being the threshold and I being the intensity that we apply. And so you can tune the spatial resolution between high and low just by decreasing
20:41
or increasing that region. And this is very useful because it can adapt the spatial resolution to the problem that you look at. And that not only works in theory, it also works in practice. So here you have something, it becomes clearer and clearer, the higher you go up with intensity. And of course, you can tune it down, you can tune it up, or you can jump
21:00
over the diffraction barrier like this. And so this is an important feature. But in the end, of course, the equation says as D goes to zero, what does that mean? Now if you have two molecules and they are very close, you cannot tell them apart because they are allowed to emit at the same time as you can see here. So what do we have to do to separate them? We turn off this one so that only one fits in
21:21
and then we can separate them by seeing them sequentially in time. Because why? Because we turn off the rest and we see only one of them in this case. And so the limit of resolution in the end is the size of the molecule. Now with organic molecules, we are not there yet. But you can, with certain molecules, sort of molecules, like defects in crystals
21:41
where you can play on and off. And that's clearly a limitation in this concept. The number of times you can play on and off, you can demonstrate very, very small regions. So like in this case, 2.4 nanometers. So it's to be taken seriously that it's a small fraction of the wavelength and eventually it will go down to a nanometer. I'm quite confident about that.
22:02
So there's also other applications. For example, these are, as I recognize, charged nitrogen vacancies, diamonds. This has implication for monadic sensing, quantum information, and so on. So there is more than biology. Now to bring it to the point, the discovery was not that you should do stimulated emission. It was clear to me there's more than
22:21
stimulated emission. Stimulated emission is just the first example. The discovery was that you can play on and off with the molecules in order to separate them. So that was obvious to me. And although stimulated emission is, of course, very fundamental because it's the most fundamental off-transition that you can imagine in a fluorophore, you can do also play out other things
22:42
like putting the molecule in the long-lived dark state. There are more states than just, of course, the first electronically excited state. Or if you have some notion of chemistry, you know that you can play with relocation of atoms like doing a cis-trans isomerization like that. And if one of them, a cis is, for example,
23:00
fluorescent, you put light in it, fluorescence of the trans doesn't emit light if it's in trans mode. You can also play on and off with cis-trans states. My difficulty is publishing this. This is why it has 95 and 94 in here. And even that was rejected. So this is a review paper where I put this in. And the reason was the following.
23:21
People didn't understand the logic, although it was clearly explained in the papers. Logic is very simple. Now don't forget, we separate by having one molecule on and the other one off. And then, of course, it's highly favorable if the lifetime of the states involved is long or relatively long. Because if this is on and neighbors off, I don't have to hurry putting in the photons quickly.
23:40
I don't have to hurry taking out the photons quickly. So the intensity, of course, that I need to do, to apply in order to make this difference in states scales, of course, inversely with the lifetime of the states involved. The longer the lifetime, the more optically by stable something is, the less intensity I need. So the threshold intensity goes down, of course, with the lifetime of the states involved,
24:00
meaning that if this gets larger by six orders of magnitude, the intensity threshold goes down by six order of magnitudes. And this is the case here because the long lifetimes, we need less light in order to get the same resolution. And so it reduces the IS and hence also the I in the equation. And in particular, switching fluorescent protons
24:21
is very, very attractive simply because there are switchable fluorescent protons that do cis-trans isomerization and you can break the diffraction barrier at much lower light levels. And so we demonstrated this after some development. And it's indicated in here, if you need little light, then you can say, oh, why should I use just one ring? I can use many of these rings or holes or so in parallel.
24:43
Why not? And so here, we applied many in parallel to do a, say, off switching by a cis-trans isomerization. And of course, you can easily paralyze the whole thing by taking a camera as a detector and having many of them. If they are further away, then the diffraction barrier is no problem. We can always separate them. And so what we did here is we imaged a living cell
25:02
with more than 100,000 donuts, so to speak, in less than two seconds at low light levels. And the reason why I'm showing you this, it's not the lens and the detector that is decisive anymore, it's the state transition. Because the state transition, the lifetime of the state determines actually the imaging modality, the contrast, the resolution, everything.
25:22
You see that the molecular transitions in the molecule become the decisive thing in this modality. So with that, I'm coming sort of to the end. So what does it take to get the best resolution? Well, let's assume you would have asked this question, well, in the 20th century. Well, the answer would have been, well, it takes good lenses,
25:41
very obviously. Because you need to focus the light as sharply as possible. So you would have to go to Zeiss or to Leica and ask them to give you the best lenses. But of course, if you separate the features just by the focusing of light, as I explained, you're here focusing here or focusing here, you're obviously limited by diffraction. Because you can't focus the light better
26:00
than to a certain extent, as determined by diffraction. So what was the solution to the problem? The solution to the problem was not to separate just by the phenomena of focusing, but to separate by playing with the molecular states. So here we separate, of course, not just by making the focus as small as possible. We don't need that, in fact. Just by having one of the molecules,
26:22
a group of the molecules in a bright state, the other one in a dark state. Two distinguishable states. I gave you some examples of states, and there are many more that you can imagine in order to play out this difference in states in order to make the separation, and that is the point. Now, the method that I talked about, STAT and the derivative of STAT, they used a beam of light in order to determine
26:42
where the molecule is on. Here it's on in the center, and here it's off. And so it's tightly controlled, so to speak, where the molecule is in the on state, where the molecule is in the off state within the sample. So this is what is a hallmark of this method called STAT and its derivatives. Now, as you know, I shared the Nobel Prize
27:01
with two people, and Eric Betzig did a method that is fundamentally, at a fundamental level, related with the STAT method, because it also used an on and off transition in order to make things separable. Now you have one molecule here, and the rest is dark. But the major difference is that only one molecule is turned on, that's very critical. And if you do only one molecule in the on state,
27:22
you have the chance of finding out its position by using the photons, of course, that come out from the molecule, okay? So you put in a pixelated detector that gives you a coordinate, and if this molecule has a specific feature, and it needs to have that, that on state, that it produces many photons in a bunch, you can also use the photons that are emitted here
27:43
in order to find out the position. You locate it here, as Steve Chu showed yesterday, you can do that very, very precisely. It gives you the coordinate, of course, but say the difference in states is absolutely required in order to make the separation. So this gives it a coordinate. Here you need many photons here, here you need many photons here.
28:02
So you always need many photons in order to find out the coordinate, why? Because one photon would go somewhere within the diffraction zone. But if you have many photons, of course, you can take the average, of course, and then you can define a coordinate very precisely. But that's just the definition of the coordinate. The actual element that really allows you
28:20
to separate features is the fact that one of the molecules here is in the on state, the rest is in the off. They're all flooded with light, but only this one is allowed to emit. They're all flooded with light, but only that group is allowed to emit, the rest is dark. So if you ask me, and as a physicist, you always want to find out the single element that really puts the thing on the scene, it's the on-off transition.
28:41
Without that, if you took that out, none of the so-called super resolution concepts would work. And so that's the upshot. So why do we separate now features that couldn't be separated before? It's very simple. Because we induce a state transition that gives us a separation. And fluorescent and non-fluorescent is just one way of playing out that state transition.
29:01
It's the easiest because, well, because you can easily disturb, so to speak, a fluorescent molecule. If I managed to do it with something else, I would have done it, don't worry. It's not that I want to do it for the life sciences. But of course, this brings me actually to the point that in the 20th century, it was all about lenses
29:20
and about the focusing of light. Today, it's about molecules. And this is how you can justify and say this is a chemistry discovery, a chemistry prize, because the molecule comes now into the focus. It's the state of the molecule that matter. But this concept can be played out, of course. You can say two different states, A and B, it could also be something absorbing,
29:41
non-absorbing, scattering, spin up, spin down. As long as it's separated states, you are within, of course, with the framework of this idea. Now, of course, Abbe's diffraction barrier. Well, the equation cannot hold. We can easily fix that, no problem. We can put in the square root factor. And then we can go down to the molecular scale.
30:01
And it means, as an insight, the misconception was that people saw that microscopy resolution is just about waves. But it's not. The microscopy resolution is about waves and states. And if you see it through the eyes of the opportunities of the states, the light microscope becomes very, very powerful.
30:20
And this is not the end of the story. There's more to be discovered. So finally, I would like to highlight those people who have important contributions to this development. Some of them left and are professors elsewhere. Some of them are still staying in my lab and working with me or have set up companies. Now, one thing, then that's the last thing
30:41
that I will mention at the end. You should not forget that it wasn't this guy who had the idea. It was this guy. Thank you very much.