 We're going to talk about how to produce hydrogen by using this fusion reactor that is always hovering above us, and how that works, Storml will show us. Please applaud him. Yeah, good morning. Yeah, good morning. All right. Adi, welcome. Thanks. Good that you're all awake. I'm a chemist. I have worked with organic chemistry, with biochemistry, and lately with physics. Physical and electrochemistry. I wrote a master's thesis at the Hamilton Center in Berlin, Wannsee. And there I dealt with this subject, how to find materials to produce, to solve the energy problems of the future. So, short agenda, how many agenda do we need per year? I'm going to go over some basics. I'm not going to dive deep. Some chemistry and physics from school will be useful. Some of you may remember the basics. Then we'll look at the well-known silicon photovoltaic cells that are spread over this camp. Then you need to store the energy somehow. I'm going to talk about that shortly. And then how to couple this way with a new class of materials. These are two diagrams from Akatec. It is a council group for academia and technology, Akatec. What is this pie chart on the left? We have only 2% solar energy, some part of renewable energy, but most of it is still from fossil fuels. And on the right side we see a diagram from the BP annual reports. BP, you know them, the refueling station, the oil producer. They write a report on energy consumption every year. And you see this spike downwards pretty well, 2008, 2009, when the world financial crisis was and the change went up back again. And now we have an increase in energy consumption of 3% per year, which is a three-year high. And on the right we see that the carbon dioxide emissions have also increased. So this climate change thing is everywhere. Everybody's talking about it, and it's time that even more people talk about it. Here's the change over the years, consumption of primary energy. The composition is on the top coal, then in orange, the renewable energies. Then we have hydroelectricity, the bright orange is nuclear energy, then gas and in green oil. The calculation is in megatons oil equivalent, MTOE, that is the usual unit. And these are 166 petawatt hours to, as a recall, peta is the one unit after terra, so 1,000 terawatt hours or 19 terawatt years per year. So you could cross that out. The earth has a rated power of 19 terawatt. Thankfully, a lot more comes from the energy from the sun, 179 petawatt, so more or less a factor 10,000 more. That's a graphic from NASA originally, the incoming radiation in yellow, and the different ways how this radiation is reflected going back to the universe or staying on the ground. 89 petawatt arrive at the ground, that's more than enough to cover all the demand. Here's a repetition. The more orange you see on the map, the more intense the sun radiation is. That is an average over the year. And the black dots are spaces with high radiation. For example, North America, the Great Basin, Nevada desert, or South America, Atacama desert, Sahara desert. And these black dots alone to cover those with solar panels would cover the whole energy demand of the world. What is actually reaching us? That's the spectral intensity of sunlight. You remember from school this theory with the black body radiation, that is the yellow curve and the orange curve is the actually measured radiation outside of the atmosphere. Inside the atmosphere it's a little bit different, that is the lower curve, you see the rainbow, it's the visible spectrum. And these notches that you can see are actually filtered by water gas in the atmosphere. Water has several absorption bands on the right side of the graphic. They absorb radiation that has already been considered in the diagram we just saw before. Which also means that the more arid your region is, the less clouds you have, the hotter it will be in the oceans on the land and this whole story will run even faster. Standardize all of this. The atmospheric mass model has been developed, we have different details, different level of details. So the first approximation is the earth. You put a plate on the earth, you see where does it go back in the atmosphere in a right angle and the angle between the sun and this reflection is the zenith angle. And then the atmosphere will be passed only once. If you go a bit more to the north or to the south, then even at 12 o'clock at Salma Salstice the sun will enter in an angle. That is the AM 1.5, that's the usual radiation force in Europe and North America. And because that's the industry, industrial countries that has been used as a standard for measurements. A bit more to the north, that would be 60 degrees, Scandinavia for example, that's AM 2, that means that the equivalent of two atmosphere thicknesses are being traversed by the radiation. And if you see the sun just above the horizon at 90 degrees in the more detailed models that is in 38-fold atmosphere penetration and of course in space, that is AM 0 because there's no atmosphere at all. So you cannot wait until Salma Salstice to make your measurements when there's absolutely no clouds. In the lab you mean you need a sun simulator. That is the device I made my measurements with. You see on the left side the control PC, that's where all the electronics is put in. In the middle is the measurement cell. And on the right, this black L is the light source. Here, let's see it in detail. The white thing is the electrochemical cell. The wires go to the measurement electronics. And on the right we have an aperture in order to go from dark to clear quickly. And on the right we have a xenon light that you also have in cars. These super clear bright bluish lights are xenon lights. Why do we take them? Well, we need to model the sun spectrum pretty efficiently. If you take incandescent lamps, then you reach 3000, 3500 Kelvin. You know the light temperature, that is exactly that data. And that orange thing is the sun spectrum. And there you see that the blue parts are just not well represented. And therefore you need that bluish light. And it is modulated with filters in order to reach this equivalent of the intensity in the cell. About solar cells, we've all seen them before. You start off with sand, arrive at silicon, and this is split into very flat wafers. And you reach these glittering polycrystal and solar cells. It's a monocrystal like in the chip industry. But in solar cells, you luckily don't need this high purity like you need on chips. In industry, you usually need about 99.999 degrees of purity. In solar industry, 99.5 are enough, so you can work with very simple processes. So how does the current come out of this? On the left-hand side, you see a model, a band model of a silicon semiconductor in the solar cell. The bluish part is the one irradiated by sun. And the red one is the actual material, which you don't see. And in between the electrical current runs. In silicon solar cells, this is 0.6 to 0.7 volts. And the irradiation intensity determines how much current you get out of the solar cell. So if there's a cloud above, it gets darker. And of course, eventually the current and voltage drop down. On the right-hand side, we see what this cell actually absorbs. The dark one is the sun spectrum that comes into the cell. The yellowish curve is the actually used part of the spectrum. In between, you see the band gap at a temperature of 300 Kelvin, which corresponds to the base state of the valence band going up to the state of the conducting band. And the gap in between is the actual band gap, which you can deduct as a voltage. So the energy is calculated by H times nu, which is the Planck-Kornem times the irradiation frequency, which is also corresponding to the color of the incoming light. So here, the orange region is the visible light up till infrared spectrum. And then there's a sharp decrease, which is due to the energy of the incoming photons not being sufficient anymore to actually bridge the gap into the conduction band. So this light simply isn't absorbed. It's going to be interesting later in the talk, because this is actually the upper limit regarding efficiency that we can get out of our solar cells, because otherwise the material simply can't catch the photons anymore. So here we have the Bohr atom model, which explains the atom as divided into the core and the shell. So here's an excitation from a second to the third shell, sending out a quantum of light. And this is exactly what we want regarding solar cells. So there's a photon coming in, and there's an energy excitation sending out, generating a voltage difference, which is what we can use in our solar cells. It's in a simple band model, in this case, Magnesia. You see that all atoms in one compound share the electrons, the so-called covalent binding, which forms the valence band. And the conduction band is only slightly above this valence band. So with only a little push in energy to the electrons, which they can already assume at room temperature, the electrons can freely move, and electronic conductivity is established. However, we choose silicon, which is a semiconductor. In a metal again, the valence and conduction bands are very close to each other, and the band gap is sufficient. However, in a nonconductor, the conduction band is too far away, so no light and no electrical energy is sufficient to get the electrons into the conduction band. They're stuck with their atom cores, and the material is nonconducting. And the average of these two is the semiconductor, where the band gap is in the region below approximately four electron volts, which roughly corresponds to ultraviolet light, and a bit lower corresponding to a yellowish shade. So silicon has a band gap of about 1.2 electron volts, which corresponds to the infrared spectrum of light. This energy is usually expressed in electron volts, which is the elementary charge times the voltage that was obtained by this charge. Up to now, everybody should have heard about this concept. Photovoltaics is known. However, what happens if we get semiconductor into direct contact with water? Photoelectrochemical cells can actually conduct electrons directly into water, so not via an outer circuit, but rather directly into water, splitting the water into hydrogen, oxygen, by that omitting an outer circuit, which would, of course, cause losses in efficiency. So the 1.23 electron volts we see is the redox potential of water, which is a minimum potential that has to be surmounted in order to start this process. Those red areas are the so-called overpotentials. So the band gap we ideally need to split water into hydrogen and oxygen would be approximately higher than 1.8 electron volts, considering the overpotentials as well. We saw so far that we obtained no material that was actually able to perform this right away, so we had to introduce some tricks. This is a schematic of a photo-electrochemical cell comprising two half-cells, so the light enters the blue layer on the right-hand side. These are very thin layers in the region of 10 to 20 nanometers, in the minimal case, or up to 100 nanometers in a thicker case, but then we have stability issues. And the rays transcend a membrane, which in this case is water-based, and arrive on the other side on the red photo-cathode. And with the light that still reaches the other red side, can start the process of splitting the water. So the electrodes kind of share the work, and only the one electrode has to surmount the work to provide the electrons for these processes. And the electrons are conducted via an outer conductor, coming back to the photo-anode. I'm sorry, they come back to the photo-cathode via this outer circuit. So, but then we also said we need higher energies than 1.2 electron volt. If we have higher energies and a larger band gap in the material, then only higher energy radiation is absorbed. Everything below it gets lost. So it's only, for example, ultraviolet or green light. So what do we do with the remaining radiation? We just send it into the next cell. So from the left-hand side, the full-spectrum light would reach the yellowish-red light. And the red light is absorbed. The yellowish-photo-catalytic material then performs the first part of the reaction, and the remaining irradiation reaches the material behind the yellowish part, which is maybe then the reddish-yellow-spectrum. And on the back side, it just reaches a normal photovoltaic cell. On the right-hand side, the image is mirrored, but in general it shows the same processes. So the full-spectrum light excites as many electrons as it can. Everything else that doesn't manage to do that shines through to an end-of-silicon material, reaches the silicon solar cell, and does its normal work. So it's a kind of tandem solar cell working together as a compound. So we're actually parting this orange region into two parts. The higher-energy part on the left-hand side is absorbed by photochemical material, and the rest still remains for the silicon solar cell. Yeah, this material on the right-hand side is marked by F203, which is simply rust. It's just corroded iron. We usually use metal oxides, because they're very practical, they're stable, they're widely available. But the problem is, of course in the periodic table of elements, there's like 50, 60 metals, and then we're running out of possibilities for materials. So what are the materials we use? Below the yellowish part, this one is marked as bismuth, vanadium, and oxygen. And if we calculate how many actual combinations of metal oxides we could have, we end up in the four-digit region. So we have a lot of experimental grounds on which to try, which materials and which combination of elements could be suitable for these processes. Because of course they all have different band gaps, they all absorb different energies, and could be more or less suitable for the aim of splitting water. Going back to energy storage, this is an excerpt from Agora Energiewende or energy transition report, marking the German-wide energy consumption over one and a half months. So the purple curve in the middle is the most interesting one. Now it's actually the pink one, which is the demand for energy. And on the 1st of May we actually entered the region that regenerative energies would have been able to actually cover the whole demand. That can be put down to different factors. So of course we had a very high sun irradiation, it was a public holiday in Germany, so with a quite lower energy demand. And for the first time we found this inversion that could have actually just been able to shut down conventional power plants and use the renewables instead. One of the applications would be simple electrolysis. Everyone might probably have seen this in their chemical essence. So we put in electricity in the water and split it into its components, hydrogen and oxygen, which works quite well in principle. Unfortunately the most effective allergic codes in this context are platinum electrodes. So platinum is the material that very much is most efficient in splitting water performing electrolysis. It's also not as rare as metals like iridium, for instance. Which for instance are also components of our modern electronic devices. These are an order of magnitude rarer. However, if we add all three concepts that I mentioned within one photoactive cell, which means generating a stack of three different semiconductor materials which absorb different colors marked by blue, green and red here in this case, which are then conducting in series or setting up a series and immersed in water, then we actually have a cell which can split water into hydrogen and oxygen by only being irradiated by light. So on my screen the video doesn't work. There's bubbles. This is actually what I want to show in this video. So the circle on the left-hand side is where hydrogen is developing. We know that's stoichiometrically. The bubbles have to be larger. So this is covered with platinum and on the other side on the right-hand side we have ruthenium oxide on the electrode which is also quite rare metal and one of the only oxides that are actually able to catalyze the oxygen reduction reaction or oxidation reaction in this case in a very good way. Oxygen needs a kind of conviction to participate in this reaction which of course we want to reduce because the overpotential times applied current is an electrical power output which we lose in this case in the reaction. I guess this everybody has seen in school as well. That's a kind of biological cell. The chlorophyll in the nature. The same principle. The first chlorophyll absorbs light of a first wavelength. It's excited. Then over this step that you see in the next it deactivates a little bit. Then on the second photosystem it is really excited and gets a harder push and the energy of the second photosystem is strong enough to push the following reaction and carry out the photosynthesis in the plant. So that would be the big picture. I hope that you have more questions than before. And I would like to answer some of those. You know the drill. There's two microphone angels on the left one on the right. Please go see them if you have questions. If you're in the stream you have lost because there's no signal angel. I'm very sorry. Hi! How is it with efficiencies and wear and tear of the materials that are being used? That's a good point. Silicon solar cells well the theoretical limit is 32-33% from thermodynamics in practice the best cells have a bit more than 20% the super expensive for military and space applications have 25% the electrolysis behind depending on how good and expensive the catalysts are 70, 80 maybe 90% and that means in the sum we have 15-20% in this stack cell where we saw the video the efficiency written next to it 9.5% that is pretty good but I mean these materials deliver efficiencies we're talking about the solar to hydrogen efficiency STH that is usually below 1% sometimes maybe 1 points or 2 points something percent there's a lot to do and the stability generally in the process the materials being corroded in general we bring on passivation layers to enhance the lifetime so in the lab we reach lifetimes of a week or something like that and usually that is already good but sometimes within minutes they get decomposed so we need to look at these thousands of donations of which material works has good efficiency and isn't decomposed within minutes another question from that side Hi I saw in the media that apparently a company has developed a prism to use the to shift the unusable part of the light spectrum into the usable part well for me that sounds like frequency doubling there are non-linear optics who of you has already seen a green laser pointer more or less everybody that is actually an infrared laser that is being sent through this non-linear optics which sends out everything that is green but it only works very selectively for a narrow spectrum I don't know if that is actually what you mean but otherwise the convert the color I don't know a process that can do that and I don't know how efficient that would be and these green laser pointers have a terrible efficiency from electrical power to light okay let's move on a small question and a big question small question is how necessarily how clean does the water need to be I guess that if you take just tap water in large quantities that after some time the catalysts will be dirty or something like this you generally don't want to clean the water with a lot of energy I guess and the other question can we imagine molecules that are specially engineered to absorb light and split water in hydrogen and oxygen I mean if we had that molecule we could maybe create a DNA strand that we can put into bacteria and then engineer bacteria in order to carry out that electrolysis or that photosynthesis that splits water I don't know if we are so far in the technology first question regarding water purity well I've only dealt with lab setups so we have ultra pure water which is called millipore has super low conductivity almost none impurities inside because the problem is we are talking about redox chemistry so if other redox active materials are implied salt, metals lead, copper they are also being reduced and oxidized electrochemically so in general you will be dropping copper or lead you can poison your material with that and in doubt you drop your efficiency talking about filters and particles you don't want them so you need to filter your stuff I can't tell you more detail about that unfortunately regarding the second question yeah there is such things that's called Gretzel cell somebody worked on it and received the Nobel Prize there was organic tolerance on a conductive support which was then excited and the electrons were lead off it there was a big hit for some time I don't know how that has developed for today it's a very special you can do that at home to make that work it's pretty simple I mean organic tolerance, something like you can do that with fruit tea actually doesn't live for a long time also here this the incoming light excites the molecules but they wear out and don't work after some time so I don't know how the practical applications look like okay 10 minutes left who wants to ask a question from the left mic there's somebody standing up this combination of solar cells and fuel cells seems only to be working in the lab there's a lot of talk about building solar cells and then a fuel cell behind it to produce just a break do you mean fuel cell or electrolysis I mean electrolysis, fuel cell comes afterwards to use the hydrogen yeah exactly that's a whole other chapter but yeah my question is that is state of the art isn't it much more inefficient or why do you want to go to this direct coupling and how far are we still in the future that's a very good question probably a couple years I had solar power or renewable power to electrolysis is being done today problem is on the one hand I put up the slide sand to silicon to solar cell it seems easy but needs a lot of energy it's a very well researched process mostly because of the chip industry that had big interest into that but the production from solar cells needs a lot of energy and the amortization times are somewhere the main of 2, 3, 4, 5 years until the solar cell has compensated, harvested the amount of energy it has cost to being produced the lifetimes are in the domain of decades 20, 25, 30 years easily and that means that after lifetime they still perform 80, 85% of their energy so one argument would be you can save energy to produce silicon from silicon dioxide the metal oxides are already energy low materials you know that iron is rusting because rost is at a lower energy level than metallic iron so the idea would be you get the raw materials more easily well silicon is one of the most frequent materials we have in most abundant materials we have in the crust of the earth but the idea is we want to produce all of that with less energy the next point is that electrolysis is only working efficiently and well if you have these expensive valuable material catalyst, noble material catalyst so you can lower the system costs by using a solar farm and then an electrolyzer and you have to maintain them that would be a possibility to save money but how well that will be in the end we will see does anyone else want to ask a question maybe to bridge some time I have a question does anyone do active research in these materials so is there any working groups that maybe simulate or just try these kinds of materials out yes, for instance my working group so indeed this is a very contemporary topic and it's one that has been done by very excellent research groups there is also the variant of quantum chemically simulating these materials on very big cluster PC to calculate the bank gaps of these materials and determine whether they actually exhibit the properties that we are looking for but actually I don't know what the state of the art in this respect is but in order to do this precisely we have to invest a lot of calculation power and energy and it could still be that theory doesn't really match reality in this respect or that the material works and has the property that we desire but then degrades and doesn't exhibit the lifetime that we wish for so it's really the literal search for the needle in the haystack we talked about directly directly electrolysis and changing the electricity into water electrolysis so I could talk about this for the whole camp but are there any alternative to actually perform this yes for instance bioenergy so for instance for instance corn crops and changing them into bi-ethanol and bi-reactors which is also the refueling station discussion this is actually being done but I don't have the total efficiency in mind however that it's not very big just a few percent and also of course we then compete against our own nutrition generation I omitted this topic on purpose for instance for instance hydroelectric storages this I omitted on purpose because it's another very big topic but for instance also then is to directly translate solar energy under water just to evaporate water and then drive turbines which is for instance then in or in the Sahara desert but otherwise I don't have any spontaneous idea about other couple processes so we have room for a couple more questions so everyone is heated up and let's thank our speaker again you have listened to the talks