Archive for the ‘photons’ Category
advancing solar 2 – more on electrons, holes, dopants and electromagnetic fields
Jacinta: So in the last post we were joking about the horrors of physicists and engineers manipulating innocent electrons and forcing them to work for us, gratis. It comes to mind that there are people who are intelligently dubious about the manipulations of scientists – Bernard Beckett, in his 2007 book Falling for science, comes to mind, as does Yuval Noah Harari in Homo deus. ‘Scientism’ was used for a while as a pejorative, especially during the debates on the values of religion ‘versus’ science…
Canto: Yeah, but – I don’t want to dwell on this issue now, except to say that the critics of science are usually not very literate on the subject. So we were talking about dopants, which are impurities that can be added to the silicon crystal lattice to mess up its fine balance, so to speak. Boron is an example – it has three electrons ready for bonding, leaving a ‘hole’, a p-type space, and presumably a loose electron to carry the charge. And then there’s phosphorus, which has five such electrons – so one to spare after bonding, which they call an n-type situation. Positive charge carriers (p-type) and negative charge carriers (n-type) is how they describe it.
Jacinta: Right, so they layer these two types together: ‘The positive holes and negative electrons migrate towards each other’. The electrons will jump into the p-type and the holes jump into the n-type [they don’t explain how holes can jump]. This causes an imbalance of charge, because now the p-type side has more negative charges, and the n-type side has more positive charges’. This apparently creates an ‘electromagnetic valve’, which allows, or perhaps forces, electrons to pass through in one direction only.
Canto: This isn’t very clear to me, but let’s continue. Maybe you have to do it, and so see it working, to get a full grasp. So, a sufficiently energetic photon enters the p-type side (the boron-doped side) of the solar cell, knocking an electron loose to float within the material. It will either recombine with a hole, and fail to create a current, or it can enter the electromagnetic field – that valve thing between the p-types and n-types, also called a depletion layer for some reason. The effect, apparently, is that it accelerates the electron into the n-type side, which of course tends to lack p-type ‘holes’, but the electromagnetic field most cruelly prevents the electron from passing back to the p-type side.
Jacinta: Yes, it’s still a bit fuzzy, but on the n-type side some ‘holes’ are somehow transported across this electromagnetic field junction, where they recombine with electrons. so one side of this junction or valve becomes negatively charged, the other positive. This creates a ‘potential difference’, aka a voltage!
Canto: Explained neatly for us as ‘The difference in electric potential between two points, which is defined as the work needed per unit of charge to move a test charge between the two points’. Just saying.
Jacinta: So, as our video-maker tells us, we can then add ‘some mental contacts and an external load circuit’ and we have created a current, presumably, as the electrons will ‘pass along the circuit to recombine with the holes on the other side’. And that’s your solar cell, apparently. But I barely understand a word.
Canto: Well, doing and seeing, as I’ve said. But there’s problem with adding this metal to the upper surface as it blocks some of the light needed for the cell to function effectively. So, problems with solutions that create problems. So engineers keep working on new shapes and materials for optimisation. They’re trying to minimise the metal coverage and electron resistance in getting into the circuit. Topology optimisation is one subject of research, using computerised algorithms.
Jacinta: And it’s fascinating but hardly surprising that this sort of research is producing shapes for solar cells that resemble leaves – which after all are like little solar cells resulting from millions of years of evolution.
Canto: Hmmm, not like ours, plants don’t use the sun to make electricity. But this quote from the video is thought-provoking:
Vascular tissue on a leaf does not perform photosynthesis. It instead brings the water that is essential for photosynthesis to the leaf and extracts the useful products, serving a similar purpose as our electric contacts – so of course plants have developed the perfect shape to optimise the energy they can absorb from the sun… However, most solar cells use a simple grid shape, as it is cheap to manufacture.
Inevitably this means an efficiency loss, measured at around 8%. So, in conclusion, a current silicon solar cell has an efficiency, under lab testing, of around 20%. The drop to 18% shortly after operating has resulted in hundreds of scientific papers, and it seems to have to do with the use of boron, as the drop didn’t occur when boron was replaced with gallium. Something to do with a ‘boron oxygen defect’, so there’s been a lot of work done on trying to reduce the ‘concentration of oxygen impurities in the silicon wafers’, caused by the Czochralski process, the standard process for silicon wafer manufacturing. Almost all silicon solar cells are made this way. Recent research using a special imaging technique found that boron oxygen molecules converted to ‘shallow acceptors’ when exposed to light:
In essence they observed the defects transforming into little electron traps that acted as recombination sites, and thus reduced the time and probability of electrons entering the circuit to do work.
It’s something I can almost grasp. And with this knowledge, engineers, whose grasp is way firmer than mine, can find some kind of fix for the problem and get that efficiency up well beyond the 20% mark.
Jacinta: Well, this has indeed been a knowledge-expanding journey. Pour qu’une chose soit interessante, il suffit de la regarder longtemps. You mentioned the depletion layer, which caught my attention. It’s a central feature of semiconductor physics, also called depletion zone, depletion region, junction region and more. The depletion zone is so called because of the depletion of carriers in the region. Charge carriers presumably. Any rate, this region, and understanding it, is key to understanding the physics of semiconductors. The Wikipedia article on what they call the depletion region is a useful supplementary to our discussion. We might explore all this further, or not, depending on our own depletion levels…
References
The mystery flaw in solar panels (video)
https://en.wikipedia.org/wiki/Depletion_region
advancing solar, the photovoltaic effect, p-type semiconductors and the fiendishness of human manipulation
how to enslave electrons – human, all too human – stolen from E4U
Canto: Back to practical stuff for now (not that integral calculus isn’t practical), and the efficiencies in solar panels among other green technologies. Listening to podcasts such as those from SGU and New Scientist while walking the dog isn’t the best idea, what with doggy distractions and noise pollution from ICEs, so we’re going to take some of the following from another blog, Neurologica, which was also summarised on a recent SGU podcast.
Jacinta: Yes it’s all about improvements in solar panels, and the materials used in them, over the past couple of decades. We’re talking about improvements in lifespan and overall efficiency, not to mention cost to the consumer. Your standard silicon solar panels have improved efficiency since the mid 2000s from around 11% to around 28% – something like a 180% improvement. Is that good maths? Anyway, it’s the cheapest form of new energy and will become cheaper. And there’s also perovskite for different solar applications, and the possibility of quantum hi-tech approaches, using advanced AI technology to sort out the most promising. So the future is virtually impossible for we mere humans to predict.
Canto: Steven Novella, high priest of the SGU and author of the Neurologica post, suggests that with all the technological focus in this field today, who knows what may turn up – ‘researchers are doing amazing things with metamaterials’. He takes a close look at organic solar cells in particular, but these could possibly be combined with silicon and perovskite in the future. Organic solar cells are made from carbon-based polymers, essentially forms of plastic, which can be printed on various substrates. They’re potentially very cheap, though their life-span is not up to the silicon crystal level. However, their flexibility will suit applications other than rooftop solar – car roofs for example. They’re also more recyclable than silicon, which kind of solves the life-span problem. Their efficiency isn’t at the silicon level either, but that of course may change with further research. Scaling up production of these flexible organic solar materials has already begun.
Jacinta: So, I’ve mentioned perovskite, and I barely know what I’m talking about. So… some basic research tells me it’s a calcium titanium oxide mineral composed of calcium titanate (chemical formula CaTiO3), though any material with the ‘perovskite structure’ can be so called. It’s found in the earth’s mantle, in some chondritic meteorites, ejected limestone deposits and in various isolated locations such as the Urals, the Kola Peninsula in Russia, and such other far-flung places as Sweden and Arkansas. But I think the key is in the crystalline structure, which can be found in a variety of compounds.
Canto: Yes, worth watching perovskite developments in the future. I’m currently watching a video from Real Engineering called ‘the mystery flaw of solar panels’, which argues that this flaw has been analysed and solutions are being found. So, it starts with describing the problem – light-induced degradation, and explaining the photovoltaic effect:
The photovoltaic effect is the generation of voltage and electric current in a material upon exposure to light. It is a physical and chemical phenomenon.
Jacinta: Okay can we get clear again about the difference between voltage and current? I know that one is measured in volts and the other in amps but that explains nout.
Canto: Well, here’s one explanation – voltage, or emf, is the difference in electric potential between one point and another. Current is the rate of flow of an electric charge at any particular point. Check the references for more detail on that. Anyway we really are in the middle of a solar revolution, but the flaw in current solar panels is that newly manufactured solar cells are being tested at a little over 20% efficiency, that’s to say, 20% of the energy input from the sun is being converted into electric current. But within hours of operation the efficiency drops to 18% or so. That’s a 10% drop in generation, which becomes quite substantial on a large scale, with solar farms and such. So this is the problem of light-induced degradation, as mentioned. So, to quote the engineering video, ‘[the photovoltaic effect] is where photons of a particular threshold frequency, striking a material, can cause electrons to gain enough energy to free them from their atomic orbits and move freely in the material’. Semiconductors, which are sort of halfway between conductors and insulators, are the best materials for making this happen.
Jacinta: That’s strange, or counter-intuitive. Wouldn’t conductors be the best for getting electrons moving? Isn’t that why we use copper in electric wiring?
Canto: That’s a good question, which we might come back to. The first semiconducting material used, back in the 1880s, was (very expensive) selenium, which managed to create a continuous current with up to 1% efficiency. And so, silicon.
Jacinta: Which is essentially what we use, in inedible chip form, in all our electronic devices. Pretty versatile stuff. Will we always have enough of it?
Canto: Later. So when light hits this silicon crystal material, it can either be reflected, absorbed or neither – it may pass through without effect. Only absorption creates the photovoltaic effect. So, to improve efficiency we need to enhance absorption. Currently 30% of light is reflected from untreated silicon panels. If this wasn’t improved, maximum efficiency could only reach 70%. So we treat the panels with a layer of silicon monoxide reducing reflection to 10%. Add to that a layer of titanium dioxide, taking reflection to as low as 3%. A textured surface further enhances light absorption – for example light might be reflected sideways and hit another bump, where it’s absorbed. Very clever. But even absorbed light only has the potential to bring about the photovoltaic effect.
Jacinta: Yes, in order to create the effect, that is, to get electrons shifted, the photon has to be above a certain energy level, which is interesting, as photons aren’t considered to have mass, at least not when they’re at rest, but I’m not sure if photons ever rest… As the video says, ‘a photon’s energy is defined by multiplying Planck’s constant by its frequency’. That’s E = h.f, where h is Planck’s constant, which has been worked out by illustrious predecessors as 6.62607015 × 10−34 joule-seconds, according to the International System of Units (SI). And with silicon, the photons need an electromotive force of 1.1 electron volts to produce the photovoltaic effect, which can be converted, apparently, to a wavelength of 1,110 nanometres. That’s in the infrared, on the electromagnetic spectrum, near visible light. Any lower, in terms of energy (the lower the energy, the lower the frequency, the longer the wavelength, I believe), will just create heat and little light, a bit like my brain.
Canto: I couldn’t possibly comment on that, but the video goes on to explain that the solar energy we get from the sun, shown on a graph, is partially absorbed by our atmosphere before it reaches our panels. About 4% of the energy reaching us is in the ultraviolet, 44% is in the visible spectrum and 52% is in the infrared, surprisingly enough. Infrared red light has lower energy than visible light but it has a wider spectrum so the total energy emitted is greater. Now, silicon cannot use light above 1,110 nms in wavelength, meaning that some 19% of the sun’s energy can’t be used by our panels.
Jacinta: Yes, and another thing we’re supposed to note is that higher energy light doesn’t release more electrons, just higher energy electrons…
Canto: And presumably they’re talking about the electrons in the silicon structure?
Jacinta: Uhh, must be? So blue light – that’s at the short-wavelength end of the visible spectrum – blue light has about twice the energy of red light, ‘but the electrons that blue light releases simply lose their extra energy in the form of heat, producing no extra electricity. This energy loss results in about 33% of sunlight’s energy being lost.’ So add that 33% to the 19% lost at the long-wavelength end, that’s 52% of potential energy being lost. These are described as ‘spectrum losses’.
Canto: Which all sounds bad, but silicon, or its reaction with photons, has a threshold frequency that ‘balances these two frequency losses’. So, it captures enough of the low-energy wavelengths (the long wavelengths beyond the infra-red), while not losing too much efficiency due to heat. The heat problem can be serious, though, requiring active cooling in some climates, thus reducing efficiency in a vicious circle of sorts. Still, silicon is the best of threshold materials we have, presumably.
Jacinta: So, onto the next piece of physics, which is that there’s more to creating an electric current than knocking an electron free from its place in ye olde lattice, or whatever. For starters, ye olde electron just floats about like a lost lamb.
Canto: No use to anyone.
Jacinta: Yeah, it needs to be forced into doing work for us.
Canto: Because humans are arseholes who make slaves of everything that moves. Free the electrons!
Jacinta: You’ve got it. They need to be forced to work an electric circuit. And interestingly, the hole left when we’ve knocked an electron out of its happy home, that hole is also let loose to roam about like a lost thing. Free electrons, free holes, when they meet, they’re happy but the circuit is dead before it starts.
Canto: This sounds like a tragicomedy.
Jacinta: So we have to reduce the opportunities for electrons and holes to meet. Such is the cruelty of progress. For of course, we must needs use force, taking advantage of silicon’s unique properties. The most excellent crystal structure of the element is due to its having 4 electrons in its outer shell. So it bonds covalently with 4 other silicon atoms. And each of those bonds with 3 others and so on. A very stable balance. So the trick that we manipulative humans use to mess up this divine balance is to introduce impurities called dopants into the mix. If we add boron, which has 3 outer electrons, into the crystal lattice, this creates 3 covalent bonds with silicon, leaving – a hole!
Canto: How fiendishly clever!
Jacinta: It’s called a p-type trick, as it has this ‘positive’ hole just waiting for an electron to fill it. Sounds kind of sexy really.
Canto: Manipulation can be sexy in a perverse way. Stockholm syndrome for electrons?
Jacinta: Okay, there’s a lot more to this, but we’ve gone on long enough. I’ve had complaints that our blog posts are too long. Well, one complaint, because only one or two people read our stuff…
Canto: No matter – at least we’ve learned something. Let’s continue to rise above ourselves and grasp the world!
Jacinta: Okay, to be continued….
References
https://www.theskepticsguide.org/podcasts
https://news.mit.edu/2022/perovskites-solar-cells-explained-0715
The Mystery Flaw of Solar Panels (Real Engineering video)
https://byjus.com/physics/difference-between-voltage-and-current/