[00:00:05]
>> So it's a pleasure to invite Today's to introduce today's speaker Professor Jiah Khan thought Ravichandran. He got his bachelor's and master's degrees in metallurgical and materials engineering at Indian Institute of Technology before getting a Ph D. in applied science and technology at U.C. Berkeley and then doing post doctoral work at Columbia and Harvard before starting in 2015 as an assistant professor in the Department of Chemical engineering and materials science as well as Electrical and Computer Engineering at the University of Southern California.

[00:00:44]
He's won a number of awards most recently being named an early career scholar by the Journal of materials research and I'm going to turn the microphone over to him. Thank you for the nice introduction and it's a pleasure to be here thanks a lot for the invitation and.

[00:01:13]
To present some of the world that my group has been working on for the past 4 or 5 years of. That I'd love the talks are just. This stock is going to focus on a particular class of materials college basketball cars and AIDS and specifically the theme of the dark will be on dimensional control of light matter interaction in these materials so before we go into the details of these materials and how they behave I want to give a very broad perspective of why we care about these materials and what is the background for this so the starting point for these systems is the idea of we're living in the semiconductor age semiconductors of powered literally many of these high performance devices that we hold in our hands every day from your phones to laptops and radios lighting devices.

[00:02:13]
And the T.V. season. That you see so I have given a few different. Case cases here so I should be transistors which are part of all the computers and also the smart phones that you see solar cells that are increasingly powering our homes by converting sun's light into electrical energy radius times of lasers that are used in close including the laser pointer that you see some of these pointers have semiconductors in there and their lives Firstly the blue lady which is a big breakthrough that won the Nobel Prize in 2014 so there was such an semiconductors is has always remained interesting enough to throw up some new.

[00:03:01]
Surprises and that has given rise to novel prices vary from. 156 men shockingly bad men and retain won the Noble Prize for a time sisters and then the latest one as you mentioned was. A Faqir monarch a motor for their inventions on gallium nitride for lighting up occasions so in between these 2 noble prices which were primarily given for a lot more applied areas that we see every day there were quite a few Nobel prizes given for a quantum hard fact and fraction of one so these are interesting physical phenomena that seem to happen in very high quality semiconductor electron gases so all in all there is a lot of scope for doing technology based impact and also basic science work in semiconductors so that seems to keep our in our research even today so what is the current status now and so many of the people in electrical engineering here and everywhere have been forecasting and also we are the end of the moves law right now so most laws something that is.

[00:04:10]
Predicted the number of transistors that we have in our computer chips has continuously increased to twice as much as a number every 18 months or so and this scaling has been very successful despite the impending end of the scaling that has been predicted for years and we are at a point where we have really reached that end and the scaling has slowed down and even the semiconductor reports roadmap studies that used to be predicting the next stage has stopped now so with the impending end of most law how can we continue to improve the computational systems that we that we use every day which are primarily powered by semiconductors so that's one part of the question.

[00:04:59]
Just to just to make. Just to make. A small joke about the situation in the 1990 S. we had a really bad loans computer of a scale transistors and that was not fast enough to run our American games or so on and so forth but in 2015 or so.

[00:05:20]
Computers became much more powerful but still is not fast enough for what we do but in any 35 if we are really optimistic about this by the most laws then I'm pretty sure we'll figure out a way to get very high performance computers and still complain that it is too slow or maybe it is several orders of magnitude faster than this and there are some clues as to what sort of computing paradigms that could get us there perhaps so I'm going to make some 80 years where continuing to work on new are improving existing semiconductors can still make an impact taking that computing as a starting point and there's an interest in developing beyond bullion logical one human computing architecture that is there in all of computers going into your market computing to analyze things that are not very easily done by conventional bullion logic and also quantum computing as a means to do high performance computing in a very efficient fast manner so these are areas where developing new materials maybe even semiconductors could play an important role of energy is another area where semiconductors continue to play an important role although many of the solid state energy conversion systems still require better and more powerful and high performance semiconductors and last but not the least lot of our communication and sense sensing up occasionally also require better semiconductors one of the important issues that are increasingly is becoming relevant for things like autonomy or scars for example is the ability to see through what is beyond the human perception which is in the.

[00:07:02]
Longer wavelengths regime of infrared and there are many areas where there is atmospheric transparency windows in the infrared but we don't have the right sort of devices and technologies or. Materials to work through that so for there are lots of applications that are emerging so in all these cases it is extremely important for us to continue to invent new materials new semiconductors that can fit in all these functionalities So with that.

[00:07:35]
With that. As a backdrop So let's think about let's try to put all the media's semiconductors that we have worked with are known for years in perspective to see what we can learn about so a couple of quantities that define semiconductors one is the carrier mobility and the the mobility is a measure of how well your carrier is transporting materials and the other aspect is carrier density is that how many carriers that you can pack in a given value material so these 2 quantities tend to go on opposing directions so they're countered contradictory to each other so as you can see some of the high mobility materials like graphene or C 5 materials tend to have extremely low current city because the ban structure itself is built such a way that they are very highly dispersive and they can only hold a small amount of charge but they can transport extremely efficiently but as you move down to the periodic move down to other materials what you notice is that the carrier density can gradually increase but that comes at an expense of mobility so you always have this issue of local area mobility high care density are high area mobility and located and city so how can you get around this one will getting us somewhere here open up new opportunities and that's the big overall question that we want to ask so to give a context to this if you look at.

[00:09:05]
If you look at the these materials there are certain things that will jump out right away many of the materials that are shown within the high carrier mobility regime tend to have connection and violence violence that have S. and P. are broadly catacomb and having this are brutal nature it's the is the basic essence that decides why I didn't see some of the mobility so high because these bans are very very dispersed over nature but as you move down as you get to transition metal based systems you move with the band conduction the kind of density goes up but at the expense of the carrier mobility but most of the divan systems that people have studied in the past has been complex oxides But these and me there's more interest in translation method I don't know it's due to the interest in 2 dimensional radios.

[00:09:57]
These radios the question is can we take these systems retain this carrier density and somehow try to push the carrier mobility so that meet some understanding of what all the different factors that affect carrier mobility one is of course the or the large this year's changed inherently limits the mobility the other aspect that also limits the mobility is the detail scattering processes that happen in your system can you control those and that will be the key answer to this puzzle so one way to think about this is so let's take this case of strong impact aid which is a very nice semiconductor material is given connection and if you take this is I'm somehow maintain the framework that this person here look at how the interaction between the.

[00:10:49]
Connection that ISIS from the B. side element and the anons is that these 2 elements tend to have very large electronegativity difference and hence the interaction between these give rise to strong scattering of electrons. In your system so now if you can somehow treating the structure and refuse this electronegativity difference between these 2 elements you have a mechanic some towards increasing the mobility of your system so one way to do this is by taking the system looking at Oxygen has very high electronegativity and moving down the periodic table you go to sulfur and selenium the local negativity drops so now this is a mechanism by which you can go for a material that has sulfur and selenium instead of oxygen in this framework where you can have much lower electronegativity difference and much better carrier mobility in your system so that's that's our general prediction and it also holds true because transition metal die child carbonates do have much higher mobility than ourselves so the question was can we find some material good slightly better carrier mobility with high care density and we think transition metal prostate cancer generates is a candidate system OK so now let's before I go into talking about these meters 1st thing is I need to define what is a pair of skates and also give some context on why the Arab material system so the 1st step is to define what is a pair of skates across cards a very nice.

[00:12:21]
Turn early material system which has a a B. X. 3 formula that's shown here where the B. element typically is something that sits on the center appearances and your element sits on the corner of the units and your X. is something that sits on the faces. Typically the band structure of this system if we were to talk about a scenario where you're taking this early transition matter elements that are shown here then your conduction band is made of the average nature of the B. side element and your relative mind is made of the R. Britton nature of the excitement now one of the interesting things that will jump out if you look at this periodic table is that the.

[00:13:04]
Candidates for this ab Excellent are all over the periodic table so essentially you can pick most of the elements from the periodic table and you can build a prettier that has the same chemical formula maybe extremely and either fall directly as a brass grade or something that is related so that's what makes prostate materials an interesting candidate to study because you have so much structural and physical chemical verity that you can build a whole slew of compounds to study and see if they can be applied for various applications so now just to give a context I'm going to start off with the well known but ask oxides.

[00:13:42]
Give you some examples to motivate by but ask how carbonates could be of interest so as I mentioned that the bank and action you slice a large denser States one way to manifest large interstates is by comparing. Complex oxide lanthanum Cobalt oxide against constant silicon silicon is the S.P. wanted material that has led to really low density of states and one way to compare density states between materials is to just measure absorption of these materials as you can see your question silicon has a very Lovecraftian confusion because it's an indirect Maggette material but even amounts of silicon turns have very low absorption coefficient but if you go to the lantern uncovered oxide which is the divine connection that doesn't go fission There's a factor of 2 or even most half an hour of magnitude higher that tells you that the given connection can give rise to a large absorption confusion so this is interesting for solar cells for example if you can make a demand system and still have some reasonable mobility then you can think of ultra thin absorbers and they can also give rise to good carrier.

[00:14:49]
Carrier separation mechanics seems. Now let's talk about few more cases this is a specially interesting if you are interested in the angular momentum of the carriers that are present in your system so one of the interesting things about the van systems is that it has radius different our battles that have very different angle a moment I'm OK So that means if you can build materials that has that has that can put carriers in radius different orbitals you can get a sign various different angle of momentum to B.'s electrons so you can take advantage of the angle momentum differences that are present for carriers in various articles.

[00:15:31]
And recently people have shown that in the case of transmission metal I didn't like I was true you can do valid Tronics in this specific case it does not do that the. Our brittle nature but it is primarily due to the. Differences in the 2 sides equal insights that are present there which gives rise to very different capons and hence you get our brittle language momentum selective execution of characters so this is especially of interest for people who are interested in quantum applications broadly defined OK so last but not the least.

[00:16:08]
I'm pretty sure people would have heard past Gates in a completely different context which is a large Helots of really taken the solar cell community by storm they came in most are 10101015 years ago and by within a short span of a decade or so they have quickly rising up on our achieve 20 to 23 percent power conversion efficiency which is even better than some of the long standing candidates such as Cap 6 and so on and so on and so forth so in a broader things basket can provide you a much larger structural and chemical to an ability which is not necessarily present and there is a transition metal compounds and also you can get this chance to achieve high mobility and I catered and see which is what we are trying to explode in the stock.

[00:17:01]
This past week dark even though we won't get to that time mobility Highgate or density part but we'll talk about the general features of these materials 1st because there's very little known about these particles so I'm going to start off with. The. Same basket structure and try to focus on a narrow set of compounds but the specific doc so the ones that we're interested in are early transition metal based systems which means you're beside has titanium the county I'm unhappy I'm and you're a site has kept him barium and strontium and you're excited for and filling in so.

[00:17:42]
Turns out these materials here and I can assume one of the few different types of crystal structures that are shown here that allows you to have the octahedron connectivity show a very specific dimensionality and that's what we're going to leverage in this perspective can show that light can interact with the differently when you're octahedron connectivity is very very different and that gives rise to readers different interesting applications so the trooper ask is defined by Connor sharing of these octahedron forming a 3 dimensional network but once you have.

[00:18:22]
Ionic sizes that do not let your system crystallize in this 3 D. network you start getting this edge sharing network which is a course a. One B. type system which has a lot more network connectivity than a true one the connectivity then you also have this hag's agonal variance where you have facial hair connectivity between the octahedron that gives rise to one of the chain like behavior so you can go from 3 dimensional all the way to one dimensional by just changing different elements that are present in your a be extreme last but not the least there are also natural simpler the structures which form 2 dimensional.

[00:19:02]
Variants of this 3 dimensional compound by just inserting an extra explainer in between this a be extremely blocks that are present so you can basically take a system make it go to dimension and also one dimension and then all these cases you should be able to see completely interesting different physical phenomenon that's what we're going to go one by one in the Star.

[00:19:26]
OK so. So let's look at this particular graph that I've shown so this is a prediction of the bang gap for these various compounds as uming rises and one of these 3 phases in principle only one of them is going to be that some random increase the ground state of the system and it's possible that you may have another phase stabilizing close to room temperature but this is a good starting point.

[00:19:55]
By Design Group at our peer predictor for us to start taking some of these compositions and try to look at it so the 1st thing that we started looking at is this one in the consult for 3 and radium the crimson for 3 because they can see that the bangle seem to match nicely with the smaller optimal value which means there is a chance of looking at single junction solar cells based on these materials so when we started working on these systems.

[00:20:22]
We knew he or she can see this made sense to look at but we don't know whether anyone has ever made it or not but turns out people have been working on this there are chemists who who synthesize the only thing that you can ever think of and they would have just struck Chile kind of president Clinton that's been.

[00:20:38]
In the $1960.00 S. there were people in Russia and also those some world done in Germany in the fifty's and more recently in the eighty's there were also people looking at these materials in India so there is this a global level but very sparse interest in trying to look at these materials so we knew we could make these materials but the issue was how quickly can you make it and in what quality.

[00:21:03]
So while we were working on these there were a couple of other reports that came out where people took the equal and oxide materials and tried to use extremely aggressive sulfur station and it's amusing serious to to get powders of these materials as you can see these are fairly successful but it's very defective material and I can see the appeal of some of these materials does not look as impressive as a typical 35 efforts look like so so we decided to go about doing this and we knew that just taking element sulfides and powders and mixing them and cooking them for a long time was not going to solve the problem so we went about taking these photos we also added a little bit of halogen So the insight was to take halogen and make it real and let it react with high highly refractory high melting point materials like the cañon and that allows us to lower the melting point of the system and the activities.

[00:22:05]
So luckily when we started doing this isn't we found that within a day of day or 2 of cooking our media started to crystallize and they looked really homogeneous visibly and we started categorizing them using the fraction. Foundered we were able to stabilize the barium recordings for green be.

[00:22:25]
The started 3 D. network connected on a shared. Face We also were able to stabilize the strength in the current cell for 3 in the needle like Chris that I showed earlier and also by heating up this material to high temperature and quenching we were able to merit stabilize the 3 dimensional CORNISH a prospective So in some sense we could even achieve 2 different phases of the same composition by just working with the way we are processing these materials so 1st generation materials are all part of this but nevertheless structurally they are looking really good.

[00:23:03]
We also did some chemical characterization of these materials using Indias and X. ray fluorescence or these and we found that consistently in a semi quantity demanded a day they had the psychometric A B X 3 conversation even globally and also locally so the most important thing is to look at optical properties of these materials so this is a busy slides I'm going to start with the right side.

[00:23:31]
So these are theoretical predictions of the bang up values for the density of change for the transit going from fired out of phase sun in the consult by B. to face which is the corner shared prosperity and barium to consult side using the density of state values that pain the early to predict what are the what is the absorption confusion and hence the Banga values for each of these 3 phases and if you look at the diffuse reflect inspector for the powders that we do these values match closely with this there is an acid because typical D.F.T. calculations always underestimate bandgap But overall the trend was very similar to what we would expect not only that when we did for the luminescent measurements we were able to identify macro particles in the samples that gave intense luminescence allowed us to confirm that the diffuse reflectance study showed the right Bangas So so the most surprising thing for us was that we made powder samples and we were ever to get very intense fear which is actually really surprising if you have been working on the upper conics of 35 if you take a P. forever and crush it down into people you see nothing because the surface examination will kill you basically So in this case is this material somehow or can survive that really been so not only that we took these systems and we tried to do quantitative luminousness to get a sense of how well they do in terms of B. external quantum efficiency and what we found was that if we use the same radiation conditions in your fast casual cell night single crystal wafers show appeal cause that can be put on.