Good morning everybody and like to thank the organizers of that series for giving me the opportunity to share and weave you a few thoughts and. So results in emerging field of organic semiconductors for optoelectronics Now when I first received the invitation and I saw this was seeming our series on advanced publication I thought well I'm probably the least qualified person to speak to students or people who are daily going into the lab or in the clean room and. Doing research and dealing with very complex fabrication I haven't done any fabrication in the past twenty years my students do. So I thought well what is it that I could share we view that would still be very useful and thin things I really care about which I think. Are important and that you should consider when you when you step in in the lab OK And so what I like to do is first share we view a few thoughts about innovation and why innovation is. Has always been so important but I think becomes even more important as as as we move or we move forward and the you know innovation the Zs we needed because it's a way of sustaining our economies to generate value not just economic value but social value some wealth and maintain our the standard of living that we enjoy right now and if you look back at innovation you know sometimes it comes that you really fast and I like that picture because it kind of illustrates you know how our innovation can really almost retire. Some legacy technology and then open up a computer completely new era with new challenges and new opportunities so innovation. Is here and. You can classify innovation in two major types and here I'm just kind of repeating all. Translating some of the thoughts that Clay Christiansen. Described in his book The Innovator's Dilemma where he kind of studied the evolution of storage devices and storage markets and you know he found that there are really two main classes of innovations there is so-called. Sustaining innovation where you make some small but important incremental improvements over existing products. We experience that currently by you know we we keep our cell phones. During a period that gets shorter and shorter because it has more gadgets and more speed and more capabilities and so on and so that's the sustaining innovation but as it is traded by the previous picture. More importantly they're also disruptive innovations the ones that you don't really see coming at the beginning because. Initially they. They cover very small markets and so the main players in any market or technology fields don't really pay much attention to these really disruptive technologies and you shouldn't because if you're a billion dollar company and you want to increase your sales. Your return by twenty percent every year you need to add two hundred million dollars of new sales. Every year and so you can do this if you start to look at this. Struct of our innovations that still correspond to two very small markets OK However as Clay Christensen described very very well in his book and as shown in the storage market an area is that these disruptive technologies they can grow very fast they grow much faster than the sustaining innovations and then at some point they can take over and the legacy technology can just disappear because of these are disruptive technologies. Now why is innovation also important and why I think as we move forward they will become even more important because if you look at some of the global trends right now. Which are driven in part We have a significant increase in population you know we are now days over seven billion people on the planet when I grew up at school I learned we were four billion. Seven from four within you know just a few decades is really impressive and if you look at the projections this will continue to grow to really impressive numbers and with that population growth there are a lot of new challenges energy energy demands and so you know a lot of other important aspects we have to find solutions to that growing energy demand while you're also making sure that we don't deplete some of the other key resources such as food and water for instance or to make sure that we develop some technology so that health care and can can remain or become more affordable and so on and of course make sure that all of this. Is done in a way where we don't really impact too much or the environment and so you know when you look at that broader context and you look at these trends these evolutionary trend. Yes you can ask yourself if for instance the force that are ongoing in generating renewable energy sources if sustaining innovations are actually enough or if you really need to focus on disruptive innovations OK So I think there's a really need for disruptive innovations. And so you know ration means many things means different things to different people that many different but in first approximation you could say OK it's a process by which you transform inventions and ideas into into products and services that have economic as well as social value economic is just one part it's over often over on for size but the social value of innovations are also very important and so these discoveries play a central role so it's really important not to be sidetracked and through remember that we need we have to make new discoveries and when you actually look at how these discoveries are made they often made by by serendipity and what is certain D.P.T. where it's a mindset it's a mindset where if you step in the lab you should be prepared. To to expect the unexpected OK or as. You put it very elegantly is that chance favors the prepared minds OK so you need no edge of course but knowledge is is not enough a discovery is really a mindset where you you have to be prepared for the unexpected not because you find the unexpected when you get into your clean room and you try to apply your process and things don't work out the way you expect they they would OK there's a lot of frustration there's a lot of. That unexpected component. To the complexity but that's where you also need to have an innovator mindset and as Churchill is famous for all of course but this is one that I think reflects quite well what an innovator mindset is is that you know this image sees the difficulty in every opportunity an optimist sees an opportunity in every difficulty so. If your process doesn't really work out the way you expected it's not necessarily a failure it could be an opportunity and it can be the beginning of a new discovery. Through timidly I think innovation. Is often OK because if you apply a way defined process and if the person who wrote the recipe did a really good job and you have ply every single step of that process and you get what you want. That's useful but it's not really innovation it's not innovation because machines and Roberts very good at doing this and they will do that much better than even the brightest students we have here at Georgia Tech and so just going in the lab and the repeating what other people have done is not necessarily what what I think as scientists and engineers. We should be doing and so what makes that discovery and the innovation. Is that emotional a relationship that we develop with what we do with our research. The samples we work with it because that artistic component is based on. Humans creative skills and imagination and without those you can't discover new things and. You can't really you know wait so I just wanted to make some of these. Share some of these thoughts with you and then that slide actually now takes me into the core of what I want to share next which has to do weave the title and the topic of my talk namely interfaces OK And you can see when you look at a box like this there are two aspects there is what's inside the box so if you're developing new materials and you processing these materials into new films and you want to optimize the physical properties such as transporting inorganic seven conductor you worry a lot and you should about what's in that film What's the most of that film and what's in there so the inside is very important but the outside is also very important on the outside other interfaces that this material will form with I.J. sent layers with electrodes with the package with a barrier coating and all of these things will timidly as we would see would have a very important impact on the performance of your device and a lot of the devices in the field of organic electronics actually relied you of course on having great materials with useful properties but at the end if you want to make something that has high performance it's at the device level and at the device level you need to worry about interference. OK And so that brings me to organic summary conductors and so I think this is an example of disruptive technology even though organic semiconductors have been around for a long time and they've been studied but now we see this is now you know prime time of deployment of some of these materials in real world applications the most mature of them is. In displays most of you probably own a cell phone that has an Active Matrix organic light emitting diode display and these are organic semiconductors that emit light and so the building blocks here are molecules OK And with these molecules you can form these solids either through solution processing techniques or by vacuum the position and again the vacuum deposition techniques are not necessarily implemented that vacuum levels that are ten a minus seven or ten A minus ten some of these vacuum techniques can be very low vacuum saw Mean or ten to minus two and so on so very very practical. And what's surprising is that you end up with something which is pretty messy. Compared to conventional inorganic semiconductors because these materials contain a lot of impurities they contain a lot of defects compared to if you take a silicon in got and you look at the level of pure read purity and Crystal unity. It's a different world OK However as you will see surprisingly these films are pretty insensitive to some of these defects because even at the microscopic level you find that molecular property has a strong component and then of course some physical properties that we measure on the microscopic scale we depend on the morphology of these films but still we can get charged transport and mobility values that are on the earth for a lot of these applications. And so the vision where you know this could be really a disruptive technology is that since the processing is done at room temperature because it's very defect tolerant you can imagine coding this on a very large scale on values kind of substrates. Plastic metal sheet and so on potentially low cost but the cost here is important in terms of Cap Ex and the overall capital investment that it would take to could these materials on the large scale remind you you know right now microprocessors are have reached a level of performance which is just amazing scaling and you know Moore's Law for the last thirty years but what's less known is what's often referred to as more second law which is that the cost of these fabs are is growing exponentially you would so on now you know is in the in the billions of dollars where here the hope and the vision is that you could do things with fairly less capital investment. So next I'd like to share examples of work that we've done and also again highlight the importance of interfaces and hopefully convince you that some of these building blocks here chose three main category of devices because they all have different function the first one the function main function these is light emissions or four for displays or solid state lighting which I think is an important part because a lot of our electricity is used in lighting and so if we can develop more efficient light sources we can make tremendous energy savings and then transistors because from Sisters are the core building blocks you can make circuits and build a lot of functionality in these devices and then the third one is on the organic food overtakes which is for power generation so let's start with all it's as I mentioned earlier of you know the this is an area that people have studied for many years but in the past few years things have really started to evolve very fast and massive. Industrial activity now and the market for. For that technology particular for small. Mobile displays and T.V.'s and the cost of these T.V.'s last year you could buy a prototype in Asia of an all it T.V. The price tag was ten thousand dollars OK. A year later it has come down now to about three thousand dollars and continues to drop so it's there and this of course our large area devices and there will be more and more displays I mean. Most probably all it displays we become the technology of choice for these plays especially that it also has many advantages in terms of power savings or source or so OK so the seminal work here goes back to the media eighty's and paper by Ching Tang and Stevens like. Eighty seven and some of the patents were filed a little bit earlier and you know at the beginning when you started talking about this technology for these plays there were of course a lot of skeptics but not here we are we have now these devices and so what's attractive about the technology is that if you look at the cross-section of a device it's fairly simple so typically you have a transparent electrode you deposit a few conducting layers one usually has a function these two to be a whole transport layer on the other side you inject electrons and then these holes and electrons meet in that layer here they generate excited states on the right communication of these Excited States leads to light emission and so you can use that for this place and so it's a good idea and so for a long time. A lot of the devices were built on that. I know that the bottom because it was glass and so it's transparent and the light can can go through but the problem with that geometry is that well first of all most of the displays that we use also require sophisticated backplane where you have all the pixels that I mean all the safety it's that control the pixels and so what you'd like to do to to optimize the you know the pixel aperture ratio is to deposit the light emitting material on top of these circuits and to have a top emitting structure rather than a bottom meaning structure where some of the circuits can get in the way and then the other advantage is that if you don't have that glass substrate anymore and the light is coming from the top you are void of some losses due to light trapping in that glass substrates through total internal reflection that usually limits the amount of light that. Goes through to the user now when you look at the current technology that are used to drive these L.E.D.S. also they are most are based on and most saw and channel type of transistors and so if you want to power the idea here and you want to control the current that goes through the food the control of the gate source voltage here you'd rather put the diode on the side between on the drain side rather than the source side and that requires also having a device where the diode the inverter or flipped where the got to these is actually at the at the bottom. So a long time. We tried values materials values architectures and we felt we just couldn't get inverted diodes or diodes to work as well as some of the more conventional ones where you finished with. That electron injecting contact and the reason for this is one widely used electron inject injection contact is based on the combination of fluoride with aluminum and so traditionally this is at the top and so the sequence for the position is that you first deposit fluoride and then you deposit aluminum on top and this was tested so when you inverse the sequence where you first deposit aluminum and then leach on Freud and then we have some of these early electron transport materials that were used by for that and it didn't work and so about ten fifteen years people thought well this just doesn't work. The other way around but again the students gave it a try and what they found is that if you use some other electron transport materials it works actually very well and so you can make a very efficient top in meeting but I've got two of the structures and you can see the performance here of these devices. Or you have to do is apply if a few bolts and on the left here you see the current versus voltage so typical diode a little bit of leakage here and if you look at the correspondence luminance as a function of voltage you can see that a voltage of about six volts you can get nearly ten thousand can they last for a square meter OK Now just to calibrate you if you're not familiar with luminance units the typical luminance of your laptop or T.V.'s is about five hundred to seven hundred they last for square meter so you know ten thousand on that I suppose for me is the is a very bright schools and the extern contemn efficiency can be as high as thirty percent in these structures which is as good or in some cases even higher than what you can achieve in devices with a conventional geometry. So. That's one of the highest efficiency you can achieve in structures like this with a medium in a meeting layer that it's fairly eyes of tropic and saw the next thing was how can we further improve this and the answer is you know you build structures where you have moved to pull hitter or drunk such that the current doesn't just floor through one diode but it flows through two diodes which means that at the same current you can get twice as much light of course the trade off is that your pride voltage has to be doubled or larger because you have a thicker structure and it makes the structure a little bit more complicated but with the current organic muti source the position systems that we use this is this is not too difficult and so here are the performance of these are muti junction devices so again the key here was in being able to get. Into facial materials that you're in for a ride to be efficient electron injecting electrodes and likewise here that compound here actually in communication with the P.C. makes a very good I know that and here you can see that the black of our the luminance and the external contemn if you can see for a single junction device if you stack and you make two you can double and now you can get even higher efficiency and here is a plot of the current efficacy much of the income that asked for AMP this is the relevant metric for display and again you double when you have to junctions and if your demise the device and you add thin layer here to improve the amount of light that's extracted or minimizing some of the cavity effects here in that structure you can build structures. Devices that have a current if you can see of two hundred on the last for OK And at at the luminance of hundred thousand come that are supposed to meter you still keep hundred condé Nast for them so that means in the future displays and future light sources can be made very efficient with these values materials and these values architectures. The second example. Offer devices I'd like to describe has to do with organic field effect transistors. Or the transistor I don't have to elaborate most of you know it's one of the core building blocks of more than the microelectronics and so here is an example of a transistor architecture that my posed of Doc young. Developed A few years ago and this is again kind of an example where Cindy P.D. had an important role and so the structure is shown here on the left there's a substrate we use glass we can use plastic they's a little passive ation layer here just to improve our decision on the glass is not very important then we first deposit some metal electrodes the So these are the source and drain electrodes then we spin code a semiconductor which is a mixture of that molecule in the Matrix so this is processed from solution and then the important part here which is a little bit unusual is the gate dielectric which is comprised of that by layer first appear for we needed a portier and then hike a dielectric layer that is processed by atomically of the position OK Initially the motivation was to use site up because site up is a fluorinated material and so it's sort of NS that. To the sort of NS that are being used to process the organics from the conductor. So you can spin coded on top without dissolving the summary conductor underneath and more importantly you get a very clean very sharp interface between these two materials that are very different and that's important in the operation of a transistor because I remind you the way to transistor operators here is that you have a conduction channel here at the semiconductor dielectric interface that is moved you lated by the value of the gate voltage so the more voltage you apply to the gate the more carriers you create in the channel and then current starts to floor here in the channel between the source and drain electrons. And so side up is great but the problem is that these materials have a low dielectric constant and so when you operate your transistor the operating voltage depends on the capacitance density of your gate dielectric so you can either increase the dielectric constant of your gate dielectric or make it very thin OK the problem when you make it very thin is that you generally have leakage and so you don't get a very good transistor and so in a way of trying to remediate to the problem with it well what if we instead of having just a single sided top layer we actually deposit the high pay dielectric on top and defect free hike a dielectric and this is what I took the position gets you get you very conformal pinhole three. High quality films and so even with fifty nine meters now you can get very good. Very low leakage in these transistors and so on the right hand side you see the the transfer characteristic of that transistor is not the greatest shape here but the BT values typically of the order of one and you see operating voltages that are typically below ten volts OK so we're pretty happy and then the question of course. It always comes up is how stable are these materials so this young went in the lab and he started testing these transistors. Day after day week after week and the To his surprise he always got exactly the same electrical characteristics these samples were not degrading it was very very strange so he started to do more detailed the stress or lifetime tests and so he first started to cycle these transistors OK so you can see here transfer characteristics over twenty thousand cycles these are samples that are you know kept in ambient conditions there is no shift in the threshold voltage which for at the time for Gannett transistors was was really unprecedented but then he kind of patients and he said Well. I don't know if the resupplying. Here. I don't have the source but anyway I can describe to you what the movie. Is showing he he took the samples and he put them in a class much Amber OK Plus much amber no protection five minutes and he took the samples out he put them back in the semiconductor power meter our eyes are and he could still see a response and then he decided to take the samples and to. Put them in immerse them in the water and the transistor was still working and so here are the detail of the results so this is the transfer characteristic of a sample when it's freshly fabricated and it's kept in an enough darkness fear it's never been exposed to the outside environment if you expose it to air after a few days you have a small change in the characteristics but. It's such rates and the it's tabulators But here is the interesting part so this. Is the transfer characteristic motored before and after. Plasma treatment five minutes in a plasma chambre and you see there is very little change and of course if you take reference samples where you have only the aluminum oxide or if you have just the side top the samples are just fried they don't work anymore and this is before and after. Immersion in water for for an hour OK So that shows you that you know organic conductors can be made very stable it kind of the pans on the structure and how you you fabricate and design that structure and going back actually to to the overall geometry of the transistor what happens is that that combination of that site top layer in addition to the aluminum oxide layer deposited by L D actually has a very good barrier property so not only does it provide a nice gate dielectric but it also protects or it voids that you have permeation of oxygen or moisture in to the film and so it doesn't degrade and that's why these devices have such a long lifetimes Now of course when you make transistors by spin coding this is great it's nice it's a proof of principle but if you want to get industries interest you also have to show that it can be manufactured using some processes that are scalable and so one of them is printing you can make you know very large area devices by injured printing so we took some of these from Sisters same geometry slightly different angle inks this is a commercial ink that you can buy from Iraq this is a small molecule that can be processed from solution that was made by our collaborators in South moderate group and you can see here the corresponding transfer characteristics of some of these transistors that have been printed on the on the plastic substrate with printed the silver. Electrode printed semiconductor and then the gates dielectric on top and printed the top gate this one I like very much because you can see this is a transistor that has nearly zero threshold voltage and now the transfer characteristic is very clean has very little contact resistance and so you know it shows that these devices not only stable in they don't just have a performance level that is comparable to a more physique on but they can be also they can be processed using. Techniques that are scalable and used in industry OK and that of course because we noticed that the sensors are now so robust in. Environment you can immerse them in the water so we develop sensors with these materials and I don't have time to go into the details but you can use them also for PH measurements so. The last part of my talk I'd like to share with you some advances that we've made in the field of organic forward takes. And so here again. You know we've been offices on interfaces and how adventures in interfaces can actually lead to better devices with higher performance but also. Highest ability and so if you look at the barriers for commissaries ation and some of the challenges of all these techniques is that they usually require high work function and lower function electrodes OK And in an L D It's relevant because you want to efficiently inject holes in it efficiently inject electrons so that you can form these Excited States and then have regulated recrimination and light emission in photovoltaic cells you want a large difference in the work function between the two electrodes because that detail minds your You're building potential and he's a. Upper limit to the open circuit voltage all the photo voltage that you can get in some of these materials and so if you take the Table of Elements and you look at the different metals there are materials out there that have a very low work function such as more magnesium but because of their work function lower function they're also highly reactive OK And so making a high efficiency device we've got some electrode these nice if you want to saw new records in efficiency but you know deploying cells that contain your business not necessarily the best approach and so there was a need to try to develop new materials or new techniques to avoid the use of highly reactive electrodes such as culture. And so let me share the next few minutes the process that you know Discovery and the serendipity. And how we came to the development of some electrodes that I believe work very well and so this all started through the work of one of my graduate students. A shim who's now graduated is a job that some saw and he was using T.R.U. to nano particles dispersed in water and to is a well known electrode and. He started using that podium or P.V.P. because it's known to be a dispersant. And so it would improve the quality of these to you two films and as he shared the early results of the solar cells he had fabricated with these two U. two. Layers. We I asked him if he had kind of optimized the concentration of that fact and that dispersant P.V.P. and he said no but I'm going to do this and so he came back and we looked at the data and he said Well it's strange because the more so fact and I pour in there the. Better the saw ourselves so it's kind of strange doesn't really make sense because the material is it's an insulator right so I say Well have you tried just the fact itself and he said why would I do this this is an insulator is nothing is going to happen and said Well that whole how do we understand the properties of the cell so very simple experiment it took a deal you would solution of P.V.P. He dipped it in to the solution and we have a Kelvin probe set up in the lab and he put the right you modified with that P.V.P. under the Kelvin probe and the Workforce and dropped from four point six to four point two We. Bustling we didn't really understand what was going on but we trust the measurement the work function really decreased so this is where. Another important part is that you can't do everything on your own you need to collaborate these are really. Truly interdisciplinary field of research and so we had the colleagues collaborators some on campus was in Corp and others or actually different institutions and one of them was all one car at the Princeton who's an expert in for two emissions spectroscopy and we called them up and said OK here is what we've we've seen and you're interested they say yeah just send me a little bit of P.V.P. and we stick I.Q. modified. Sample into my. U.P.S. set up and we'll see what happens and this is what he muttered basically these are on the left here is the onset of the secondary for two electrons. Of. U.P.S. experiment and you can see that the bear I.Q. that onset is that this energy and once you put P.V.P. on top you have that shift here which is an indication of corresponding shift of the vacuum level which can be understood by the formation of a surface dipole OK. So now we started to understand what was causing that change in work function and then we went back and talk to our colleagues who are chemists and say well what could possibly be happening here and of course one suspect was that this isn't about I might hear about the that the nitrogen which has donating properties and so we started to hypothesize and then we went and we picked. Which is kind of the Bible for chemists where you have all these compounds and we started looking for compounds that contain a lot of I means OK and we ordered like a dozen and among these materials we had these. Materials that are shown here so these branched. Materials containing a lot of our means we call it these one these. And we started doing the same thing we just dipped two electrodes into that solution Mosher the changes with the Kelvin probe send them to Princeton they murdered the shifts of the vacuum levels but now these shifts are not just tens of yeas the shift was huge over evolved OK and what the data show here also is that it doesn't just work on night your but you can do the same we've brought you take gold which is a very stable high work function electrode you dip it into a dilute solution of that material and zoom you end up with something which has a work function which is one and a half wall to your OK so you can turn stable how you were functional ECT frauds into functional electrodes with values that are actually comparable thought to some of these are very unstable materials the other nice part is that on the top here you see ph one thousand this is the the name of a conducting polymer P.S.S. So this is a plastic electrode so you can apply the same strategy on an organic electrode. And so. Here are the data you can see that if you take I.Q. you were treated with the work function drops from about four point seven down to three point six It's stable as a function of temperature for temperatures up to two hundred degree C. which is plenty because these are the typical temperature ranges that we use when we were we've organic semiconductors and if you look at the and for mental stability if you expose these films to air thousands of hours are you have very little change in the workforce and saw very good the stability so the first step was of course to validate that that change in work function and the ability to produce function electrodes was working in sort of cells so here's on the left a typical architecture here to structure the glass which typically is used as a whole collecting electrode but if you modify it with that I mean painting or it becomes a lower puncture electrode so it becomes an electron collecting electrode than an active layer top hole collecting electrode here. Makes fairly nice sort ourselves with park with version efficiencies larger than than than six percent but remember the past that was also surprising was the fact that we could change the work function of conducting pullers not conducting polymers such as Peter P.S.'s are usually known as they are they are paedo ped and they have a higher work function values here but now here with the possibility to reduce the work function of these conducting we had a function and function on the king points which means that we were able to fabricate a completely plastic solar cell so the substrate is plastic the electron collecting electrode is that modified P.S.S. lower function conducting polymer you. Supreme Court Act if they are on top and you spin codes another function electrode and so all or plastic cell the details are shown here if you should see these are comparable to devices on glass with metal electrodes maybe a little bit less because of more serious resistance but that can be these are details that can be optimized OK. So initially the cells were fabricated using spin coding but the spin coding Peter P.S.S. on top of active layer that process from the very needed a typical organic solvent was a little bit challenging and so again in collaboration with people at Princeton we started to look for other ways of fabricating these cells rather than coding and the one of them was. The nation dry transfer which is you know something that's that used a lot in the industry and so in the next example that I'm showing you here we started with a lower function but I'm electrode but then the active layer rather than being spin coated was first fabricated on an independent substrate kind of a stamp and then it was transferred onto the substrate and the electron and then the top electrode here to the top of Peter P.S.S. L L here means that these are dry transfer I mean it layers was transferred on top and we can make saw ourselves that not only operate very well but we've unprecedented dynamic range. Nowadays we take even the commercial silicon solar cells they operate very well under in outdoor applications under a bright sun condition or diffuse light but you need fairly high injections you need the high insulation because when you the way the cells are fabricated. When you put you put the grids on the electrodes you introduce some defects and so you have what's called the parasitic resistance and the low shunt resistance and that resistance actually limits the performance of these solar cells and the light level this is why is difficult to make solar cells that actually convert light in those applications into into power OK but because of the transfer I mean Nation of these films and a corresponding with duction in defects we could get cells that have very high shunt resistance in the hundred may go on square centimeter which is you know ten thousand times higher than what we're typically muttering in conventional cells which means that we could fabricate also our cells that actually operate from food or Sun conditions that once earned over six orders of magnitude OK So you take a neutral density filter or do you one that reduces your incident light by by ten percent OK one order of magnitude if you stack six filters like this you have to work really hard to get to the conditions where you have that little light in your lab and you still get open circuit voltage of the order of point two in the in the cells and so here you can see the photo will take dynamic range when the light level is very by six orders of magnitude. We've recently taken some of these samples and we evaluated their performance as photo detectors and I can't give you the details because we've yet to file patents and so on but we have shown that these devices can have a linear dynamic range of ten orders of magnitude which means that they are better than some of the best silicon photodiodes that are available. So stay tuned. Anyway so the way you can improve the performance of light emitting diodes is making muted junction devices the same is true here we've organic solar cells so you can make stacks where you can connect these cells in series and the advantage is that you you know you can add up the voltages and you can harvest different parts of the spectrum if you use materials that have different optical properties and here again that I mean containing polymer as a surface modifier turned out to be a very good process to develop these interconnection layers that connect these two cells so going back to the previous architecture you see here the challenges in that layer here the so-called charge recombination layer that connects these two cells you don't want to introduce optical losses you don't want to introduce resistive. So resist series resistance and so here we could do this by having just a very thin conducting polymer layer that on one interface is a modified by that I mean containing material and if you combine these two layers you can get not on themselves with efficiencies higher than eight percent and fill factors that are beyond seventy percent OK this is the the recall champion sellers love we are so this is a summary of. You know that technique he shows that it's really universal in the fact that it works with a very broad. Range of different electrodes from metals to conducting polymers to values oxide even graphene. And so next. Next few minutes I'd like to also. Share with you kind of a small step of something which I think we've become more and more important as just kind of stimulate your intellect and and I think get us thinking more about full life cycle assessment in everything that we do because if you look at today's use of energy is residential use commercial use but a lot of energy is used at the industrial scale and so how much energy is actually used to produce something for the whole lifecycle of the product including the recycling I think needs to be evaluated when we compare different technologies and not just a metric which is like the dollars per watt that the module level when we compare different Pete acknowledges OK so the carbon footprint until now is irrelevant but I think it's becoming center stage and so we really need to worry about the environmental footprint of what we do and how much energy is actually used to produce these things are just you know in parents' S.S. It's interesting when you soft looking at the life cycle assessment of existing legacy technologies such as City Con Crystal in silicon You know these make very nice cells and Crystal and silicone is great there's nothing wrong with the ploughing more P.V. based on City Con But if you look in the long term in terms of sustainability and you think about how much energy is used to produce these City Con and how much energy is produced to go with these materials at very high temperatures and the ease at which you can actually make these new materials and dope them doping is just mixing two compounds at room temperature compared to you know heating things up at a thousand degrees for four minutes or hours OK and saw this is. Work that we did a few years ago and after we published the our work on the plastic solar cell in science it received some publicity and people started to cry. Me up and say What are you thinking you want to make plastics saw ourselves but plastics are horrible they are you know they are. At the origin of the pollution of oceans and so on so why would you want to massively deploy more plastics and basically try to solve an Andrew mantle problem by creating another one OK So that was kind of. A good stimulation and we started thinking about well what could we possibly do to further reduce the footprint of some of these solar cells and so if you look at the cross-section of these cells right now the active layers of very thick typically hundred two hundred one of these so it what makes most of the cell is actually the substrate because you need something that has the mechanical properties so that you can manufacture it and handle it deploy it and so we thought well can we actually replace some of these synthetic plastics substrates with natural materials and this is where again by serendipity I learned about. Those nano crystals at a dinner that the I.P.S. T. had organized years ago and I learned about these materials and you start from wood and you know all that process down here where you have these fibers is well known because this is the pulp that we use to produce paper but if you go further and you you continue to break down the paper pulp into the smallest substitutions you can. Fabricate the cellulose none of crystals and you can make transparent optically transparent substrates from these materials and so we fabricated our cells on these C.N.C. substrates and because they are made from these very small constituents you can. Exorcise which are very smooth very plain are much smoother than paper we try to put on paper is very difficult because the paper is too rough to fabricate these organic structures and so we could make cells at the time of population efficiency of about three percent well but what's really interesting is the ease at which you can actually recycle all recover these solar cells fabricated on the sea and substrate so here's a movie that shows that you know within ten minutes or less if you if you steer and shake the substrate none of cellulose substrate dissolves in water at room temperature if you filter it out you can separate the. Rose That's in the sort of and from. The organic material that. Absorbs the light and then if you wash the what's left you can see that it changes color and the color here is because you dissolving the active layer that is used in the soft cell and so you can recover the active polymer in another solvent and then at the end you can also recover some of the metal nano particles coming from the electrodes so within a few minutes at room temperature you can take these cells and you can you can separate them in their constituents OK And so this is just a recent work we've done in using that to transfer I mean nations so that now we can have higher efficiencies of the order of four percent and also recently we've shown that you can fabricate light sources you can make organic light emitting diodes on some of these substrate so basically improve paper OK All right so that brings me to my last slide I don't really have a conclusion because I think this is all a work in progress I hope I convinced you that you know these are small steps. There are still there's still plenty of room for additional science engineering and innovation and so before I and I'd like to also acknowledge the students. The students and post docs who've and collaborators over worked and made some of these results possible without them none of what I was showing today would have been possible and so we have this I like to thank you for your attention. For. Your. Work. Well you write the morphology on the control of the morphology are very critical in many aspects OK but you can either do that in situ in a cell where you fabricated the layer by conventional you know all additive spin coding or you can do the same processing in the film before you transfer OK. And so the transfer itself I guess the there limitations with the Trans Fair is that not the old materials are from. Parable with the same ease because it depends on the the softness or the level of crystal in India of some of these layers so that has to be you know tuned a little bit I would want to imply that the transfer lamination is straightforward this is sometimes a little bit tricky but at least this is a proof of principle that it can be done. On one. Of the. Plate. Well it's a good point things are a little bit more complicated that than what I implied And so what. What happens also and you know we we've shown that in our paper is that at the interface once you put. The semiconductor in contact with the low road after the modification you have doping at that interface of the semiconductor and that also contributes to hire an efficient injection I mean the injection is you know. You see it in every single diode where you do that modification it's a proven very efficient way of injecting electrons which you can describe as a lorry effective barrier OK. But you're right what really matters is the combination of the electrode with the semiconductor that you put on top. And then it's that effective barrier for charge injection and it's not just the work function of the material that matters but there's also. There are you know physical processes happening in the semiconductor itself once you make the contact and so that's that's a critical part of. It. So. Hard. So. Well. As light. Passes. Well but what you see is that you know the. The applied voltage. Is significantly reduced and you know. Maybe it works.