Great poster and video explanation Michele! Just curious about how stable OPVs are and how long the films of organic materials last before not working any more?
Judges’ Queries and Presenter’s Replies
Presentation Discussion
-
-
Thank you!
Great question – in general, OPVs are less stable than traditional PV technologies. The general premise for using organic materials for electronic devices like solar cells is that the lower efficiency and lifetime of devices made with organic materials may be more than compensated for by the much lower cost of materials/production for OPVs. It has been estimated (see ref below), that OPVs can begin to compete with other PV technologies once they achieve power conversion efficiencies > 7% and lifetimes greater than 7 years. Both of these markers have since been approached or exceeded (see second ref), and so there is some cause to be optimistic about this technology.
The stability of any given OPV device is going to be highly system-dependent. The dominant factor in many cases is actually the stability of the top electrode, commonly Al, which is easily oxidized. For this reason, inverted solar cell architectures, for which the top electrode is a higher-workfunction and thus more stable material, tend to have very promising lifetimes.
Hope this answers your question. Also, how have you been?! I haven’t seen you since you were at UC Santa Barbara! You’re already a post-doc now?
1. Dennler, G.; Scharber, M. C.; Brabec, C. J. Polymer-Fullerene Bulk-Heterojunction Solar Cells Adv. Mater. 2009, 21, 1323-1338.
2. Peters, C. H.;Sachs-Quintana, I. T.;Kastrop, J. P.;Beaupre, S.; Leclerc, M.; McGehee, M. D. High Efficiency Polymer Solar Cells with Long Operating Lifetimes Adv. Energy Mater. 2011, 1, 491-494.
-
Hi Michele:
That answers my question! Thanks!
I’ve been doing well, I actually just finished my PhD at the beginning of May looking at the atmospheric transport of pathogens and started a post-doc in the Department of Civil and Environmental Engineering (big change from biochemistry when I was at UCSB) looking at virus stability in the atmosphere. Hopefully, you have been doing well. I am guessing you and others we started UCSB with are starting to think about finishing up? Hope all else is well in warm and sunny Santa Barbara!
-
That’s awesome – I’m so glad it sounds like things are going well for you. Yes, starting to think about what’s next! Thanks again for the message. Take care!
-
After all these years of lab work, it is wonderful to get to see your work and revel at the potential benefits you may yet introduce to us all.
-
Thank you, Jodi!
-
It seems like you think the i-layer mostly plays a role in morphology, is that right? In inorganic solar cells, we use intrinsic layers often as tunnel layers, where there is less carrier recombination. That is, it plays an electronic role.
-
Hi Prof. Opila,
Well, I think that the i-layer or any other part of the active layers plays an electronic role, but that the particular morphologies in the i-layer with these two acceptor materials, PCBM and PCBNB, also play an important role.
Organic solar cells that employ a bulk heterojunction device strategy (as opposed to a bilayer architecture) rely on the particular morphology of the materials in the active layer in order to maximize surface area but also maintain percolating networks for charge carrier transport. Morphological effects can significantly effect the device performance.
That’s an interesting comparison to inorganic PV – thanks!
-
Further posting is closed as the event has ended.
Ananth Iyer
Faculty: Project Co-PI
Do you expect to use your data be able to build a statistical model of the surface structure to optimize its performance ?
Michele Guide
Dear Prof. Iyer,
Thank you for your question! I would say that the kind of analysis shown in this study is best suited for a qualitative comparison of the surface structure and device performance. One of the goals of my graduate work is to make pc-AFM a more robust technique so that more quantitative analysis can be performed and so it can be an even closer analog to bulk solar cell testing.
Another challenge for modeling this kind of structural data is that it is difficult to standardize what an ideal morphology would be. In the OPV literature, there are some trends for what constitutes a favorable morphology, however, there is still significant debate. Additionally, what may be a good morphology for one set of donor and acceptor materials may be a poor morphology for another pair of materials. As a result, we must evaluate the morphology of our system in the context of the other properties of the system, using the cumulative morphological intuition of the OPV community as a guide.
Paulette Clancy
Faculty: Project Co-PI
Are you surprized at the significant difference in the surface aggregation that you found for PCBM and PCBNB, which are themselves so chemically similar? What is driving this difference and what does it mean for the optimal design for n-type fullerenes?
Michele Guide
Dear Prof. Clancy,
You bring up an important point. The chemical structures of PCBM and PCBNB are very similar, and yet they result in dramatically different solid-state structures. I think that this is a testament to the complexity of the solid-state morphology of solution-processed organics, which can be greatly affected by both solution and solid state interactions of the materials with each other and with solvent molecules.
Qualitatively, PCBNB is more soluble (in solution) than PCBM, owing to its longer alkyl chain. It is possible, then, that PCBNB may also be better at inhibiting large-scale BP crystal formation in the solid-state, although the exact mechanism of this have not yet been determined.
The tricky thing about designing the “optimal” materials for OPV applications is that one must design on the system level: taking into consideration both the donor and acceptor and their interactions with each other. There remains so much that is unknown about these interactions, and so there are no established formulae to follow to design materials with the right properties to obtain an optimal OPV device. This is a big topic of discussion in the organic electronics field; I would be happy to discuss it further.
Paulette Clancy
Faculty: Project Co-PI
Thank you, Michele. A thoughtful response. Actually, I am quite interested in this point and we should perhaps touch base to discuss the interests of my lab and yours. My email address is Paulette.Clancy@cornell.edu
Michele Guide
Hi Prof. Clancy,
Thank you again for your question. Absolutely, I will get in touch with you via email.
Ranjit Koodali
Faculty: Project Co-PI
Could you share an insight into how or why you came up with the idea of using PCBNB? Based on the results obtained with PCBNB, What other PCB like molecules do you wish to investigate in the future?
Michele Guide
Dear Prof. Koodali,
Thank you for your question. Our interest in PCBNB was sparked by our collaboration with Mitsubishi Chemical Company through the Mitsubishi Chemical Center for Advanced Materials (MC-CAM) at UCSB. Researchers in Japan had recently developed this high efficiency p-i-n system with benzoporphyrin (the second reference in my poster, Matsuo et al), and found that the chemical structure of the acceptor (PCBM and PCBNB as well as other fullerenes developed by Mitsubishi) used in this system significantly influences the device performance.
So from the starting point that the use of PCBM and PCBNB leads to different macroscopic effects on the devices, we set out to understand what was causing that difference. We could rule out the likelihood of it being a difference in the frontier energy levels of the acceptors because they are quite similar. Therefore, we suspected that nanoscale morphological differences may be at the heart of the issue. We found c-AFM and pc-AFM to be a powerful tool for investigating the nanoscale morphologies and photovoltaic properties of this system.
While fullerene derivatives are very special and fascinating molecules, I am actually interested in exploring non-fullerene acceptors for OPV applications in the future. Fullerene derivatives have quite a high embodied energy compared to other organic molecules, and I think there is much promise in other electron accepting materials that can be made from less energy intensive and less expensive synthetic procedures. Of course, fullerene derivatives are ubiquitous in the OPV literature, as a result of their excellent electronic properties, frontier energy levels, and ability to form favorable nanoscale morphologies with a wide range of donor materials. It will be a challenge to design OPV systems with non-fullerene acceptors that can match what has been accomplished with fullerenes to date.
Additionally, I consider the study shown in my poster to be a first step towards developing the pc-AFM technique itself for the robust characterization of OPVs. If we can develop low workfunction AFM probes for pc-AFM and better understand the nature of charge transport and photoconductivity in a pc-AFM experiment, we may be able to make pc-AFM an even more powerful technique for understand OPVs, and an even closer analog to bulk OPV testing.
Ranjit Koodali
Faculty: Project Co-PI
Thank you very much for your very thoughtful response, Michele
Michele Guide
My pleasure.
Matthew Yates
Faculty: Project Co-PI
Can anything be gleaned from the nanoscale morphology to predict device lifetime and stability?
Michele Guide
Dear Prof. Yates,
Thank you for your question. I wouldn’t claim that nanoscale morphology could be used as a comprehensive tool to predict device lifetime and stability, however, it could be used as an indicator for a couple issues relating to stability.
First, certain OPV systems have unstable morphologies: AFM can be used to monitor the nanoscale morphology of these systems. If the morphology noticeably changes with time, that would probably go hand-in-hand with shorter device lifetime.
Secondly, some pc-AFM setups (such as in our lab) use a white light with a maximum intensity of ~300 “suns”. This can be used to accelerate the aging process for a sample. Indeed, for some systems, we find relatively fast decay of the photocurrent measured as the sample is continuously exposed to such high intensity light. Other systems are more impervious to the high light intensity. I hope this answers your question.
Ian Harrison
Faculty: Project PI
Does the relative energetic ordering of the PCBNB and PCBM electronic energy levels have any anticipated impact on the iV curves? If so, why and how are you ultimately able to ascribe the main differences in the iV curves to morphological effects alone? [You mentioned that the difference in frontier orbital energies was small but at what magnitude is the difference completely negligible?]
Michele Guide
Dear Prof Harrison,
Thank you for your question. You are correct in pointing out that the energetic order of the acceptor materials can impact the device performance. For instance, a difference in the crystallinity of the PCBM and PCBNB phases could result in different density of states of these two materials, which can, in turn, impact device performance.
In this study, both acceptor materials in the i-layer are annealed above their glass transition temperatures and may likely exhibit crystallinity (although we haven’t measured this specifically in my system). It is thus unlikely that the energetic ordering in the two materials are sufficiently different as to affect the J-V characteristics. While we cannot claim that the morphological differences we observe in the two systems account for all differences in device performance, the stark differences in morphology and the relative similarity of the materials in other respects lead us to conclude that the morphology plays a significant role in the performance of these systems.
You also bring up a good point about the difference in the frontier orbital energies. While the measurements we reference for the LUMO values of the acceptors yield the same number, it is very difficult to determine these energy levels with high certainty, and so there could be some variation in the true energy levels. This error could be on the order of the difference in the open-circuit voltage in the two systems (from 0.44 V to 0.60 V), however, it does not account for the differences in fill factor and short-circuit current density. I hope this addresses your question – thanks again.
Ian Harrison
Faculty: Project PI
Thanks!