Final Closeout report for grant FG36-08GO18018, titled: Functional Multi-Layer Solution Processable Polymer Solar Cells

Abstract

The original objectives were: (1) Develop a method to deposit multiple conjugated polymer film layers and avoid the problem of dissolution from mutually solubility; (2) Use this deposition method to develop multi-layer polymer based solar cells with layers that are function specific; (3) characterize these layers and devices; (4) develop electrical and optical models that describe and predict the properties of the multi-layers; and (5) Ultimate efficiency goals are {approx}6.75% with J{sub sc} = 12 mA/cm{sup 2}, FF = 0.75, and V{sub oc} = 0.75. The question of whether photovoltaic (PV) cells will ever be able to replace fossil fuels as the main provider of electrical power is not just a question of device efficiency; it is a question of how much power can be provided for what price. It has been well documented that PV devices at 10% power efficiency can provide for all of the world's power needs without covering too much of the earth's surface. Assuming desert like cloud coverage, an area equivalent to the land area of Texas would have to be covered. However, it has also been shown that using the current state-of-the-art silicon devices, the price-per-Watt will never be low enough to be economically feasible for large-scale development. Solution-processable PV devices based on polymers are a very attractive alternative to traditional Silicon PV because this technology is much lower in materials cost and in environmentally toxic waste production. Solution-based polymers can be rapidly deposited using printing technologies and are compatible with light-weight flexible substrates that can increase the range of available PV applications. In the past ten years, the efficiency of polymer based PV devices has increased from {approx}1% to over 10%. The highest efficiency organic solar cells are based upon a single layer than consists of a mixture of donor and acceptor moieties. This one layer has multiple optical and electrical functions, so the design of a single heterojunction layer is based upon the idea of balancing good and bad properties within a single film. This proposal addresses the idea that the use of multiple layers that have differing electrical and optical functions could lead to greater efficiency because fewer materials compromises must be made. This idea is not new, multiple functional layer have been successfully used in cross-linked OLED's and organic small molecule evaporated PV devices. The main reason that multiple layers of polymers are not commonly deposited is that most conjugated polymers are mutually soluble in the same solvents. The work outlined in the proposal was intended to develop a new deposition strategy that would allow multiple organic layers to be deposited from solution using spin coating. The deposition method that we proposed was successful, sometimes, but ultimately not reliable. Instead we focused on more reliable methods to implement doping along the interface between layers. This work has been very successful. We found that using PEDOT:PSS, the PSS would form a surface layer of {approx}2-3 nm thickness that would mix with and electrochemically react with P3HT upon heating. This mechanism is also a crosslinking reaction in that H{sub 2} is released and permanent new bonds are formed. Using the Plextronics Inc. replacement to PEDOT:PSS, for which there are no mobile dopants, we were able to show that a second and different mechanism can be used to p-type dope organic materials. We are currently working with Plextronics to develop a new product. Finally we produced n-type doping of a thin fullerene layer near the cathode also using a self-assembly method. Low work function metals will diffuse into the BHJ layer and dope the fullerene upon heating. This doping also affects the vertical segregation of BHJ materials in a predictable way. We accomplished all of the scientific goals that we set out in the proposal written in May 2007. Some of the methods we proposed were not fully successful, but we did come up with better methods to achieve the same goal. We did not achieve the efficiency goal for this proposal. We did increase V{sub oc} and explain why, but the same effect did not affect FF or J{sub sc} as we had proposed.