"Morphology and recombination in P3HT:PCBM organic solar cells."
The general aim of this thesis work has been the study of the relation between morphology and recombination in polymer:fullerene solar cells, since both are crucial factors governing the photovoltaic parameters. The polymer:fullerene combination P3HT:PCBM has been chosen as the model system for this investigation since it is a well known material system and its morphology and crystallinity (e.g. fiber contents) can be varied.
Non-geminate recombination of charge carriers has received here particular attention as it has been proven, in the last years, to be one of the main loss pathways in organic solar cells. The open circuit voltage (VOC) as well as the fill factor (FF) are mainly shaped by non-geminate recombination and by the relative weight of direct and trap-assisted recombination in the system studied.[70-72] In some polymer:fullerene blend it has been shown that the non geminate recombination is so strong to affect even the short circuit current (JSC) while in P3HT based solar cells the sweep out of the charges is faster than recombination reducing the current loss for voltages close to zero.
To measure the mobility and the recombination behavior of solar cells, next to the already available measurement techniques at IMO, in this PhD-work several additional techniques have been introduced. For mobility measurements Charge Extraction by Linearly Increasing Voltage (CELIV) and photo-CELIV were introduced. They are relatively easy measurements to carry out but powerful as they allow the study of mobility in a real solar cell configuration. Other measurements techniques require to build samples with a different structure compared to a real solar cell (e.g. Space Charge Limited Current SCLC requires appropriate electrodes) or they provide the value of mobility in a different direction compared to the normal percolation path of a solar cell (e.g. Field Effect Transistor addresses planar transport instead of vertical transport). The main drawback of CELIV is that is not possible to assign the measured mobility to holes or electrons. A recent work assigned this mobility value to an ambipolar mobility (dominated by the main carrier). Our measurements on P3AT:PCBM systems compare well with SCLC measurements on hole only diodes suggesting the CELIV mobility in this system probes mainly the hole mobility.
To measure non-geminate recombination, Transient Photovoltage (TPV) and Transient Photocurrent (TPC) techniques were introduced, next to the above mentioned photo-CELIV. Using TPV and TPC is of particular interest as they are performed biasing the solar cell with a white light. This allows to probe the lifetime of carriers and the number of generated carriers at different quasi-Fermi level splitting.
Next to the general aim of the thesis formulated above, four specific research questions have been addressed in this work. In the following, from the experiments and the measurements performed, a summary is given of the answers obtained in this work to the formulated research questions:
I. Is there a connection between the non-geminate recombination, the morphology of the active layer of solar cells and the amount of band gap trap states?
By varying the mass fraction of highly crystalline nanofibrillar P3HT to the total P3HT content in P3HT:PCBM solar cells it has been possible to observe that decreasing the P3HT fiber fraction delivers a higher apparent recombination order. This phenomenon was assigned to the increasing of sub-band gap trap levels caused probably by a decreased conjugation length in less crystalline materials. The increase of the trap concentration was proved by admittance spectroscopy measurements. A correlation between the fraction of crystalline P3HT nanofibers, the apparent recombination order and the trap concentration in the band gap has been found. The apparent recombination order is higher than 2 also for a solar cell with a fiber fraction of 1. We suggested that this could be probably due to the lack of crystallinity of the fibers themselves, which was estimated to be 65%. The estimated degree of crystallinity of 65% for P3HT nanofibers suggests that within 100% crystalline P3HT the recombination order would be 2, a fully direct recombination with no trap assisted recombination contribution.
II. Which is the effect of changing the molecular weight of the donor polymer? How do the photovoltaic parameters evolve? How is the crystallinity affected and how this reflects on non-geminate recombination?
After having looked to the effect on recombination by changing the fiber content in P3HT:PCBM solar cells, we had a look to the effect of changing the molecular weight. Recycling GPC was used as a convenient technique to obtain different Mn polymer fractions from a single master batch of P3HT. The fractions showed an almost identical polydispersity and regioregularity, while in previous works it has been shown to be difficult to achieve this result. Therefore disentangling solely the effect of Mn in previous work was difficult as the influence of polydispersity and regioregularity was not fully taken into account. We saw in this work that varying Mn has an effect on the crystallinity of the polymer, delivering a red shift of ECT with decreasing Mn. This shift has been proven to agree well with the shift in VOC, supplying an explanation for the observed reduction of VOC with decreasing the Mn. The Mn also affects the short circuit current delivered by the solar cells, mainly due to an increased light absorption for lower Mn P3HT, despite a decrease of CELIV mobility. The apparent recombination order follows the crystallinity of the polymer fraction and seems to affect the fill factor. The charge density dependence of the recombination coefficient leads to different recombination current at different charge density (illumination) and therefore to different fill factors.
III. What happens when a solar cell device is degraded under continuous light conditions and how does the degradation relate to the band gap trap states and recombination?
The stability of organic solar cells is a critical factor which needs to be considered in view of a future commercialization of this technology. We investigated the photo-degradation mechanisms under inert atmosphere of P3HT:PCBM solar cells, keeping an eye on the role of non-geminate recombination. We found that over a period of 1000 h the photo-degradation has different stages. In the first 250 h of illumination the devices exhibited an increase of p-type dopant concentration and a tenfold increase of the trap density. The alteration of these properties leads to a decrease of the mobility and a non-homogenous electric field, explaining the decrease in the short circuit current. The apparent recombination order starts to increase, probably due to the increased traps density and therefore of the trap assisted recombination causing a loss in VOC. A rather stable zone for device properties and electron transport parameters follows until 700 h illumination. After this period the fill factor strongly reduces attributed to the reduced surface recombination velocity at the contacts.
IV. Can the description of recombination kinetics in fully organic bulk heterojunction solar cells be extended to hybrid organic:inorganic solar cells?
Making use of impedance spectroscopy measurements under varying light intensity we proved that for hybrid solar cells based on ZnO nanorod arrays and P3HT, the mechanism governing the VOC is non-geminate recombination. The same approach used to describe recombination kinetics in fully organic P3HT:PCBM devices can be generalized and extended to hybrid solar cells. The working principles of photo excitation and recombination in hybrid solar cells seems the same as the one found for the model system P3HT:PCBM. This finding is achieved by extracting voltage dependent charge carrier densities and lifetimes by means of impedance spectroscopy, used as input parameters for a recombination model developed for purely organic solar cells. The predicted values for VOC are accurate, therefore it is possible to extend the validity of the recombination model towards hybrid solar cells and, consequently, towards a more general applicability.
There are still a lot of research questions to address in the field of organic photovoltaic. In the last years, models have been presented to describe the non-geminate recombination in organic bulk heterojunction solar cells deriving from amorphous semiconductors. These models take into account the recombination of free with trapped charge carriers, beyond the recombination of free with free charges. The trapped charges are usually modeled with an exponential tail of trapped states tailing in the bandgap, like the simple model presented in paragraph 18.104.22.168. Anyway the real shape of these trapped states is still unclear. Gaussian distribution have been proposed as well as a combination of exponential and Gaussian distributions. The shape of trap states density influences the apparent recombination order. It is important, therefore, to know this shape and to know the way the morphology of the film influence the shape, in order to control non-geminate recombination. It has been proposed, also, that the high recombination order cannot be explained solely by the charge carrier concentration dependent charge carrier mobility. If a charge carrier is trapped within one material phase far from the interface between the two phases, it has to be first released before it can recombine. The latter is an energy activated process and leads to a delay in recombination.[133,134] This delay results in the experiment as a reduced charge carrier decay in time and hence in an increased charge carrier recombination order. It is important to know if this phenomenon is present in organic BHJ.
The charge transfer state is commonly indicated as the intermediate state for two charges when they recombine. Therefore absorption/emission from this state is connected with the non-geminate recombination. Quantitative investigations on the origin and impact of radiative and nonradiative recombination on the solar cell performance have been started only recently. It is, therefore, still an open field interesting to investigate to finally improve the devices performance.
Regarding the degradation of solar cells, it has been shown in our institute that PCBM in solar cells dimerize under illumination. This effect counteracts the phase segregation of P3HT and PCBM when exposed to thermal degradation. This phenomenon was not taken into account before and needs to be studied. In particular, in the framework of this thesis work, it would be interesting to study its influence on non-geminate recombination.