Numerical Simulations On Organic Solar Cells

Environment and energy issues have been raising the difficulties of long-time living for human beings over the last few decades.As the most powerful energy source,solar energy should and must play an important role in achieving sustainable development in the near future.Inorganic silicon based photovoltaic has been well developed both theoretically and practically,and shows very high power conversion efficiency over 20% for single junction cells.However,the major problems for silicon industry are the air pollution and high production cost,which cannot be mitigated easily.It's part of the reason to develop new technique for solar energy harvest.As a novel candidate,organic photovoltaic has attracted lots of attentions in the last few decades,owing to the advantages of flexibility,low-cost,light-weight and large area achievability.Recently,the development of non-fullerene acceptor has boosted the power conversion efficiency to almost 15% compared to very first one(lower than 1%),which paves the way to future mass production.In the meantime,physicists have been working very hard to understand the unusual behaviour and develop new concepts to differentiate from inorganic cells.In this thesis,we try to understand the macroscopic device behaviour by means of driftdiffusion modelling supported with experimental data.In the background part,we provide a detailed description of the most fundamental theory and model in the field of photovoltaics,which allows us to further extend to the specific organic case.The first result chapter will focus on the effects of charge transport mobility,based on the well-known Shockley-Queisser theory.Starting from the ideal cell,we are showing how the device performance will be changed under low(10-4 cm2 V-1 s-1)and high mobility(104 cm2 V-1 s-1)values with different contact and recombination properties.High mobility device always gives us high short circuit current,but can be very sensitive to contact properties,whereas low mobility device is less sensitive to contacts,but limited more by the poor charge transport and collection efficiency.Generally speaking,high mobility device has more potential to reach the Shockley-Queisser limit,while low mobility cannot reach that limit,but still have change to reach over 10% with low recombination coefficient and ideal contacts.The second result chapter will be focusing on the prediction of radiative efficiency limit based on optoelectrical reciprocity relation.Ten different blends will be used as the research targets,including some classical donors,like P3 HT and PTB7,and novel non-fullerene acceptors,ITIC and C8-ITIC.The prediction is based on the reciprocity relation between absorption and emission of the blends,which can be used to calculate the radiative losses.Combined with the transport equations,we are able to first of all simulate the ideal case without transport losses,then the mobility limited case.The results indicate that non-fullerene acceptor has more potential,but might be limited more by transport problems.The small difference between absorption band gap and effective band gap in non-fullerene case is thought to be the reason for largely reduced fill factor and efficiency with higher recombination coefficient,which might be the reason why most non-fullerene acceptor devices show much lower fill factor than fullerene based devices.In the very last chapter,we are going to focus on the most sensitive parameter fill factor.From an easy survey on the relation between mobility and fill factor,we find that there might be an optimum mobility value,which is also not high.Chapter ? gives the possible reason why low mobility devices still provide with high fill factor over 80%.In this part,we are focusing more on how to achieve high fill factor and understand it in particular in smallmolecule donor based solar cells.With a good collaboration with Prof.Wang from Beijing Institute of Technology,we developed a series of novel high efficiency small-molecule donors.Using a fancy technique called “solvent vapour annealing”,we are able to change the blend morphology in a large range,which ends up with largely changed device performance as well.Using BIT-4F as the hero,we reported a record high fill factor over 78%.Transient photovoltage measurements supported with classical Onsager-Braun model confirms that the largely reduced geminate recombination should be responsible for the improved fill factor.Further simulation based on Onsager-Braun model proceeding with drift-diffusion modelling suggests a similar trend in the picture of “fill factor versus hole mobility” from the previous survey.Combined simulation and survey suggests that the optimum mobility value might be around 5×10-3 cm2 V-1 s-1 for organic photovoltaics.