A new copolymer (PC20BDTDPP) based on modified benzo[1,2-b:3,4-b']dithiophene (BDT) unit as donor and diketopyrrolopyrrole (DPP) unit as acceptor was synthesized. The bulkier alkoxy on BDT and alkyl on DPP were used to increase the solubility of the resulting polymers. It was striking that Power conversion efficiency (PCE) of polymer solar cells fabricated with PC20BDTDPP as the electron donor blended with [6,6]-phenyl-C71- butyric acid methyl ester (PC71BM) as the electron acceptor was improved from 0.72% to 1.6% by adding 3% of 1,8-diiodooctane as the solvent additive into o-dichlorobenzene. The enhanced PCE was attributed to optimized surface morphology and packed polymer chains leading to better phase separation morphology by the solvent additive. Furthermore, owing to its very narrow band gap, the synthesized polymer demonstrates a great potential for tandem or parallel-like solar cells.
Krylov subspace algorithms are important methods for solving linear systems. In order to efficiently solve large-scale linear systems, parallelism techniques are often applied. However, parallelism often enlarge the non-associativity of floating-point operations, which can lead to non-reproducibility of the computations. This paper compares the performance of the parallel preconditioned BiCGSTAB algorithm implemented with two different libraries (ExBLAS and ReproBLAS) that can ensure the reproducibility of computations. To address the effect of the compiler, we explicitly utilize the FMA instructions. Finally, numerical experiments show that based on two BLAS implementations, the BiCGSTAB algorithms are reproducible. By contrast, the BiCGSTAB algorithm based on ExBLAS is more accurate but more time-consuming than the one based on ReproBLAS.