Synthesis And Photovoltaic Performance Of Polymer Acceptors And Block Copolymers Based On Y-series Structure

Over the last three decades,solution-processed organic solar cells（OSCs）have attracted much attention due to the advantages of low cost,light weight,and their potential application in semi-transparent and flexible devices.At present,the maximum PCE of single junction OSCs based on Y-series small molecule acceptors has exceeded19%,yet the stability issues of this kind of OSCs device remain challenging.At present,the most wildly used acceptors are small molecule non-fullerene materials.Compared with small molecular materials,polymer materials are intrinsically more stable in chemical structure and film morphology.In this work,we researched on polymer acceptors based on Y-series structure which have demonstrated high performance in small molecule system.Then,this kind of polymer acceptors were used as segments to construct block copolymer containing both donor and acceptor blocks,which were further applied to single materials organic solar cells（SMOSCs）.We aimed to improve the device stability via material system optimization while maintaining high PCE of Y-series materials.According to this strategy,the main works were summarized as follows.（1）A series of terpolymer acceptors PYEx（x=1,2,3,4）were synthesized by introducing ester substituted thiophene（EST）into polymer acceptors which were polymerized by Y-series small molecule.With the increase of EST content,the lowest unoccupied molecular orbital energy level of polymer acceptors decreased slightly,and the charge transfer efficiency in the blend film was improved,as a result of which,there was a trade-off between open circuit voltage(V_OC)and short circuit current(J_SC).On the other hand,the introduction of EST units can effectively adjust the orientation ofπ-πstacking and the packing order of lamellar stacking.Finally,the polymer acceptor with 20%EST obtained the highest PCE of 13.57%in all polymer solar cells.In addition,all PYEx based devices achieved PCE over 13%,indicating high tolerance to the content of EST.（2）The monomers of Y-series polymer acceptor and polymer donor were used to construct conjugated block copolymer PBDB-T-b-PYT containing both donor and acceptor block via stepwise Stille polymerization.Intermolecular charge transfer and intramolecular charge transfer can coexist in the block copolymer films,resulting in highly efficient exciton dissociation.The SMOSCs based on PBDB-T-b-PYT has achieved a high PCE up to 11.32%,and the PCE certified by the Chinese Academy of metrology is 10.8%.Besides,compared with the corresponding all polymer bulk heterojunction OSCs,the SMOSCs device based on PBDB-T-b-PYT exhibited higher stability and lower energy loss.（3）To investigate the batch-to-batch difference of conjugated block copolymers,a series of block copolymers PB-b-PY-x（x=1,2,3,4）with different donor block lengths were synthesized by controlling the polymerization time.With the increase of block length,the absorption intensity of donor block increases,the mobility of block copolymer is improved,and the lamellar stacking of films gets more ordered.Finally,compared with other block copolymers,the copolymer PB-b-PY-4 which has the longest donor block achieved the highest PCE of 13.05%in SMOSCs and higher photostability were also observed.（4）In order to further broaden the absorption range of PBDB-T-b-PYT and adjust its aggregation behaviors,a series of block copolymers PB-b-PYCl-x（x=1,2,3）with different chlorine substitution positions were synthesized.The chlorinated block copolymer has more red shifted absorption,and the short-circuit current of the corresponding device is increased.Among these polymers,the PB-b-PYCl-2 film has the highest and most balanced charge mobilities.Consequently,compared with other block copolymers,SMOSCs based on PB-b-PYCl-2 achieved the highest PCE of13.42%.Besides,the PB-b-PYCl-2 film also showed high thermal stability.（5）Based on the above-mentioned work on material design,synthesis and their photovoltaic performance from which we explored the strategy of designing high-performance materials,we were further thinking about how to improve the probability of designing high-performance materials in lab.To explore auxiliary tools for materials development,a series of machine learning models were built by five algorithms:linear regression,multinomial logistic regression,random forest,boosted regression tree and artificial neural network,to predict the PCE of donor/acceptor（D/A）pairs in OSCs.The correlation coefficient（r）between the predicted value of RF and BRT model and the actual value have exceeded 0.7.We further conducted high-throughput virtual screening through RF and BRT models on 32 million D/A pairs which were automatically generated.According to the prediction results,6 D/A pairs were synthesized and characterized in OSCs.Based on the new D/A pairs,PCE up to 15.7%was achieved,which demonstrated that the machine learning model is very helpful for developing high-performance photoactive materials.By comparing the experimental results with the prediction results,we find that among these models,RF model is the most qualified model,which can effectively carry out high-throughput virtual screening of photovoltaic materials,offering guidance on materials design and D/A counterpart selection.