Multi‐Fidelity High‐Throughput Optimization of Electrical Conductivity in P3HT‐CNT Composites

Combining high‐throughput experiments with machine learning accelerates materials and process optimization toward user‐specified target properties. In this study, a rapid machine learning‐driven automated flow mixing setup with a high‐throughput drop‐casting system is introduced for thin film preparation, followed by fast characterization of proxy optical and target electrical properties that completes one cycle of learning with 160 unique samples in a single day, a >10× improvement relative to quantified, manual‐controlled baseline. Regio‐regular poly‐3‐hexylthiophene is combined with various types of carbon nanotubes, to identify the optimum composition and synthesis conditions to realize electrical conductivities as high as state‐of‐the‐art 1000 S cm−1. The results are subsequently verified and explained using offline high‐fidelity experiments. Graph‐based model selection strategies with classical regression that optimize among multi‐fidelity noisy input‐output measurements are introduced. These strategies present a robust machine‐learning driven high‐throughput experimental scheme that can be effectively applied to understand, optimize, and design new materials and composites.

[1]  Alán Aspuru-Guzik,et al.  Designing and understanding light-harvesting devices with machine learning , 2020, Nature Communications.

[2]  J. Min,et al.  Machine learning for accelerating the discovery of high-performance donor/acceptor pairs in non-fullerene organic solar cells , 2020, npj Computational Materials.

[3]  Qianxiao Li,et al.  Two-step machine learning enables optimized nanoparticle synthesis , 2020, npj Computational Materials.

[4]  Reiner Sebastian Sprick,et al.  A mobile robotic chemist , 2020, Nature.

[5]  A. Ludwig,et al.  Combinatorial Synthesis and High-Throughput Characterization of Microstructure and Phase Transformation in Ni–Ti–Cu–V Quaternary Thin-Film Library , 2020 .

[6]  Brian L. DeCost,et al.  A high-throughput structural and electrochemical study of metallic glass formation in Ni-Ti-Al. , 2019, ACS combinatorial science.

[7]  Alán Aspuru-Guzik,et al.  Beyond Ternary OPV: High‐Throughput Experimentation and Self‐Driving Laboratories Optimize Multicomponent Systems , 2019, Advanced materials.

[8]  S. Tolbert,et al.  Dopant-Induced Ordering of Amorphous Regions in Regiorandom P3HT. , 2019, The journal of physical chemistry letters.

[9]  Alfred Ludwig,et al.  Discovery of new materials using combinatorial synthesis and high-throughput characterization of thin-film materials libraries combined with computational methods , 2019, npj Computational Materials.

[10]  In Hwan Jung,et al.  Enhanced Thermoelectric Performance of Conjugated Polymer/CNT Nanocomposites by Modulating the Potential Barrier Difference between Conjugated Polymer and CNT , 2019, ACS Applied Electronic Materials.

[11]  Stephen Wu,et al.  Machine-learning-assisted discovery of polymers with high thermal conductivity using a molecular design algorithm , 2019, npj Computational Materials.

[12]  A. Aspuru-Guzik,et al.  Self-driving laboratory for accelerated discovery of thin-film materials , 2019, Science Advances.

[13]  G. Park,et al.  Modeling the electrical resistivity of polymer composites with segregated structures , 2019, Nature Communications.

[14]  S. Harding,et al.  Probabilistic neural network identification of an alloy for direct laser deposition , 2019, Materials & Design.

[15]  Jun Ye,et al.  Machine learning enables polymer cloud-point engineering via inverse design , 2018, npj Computational Materials.

[16]  A. Bollinger,et al.  Combinatorial synthesis and high-throughput characterization of copper-oxide superconductors , 2018, Chinese Physics B.

[17]  Pawan Kumar,et al.  Polymer morphology and interfacial charge transfer dominate over energy-dependent scattering in organic-inorganic thermoelectrics , 2018, Nature Communications.

[18]  E. Hüllermeier,et al.  Multi-target prediction: a unifying view on problems and methods , 2018, Data Mining and Knowledge Discovery.

[19]  Ruipeng Li,et al.  Influence of dopant size and electron affinity on the electrical conductivity and thermoelectric properties of a series of conjugated polymers , 2018 .

[20]  Shinji Nagasawa,et al.  Computer-Aided Screening of Conjugated Polymers for Organic Solar Cell: Classification by Random Forest. , 2018, The journal of physical chemistry letters.

[21]  B. D. Conduit,et al.  Probabilistic design of a molybdenum-base alloy using a neural network , 2018, ArXiv.

[22]  B. D. Conduit,et al.  Design of a nickel-base superalloy using a neural network , 2017, ArXiv.

[23]  S. Tolbert,et al.  The Effects of Crystallinity on Charge Transport and the Structure of Sequentially Processed F4TCNQ‐Doped Conjugated Polymer Films , 2017 .

[24]  Wei Chen,et al.  An ab initio electronic transport database for inorganic materials , 2017, Scientific Data.

[25]  Anubhav Jain,et al.  Effective mass and Fermi surface complexity factor from ab initio band structure calculations , 2017, npj Computational Materials.

[26]  Zhiqiang Fan,et al.  Straightening Single-Walled Carbon Nanotubes by Adsorbed Rigid Poly(3-hexylthiophene) Chains via π–π Interaction , 2016 .

[27]  C. Uher,et al.  Highly anisotropic P3HT films with enhanced thermoelectric performance via organic small molecule epitaxy , 2016 .

[28]  S. Tolbert,et al.  Overcoming Film Quality Issues for Conjugated Polymers Doped with F4TCNQ by Solution Sequential Processing: Hall Effect, Structural, and Optical Measurements. , 2015, The journal of physical chemistry letters.

[29]  Atsuto Seko,et al.  Prediction of Low-Thermal-Conductivity Compounds with First-Principles Anharmonic Lattice-Dynamics Calculations and Bayesian Optimization. , 2015, Physical review letters.

[30]  Zoubin Ghahramani,et al.  Probabilistic machine learning and artificial intelligence , 2015, Nature.

[31]  Y. Furukawa,et al.  Electronic and vibrational spectra of positive polarons and bipolarons in regioregular poly(3-hexylthiophene) doped with ferric chloride. , 2015, The journal of physical chemistry. B.

[32]  J. Bell,et al.  An efficient hole transport material composite based on poly(3-hexylthiophene) and bamboo-structured carbon nanotubes for high performance perovskite solar cells , 2015 .

[33]  F. Mjalli,et al.  Effect of water on the thermo-physical properties of Reline: An experimental and molecular simulation based approach. , 2014, Physical chemistry chemical physics : PCCP.

[34]  H. Kataura,et al.  Giant Seebeck coefficient in semiconducting single-wall carbon nanotube film , 2014, 1401.7469.

[35]  Luís Torgo,et al.  SMOTE for Regression , 2013, EPIA.

[36]  Taylor D. Sparks,et al.  Data-Driven Review of Thermoelectric Materials: Performance and Resource Considerations , 2013 .

[37]  M. Martín-González,et al.  Thermoelectric composites of poly(3-hexylthiophene) and carbon nanotubes with a large power factor , 2013 .

[38]  Grigorios Tsoumakas,et al.  Multi-target regression via input space expansion: treating targets as inputs , 2012, Machine Learning.

[39]  Nando de Freitas,et al.  A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning , 2010, ArXiv.

[40]  J. Grossman,et al.  Self-assembly and its impact on interfacial charge transfer in carbon nanotube/P3HT solar cells. , 2010, ACS nano.

[41]  J. Joo,et al.  Light-emitting color barcode nanowires using polymers: nanoscale optical characteristics. , 2010, ACS nano.

[42]  Nir Friedman,et al.  Probabilistic Graphical Models - Principles and Techniques , 2009 .

[43]  Yang Xu,et al.  Electrical, Optical, and Morphological Properties of P3HT-MWNT Nanocomposites Prepared by in Situ Polymerization , 2008 .

[44]  Young Hee Lee,et al.  Absorption spectroscopy of surfactant-dispersed carbon nanotube film: Modulation of electronic structures , 2008 .

[45]  B. Jousselme,et al.  Elaboration of P3HT/CNT/PCBM Composites for Organic Photovoltaic Cells , 2007 .

[46]  M. Knupfer,et al.  Electronic properties of barium-intercalated single-wall carbon nanotubes , 2004 .

[47]  H. Sirringhaus,et al.  Effect of interchain interactions on the absorption and emission of poly(3-hexylthiophene) , 2003 .

[48]  V. C. Moore,et al.  Band Gap Fluorescence from Individual Single-Walled Carbon Nanotubes , 2002, Science.

[49]  Jandeleit,et al.  Combinatorial Materials Science and Catalysis. , 1999, Angewandte Chemie.

[50]  P. Schultz,et al.  Combinatorial approaches to materials science , 1998 .

[51]  B. Singh.  A Mechanical Pipette , 1949, The British journal of venereal diseases.

[52]  Nando de Freitas,et al.  Taking the Human Out of the Loop: A Review of Bayesian Optimization , 2016, Proceedings of the IEEE.

[53]  J. Friedman,et al.  Predicting Multivariate Responses in Multiple Linear Regression , 1997 .