서브메뉴
검색
Enabling Data-Driven Experimentation for High-Performance Polymer Thin Film Formulations
Enabling Data-Driven Experimentation for High-Performance Polymer Thin Film Formulations
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20260202105553
- ISBN
- 9798263397227
- DDC
- 620.11
- 저자명
- Liu, Aaron Li.
- 서명/저자
- Enabling Data-Driven Experimentation for High-Performance Polymer Thin Film Formulations
- 발행사항
- [Sl] : Georgia Institute of Technology, 2023
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2023
- 형태사항
- 204 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 87-05, Section: B.
- 주기사항
- Advisor: Grover, Martha;Meredith, Carson;Reichmanis, Elsa.
- 학위논문주기
- Thesis (Ph.D.)--Georgia Institute of Technology, 2023.
- 초록/해제
- 요약Polymer thin films are a ubiquitous class of materials, as they demonstrate unprecedented performance in countless modern applications spanning electronics, coatings, composites, clean energy, packaging, and more. However, their final formulations are time-consuming to optimize through trial-and-error, as they are generated from a myriad of component choices and processing histories. While the advent of data science approaches has uncovered the promise of polymer informatics to accelerate new developments in polymer research, several challenges exist in adopting data-driven approaches effectively for the experimentation of polymer thin films. A foremost challenge is the low availability of experimental data, pertaining to the relevant process-structure-property relationships, that would yield the requisite knowledge necessary to construct precise models. "Small data" is an inherent problem for polymer thin film formulations; because their figures of merit are performance-based and often application-specific, parameter spaces are large, and reported data is inconsistent and sparse. To bridge the small data gaps that preclude the broader adoption of polymer informatics, this body of work details a series of case studies, spanning polymer stabilizers, composite blends, and organic electronics, addressing objectives that exemplify the broad challenges associated with accelerated development of polymer formulation technologies.The chapters within this dissertation address objectives that exemplify the broad challenges associated with enabling data-driven development of polymer formulations. First, when small datasets are inevitable, the incorporation of "small data analytics" approaches on small datasets provides a crucial foundation in extracting preliminary domain knowledge and informing future experimental work toward richer datasets. This thrust is demonstrated through a case study focusing on polymer stabilizer candidate discovery through descriptor calculation, domain knowledge extraction, and machine learning on a small dataset extracted from a patent. Second, while generating a larger amount of data through automated approaches is a prevalent area of interest for materials and polymer informatics, process constraints must be overcome to enable a broader adoption of high-throughput experimental methodologies. A major contribution to this goal is detailed in the design and implementation of a composition gradient thin film methodology that can handle systems at a variety of volume scales and temperature ranges, demonstrated through the interrogation of process-structure-property relations in polypropylene/polystyrene and poly(3-hexylthiophene-2,5-diyl)/polystyrene blend formulations. Finally, evolving polymer informatics to "big data" requires the development of robust data models and database structures that are representative of the complex, realworld parameter spaces associated with sample records, such that their provenance and processing details can be fully captured in a given materials sub-domain. The final chapter demonstrates a data management model for organic thin film transistors and implements it as an example system in polymer-based electronics, with the goal of promoting data management practices across other domains. Overall, this body of work features the rational integration of small-data analytics, high-throughput experimentation, and database design as key factors in enabling holistic data-driven methodologies for experimental materials development, particularly for polymer thin films.
- 일반주제명
- Materials research
- 일반주제명
- Semiconductors
- 일반주제명
- Protective coatings
- 일반주제명
- Genomes
- 일반주제명
- Data science
- 일반주제명
- Electronics
- 일반주제명
- Effluents
- 일반주제명
- Expenditures
- 일반주제명
- Thin films
- 일반주제명
- Viscosity
- 일반주제명
- Additives
- 일반주제명
- Solvents
- 일반주제명
- Design
- 일반주제명
- Reynolds number
- 일반주제명
- Digital technology
- 일반주제명
- Condensed matter physics
- 일반주제명
- Fluid mechanics
- 일반주제명
- Genetics
- 일반주제명
- Industrial engineering
- 일반주제명
- Materials science
- 기본자료저록
- Dissertations Abstracts International. 87-05B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008260126s2023 us c eng d■001000017360596
■00520260202105553
■006m o d
■007cr#unu||||||||
■020 ▼a9798263397227
■035 ▼a(MiAaPQ)AAI32315796
■035 ▼a(MiAaPQ)GeorgiaTech75129
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a620.11
■1001 ▼aLiu, Aaron Li.
■24510▼aEnabling Data-Driven Experimentation for High-Performance Polymer Thin Film Formulations
■260 ▼a[Sl]▼bGeorgia Institute of Technology▼c2023
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2023
■300 ▼a204 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 87-05, Section: B.
■500 ▼aAdvisor: Grover, Martha;Meredith, Carson;Reichmanis, Elsa.
■5021 ▼aThesis (Ph.D.)--Georgia Institute of Technology, 2023.
■520 ▼aPolymer thin films are a ubiquitous class of materials, as they demonstrate unprecedented performance in countless modern applications spanning electronics, coatings, composites, clean energy, packaging, and more. However, their final formulations are time-consuming to optimize through trial-and-error, as they are generated from a myriad of component choices and processing histories. While the advent of data science approaches has uncovered the promise of polymer informatics to accelerate new developments in polymer research, several challenges exist in adopting data-driven approaches effectively for the experimentation of polymer thin films. A foremost challenge is the low availability of experimental data, pertaining to the relevant process-structure-property relationships, that would yield the requisite knowledge necessary to construct precise models. "Small data" is an inherent problem for polymer thin film formulations; because their figures of merit are performance-based and often application-specific, parameter spaces are large, and reported data is inconsistent and sparse. To bridge the small data gaps that preclude the broader adoption of polymer informatics, this body of work details a series of case studies, spanning polymer stabilizers, composite blends, and organic electronics, addressing objectives that exemplify the broad challenges associated with accelerated development of polymer formulation technologies.The chapters within this dissertation address objectives that exemplify the broad challenges associated with enabling data-driven development of polymer formulations. First, when small datasets are inevitable, the incorporation of "small data analytics" approaches on small datasets provides a crucial foundation in extracting preliminary domain knowledge and informing future experimental work toward richer datasets. This thrust is demonstrated through a case study focusing on polymer stabilizer candidate discovery through descriptor calculation, domain knowledge extraction, and machine learning on a small dataset extracted from a patent. Second, while generating a larger amount of data through automated approaches is a prevalent area of interest for materials and polymer informatics, process constraints must be overcome to enable a broader adoption of high-throughput experimental methodologies. A major contribution to this goal is detailed in the design and implementation of a composition gradient thin film methodology that can handle systems at a variety of volume scales and temperature ranges, demonstrated through the interrogation of process-structure-property relations in polypropylene/polystyrene and poly(3-hexylthiophene-2,5-diyl)/polystyrene blend formulations. Finally, evolving polymer informatics to "big data" requires the development of robust data models and database structures that are representative of the complex, realworld parameter spaces associated with sample records, such that their provenance and processing details can be fully captured in a given materials sub-domain. The final chapter demonstrates a data management model for organic thin film transistors and implements it as an example system in polymer-based electronics, with the goal of promoting data management practices across other domains. Overall, this body of work features the rational integration of small-data analytics, high-throughput experimentation, and database design as key factors in enabling holistic data-driven methodologies for experimental materials development, particularly for polymer thin films.
■590 ▼aSchool code: 0078.
■650 4▼aMaterials research
■650 4▼aSemiconductors
■650 4▼aProtective coatings
■650 4▼aGenomes
■650 4▼aData science
■650 4▼aElectronics
■650 4▼aEffluents
■650 4▼aExpenditures
■650 4▼aThin films
■650 4▼aViscosity
■650 4▼aAdditives
■650 4▼aSolvents
■650 4▼aDesign
■650 4▼aReynolds number
■650 4▼aDigital technology
■650 4▼aCondensed matter physics
■650 4▼aFluid mechanics
■650 4▼aGenetics
■650 4▼aIndustrial engineering
■650 4▼aMaterials science
■690 ▼a0389
■690 ▼a0800
■690 ▼a0611
■690 ▼a0204
■690 ▼a0369
■690 ▼a0546
■690 ▼a0794
■71020▼aGeorgia Institute of Technology.
■7730 ▼tDissertations Abstracts International▼g87-05B.
■790 ▼a0078
■791 ▼aPh.D.
■792 ▼a2023
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17360596▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


