Mapping Correlated Environmental Landscapes in the Fluorescence Response of a Pyrenylated Charge‐Transfer Probe via Data‐Driven Nonlinear Regression
- 1Department of Chemistry Birla Institute of Technology and Science Pilani Hyderabad Telangana India
- 2Department of Electrical and Electronics Engineering Birla Institute of Technology and Science Pilani Hyderabad Telangana India
Abstract
Here, we report the design and detailed photophysical investigation of a pyrene aniline‐based donor–acceptor probe (compound 1) that displays pronounced sensitivity to multiple environmental factors. Steady‐state and time‐resolved fluorescence studies reveal strong solvatochromism, viscosity‐enhanced emission following the Förster Hoffmann relationship over a broad range (≈1.9–100 cP), concentration‐dependent aggregation effects, and thermally activated fluorescence quenching without changes in the emissive electronic state. Dynamic light scattering and density functional theory calculations establish a sterically twisted donor–acceptor architecture that supports intramolecular charge transfer while remaining highly responsive to microenvironmental constraints. Building on this experimentally validated framework, machine learning (ML) was employed as a computational extension of fluorescence spectroscopy to reconstruct continuous fluorescence–environment relationships from discrete datasets. Nonlinear regression models, particularly random forest and Gaussian process regression (GPR), accurately capture parameter‐specific response topologies, outperforming linear approaches. Correlated viscosity with temperature and viscosity with concentration intensity maps reveal regime‐dependent fluorescence behavior that cannot be inferred from single‐parameter trends alone. Point‐wise experimental validation confirms agreement in relative intensity ordering under combined parameter variation. Together, this work demonstrates a data‐curated strategy for integrating supramolecular photophysics with ML, enabling physically consistent interpretation of complex fluorescence responses without extrapolation beyond experimental control.
Related
- Weak-Form Regression Recovers the Delayed-Fluorescence Rate Law from Flash-Lamp Decays with Near-Optimal Precision
- Selective Hg2+ Detection Using an Aza-BODIPY-Derived NIR Fluorescence Sensor: Environmental and Biological Applications
- Applying landscape and jurisdictional approaches for nature positive outcomes in beef-producing landscapes.
- Electron Density of Delocalized Bonds from Correlated One-Particle Density Matrices
- A linear-scaling explicitly correlated local natural orbital CCSD(T) approach
BibTeX
@article{barkale2026mapping,
title = {{Mapping Correlated Environmental Landscapes in the Fluorescence Response of a Pyrenylated Charge‐Transfer Probe via Data‐Driven Nonlinear Regression}},
author = {Harshal V. Barkale and Anakhi Hazarika and Nilanjan Dey},
journal = {Advanced Materials Interfaces},
year = {2026},
month = nov,
doi = {10.1002/admi.70676},
url = {https://doi.org/10.1002/admi.70676},
note = {CC BY},
}
RIS
TY - JOUR
TI - Mapping Correlated Environmental Landscapes in the Fluorescence Response of a Pyrenylated Charge‐Transfer Probe via Data‐Driven Nonlinear Regression
AU - Harshal V. Barkale
AU - Anakhi Hazarika
AU - Nilanjan Dey
PY - 2026
DA - 2026/11/06
JO - Advanced Materials Interfaces
DO - 10.1002/admi.70676
UR - https://doi.org/10.1002/admi.70676
AB - Here, we report the design and detailed photophysical investigation of a pyrene aniline‐based donor–acceptor probe (compound 1) that displays pronounced sensitivity to multiple environmental factors. Steady‐state and time‐resolved fluorescence studies reveal strong solvatochromism, viscosity‐enhanced emission following the Förster Hoffmann relationship over a broad range (≈1.9–100 cP), concentration‐dependent aggregation effects, and thermally activated fluorescence quenching without changes in the emissive electronic state. Dynamic light scattering and density functional theory calculations establish a sterically twisted donor–acceptor architecture that supports intramolecular charge transfer while remaining highly responsive to microenvironmental constraints. Building on this experimentally validated framework, machine learning (ML) was employed as a computational extension of fluorescence spectroscopy to reconstruct continuous fluorescence–environment relationships from discrete datasets. Nonlinear regression models, particularly random forest and Gaussian process regression (GPR), accurately capture parameter‐specific response topologies, outperforming linear approaches. Correlated viscosity with temperature and viscosity with concentration intensity maps reveal regime‐dependent fluorescence behavior that cannot be inferred from single‐parameter trends alone. Point‐wise experimental validation confirms agreement in relative intensity ordering under combined parameter variation. Together, this work demonstrates a data‐curated strategy for integrating supramolecular photophysics with ML, enabling physically consistent interpretation of complex fluorescence responses without extrapolation beyond experimental control.
SN - 2196-7350
ER -