Supporting Information

Input Structure Preparation

The input files used for geometry optimization of urea and DCDHF-Me2 chromophores were prepared using Avogadro, an open-source molecular builder and visualization tool.^1,2 The general workflow for generating these input files involved building molecular structures using Avogadro’s graphical interface, performing pre-optimization with the Universal Force Field (UFF) within Avogadro, and exporting the structures as input files for subsequent quantum mechanical geometry optimization.

Urea Input Structure

The urea input file (urea.in) contains 8 atoms including the carbonyl group and two amino groups. The input structure was created to match the standard geometry of urea with appropriate bond lengths and angles. The original content of the Avogadro-generated file is shown below:

8
	Energy:    -115.2781918
N          0.20371        0.16819       -0.10412
C         -1.06311        0.59925        0.05445
N         -1.98034       -0.38739        0.03747
O         -1.35312        1.77482        0.19568
H          0.45121       -0.80639       -0.07401
H          0.93588        0.85986       -0.01745
H         -2.95249       -0.11457        0.08390
H         -1.74996       -1.34295       -0.17591

When using this input file with the optimize.py script, the file is placed in the input_structures directory.

DCDHF-Me2 Input Structure

The DCDHF-Me2 structure was created based on the donor-acceptor chromophore described by Lu et al.³ The molecular structure was built in Avogadro based on the molecular connectivity reported in the literature. The Avogadro-generated file contains 39 atoms and is shown below:

39
	Energy:      44.1010644
N          1.30016        1.74158        1.49069
C          0.93304        1.74035        0.13814
C         -0.18841        2.46517       -0.28855
H         -0.77438        3.03805        0.42543
C         -0.56453        2.52482       -1.63577
H         -1.43139        3.12966       -1.89443
C          0.16763        1.85097       -2.61881
C          1.28057        1.11677       -2.21283
H          1.87402        0.54966       -2.92382
C          1.66014        1.07514       -0.85989
H          2.53739        0.48626       -0.60735
C          0.22097        1.41135        2.42742
C          2.61408        1.22486        1.86436
H         -0.58363        2.15296        2.40671
H          0.59508        1.39787        3.45757
H         -0.19669        0.42145        2.21094
H          3.40721        1.68558        1.26470
H          2.66022        0.13550        1.76004
H          2.83404        1.47310        2.90897
C         -0.23019        1.98691       -4.02523
C         -1.44605        1.73833       -4.53771
C         -2.51721        1.11841       -3.84474
C          0.65874        2.54589       -5.10396
N         -3.38646        0.59284       -3.28407
O         -0.16750        2.51734       -6.30524
C         -1.47058        2.19592       -5.91182
C          1.90159        1.71406       -5.42867
C          1.03796        4.00891       -4.85646
H          1.65172        0.65234       -5.53864
H          2.67619        1.81370       -4.66232
H          2.33766        2.02717       -6.38495
H          0.14610        4.62934       -4.70756
H          1.55794        4.42529       -5.72711
H          1.68557        4.11961       -3.98083
C         -2.50517        2.43011       -6.74475
C         -3.86482        2.29156       -6.35795
C         -2.25639        2.93810       -8.05356
N         -4.97873        2.21982       -6.04165
N         -2.06490        3.35942       -9.11767

The structure is placed in the input_structures directory for processing with the optimize.py script.

Hardware and Software Environment

All calculations were performed on a Linux system (Ubuntu noble/24.04) with 15 GB total memory (2.8 GB used, 1.3 GB free, 11 GB buffer/cache), 4.0 GB swap space, and 8 CPU cores. The computational environment utilized Python 3.10, Psi4 1.7, and Miniconda for Python package management.

The AI-assisted development of the computational tools was performed using Claude 3.7 Sonnet.⁴

References

Hanwell, M. D.; Curtis, D. E.; Lonie, D. C.; Vandermeersch, T.; Zurek, E.; Hutchison, G. R. Avogadro: An Advanced Semantic Chemical Editor, Visualization, and Analysis Platform. Journal of Cheminformatics 2012, 4 (17). https://doi.org/10.1186/1758-2946-4-17.

Avogadro: An Open-Source Molecular Builder and Visualization Tool, 2022. http://avogadro.cc/.

Lu, Z.; Liu, N.; Lord, S. J.; Bunge, S. D.; Moerner, W. E.; Twieg, R. J. Bright, Red Single-Molecule Emitters: Synthesis and Properties of Environmentally Sensitive Dicyanomethylenedihydrofuran (DCDHF) Fluorophores with Bisaromatic Conjugation. Chemistry of Materials 2009, 21 (5), 797–810. https://doi.org/10.1021/cm8026797.

Anthropic. Claude 3.7 Sonnet, 2025. https://www.anthropic.com/claude.