Data for: Stereoregular radical polymers enable selective spin transfer - computational studies (Data S1 and S2) and crystal structure of M1
Data files
Mar 12, 2025 version files 999.94 KB
-
2324183.cif
704.46 KB
-
boltz_SRR2.p
41.40 KB
-
boltz_SRS2.p
60.15 KB
-
boltz_SSS2.p
45.39 KB
-
magnetic_persistence.ipynb
145.14 KB
-
README.md
3.41 KB
Abstract
Spintronics offers a promising avenue for surpassing the performance and energy efficiency limits of conventional electronic devices. However, existing spintronic materials, including metals and doped conjugated polymers, face intrinsic stability and performance challenges. To address these limitations, we employed computational methods to investigate the electronic structure, spin transport properties, and crystallographic order of a stereoregular radical polymer. Computational simulations in different stereoregularities were utilized to model spin-spin interactions, charge delocalization, and long-range order within the polymer backbone. Crystallographic data analysis provided insights into the molecular packing of the monomer (M1) and the role of stereochemistry in controlling spin alignment. Our findings highlight how stereoselective polymerization enables persistent radicals in each repeat unit to support long-range spin transport without conventional doping. This computationally guided approach underscores the potential of stereoregular radical polymers as a novel platform for next-generation spintronic devices and quantum information processing. The computational results support the hypothesis that molecular-level alterations in polymer stereochemistry are critical for controlling spin-spin interactions and alignment. An additional file containing the crystal data, pre-reported and uploaded to CCDC (Cambridge Crystallographic Data Centre) is attached - composed by Cole C. Sorensen under the supervision of Frank A. Leibfarth. The original works of Data S1 and S2 were composed by Andrew Marquardt under the supervision of Brett M. Savoie.
https://doi.org/10.5061/dryad.44j0zpcr2
Description of the data and file structure
Computational results (Data S1 and S2) and crystal data of M1.
- .p files are binary files formatted using the pickle module, a built-in Python package. These files are optimized for efficient data storage and can be readily accessed using the pickle module. When loaded, they are automatically interpreted as Python dictionaries, providing a structured and convenient means of storing and retrieving data.
- The raw .p files given correspond to each trimer conformer ensemble tacticity analyzed during the computational section of the results. They provide the raw data used to determine Boltzmann distributions of radical chain alignment (as defined by the dot product) and radical separation. Dictionary headings are as follows, with conformer sequentiality conserved across all data sets/lists:
P: the Boltzmann probability of each conformer as calculated from the distribution partition function, organized by decreasing probability
S(x)(y): the separation between the radical COM (what is meant by COM in this context should be defined in the paper text) of pendants (x) and (y), where x and y are integers sequentially given to the pendants in the trimer.
D(x)(y): the dot product between vectors (vector definition used consistent atoms, this is defined in the paper text also) defining the pendants (x) and (y), where x and y are integers sequentially given to the pendants in the trimer. - magnetic_persistence.ipynb is a python notebook (which can be run in jupyter or other .ipynb software) that calculates the magnetic persistence of a given conformer. It takes the .p files described above as input data and analyzes the dot product distribution to determine the expected segment length (in pendant monomers) within the polymer at which pendants are sufficiently aligned that the segment can be considered to exhibit magnetic effects.
- .cif (Crystallographic Information File) is a standard text-based file format used for storing and exchanging crystallographic data. Same file is uploaded and published in the database entries in Cambridge Structural Database (CSD) with a CCDC#: 2324183.
Files and variables
File: boltz_SRR2.p
Description: Data S1 - 1
File: boltz_SRS2.p
Description: Data S1 - 2
File: boltz_SSS2.p
Description: Data S1 - 3
File: magnetic_persistence.ipynb
Description: Data S2
File: 2324183.cif
Description: Crystal Data of M1 / CCDC#: 2324183
Access information
Other publicly accessible locations of the data:
- https://github.com/Savoie-Research-Group/papers
- https://www.ccdc.cam.ac.uk/structures/?gad_source=1&gclid=CjwKCAiA2cu9BhBhEiwAft6IxPrphvEACJ89VGXlIjIXRg3I9OeGBD4EO-1psQuDFgADwbNk11gSphoCCQ8QAvD_BwE
Data was derived from the following sources:
- Conformer-Rotamer Ensemble Sampling Tool (CREST), GFN2-xTB, OpenBabel quantum chemistry file conversion package
Three trimers were analyzed using the Conformer-Rotamer Ensemble Sampling Tool (CREST) and GFN2-xTB semi-empirical potential. The trimers corresponded to the heterotactic (m,r), syndiotactic (r,r), and isotactic (m,m) triads that map to the atactic, syndiotactic, and atactic polymerization scenarios. The universal force field (UFF) in the OpenBabel quantum chemistry file conversion package was used to generate an initial optimized geometry from stereochemically-specific SMILES strings for each triad, which was then optimized using xTB before being subjected to conformational sampling via CREST using the most comprehensive default setting. The initial geometries and outputted conformers were analyzed to ensure no incorrect bond rearrangements or intramolecular radical dimerization had occurred. Separation and alignment analysis was performed on all conformations. Each triad has two pairs of nearest neighbors (i.e., the first and second radicals, and the second and third radicals) whose alignment and separation were parsed and reflected in the histograms. Conformer weights within the distributions were calculated via Boltzmann probabilities using the relative energy calculated by CREST for each conformer.
The alignment autocorrelation decay was calculated for each tacticity using the alignment data from the corresponding triads. A virtual polymerization was performed by Boltzmann sampling (i.e., selecting pairs of alignment angles based on the energy of the corresponding triad conformer) the triad statistics 250,000 times, producing a sequence of 500,000 pendant radical alignments. The alignment decay was calculated from these radical sequences using each radical as an independent origin and averaging the alignment dot products with respect to separation along the sequence. This procedure resulted in 499,900 (i.e., 500,000-100) decay curves that were averaged over to yield. The persistence length was reported as the separation after which the alignment fell to e-1. A Jupyter notebook illustrating these calculations is distributed with this work.
