FUNPOLYMER · Diffusion NMR and artificial intelligence
Molecular weight prediction in any solvent and at any concentration*
DiffAtOnce is a platform developed within the FUNPOLYMER project to process diffusion NMR experiments and estimate polymer molecular-weight distributions. It combines Arrabal–Fernández scaling, universal calibration, advanced numerical algorithms, inverse Laplace transforms and artificial intelligence.
* Solvent and concentration independence through a calibrated model. The 2025 paper studies linear, low-dispersity PS and PPG and concentrations from 1.5 to 150 mg/mL. Extrapolation to other systems requires reviewing calibration and applicability. Paper and conditions ↗
FUNPOLYMER addresses fast, traceable determination of molecular-weight distributions of polymers in solution by diffusion NMR, as an alternative to classical size-exclusion chromatography (SEC/GPC). Its central advance is to incorporate solvent and concentration dependence through the universal \( \kappa(M_w) \) and \( D\eta_{1/\infty}(M_w) \) curves.
Reduce measurement time and solvent consumption, aiming to maintain or improve accuracy against SEC/GPC and move beyond the requirement to always work at very high dilution.
The objective is molecular-weight estimation independent of solvent and sample concentration, within the model’s applicability domain.
Integrate advanced ILT algorithms, universal calibration (UCC) and the \( \kappa(M_w) \) and \( D\eta_{1/\infty}(M_w) \) curves, alongside extended diffusion NMR (ediffNMR) and artificial intelligence.
One platform connecting experimental decays with diffusion and molecular-weight distributions that can be reviewed.
DiffAtOnce: a research platform built around Arrabal–Fernández scaling and the universal curves developed within FUNPOLYMER.
This scientific foundation now supports five independent products: DiffAtOnce Legacy, DiffAtOnce DM, DiffAtOnce Continuum, DiffAtOnce Prime and ResinAtOnce.
Discover FUNPOLYMER · Methodological paper · European Polymer Journal (2025) · Equations and methodology
Numerical, physical and AI tools for the complete workflow, from diffusion acquisition to molecular-weight distributions and comparison with SEC/GPC. The model incorporates solvent and concentration effects instead of restricting analysis to a single experimental condition.
A library of inverse Laplace transform methods tailored to diffusion NMR: discrete and continuous approaches with advanced regularisation.
Method, grid and regularisation choices are part of the interpretation; fits and residuals should accompany the distributions.
Viscosity is treated as an explicit parameter. UCC and the \( \kappa(M_w) \) and \( D\eta_{1/\infty}(M_w) \) curves connect diffusion with molecular weight.
The scientific programme considers families including PS, PPG, PMMA, PE, dextran and polyisoprene. Calibration and supporting evidence are reviewed for each family; they do not all automatically share the same validation domain.
Deep learning and high-performance computing to accelerate large diffusion-NMR datasets and explore ultrafast and time-resolved methodologies.
Beatrix adds contextual help, scientific rules and documentary consultation in ResinAtOnce. These assistance functions are distinct from AI models dedicated to numerical reconstruction.
Developed at the University of Almería within FUNPOLYMER (PID2021-126445OB-I00), in collaboration with the Institute of Physical Chemistry, Polish Academy of Sciences, Nuclear Hyperpolarization of Molecular Systems and Nanomaterials group.
The collaboration described by the project brings together Dr Tomasz Ratajczyk (group leader), Dr Mateusz Urbańczyk (main contact) and PhD student Marek Czarnota.
A shared purpose: making diffusion NMR more accessible, reproducible and interpretable.
Team and collaborations · Publications and scientific outputs
THE DIFFATONCE FAMILY
Choose the focus that fits your system and research workflow. ResinAtOnce does not replace the platform: it is one of its applications.
Learn diffusion by doing
Training · Web and Windows
Bilingual course with objectives, causes and procedures, two-paper derivations, examples, exercises and diffusion NMR simulations. Local Beatrix in the WPF download.
Open Academy or download WPF with Beatrix ↗ES / EN · 1.2.0The original idea, renewed
Classic diffusion · Windows
The original workflow with peak selection, attenuation, D, Rh, shape, microfriction and apparent molecular mass. Light and dark themes, ES/EN help, PDF, PostScript, EPS and Excel reports.
Download Legacy and read the guide ↗Windows x64 · 2.9.1Discrete molecules
Discrete molecules
A product line for discrete molecular systems and interpretation of their diffusion coefficients.
Product line introduced · distribution to be announcedPolymers & macromolecules
Polymers and macromolecules
A product line focused on continuous distributions, dispersity and diffusion-NMR characterisation of polymers and macromolecules.
Product line introduced · distribution to be announcedAdvanced diffusion
General-purpose · Windows / WPF
Advanced diffusion-NMR workspace: DOSY/MwOSY, attenuations, calibration, distributions, alignment and Beatrix. Experimental screenshots, guide and FAQ.
See Prime, screenshots and help ↗Windows · 0.14.1 · researchResin analysis
Resins · independent application
The dedicated resin application: spectra, ramps, internal standards, components, Beatrix and traceable results.
See application and screenshots ↗Windows · research versionPrime and ResinAtOnce are presented with their user documentation. DM and Continuum remain introduced product lines. Legacy provides a Windows download with per-installation access approval.
Software, code and availability →NEW TOUR / PRIME 0.14.1
DOSY, MwOSY, attenuations, standards and Beatrix. Current experimental screenshots, ultrafast D–T₂ with TRAIn2D/FISTA, SPEN sources, a complete guide and FAQ.
Explore Prime →User guide →
DIFFATONCE LEGACY · 2.9.1
The spectrum at the heart of your work: track resonances, review ln(I/I₀), calculate D and Rh and explore shape and apparent mass. Probes and calibration, traceable projects and reports in a Windows WPF application.
FROM EXPERIMENT TO REPORT
Spectra, ramps and acquisition parameters.
Probe, sequence, gradient, temperature and standards.
ILT methods, distributions and molecular models.
Results, decisions and traceable reports.
INSIDE THE APPLICATION
A look at the real interface, from signal selection to interpretation. Select a screenshot to enlarge it.
DEMO · v0.50.1 Real interface with simulated data for this demonstration. These are not experiments or validation results.
Inspect the spectrum in ppm, mark signals and review internal standards before calculation.
Review attenuation, fit, diffusion and apparent molecular-weight distributions, alongside the results table.
Explore the map, decompose the projection and review polymer and resin-oriented suggestions, with limitations and sources.
Beatrix guides; the researcher decides. Suggestions do not replace scientific validation or change parameters automatically.
Complete help: 16 chapters with screenshots → Explore the full workflow →WATCH, PAUSE, REPEAT
Six workflows with separately narrated Spanish and English movies, in the application and on the web. Page language selects the voice, interface and timeline; CC offers ES/EN captions. Chapters and downloadable steps. No autoplay or external providers.
DIFFATONCE LEGACY · WEB / WPF
Spectra and local maxima, calibration and R², diffusion, shape and reports. Six guides with narration, captions and chapters.
DIFFUSION MOTION LAB · ES/EN
16 interactive scenes and 32 videos to explore diffusion NMR: calibration, gradients, ILT inversion, ultrafast and SPEN-DOSY. Change the parameters and see what happens.
Open the animation laboratory →DIFFATONCE / MOLECULAR DIFFUSION
