Sign in

Michael Marthaler 🇪🇺

@michaelmarthaler.bsky.social
61 followers 30 following 390 posts

Spectroscopy - NMR Spectroscopy - Quantumcomputing hqspectrum.cloud.quantumsimulations.de

PostsRepliesMedia
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 22h
The key test: can a quantum computer simulate a full pulse sequence where classical methods no longer provide the needed accuracy, speed, or scalability? I believe yes. We’ll soon share spin Hamiltonians so others can test classical solvers.
000
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 22h
A genuine advantage requires more than replacing one solver: accurate molecular-to-Hamiltonian mapping, an experimental NMR workflow, suitable time-evolution algorithms, rigorous classical baselines, and hardware within error and coherence limits.
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 22h
TOCSY is especially interesting because its transfer pathways probe correlations across a spin system. COSY could also benefit when dense coupling networks become difficult for classical simulation, such as sterane spin-connectivity graphs.
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 22h
What would quantum advantage in NMR look like? A promising target is simulating pulse sequences—not just static spectra. The goal is to capture full experimental dynamics in chemically relevant spin systems.
What would quantum advantage in NMR look like? A promising target is simulating pulse sequences—not just static spectra. The goal is to capture full experimental dynamics in chemically relevant spin systems.
110
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 09/10/2026
We now call “Radicals” Electron Spectroscopy, reflecting our focus on singlet–triplet splittings. The use case remains strong, but quantum advantage will require disciplined engineering, rigorous benchmarks, and updated assumptions. www.youtube.com/watch?v=zJdlp0Rb3cE
000
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 09/10/2026
For multireference chemistry, we benchmarked the Active Space Finder for excited-state calculations. Active-space selection is a key friction point for robust, automatable workflows; the paper clarifies what works and where limitations remain. arxiv.org/abs/2511.05732
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 09/10/2026
For NMR, our detailed classical benchmark shows how far state-of-the-art methods can go. It remains promising: the mapping is natural and the dynamics are genuinely quantum. The open question is where quantum computers are truly needed. github.com/HQSquantumsimulations/hq…
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 09/10/2026
Update on our ITBQ framework and use-case classification paper, “What is a good use case for quantum computers?” The three use cases remain strong candidates—but benchmarking is essential to distinguish promising ideas from genuinely advantageous ones. quantumsimulations.de/paper-use-cas…
Update on our ITBQ framework and use-case classification paper, “What is a good use case for quantum computers?” The three use cases remain strong candidates—but benchmarking is essential to distinguish promising ideas from genuinely advantageous ones.
https://quantumsimulations.de/paper-use-cases
110
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 08/10/2026
The RPA-based method doesn’t yet show quantum advantage: its classical implementation scales quartically. But it achieves respectable accuracy for 14 singlet states, with TD-DFT-like timings, and offers a promising first step.
010
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 08/10/2026
We retain strongly coupled bosonic modes as oscillators coupled to the active space, while tracing out the rest. These approximations make the model suitable for available quantum hardware.
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 08/10/2026
For UV/Vis spectroscopy, we need molecular excited states. Our approach starts from Hartree–Fock, keeps a minimal HOMO–LUMO active space, and represents remaining orbital interactions as bosonic excitations, reducing model complexity.
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 08/10/2026
Can a model designed for a quantum computer compete with classical approaches in quantum chemistry? At HQS, we’re exploring that through quantum utility: useful, reliable results—even before quantum advantage.
Can a model designed for a quantum computer compete with classical approaches in quantum chemistry? At HQS, we’re exploring that through quantum utility: useful, reliable results—even before quantum advantage.
110
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 07/10/2026
The clearest distinction is the ~1.3 ppm multiplet from butanol’s extra methylene group. A 500 MHz ¹H NMR spectrum tracks the shift from pure propanol to pure butanol across concentration ratios.
010
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 07/10/2026
A simple propanol–butanol mixture in D₂O shows a key challenge: corresponding signals overlap because similar nuclei in both compounds have very similar chemical shifts.
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 07/10/2026
Its applications span industries—from verifying drug quality to improving food safety and supporting biological analysis.
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 07/10/2026
qNMR turns NMR signal integrals into concentration measurements: because integrals are proportional to the number of contributing nuclei, it can quantify components in mixtures.
130
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 06/10/2026
Overlapping signals make mixture analysis more difficult, especially in ¹H NMR. Solvent, temperature, ion concentration, acid–base equilibria, and other reactions can also shift signals and affect quantitative analysis.
000
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 06/10/2026
Although ethanol and 2-propanol share physical properties, their characteristic NMR signals can help identify each component. Spectral regions may be readily assigned when signals are separated.
110
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 06/10/2026
This matters for identifying and quantifying alcohols used in disinfectants and hand sanitizers. Incorrect compositions or concentrations can affect antimicrobial effectiveness, safety, and performance.
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 06/10/2026
NMR spectroscopy can distinguish ethanol (CH₃CH₂OH) from 2-propanol ((CH₃)₂CHOH) in mixtures through their distinct proton environments, chemical shifts, and coupling patterns.
NMR spectroscopy can distinguish ethanol (CH₃CH₂OH) from 2-propanol ((CH₃)₂CHOH) in mixtures through their distinct proton environments, chemical shifts, and coupling patterns.
131
Reposted by Michael Marthaler 🇪🇺
HQS Quantum Simulations @hqsquantum.bsky.social · 06/10/2026
Higher-order OTOCs carry information beyond the standard NMR spectrum: k=1/2 order exactly reproduces the spectral function, each successive order generates new peaks at all pairwise and higher-order combinations of the transition frequencies! Register here: quantumsimulations.de/summit-26
032
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 05/10/2026
The pipeline: 2D/SMILES → 3D structures → conformers → quantum chemistry and weights → NMR shieldings/J-couplings → spin Hamiltonian → spectra. Mixtures combine spectra and fit experiment, making NMR benchmarkable and industrially relevant.
010
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 05/10/2026
NMR is compelling: one workflow supports structure determination and mixture analysis. Psoralen and Angelicin share a molecular formula yet have clearly distinct spectra; Bakuchicin is simulated-only in the figure.
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 05/10/2026
Our paper, “What is a good use case for quantum computers?” introduces ITBQ: Identify an industry problem, Transform it into a quantum-ready formulation, Benchmark with strong classical methods, then Show Quantum Advantage. arXiv:2506.15426v2
110
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 05/10/2026
Many “quantum use cases” fail before hardware limits: the workflow from real industrial input to a quantum-ready Hamiltonian—and a fair classical benchmark—is missing.
Many “quantum use cases” fail before hardware limits: the workflow from real industrial input to a quantum-ready Hamiltonian—and a fair classical benchmark—is missing.
141
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 04/10/2026
This quick NMR lesson refreshes the concept—and offers a broader reminder: difficult problems sometimes need a change in perspective. Want the essence without the derivation? Jump to the last slide.
000
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 04/10/2026
These problems can involve widely separated time scales, creating challenges for numerical simulation. Using the interaction picture properly helps overcome them.
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 04/10/2026
At HQS, we use the interaction picture to simulate effects such as other isotopes on spectra measured in proton NMR experiments.
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 04/10/2026
Remember the interaction picture from graduate quantum mechanics? Often seen as advanced or intimidating, it is actually a powerful way to simplify important problems in particle physics, nuclear magnetic resonance, and beyond.
120
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 03/10/2026
TPPO has too many hydrogen nuclei for naive time evolution. HQS is developing tools to study heteronuclear decoupling in challenging molecules like TPPO—important for interpreting experiments. Stay tuned!
010
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 03/10/2026
At 80 MHz, TPPO’s spectrum looks completely different when phosphorus couplings are artificially removed from the molecular Hamiltonian. But what happens with a real decoupling sequence?
110
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 03/10/2026
In our NMR Benchmarking paper [link to the NMR Benchmarking paper], TPPO’s high symmetry and strong coupling to phosphorus require special care to simulate correctly.
100
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 03/10/2026
What happens when heteronuclear coupling is so strong that it completely reshapes a proton NMR spectrum? Triphenylphosphine oxide (TPPO) offers an unusually challenging example.
What happens when heteronuclear coupling is so strong that it completely reshapes a proton NMR spectrum? Triphenylphosphine oxide (TPPO) offers an unusually challenging example.
141
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 02/10/2026
I am sorry for the delay. I have a life. Anne Fausto-Sterling is a Sexologist and not a Biologist. When writing the paper she "had intended to be provocative, but I had also written with tongue firmly in cheek". Her own words from here: www.researchgate.net/publication/...
researchgate.net
(PDF) The Five Sexes, Revisited
PDF | On Jul 31, 2013, Anne Fausto‐sterling published The Five Sexes, Revisited | Find, read and cite all the research you need on ResearchGate
100
Reposted by Michael Marthaler 🇪🇺
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 29/07/2026
the best comic strip I have read in quite while
021
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 02/10/2026
Applicable to all radical types and correlated methods, this is a direct use of HQS Spin Mapper’s Spin Finder. Read “Understanding Radicals via Orbital Parities”: arxiv.org/abs/2404.18787 May the parity be local.
000
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 02/10/2026
It also produces spin-like orbitals that visualize excess spin. For disjoint (zwitterionic) radicals, they localize automatically, revealing whether spin centers are separable or intrinsically entangled.
110
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 02/10/2026
Built from the one- and two-electron reduced density matrices (1-RDM and 2-RDM), parity analysis gives a global, quantitative measure of radical character—without ad hoc assumptions.
110
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 02/10/2026
There is no chemistry without radicals. But how do we distinguish a radical from a closed-shell molecule when both may have singlet wave functions? HQS introduces orbital parity analysis to classify radical character.
There is no chemistry without radicals. But how do we distinguish a radical from a closed-shell molecule when both may have singlet wave functions? HQS introduces orbital parity analysis to classify radical character.
141
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 01/10/2026
FYI, about "not producing gametes", did you actually read what I wrote? Of course we can always say around which Gamete production a body is organized, even if the production is not happening.
110
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 01/10/2026
Is your claim of the existence of 5 sexes based on this amazing piece of performance art: nyaspubs.onlinelibrary.wiley.com/doi/10.1002/... ?
200
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 01/10/2026
Actually, the medical term for intersex is DSD (Disorders of Sex Development). It is used because there are only two sexes, defined by the size of the gametes (small or large) that the body is built to produce. And yes, this also applies even if no gamete production occurs.
130
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 01/10/2026
HQS post-processes conformer ensembles to recover degenerate structures without user input. The resulting simulation reproduces recognizable multiplets, though a small methyl artifact remains. Experimental data: Merck KGaA, Darmstadt, Germany. What is your NMR experience?
020
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 01/10/2026
Modern workflows sample conformers using tools such as CREST and ORCA. Propofol’s ensemble is highly degenerate: methyl rotations and OH flips alone create a factor of 162. Grouping rotamers improves the simulation, but incomplete ensembles remain a challenge.
120
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 01/10/2026
Optimizing one propofol geometry and calculating its NMR parameters produces a poor simulation. Molecules rapidly interconvert between conformations during an NMR experiment, so the observed parameters are averages over those conformations.
110
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 01/10/2026
How can we calculate NMR parameters to simulate an organic molecule’s spectrum? Propofol looks simple, but predicting its chemical shifts and J-couplings accurately is not. Quantum chemistry alone is only the starting point.
How can we calculate NMR parameters to simulate an organic molecule’s spectrum? Propofol looks simple, but predicting its chemical shifts and J-couplings accurately is not. Quantum chemistry alone is only the starting point.
141
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 01/10/2026
That's the type of post that definitely will play well on Bluesky!
020
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 01/10/2026
You would assume that if Erin uses AI, the article would be more accurate.
020
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 30/09/2026
The workflow scales to Friedelin (50 spins) and is especially useful on benchtop NMR, where overlap makes peak-by-peak tweaking ambiguous. Use HQSpectrum Web UI or Python via HQS Spectrum Tools (HQStage). Data: KIT; thanks Sören Lehmkuhl.
010
Michael Marthaler 🇪🇺 @michaelmarthaler.bsky.social · 30/09/2026
3) If needed, refine individual chemical shifts for solvent, concentration, temperature, and remaining model bias—while keeping the predicted coupling pattern. Recalculate the spectrum; this is more than sliding peaks. Automatic, semi-automatic, or manual.
110