Benjamin Mallada

🧵 Using nc-AFM, we've visualized #cyclodextrins (CDs) with submolecular resolution— molecules made of glucose units with the remarkable ability to "trap" other molecules. 1/10 (paper in Nat Comm: rdcu.be/dYWOL)

@lidlespana cringe

Fart & Furious. Ricos en prrrrrrrrroteínas.

@OrlandoMejaRiv1 Hacedor de Estrellas (Stapledon). Nunca aparece en las listas españolas, supongo que por las escasísimas veces que se han editado. Prácticamente todos los grandes la han alabado (Clarke, Lem, etc9

No puedo dejar de nombrar otras 5 novelas de ciencia ficción del siglo XX que seguirían luego de mi top 5. Son: El fin de la infancia (Clarke), Limbo (B. Wolfe), Neuromante (Gibson), La quinta cabeza del Cerbero (G. Wolfe) y Maestro Cantor (Scott Card).

@philipturnerar What about the tilt of the surface itself ? It can vary between microscopic regions of the sample. Terraces are flat, but they almost never grow perfectly normal to the tip axis. Even a 0.1° tilt, in 1 um ,will deliver almost two nanometers of height difference

X/Y/Z feedback entirely. I feel the only time PI Z-feedback is ever needed, is bringing the tip to the surface the first time, and suppressing long-term thermal drift. XY feedback would keep the tip still while sitting idle between tip registrations and atom placements. [9/n]

The limiting factor in creepless STM: PI integral gain equals eigenfrequency / Q-factor. So if the scanner resonates at 5 kHz and Q = 100, you can only do interesting things at 50 Hz or slower. Even if pixels are coming out at 1-10 kHz because of the current sensor's speed. [1/n]

@igor_roncevic @ManuelMelleLab Congrats! I saw a bit of it in a talk by Florian and it was quite curious

We did some pretty twisted stuff and Science liked it: science.org/doi/10.1126/sc…

Beautiful result

We found (theoretically) another case of F tunnel! Now, for the same price, including a fluorine halogen bond. "Fluorine Ping-Pong Tunnelling in Cages" New J. Chem. - Festschrift for Giuseppe Resnati doi.org/10.1039/D5NJ03…

@observie The analogy goes very far... for instance, the same mechanical control of an atomic force microscopy sensor, a mechanical fork or cantilever, to keep it in resonance and obtain the interaction between tip and sample, is the same as AM/FM demodulation of radio signals.

Wow, what a great analogy: a transformer is to voltage, current, and power what a gear is to torque, speed, and mechanical power.

Imagine the potential N layered sandwiches of materials wildly differing in their electronic properties...

Congratulations to @HLAGroupOx (!) On surfaces, we have seen cyclocarbon rings created with clever tip-induced strategies (tldr: removing atoms from a precursor) Now…48 carbons stabilized with catenates in solution. phys.org/news/2025-08-c…

Sometimes intuition is enough, but steric repulsion was hard to visualize for me. I used to eyeball it from simulations, and now it’s right there to see...cool work!

@InnaVishik I like to think that strange properties of matter live within the inclusions of random minerals with strange stoichiometries. Perpetually hidden in collections or Earth's crust

Sometimes, at the interface of two different materials, you can get entirely new behavior not found in either material. The ‘classic’ example is LaAlO3 and SrTiO3, two semiconductors. When you put them together, at the interface you get a two-dimensional electron gas (a metal) which becomes superconducting at low temperature (and some magnetism too). There are many other examples of new phenomena at heterointerfaces, and the combinatoric possibilities are vast.

Great read! Mass selection + landing energy control is crucial to avoid any fragments on surfaces that interrupt or hinder interesting reactions on surfaces. But also enables using gigantic precursors or biomolecules into surfaces. Maybe I'll show soon some examples of this...

@kurtilaszlo I didn't know about the potential of N4 at all until I read this post, a cool topic, never heard of it. Congrats for the award!

Honored to be named a 2025 Ross M. Brown Investigator! browninstitute.caltech.edu We’re taking on a grand challenge: building stable, neutral allotropes of nitrogen (N₄) as clean, ultra energy-dense fuels. No combustion & emissions. Just on-demand decomposition to pure hot N₂ gas.

@ZimingLiu11 Cool! The statistical nature of LLMs seems a nice 'esoteric' niche to check for thermodynamics-like laws

Interested in the science of language models but tired of neural scaling laws? Here's a new perspective: our new paper presents neural thermodynamic laws -- thermodynamic concepts and laws naturally emerge in language model training! AI is naturAl, not Artificial, after all.

Maybe the shortest introduction to the Ising model I have ever seen

@PaesaniLab @ChemRxiv Nicely written! Any plan to extend it to their typical alkali 'friends'?

🚨 New paper alert! 🚨 What do halide ions really do to #water? Find out on @ChemRxiv: 👉 doi.org/10.26434/chemr… In our latest study, we use our MB-nrg data-driven many-body potentials to unravel how F⁻, Cl⁻, Br⁻, and I⁻ reshape the structure and dynamics of liquid water — from hydration shells to IR spectra. 🖥️ Fluoride stands out: strong, directional H-bonds, slow water exchange, major spectral shifts. 🖥️ Heavier halides? Minimal disruption, even in the first shell. These insights not only challenge the classic “structure maker/breaker” view but also set a new benchmark for modeling ion hydration at the molecular level — all without empirical fitting. 🏄‍♀️ Big thanks to @NSF for funding and to @ACCESSforCI and @NERSC for computational resources! @UCSanDiego @UCSDPhySci @UCSDChemBiochem @HDSIUCSD @SDSC_UCSD

@philipturnerar Is this also including the almost by 'default' 1 pF capacitance of having a short wire attached to an amplifier? (typical number from Chen and elsewhere)

Electromagnetic interference. That's one of the reasons I felt like giving up on solving piezo creep. In an ideal world, electrical components behave how they would in isolation. In an analog world, they're all coupled to each other and interfere with each other.

These analog systems are always full of ingenuity 'how to embody mechanical computation into 3D printed objects, i.e., without electronic sensors, actuators, or controller'. The video in the thread is really cool.

This paper is mind-blowing. I'll have to read it several times I think, msybe on holiday. It's one of those refreshing speculative ones.

@GeueNiklas @pagelgroup congratulations

Thrilled to hear that the German Research foundation funded my proposal "Elucidating the Fragmentation Pathways of Glycans for Clinical Glycomics"! Looking forward to pursue gas-phase glycoscience as a Walther-Benjamin fellow at FU Berlin, mentored by @pagelgroup .