The NSF MRSEC program - end of an era?

Are we witnessing the irreversible deconstruction of an historically successful NSF program, not because of any clear strategy or planning but instead because of the cumulative impact of many forces?

The National Science Foundation has historically supported a mixture of individual investigator (or small team) funding opportunities and large center grants.  The center programs are meant to bring together collaborative teams of researchers to tackle sets of research questions that require a larger scale approach - a hugely emphasized review criterion of those centers is always the question, "Is the proposed work really a center, in the sense of being a coherent effort larger than the sum of its parts, or does it instead read like a collection of loosely connected individual projects?"  Center programs are also a way that NSF has supported Research Experience for Undergraduates programs, providing crucial gateways into real science and engineering research and training for hundreds of students per year.  Similarly, center programs have contributed toward building up networks of facilities with specialized research infrastructure and capabilities.  Apart from topical centers that come and go (e.g., the nano centers in the early 2000s, the quantum centers now), center programs include Engineering Research Centers, Science and Technology Centers, and the less applied Physics Frontier Centers (Division of Physics), Centers for Chemical Innovation (Division of Chemistry), and the flagship center program of the Division of Materials Research, the Materials Research Science and Engineering Centers (MRSECs).

The MRSEC program has such a long history (going back to the 1960s) that it has a reasonably good wikipedia page telling the story of its origins and evolution.  In brief:  There are typically around 20 MRSECs at any one time since the program was shifted into its present form 31 years ago.  Each center award is for six years, renewable.  There is a national competition every three years, during which half of the existing centers are up for renewal (though depending on reports by visiting committees, some centers may be recommended not to submit a renewal proposal).  Research in MRSECs is organized into "Interdisciplinary Research Groups" (IRGs), as well as some seed projects.  Once upon a time, a MRSEC could have as many as five IRGs.  Over the last 15 years or so, NSF budgets have largely been flat, and DMR has tried to balance the demands of individual investigator grants vs. center grants, along with knowing that underfunding a project can be worse than not funding it at all.  As a result, the number of funded IRGs has decreased steadily, so that a typical MRSEC now has two or three IRGs (usually two).  

I won't bore you with details about how the funding competitions work.  The short version:  universities send in preproposals consisting of draft IRGs + the other center components (facilities, education, management, etc.); IRGs are reviewed in panels, and then universities are invited or not to the full proposal stage.  Suffice it to say, every three years there is a national competition that consumes many many hours of effort.  Landing a MRSEC is both a sponsored research award and a point of pride.  Typically maybe 80% of the cohort up for renewal make it, with new starts making up the balance, so that there is some rotation among institutions.   Like all NSF programs, the peer review efforts are done for free by the community.

The MRSEC program has had a big impact over the years, at minimum in the number of people trained and supported through these efforts.  The National Research Council/NAS did a study of the MRSEC program back in 2007 that is publicly available here, if you're interested.  No program is perfect, and I don't want to argue about whether the balance of, e.g., reporting requirements vs. research dollars is right, etc.  However, I think the large majority of materials researchers in the US would say that the MRSEC program is a mainstay that has been a key pipeline of people into the field, both in academia and industry.

This year, I worry that we are watching likely irreversible harm to the program, and not by the voluntary choice of anyone at NSF.   The budgetary uncertainty is crushing right now.  It is unclear when and how the agency will be making awards, and how much funding they will have (since you can't actually plan based on the possibility of a continuing resolution in the absence of actual budget bills that can pass congress).  As a result, after the preproposal phase in the current competition, NSF revised their guidance, so that instead of the typical "8-10 awards" expected (see here), now they say to expect "2-5 awards".  That's because they're doing contingency planning assuming a cut in the program from $27M to $15M.  This means that there is a real possibility that the total number of existing MRSECs could be cut by 40% at a stroke, and the next cycle of the competition will be due to start in 2028, with little reason to think that budgets will be any better or smoother by then. 

There will almost certainly never be a return to "normal", for multiple reasons, including the general evolution of all funding programs with time.  The end result of the current shakeup may also have some positive outcomes in terms of new approaches.  That said, it sure feels like paths are being set by circumstances, not considered choice.  I would say that the path of this program is a question that should be addressed by the NSF Math and Physical Sciences advisory committee, but of course that was disbanded back in April, along with 11 others.  You might imagine asking the National Academies for thoughts on this, but I gather anecdotally that is not happening much at all anymore either.  

I'm writing because I hope someone more influential than me can report on this.  At a time when "materials" are clearly of major importance to US competitiveness (e.g., they are clearly relevant to multiple priority areas of the Genesis Mission), is anyone thinking about the impact of the trajectory we are on here?

(Back to science soon, hopefully.)

Taking stock: some federal science news

Some general science news:

•  The New York Times ran an interactive article this week that shows what we all know.  This past year was a very bizarre funding environment.  The article focuses on NIH and NSF, but the major points are generalizable.  The combination of circumstances (DOGE, general administrative turmoil, uncertainty and legal cases about indirect costs, the lack of a real budget followed by a late continuing resolution, plus the government shutdown and continued lack of real budgets) has been extremely disruptive, resulting unquestionably in less science and engineering research being funded by the US government than in many years.  
• Conversations I've had with program officers at two agencies have conveyed that everyone thinks it is very likely that there will be another shutdown in January, when the present spending authority expires.  To put that another way, there is very little confidence that actual spending bills appropriating real budgets for NSF, DOE, NIH, etc. will pass the House and Senate, with some reconciled conference version getting filibuster-proof support in the latter, before then.  This uncertainty means that right now it's going to be nearly impossible for the NSF, for example, to make much in the way of awards in the meantime, since they have no budget and can't plan on a year-long continuing resolution.  
• There has been an executive order announcing the Genesis Mission, which is going to be a large federal AI+science project.  The goal is to "accelerate the AI and quantum computing revolution and to double the productivity and impact of American science and engineering within a decade", according to undersecretary of energy Dario Gil.  Broadly, the plan is to have AI/ML agents developed (presumably by private contractors or private/public partnerships) and trained on vast datasets (ones already in existence in, e.g., national labs and public repositories).  At the same time, a list of Grand Challenges will be defined (within the next 60 days), with the idea that these AI agents will be used to address these (and demonstrating application of the AI "Platform" toward at least one challenge within 270 days).  Any stated support for science and engineering research is welcome.  I hope that this ends up bearing fruit in terms of real research advances, and that university researchers can contribute effectively. (I worry about a framework for massive taxpayer-funded financial support of for-profit AI companies, privatizing financial/IP benefits from publically funded datasets.  Of course, I worry about a lot of things.  Ask anyone who knows me.).   Ideas about grand challenges would be fun to discuss in the comments.   
• We had a great physics colloquium this week from Steve Fetter at the University of Maryland about the continuing threat of nuclear weapons.  Very sobering.  One fact that I gleaned:  In terms of missile defense, the Next Generation Interceptor is likely to cost $660M per interceptor.   That is something like 50 times the cost of a Russian ICBM.  Something else to bear in mind:  The Houston Food Bank, one of the largest and most effective in the US, has an annual budget of about $64M.  The amount of resources consumed by nuclear arms since 1945 is just staggering.

What is the orbital Hall effect?

In the course of thinking about how best to revise my too-math-infused post about quantum geometry, I realized that writing about the orbital Hall effect lays nice groundwork.  

I've previously written about the spin Hall effect (SHE), in which a charge current \(\mathbf{j}_{\mathrm{c}}\) directed along \(\hat{\mathbf{x}}\) generates a net flow of \(\hat{\mathbf{y}}\)-directed spin angular momentum along the \(\hat{\mathbf{z}}\) direction.  This is a consequence of spin-orbit coupling, and it was first predicted in 1971 with a major revival sparked in 1999.  Electrically generating angular momentum currents has proven very useful, leading to many ideas about magnetic memory devices.  Microscopically, it's not easy to develop an intuition about the SHE, though as a spin-orbit effect, it is expected to be much stronger in heavier metals, since the spin-orbit coupling in atomic orbitals scales like \(Z^{4}\), and electronic bands in solids are built from those orbitals.  

That fact, that the electronic bands originate from atomic orbitals, is something that can get lost in a Bloch wave/nearly-free electron treatment of electronic structure.  In the orbital Hall effect, this idea is paramount.  This was explained clearly in this PRL (arXiv here).  The little \(p\)-orbitals are drawn on top of the \(k_{x}-k_{y}\) plane, to illustrate the idea that the electronic states in \(\mathbf{k}\)-space have different orbital content, depending on \(\mathbf{k}\).   The blue circle represents the "Fermi disk", with \(\mathbf{k}\)-states inside the circle occupied, and \(\mathbf{k}\)-states outside the circle empty.  

Adapted from Fig. 1 here.

When no electric field is applied, the Fermi disk is centered on \(\mathbf{k} = 0\); there is no net current, and there is no net orbital angular momentum once all the filled states are considered.  When an electric field is applied in the \(+x\) direction, though, the Fermi disk is shifted away from the origin in the \(-x\) direction (because of our convention that electrons are negatively charged).  Now adding up the \(z\)-directed orbital angular momentum contained within the Fermi disk, there is net \(+z\) orbital angular momentum carried by states with positive \(k_{y}\), and net \(-z\) orbital angular momentum carried by states with negative \(k_{y}\).  So, for this orbital texture, a charge current \(\mathbf{j}_{\mathrm{c}}\) directed along \(+\hat{\mathbf{x}}\) generates a net flow of \(\hat{\mathbf{z}}\)-directed orbital angular momentum along the \(+\hat{\mathbf{y}}\) direction.  Charge current generates a transverse flow of orbital angular momentum, entirely due to the way atomic orbitals come together to make Bloch states in \(\mathbf{k}\)-space, independent of any spin-orbit physics.  That's why the orbital Hall effect has been inferred experimentally in several materials with weak spin-orbit effects, like chromium and titanium.

These effects can be large, and orbital Hall physics plus some \(\mathbf{L}\cdot\mathbf{S}\) coupling may be responsible for some of the results labeled as spin Hall.  See here for a discussion.  Electrically pumping around angular momentum through orbital and spin Hall effects, and their inverses, is the idea behind a variety of device concepts for memory (e.g. here) and logic.  Fun stuff.

Quantum geometry - some intuition

There has been a great growing interest in quantum geometry in recent years.  Last week, I heard an excellent talk by Raquel Queiroz about this that gave me a more physically intuitive interpretation  of this topic.  The more formal write-up is in this preprint from this past April, which I'd missed at the time.

Caution:  Math incoming.  I will try to give a more physical picture at the end.  I know that this won't be very readable to non-experts.    

As I've written before,  (e.g. here and a bit here), the electronic states in crystalline solids are often written as Bloch waves of the form \(u_{n\mathbf{k}}(\mathbf{r})\exp(i \mathbf{k}\cdot \mathbf{r})\), where \(u_{n\mathbf{k}}(\mathbf{r})\) is periodic in the spatial period of the crystal lattice.  For many years, the \(\mathbf{k}\) dependence of \(u_{n\mathbf{k}}(\mathbf{r})\) was comparatively neglected, but now it is broadly appreciated that this is the root of all kinds of interesting physics, including the anomalous Hall effect and its quantum version.  

We can compute how much \(u_{n\mathbf{k}}(\mathbf{r})\) changes with \(\mathbf{k}\).  The Berry connection is related to the phase angle racked up by moving around in \(\mathbf{k}\), and it's given by \( \mathbf{A}(\mathbf{k}) = i \langle u_{n\mathbf{k}}| \nabla_{\mathbf{k}}| u_{n\mathbf{k}} \rangle \).  One can define \(\mathbf{\Omega} \equiv \nabla \times \mathbf{A}(\mathbf{k})\) as the Berry curvature, and the "anomalous velocity" is given by \(-\dot{\mathbf{k}}\times \mathbf{\Omega}\).  

If we worry about possible changes in the magnitude as well, and \( |\langle u_{n\mathbf{k}}| u_{n\mathbf{k+dk}} \rangle |^{2} = 1 - g^{n}_{\mu \nu}dk_{\mu}dk_{\nu}\) plus higher order terms.  The quantity \(g^{n}_{\mu \nu}\) is the quantum metric, and it can be written in terms of dipole operators:  \(g^{n}_{\mu \nu}= \sum_{m\ne n}\langle u_{n,\mathbf{k}}|\hat{r}_{\mu}|u_{m \mathbf{k}}\rangle \langle u_{m,\mathbf{k}}|\hat{r}_{\nu}|u_{n \mathbf{k}}\rangle\).  The quantum metric quantifies the "distance between" the Bloch states as one moves around in \(\mathbf{k}\).  

That last bit is what I really learned from the talk.  Basically, if you try to consider electrons localized to a particular lattice site in real space, this can require figuring in states in multiple bands, and the matrix elements involve dipole operators.  The quantum geometric tensor \(g_{\mu \nu}\) quantifies the dipole fluctuations in the electronic density.  You can define a lengthscale \(\ell_{g}\equiv \sqrt{\mathrm{Tr} g}\), and this can tell you about the spatial scale of polarization fluctuations relative to, e.g., the lattice spacing.  Metals will have essentially divergent fluctuation lengthscales, while insulators have nicely bound charges (that give peaks in the optical conductivity at finite frequency).   The quantum geometry then influences all kinds of experimentally measurable quantities (see here).  

Neat stuff.  Someday I'd like to return to this with a nice cartoon/animation/presentation for non-experts.  The idea that there is so much richness within even relatively "boring" materials still amazes me.

Vortices everywhere

The 2026 APS Oliver E. Buckley Prize in condensed matter physics was announced this week, and it's a really interesting combination of topics that, to a lay person, may seem to be completely unrelated.  

Fig. 1 from this follow-up PRB.

On the one hand, John Reppy (at age 94!) and Dave Bishop were honored for their work examining the properties of vortices in thin films of superfluid helium-4.  Relevant papers include this one from 1977, where they used a torsion pendulum coated with the helium film to examine the transition between normal and superfluid.  When the helium becomes a superfluid, it has (at low speeds) no viscosity, so it no longer has to rotate with the torsion pendulum; this means the rotational moment of inertia goes from that of (pendulum+helium) to just (pendulum), and the period of the oscillations increases.  Really detailed measurements of the oscillations and their damping allowed Reppy and Bishop to compare with models of the superfluid transition based on work by Kosterlitz and Thouless (and Berezinskii).  See the image for a diagram of the experimental setup - very clever and intricate.  

The key idea here is the role of vortices.  Superfluidity in helium is described by an order parameter that looks like a wavefunction - it has an amplitude, \(\Psi_{0}\), and a phase \(\phi\), so that \(\Psi(\mathbf{r}) = \Psi_{0} \exp(i \phi)\).   That order parameter is supposed to be single-valued, meaning if you go around a closed loop of some kind, that phase will either remain the same or ramp by some integer multiple of \(2\pi\).  The gradient of the phase is related to the velocity of the superfluid, so if the phase winds by \(2\pi\), that implies there is a circulation of flow and orbital angular momentum that has to be an integer multiple of \(\hbar\).  In the BKT theory, the demise of the superfluid phase as the system is warmed happens through the creation and unbinding of vortex-antivortex pairs.

On the other hand, the other recipients of the Buckley Prize were Gwendal Fève and Mike Manfra for their work (experiments here and here) regarding the braiding statistics of anyons in fractional quantum Hall systems.  I'd written about anyons here.  For electrons in 2D, the wavefunctions of excitations of the fractional quantum Hall system look like vortices.  The phase of the electronic wavefunction can wind due to circulation, and because electrons are charged, the phase can also wind due to magnetic flux attached to the little whirlpool.  It's the combination of these phase effects that can lead to those excitations acting like anyons (so that when two are physically swapped or braided around one another, the wavefunction picks up a phase factor that is not just the \(+1\) of bosons or the \(-1\) of fermions).  

As my friend Dan Arovas pointed out, there was a hope back in the early 1980s that perhaps vortices in superfluid helium would also act like anyons and have fractional statistics.  However, this paper by Haldane and Wu disproved that possibility.  

Vortex shedding, from here.

Because of the relationship between quantum phase winding and actual flow of (density) currents, vortices show up in lots of places in hard condensed matter physics.  Classical vortices are also physically nontrivial objects - they're topological and often seem to have very counterintuitive properties and motions.  Heck, Lord Kelvin was so taken by this that he thought (pre-quantum) that maybe everything is really vortices of some kind.  

Perhaps it is fitting that I am posting this on the 85th anniversary of the Tacoma Narrows bridge collapse.  That classic civil engineering failure was caused by vortex shedding by the bridge coupling to its torsional resonance frequency.  Vortices can have big consequences!  

Science journalism - dark times

At this point it's old hat to decry the problems facing traditional news media.  Still, it is abundantly clear in our late stage capitalist society that there has been a collective business decision over the last 20+ years that, like local newspapers and television news, real science journalism is not a money maker.   Just a few examples:  Seventeen years ago, CNN cut its entire science, technology and environment reporting team.  In 2022, Popular Science ceased publication.  In 2023, National Geographic laid off their staff writers.  Last week, the Wall Street Journal laid off their science and health reporters.  

I have it on good authority that there is now only one science reporter left at the WSJ.  One, at a time when science and technology are more critically important to our rapidly changing society than ever, and there is enormous tumult in the US and elsewhere about how science is or is not supported and is or is not factoring into policy decisions.  All of this is happening at a time when public trust in science is falling.  (Check out this from Science Friday.)  

(updated for context) Leaving aside professional science outlets (the news sections of Science, Nature, and society publications like Physics Today, C&EN, Physics World, Chemistry World), there are some good publications out there, like Quanta and Nautilus (both founded by nonprofits). There are outstanding public writers of science, like Philip Ball, Helen Czerski, Katie Mack, Ethan Siegel, and many others (apologies for the incompleteness of this list).  There are some excellent freelance journalists.  The internet also means that there are many opportunities for great engagement.  For example, the videos from 3blue1brown are uniformly outstanding.  However, there are no filters, and the temptation to be click-baity or sensationalistic is problematic.  

I have no solutions to offer, except that I encourage you to support good science journalism and reporting when you see it.  It's important.

Interesting preprints: chirality-induced spin selectivity + quantum gravity

This continues to be a very busy time, but I wanted to point out two preprints that caught my eye this week.  Their subjects are completely disparate, but they stand out as essentially reviews written in a much more conversational tone than the usual literature.

The first is this preprint about chirality-induced spin selectivity, a subject that I've mentioned before on this blog.  There is now an extensive body of evidence (of varying quality) that there is a connection between structural chirality of molecules and their interactions with the spin angular momentum of electrons.  This includes monolayers of chiral molecules leading to net spin polarization of photoemitted electrons (here), a lot of electronic transport experiments involving chiral molecules and magnetic electrodes that seem to show spin-dependent transmission that is absent with achiral molecules, and even a chirality dependence of molecular adsorption kinetics on magnetic surfaces (here).  The preprint is a provocative discussion of the topic and possible mechanisms, and the importance of precision in the description of the various phenomena.

On a completely different topic, this preprint is a fun discussion about quantum gravity (!) and how condensed matter ideas of "the vacuum" can lead to insights about how quantum mechanics and gravity might need to play together.  One fun bit early on is a discussion of something I like to point out to my undergrad stat mech students:  A single hydrogen atom in a very very large box will apparently (if the usual stat mech formalism of partition functions is valid) be spontaneously ionized, even when the box (which presumably functions as a reservoir at temperature \(T\)) and atom are at temperatures faaaaaar below the energy scale for ionization.  This is discussed nicely in this 1966 article in the Journal of Chemical Education.  Anyway, I thought this was an interesting discussion from three condensed matter theorists.

Postdoctoral opportunity in materials

The Rice Advanced Materials Research Institute is having its 2025-2026 competition for prestigious postdoctoral fellowships - see here:  https://rami.rice.edu/rami-postdoctoral-fellowship-program  .

If you are interested and meet the criteria, I'd be happy to talk.  I have some ideas that lean into the materials for electronics direction, and other possibilities are welcome.  

2025 Physics Nobel: Macroscopic quantum tunneling

As announced this morning, the 2025 Nobel Prize in Physics has been awarded to John Clarke, Michel Devoret, and John Martinis, for a series of ground-breaking experiments in the 1980s that demonstrated macroscopic quantum tunneling. 

For non-experts: "Tunneling" was originally coined to describe the physical motion of a quantum object, which can pass through a "classically forbidden" region.  I've written about this here, and here is an evocative picture. Suppose there is a particle with a certain amount of total energy in the left region.  Classically, the particle is trapped, because going too far to the left (gray region) or too far to the right (gray region) is forbidden:  Putting the particle inside the shaded regions is "classically forbidden" by conservation of energy.  The particle bounces back and forth in the left well.  If the particle is a quantum object, though, it is described by a wave function, and that wave function has some non-zero amplitude on the far side of barrier in the middle.  The particle can "tunnel" through the barrier, with a probability that decreases exponentially with the height of the barrier and its width.

Fig. 2 from here

Clarke, Devoret, and Martinis were working not with a single particle, but with electrons in a superconductor (many many electrons in a coherent quantum state).  The particular system they chose was a Josephson junction made from an oxide-coated Nb film contacted by a PbIn electrode with a dc current flowing through it.  Instead of an x coordinate of a particle, the relevant coordinate in this system is the phase difference \(\delta\) of the superconducting wave function across the junction.  There is an effective potential energy for this system called a "washboard" potential, \(U(\delta)\), as in this figure.  At the particular DC current, which tilts \(U(\delta)\), the system can transition from one state (\(\delta\) bopping around a constant value, no voltage across the junction) to a state where \(\delta\) is continuously ramping (corresponding to a nonzero voltage across the junction).  The system can get thermally kicked from the zero voltage state to the nonzero voltage state (thermal energy doinks it over the barrier), but the really interesting thing is that the system can quantum mechanically tunnel "through" the barrier as well.

This idea, that a macroscopic (in the sense of comprising many many electrons) system could tunnel out of a metastable state like this, had been investigated by Amir Caldeira and Tony Leggett in this important paper, where they worried about the role of dissipation in the environment.  People tried hard to demonstrate this, but issues with thermal radiation and other noise in the experiments were extremely challenging.  With great care in experimental setup, the three laureates put together a remarkable series of papers (here, here, here) that showed all the hallmarks, including resonantly enhancing tunneling with tuned microwaves (designed to kick the system between the levels shown in panel (d) of the figure above).  

This was an impressive demonstration of controllable, macroscopic quantum tunneling, and it also laid the foundation for the devices now used by the whole superconducting quantum computing community.  

ACS National Nanotechnology Day webinar, Thursday Oct 9

Time for a rare bit of explicit self-promotion on this blog.  This coming Thursday, October 9, as part of the American Chemical Society's activities for National Nanotechnology Day (Why October 9?  In US convention, Oct 9 = 10/9, and 10^-9 m = 1 nm.  Look, it wasn't my idea....), I'm speaking in a free webinar titled "Illuminating the Nano Frontier", with Prof. Dongling Ma of INRS in Quebec.  The event is 11am-12:30pm EDT, and there will also be a recording for people who are unable to watch it live.  Should be a fun event. 

Update:  Here is the link to the webinar recording.  It's free and open-access.

Annual Nobel speculation thread

It’s that time of year again.  The physics Nobel will be announced next Tuesday, and the chemistry prize on Wednesday.  Who will it be this time?  Please speculate in the comments.  As is my annual futile tradition, I will put forward that the physics prize could be Aharonov and Berry for geometric phases in physics (even though Pancharatnam is intellectually in there and died in 1969).  This is a long shot, as always.  Last year was neural networks.  Astro is probably “due”, but who knows.  On the chem side, last year was computational protein design and AlphaFold.  

Fluid mechanics of electrons

Condensed matter physicists often speak of the "electronic fluid" in conductors, and we use a lot of vocabulary that makes analogies between the motion of liquids and the flow of charge (current, vortices, Fermi liquids, and in the UK vacuum triodes were called "valves").   In recent years, there has been a lot of talk about electron flow in the hydrodynamic regime (e.g. this highly cited paper and this recent review and this preprint).  In this regime, electron-electron scattering is strong, and it makes sense to think less about a single-electron picture and more like the electrons acting as a viscous fluid, which can have vortices and a velocity profile transverse to a channel.  (The boundary conditions at the channel walls are a subtle question for sure.)  

Adapted from Fig. 1 of this preprint

This preprint from last week is a great example of pushing this similarity to new limits.  Through a combination of electronic transport measurements and scanning Kelvin probe microscopy, the authors observe what really looks like a hydraulic jump in the flow of electrons through a constriction.  I've written about hydraulic jumps before, here and here.  They're the incompressible flow analog of a standing shockwave, when local fluid velocity goes from supersonic (relative to some sound-like excitation) to subsonic.  Very cute.

(I hope to write more soon, about the role of fluctuations in condensed matter and nanoscale physics.)

H-1B visas and academia

Sorry to talk politics.

In now-classic fashion, there was a just-before-close-of-business-Friday announcement by the executive branch yesterday of a "proclamation" that would seriously up-end the H-1B visa program.  (I have no idea why some of these things are called "executive orders" and some are called "proclamations".)  The short version (see this summary by an immigration law firm):  After midnight EDT tonight, for at least the next year, getting a new H-1B visa for someone outside the US will require the employer to pay a $100K fee.  If someone is a current H-1B visa holder and they are presently outside the US after tonight, it sounds like that fee may need to be paid to get back into the US, if this stands. Update:  The USCIS has issued a memo this evening (!) clarifying that the $100K fee only applies to new H-1B petitions, not present visa holders or those who have already filed.  

The history of this is complicated.  A standard concern is that this program can be and has been abused by tech companies, for example, who have been able to bring skilled workers in certain sectors into the US and pay them comparatively low wages.  Thus, runs the argument, these visas depress wages for domestic workers and make it harder for domestic workers to be competitive for such jobs.   While these are non-immigration visas, certainly a decent fraction of H-1B holders have gone down the path of applying for permanent residency or citizenship.  Personally, I think that the overall economic and cultural benefit to the US is hugely positive, though this being the internet I'm sure someone will argue this in the comments.

Like a large fraction of the Friday afternoon proclamations or executive orders, this is a chaos grenade for academia.  The large majorities of articles in the media about this issue do not point out that the H-1B process is widely used by universities to bring international scholars (faculty members, postdocs, research scientists) into the US.   If every H-1B for someone presently outside the US is now going to cost the sponsor $100K up front, this would be extremely disruptive.  The situation in academia is distinctly different from that in high tech industry - the arguments about wage suppression are not nearly as relevant. 

For the last 70+ years, the US has reaped enormous economic, societal, and national security benefits from being a global destination for top scientists, engineers, and scholars.  No system is perfect, but destroying all of this without any realistic plan to replace it is just self-defeating.  When the secretary of commerce says "the gold [$2M for sponsorship] and platinum [$5M for sponsorship] cards would replace employment-based visas that offer paths to citizenship, including for professors, scientists, artists and athletes", that's disconnected from reality for the world of professors and scientists.  It's not hard to envision that a dedicated visa class (like the F1 visas for students) exempt from crazy fees could be created specifically for PhD-level scholars and researchers, but this would require an actual plan from Congress as well as support for the idea.  

As I've said about other topics, don't panic just yet.  It seems certain that there will be lawsuits filed about this bright and early on Monday morning.  Like many other research-related issues (e.g. slashed indirect cost rates), this will very likely be tied up in court for years, and the biggest fees in the near term will go to lawyers.  Still, it's unclear what the status quo will be while those legal arguments are waged, and the executive branch does have a lot of latitude in the US system.  Stay tuned, and if you are in a position to do so, make your voices heard to your legislators.

(US budget update:  the actual spending bills before congress largely ignore the presidential budget recommendation and its brutal cuts.  Of course, there is a showdown brewing about a continuing resolution since it's unclear whether these bills can pass congress.  It's also unclear whether agencies would actually spend their appropriated budgets.  As before, this is a marathon, not a sprint.)

DOE Experimental Condensed Matter Physics PI Meeting 2025 - Day 3 and wrap-up

 A few more interesting tidbits from the concluding half-day of the DOE ECMP PI meeting:

• Dmitri Basov showed some of the remarkable experiments enabled by layers of MoOCl2, which in the IR is an intrinsically hyperbolic optical material.  This material has unusual plasmonic properties considering its high resistivity.  These include peculiar cavity effects such as modifying superfluid density of a proximally coupled superconductor.
• Leonid Butov explained some remarkable evidence for superfluidity of indirect excitons excited in the moire bilayer of MoSe2/WSe2.  Low temperature mean free paths of these objects can exceed hundreds of microns (!).
• Cui-Zu Chang showed evidence that truly stoichiometric FeTe is actually a superconductor with a critical temperature of about 13.5 K, rather than the usual thinking that it is an antiferromagnetic metal. Apparently an extra 2% of interstitial iron is enough to kill superconductivity and induce AFM order.
• James McIver presented an example of how nonlinear optical effects in an optically driven (Floquet) Weyl semimetal seem to vary linearly with driving field - anomalously strong.
• Dmytro Bozhko showed a really neat technique, using Brillouin light scattering to map out the dispersion of phonons and magnons in YIG, and to extend this approach with a special hollow-core optical fiber to low temperatures with the motivation of probing magnon superfluidity in a particular antiferromagnetic insulator.
• Ray Ashoori used his characteristically pretty quantum capacitance measurement technique to examine the density+displacement field+magnetic field phase diagram of 5-layer rhombohedral graphene, revealing some surprising fractional Chern insulator states.
• Claudia Ojeda-Aristizabal discussed some mesoscopic transport measurements in bilayer graphene, where an adsorbed layer of spin-containing CuPc molecules seems to affect both decoherence and the trigonal warping contribution to it (related to intervalley scattering). 
• Feng Wang and You Zhou both discussed recent measurements looking at Wigner crystals and their properties in 2D TMDs, through a variety of means.
• Liuyan Zhao showed some very rich physics obtained in studies that moiré stack bilayers of the van der Waals insulating magnet CrI3.  

Unfortunately I missed the last talk because of the need to head to the airport.  Overall, the meeting was very good.  Program PI meetings can tend to become less about telling coherent scientific stories and more about trying to show everything someone has done in the last three years.  This meeting avoided that, with clear talks that generally focused on one main result, and that made it much more engaging.  As good as tools for virtual gatherings have become, there really is no substitute for an in-person event when you can just talk to someone by the coffee about some new idea.

DOE Experimental Condensed Matter Physics PI Meeting 2025 - Day 2

It was another very full day.   I had to pop in and out to attend to some things so I didn't get everything, but here are some physics items I learned:

• Dillon Fong introduced me to a technique I didn't know about before, x-ray photon correlation spectroscopy (see this paper).  You can look at time correlations of x-ray speckle near a particular Bragg spot and learn about dynamics and kinetics of transitions and materials growth.  Very cute.
• Charles Ahn presented work on high magnetic field superconductivity in Nd(1-x)Eu(x)NiO2, and I learned about the Jaccarino-Peter effect, in which an external magnetic field can counter the interaction between magnetic dopants and the conduction electrons.  This leads to "reentrant" superconductivity at high magnetic fields. 
• Danny Phelan showed that you can have two different crystal structures for La3Ni2O7, one that is stacked bilayers ("2222"), and one that is stacked monolayer/trilayer ("1313").  
• Ian Fisher talked about using the elastocaloric effect (rapidly and therefore adiabatically stretch or compress a material, leading to a change in its temperature) to identify phase transitions, since the effect is proportional to \( (\partial S/\partial \epsilon)_{T}\), the change in entropy with strain.
• Dan Dessau presented an interesting analysis of data in cuprates suggesting a form for the electronic self-energy that is called a power law liquid, and that this analysis implies that there is not a quantum critical point under the middle of the superconducting dome.
• Jak Chakhalian showed that epitaxially growing an iridate Weyl semimetal directly on top of insulating Dy2Ti2O7 spin ice leads to a dramatic anisotropic magnetoresistance at high in-plane fields that identifies interesting previously unknown physics.
• Daniel Rhodes showed some pretty work on superconductivity in T_d-MoTe2.  This material is extremely air-sensitive, and all of the device fabrication has to be done with great care in a glovebox.  This led to the following exchange.  Audience question: "It is notoriously difficult to make electrical contact to this material.  How did you do this?"  Answer: "Through tears and blood."  This was followed by a serious answer that concluded "The glovebox is always the problem."

DOE Experimental Condensed Matter Physics PI Meeting 2025 - Day 1

That was a full day.  Here are some things I learned, beyond the fact that the ballroom here is clearly kept at about 15°C by default.  (Apologies for not getting everything....)

• About 40% of the DOE ECMP program is related to 2D materials these days.
• Long Ju showed some interesting work trying to understand rhombohedral (ABC-stacked) 5-layer graphene encapsulated by hBN.  Trying to get rid of moiré effects from the hBN/graphene interfaces leads not to more robust quantum anomalous Hall response, but instead leads to very peculiar superconductivity that survives up to very large in-plane and moderately large out-of-plane magnetic fields. This happens in the same regime of charge and gate that would otherwise show QAH.  Looks like some kind of chiral superconductivity that may be topological.
• Andrea Young, meanwhile, in fewer layer rhombohedral systems, showed experiments pointing to superconductivity happening at the verge of a canting transition, where spins are reorienting.
• Eva Andrei gave a nice talk looking at the variety of states one can get when interfacing moiré systems with other moiré systems, and explaining what is meant by intercrystals.  
• Gleb Finkelstein showed how a measurement intended to look at shot noise instead became a very cute noise thermometry probe of thermal transport at the boundary between (graphene) quantum Hall currents and a superconducting electrode.
• Xiao-Xiao Zhang showed a really cute experiment, where the resonance of a drumhead made from an atomically thin film of MnPS3 convey information about magnetic transitions in that material as a function of magnetic field.
• Dan Ralph gave a nice talk about the challenges of electrically generating currents of properly oriented spins to drive magnetic switching in films magnetized perpendicular to the film plane, for spin-orbit torque memories (and fundamental understanding).
• Philip Kim gave a great overview of some remarkable results in electronic interferometers made on graphene, in which telegraph noise shows signatures
• Lu Li spoke about recent measurements showing magnetic oscillations and specific heat signatures of possible neutral fermions in a kagome lattice Mott insulator.
• Xavier Roy talked about CeSiI, a 2D material that is also a heavy fermion metal.  This and its related compounds look like a fascinating family of (unfortunately extremely air sensitive) materials. 
• Harold Hwang gave a great overview of recent work in nickelate superconductors, highlighting the similarities to the cuprates as well as the profound differences (like how electronic configurations other than d9 can also lead to superconductivity).

DOE experimental condensed matter PI meeting, + other items

This week I am attending the every-two-years DOE Experimental Condensed Matter Physics PI meeting.  Previously I have written up highlights of these meetings (see here, here, here, here, here), though two years I was unable to do so because I was attending virtually.  I will do my best to hit some high points (though I will restrict myself to talking only about already published work, to avoid any issues of confidentiality).  

In the meantime, here are a couple of topics of interest from the last couple of weeks.  

• I just learned about the existence of Mathos AI, an AI product that can function as a math solver and calculator, as well as a tutor for students.  It is pretty impressive.
• I liked this historical piece about Subrahmanyan Chandrasekhar (he of the “Chandrasekhar limit”, which describes the degeneracy physics + gravitation that limits the upper size of compact stellar objects like white dwarfs and neutron stars before they collapse into black holes) and his interactions with Stephen Hawking.  It's pretty humanizing to see an intellectual giant like Chandra sending a brief letter to Hawking in 1967 asking for advice on what to read so that Chandra can understand Hawking’s work on singularities in cosmology.  Hawking’s handwritten response is clear and direct.
• In an online discussion about what people will do if Google decides to stop supporting Google Scholar, I was introduced to OpenAlex.  This seems like an interesting, also-free alternative.  Certainly worth watching.  There is no obvious reason to think that Google Scholar is going away, but Alphabet has retired many free products, and it’s far from obvious how they are making any money on this.  Anyone from Google who reads the blog, please chime in.  (Note to self:  keep regularly backing up this blog, since blogger is also not guaranteed future existence.)

25 years of Nano Letters

Back in the dawn of the 21st century, the

American Chemical Society founded a new journal, Nano Letters, to feature letters-length papers about nanoscience and nanotechnology.  This was coincident with the launch of the National Nanotechnology Initiative, and it was back before several other publishers put out their own nano-focused journals.  For a couple of years now I've been an associate editor at NL, and it was a lot of fun to work with my fellow editors on putting together this roadmap, intended to give a snapshot of what we think the next quarter century might hold.  I think some of my readers will get a kick out of it.  

Learning and AI/LLMs - Why do we need to know or teach anything anymore?

The fall semester is about to begin at my university, and I'm going to be teaching undergraduate statistical and thermal physics.  This is a course I've taught before, last full term in 2019, and the mass availability of large language models and generative AI tools have changed the world in the interim.  We've all seen the headlines and articles about how some of these systems can be very good at solving traditional homework and exam problems.  Many of these tools are capable of summarizing written material and writing essays that are very readable.  Higher education is wrestling with the essential question:  What is the right working relationship between students, teachers, and these tools, one that benefits and actually educates students (both about subject matter and the use of these tools)?  Personalized individual AI tutoring seems like it could be great for teaching huge numbers of people.  Conversely, if all we are doing is teaching students to copy-paste assignments into the homework-answer-machine, clearly we are failing students at multiple levels.  

The quote in the image here (from Kathy Hepinstall Parks) is one that I came across this week that originates in the FAQ from a writers workshop.  For my purposes I could paraphrase:  Why should we learn physics (or any other science or engineering discipline) when a machine already knows the formalism and the answers?  On some level, this has been a serious question since the real advent of search engines.  The sum total of human knowledge is available at a few keystrokes.  Teaching students just rote recall of facts is approaching pointless (though proficiency can be hugely important in some circumstances - I want a doctor who can diagnose and treat ailments without having to google a list of my symptoms.).

My answer to this question is layered.  First, I would argue that beyond factual content we are teaching students how to think and reason.  This is and I believe will remain important, even in an era when AI tools are more capable and reliable than at present.  I like to think that there is some net good in training your brain to work hard, to reason your way through complicated problems (in the case of physics, formulating and then solving and testing models of reality).  It's hard for me to believe that this is poor long-term strategy.  Second, while maybe not as evocative as the way creative expression is described in the quote, there is real accomplishment (in your soul?) in actually learning something yourself.  A huge number of people are better at playing music than I am, but that doesn't mean it wasn't worthwhile to me to play the trumpet growing up.  Overworked as referencing Feynman is, the pleasure of finding things out is real.  

AI/LLMs can be great tools for teachers.  There are several applet-style demos that I've put off making for years because of how long it would take for me to code them up nicely.  With these modern capabilities, I've been able to make some of these now, in far less time than it would otherwise have taken, and students will get the chance to play with them.  Still, the creativity involved in what demos to make and how they should look and act was mine, based on knowledge and experience.  People still have a lot to bring to the process, and I don't think that's going to change for a very long time.

20 years of Nanoscale Views, + a couple of things to read

Amazingly, this blog has now been around for more than twenty years (!) - see this first post for reference from June of 2005, when I had much less gray hair and there were a lot more science blogs.  Thanks to all of you for sticking around. Back then, when I debuted my writing to my loyal readers (all five of them at the time), I never thought I'd keep this up.  Some info, including stats according to blogger:

• Total views: 8.3M
• Most views in one day, this past May 31, with 272K
• Top two most-viewed posts are this one from 2023 with a comment thread about Ranga Dias, and this one from 2009 titled "What is a plasmon?"
• Just a reminder that I have collected a bunch of condensed matter terms and concept posts here.
• I've also written some career-related posts, like a guide to faculty job searches, advice on choosing a graduate school, needs-to-be-updated advice on postdoc positions, etc.
• Some personal favorite posts, some of which I wish had gotten more notice, include the physics of drying your hands, the physics of why whiskey stones aren't as good as ice to cool your drink, materials and condensed matter in science fiction, the physics of vibranium, the physics of beskar, the physics of ornithopters, and why curving your pizza slice keeps if from flopping over.  I'm also happy with why soft matter is hard, which was a well-viewed post.
• I also like to point out my essay about J. Henrik Schön, because I worry that people have forgotten about that episode.

Real life has intruded quite a bit into my writing time the last couple of years, but I hope to keep doing this for a while longer.  I also still hope one day to find the right time and approach to write a popular book about the physics of materials, why they are amazing, and why our understanding of this physics, limited as it is, is still an astonishing intellectual achievement. 

Two other things to read that I came across this week:

• This post about Maxwell's Demon from the Skull in the Stars blog (which has been around nearly as long as mine!) is an excellent and informative piece of writing.  I'm definitely pointing my statistical and thermal physics undergraduate class to this next month.
• Ross McKenzie has a very nice looking review article up on the arXiv about emergence. I haven't read it yet, but I have no doubt that it will be well-written and thought-provoking.