✓ Detailed balance and maximum entropy in non-equilibrium systems.
Detailed balance has been believed to be the property only of systems in thermal equilibrium (see, `proof' by M.J. Klein 1955). In three recent papers, one published, one submitted and one in preparation, we show theoretically and support experimentally that steady states of quasi-statically sheared planar granular systems, which are very far from equilibrium, also exhibit this property. In another paper, we show that the dynamics of these systems maximise the configurational entropy of the irreductible loops of their contact networks, subject to stability constraints. These derivations and observations are siginificant. Firstly, they uncover fundamental principles that underlie the general dynamics of driven granular systems. Secondly, they suggest that the distinction between equilibrium and steady states needs to be revisited. They also support Sir Sam Edwards's original insight that granular systems can be studied with statistical mechanics. We keep exploring the ramifications of these general principles.
✓ The relations between the physics endeavour and reality.
We do not perceive reality as it is. Rather, we have our brain's interpretation of it through the senses. The brain's interpretation is tailored only to enable survival and continuation of the species. To this end, it evolved to allow us the best tools to manipulate reality, not really understand it. The corrolary of this realisation is rather off-putting: what we physicists do is not construct better models of reality, but better algorithms to manipulate it. Namely, if we do that experiment, this is what would happen. Not as glorious an endeavour as I thought when I started out to become a theoretical physicist.
✓ An `uncertainty principle' in granular mechanics.
Below is an excerpt on this issue from our review chapter, "Granular Systems", in The Oxford Handbook of Soft Condensed Matter (Oxford University Press, Oxford, UK, 2015):
"...these calculations are idealised and taking into consideration the redundancy is problematic. In real systems in mechanical equilibrium, exact redundancy is possible only when a grain has exactly two force-carrying contacts, in which case they must be equal, opposite and align along the line between the contacts. For grains with three contacts or more, it is impossible to determine the force directions in general with absolute accuracy and, in particular, whether those coincide exactly at a point. At most one can say that the forces apply very small torques to the grain because the `region of coincidence' (the shaded region in figure 5.10b) is small. Consequently, except for two-contact grains, the equations are never really redundant - they are at most negligible. It follows that the value of $z_c$ depends to some extent on the level of accuracy of the measurement of forces; the more accurate the measurement, the more torque balance equations need to be included and the higher the value of $z_c$.
This sounds strangely like an `uncertainty principle' - the more accurate the measurement the more equations one needs to determine the stresses. This `uncertainty principle' is also a manifestation of the inherent relation between the structure on the granular level and the stress field that the assembly supports ...".
✓ A-thermal (or particulate) statistical mechanics appears to give intriguing insight into the nature of reality, time and our perception of these two. I am not referring to entropy and the arrow of time, but rather insight into time's existence in the first place.
✓ "Experimental evidence of detailed balance in granular systems", Sun, Wang, Wang, Blumenfeld, Zhang, corroborates experimentally our 2020 theoretical derivation of detailed balance in out-of-equilibrium sheared granular systems, a phenomenon supposedly proven by Klein in 1955 to be impossible. Surprisingly, the experiments reveal that detailed balance is satisfied even in much less dense systems than theoretically predicted. The textbooks on detailed balance may need to be revisited.
✓ "Friction-controlled entropy-stability competition in granular systems", Sun, Kob, Blumenfeld, Tong, Wang, Zhang, squeaked into Phys. Rev. Lett. on the last day of 2020. In this paper, we present an analysis of the structural characteristics of two-dimensional granular assemblies, constantly rearranging under shear, and show that underlying the dynamics is an inherent competition between entropy and mechanical stability. This competition can be controlled by intergranular friction: the higher the friction the higher the entropy. We show that, for high friction, a straightforward maximum-entropy calculation predicts the distribution of cell orders to a very good accuracy. We also demonstrate the reduction in entropy with decreasing friction. Very large cells, up to order 30 in such systems, are short-lived, implying that our quasi-static granular system is not glassy but liquid.
✓ "Structural Evolution of Granular Systems: Theory", Wanjura, Gago, Matsushima, Blumenfeld, appeared in Granular Matter. In this paper, we formulate a theory to quantify the evolution of the structure of granular matter under external loading, and test its predictions successfully against numerical simulations. Astonishingly, we find that, although far from equilibrium, the steady states of very dense quasi-static such dynamics satisfy detailed balance!
✓ "Sink-rise dynamics of horizontally oscillating active matter in granular media: Theory", Liu, Ran, Blumenfeld, submitted to Phys. Rev. Lett.. An intermediate step to modelling behaviour of active matter is understanding interactions of active matter with inanimate matter, often also leading to rich behaviour. We present a range of simulations of the interaction of a self-energised sphere with 3D granular medium and develop a first-principles theory to describe the observed phenomena. While oscillating horizontally, the active sphere rises against gravity or sinks, depending on the oscillation amplitude and frequency. We identify the competing mechanisms driving these dynamics. Below a critical speed, a jammed stagnant zone builds up ahead, leading to rising, while above this speed, the sphere fluidises its surounding medium and sinks. The duration of the rising and sinking phases depend non-trivially on the amplitude and frequency, leading to an intricate nonlinear dynamics. A first-principles equation of motion is developed for the time-dependent depth and its solutions agree well with the simulations.
✓ "Statistical mechanics of high table conversations", in preparation. This is an interesting application of statistical mechanics methods to social situations. In my college (Gonville & Caius, Cambridge), formal dinner is held (in normal times) in the Hall, with fellows and their guests sitting at the high table, while students fill much of the rest of the Hall (see photo). Conversations at the high table have to be conducted above the noise, which is not negligible, not least because our college allows students to bring alcohol to dinner. High table conversations are then constrained by the distance between seats. Owing to the noise (and the relatively high average age) conversations around the high table can take place only between nearest or next nearest diners. In this work, we model the statistical mechanics of such a situation. In particular, we aim to determine the optimal conditions for the most hospitable dinner, i.e. one where as many conversations as possible take place.
✓ "The unusual problem of upscaling isostaticity theory for granular matter", Blumenfeld, appeared in the special issue of Granular Matter in memorium of Bob Behringer. In this paper, following some thoughts about the late Bob, I describe a method to coarse-grain the stress equations in solid granular media from the grain-scale up to the continuum.
✓ Nominated a "High Level Foreign Talent" in China and a Distinguished Visiting Professor in Hunan Central South University (comes with a long-term visa).
✓ "Support of Modified Archimedes' Law Theory in Granular Media", Feng, Blumenfeld, Liu, has been published in Soft Matter. In this paper, we elucidate the different mechanisms governing penetration of solid objects into granular materials and show that a generalised Archimedes law, expressed as a function derived frim first-principels, unifies the resistance behaviour of real (rough) granular materials, frictionless spheres, and water!
✓ "Force-based three-dimensional model predicts mechanical drivers of cell sorting" Revell, Blumenfeld, Chalut, has been published in Proc. Roy. Soc. B: Biological Sciences. In this paper, we elucidate the roles of physical mechanisms in stem cell sorting in the embryo.
✓ "Stress Tensor for Dense Granular Flow in Plug-Free Regions", Schwartz, Blumenfeld, has been published in Phys. Rev. E. In this paper, which has taken 6 years to complete, we derive frm forst principles the rheology of dense plug-free particulate fluids and put the solid-friction-based rheology on solid fundamental basis.
✓ Clara Wanjura has received her masters degree.
✓ "Equally probable positive and negative Poisson's ratios in disordered planar systems", Verstreken, Chalut, Blumenfeld, has been published in Soft Matter. In this paper, we show that disordered assemblies of isostatically linked compliant triangles are, surprisingly, equally likely to have negative and positive Poisson's ratios.
✓ Delivering 4 lectures at the "Summer School on Soft Matter and Biophysics" on stress field theory in particulate media. Expect lively informal discussions. All welcome.
✓ I now have a paper that I cannot read (!) "Numerical simulation of a spinning sphere moving in granular matter" , Qi, Ran, Liu, Tang, Blumenfeld, Acta Physica Sinica.
✓ Chairing session K48 of "Athermal Systems and Statistical Mechanics" at the APS meeting, March 5-9, LA, USA. Please join us there.
✓ "Archimedes' law explains penetration of solids into granular media: modelling and experimental support", Kang, Feng, Blumenfeld, Liu, has been published in Nature Communications. In this paper, we show that the process of quasi-static object penetration into dense granular matter is governed by a simple, albeit universal, Archimedes' law.
These pages have grown uncontrollably from a modest beginning in 1993, when a home page was a rarity. I cannot pinpoint when exactly over the years it was that I lost the fight against this unstable self-(dis)organised growth. I was probably never in control.
You may notice that the pages are written in the old basic HTML format. This is because none of the current fancy new softwares, which essentially do all the nitty-gritty work, existed. Hence, there are no fancy java scripts, flashing lights and colours, background music, or automatically starting videos. This may be one of the last bastion of the HTML purists, uncluttered by sights or sounds.
As my friend Peter once put it:
"This is about delegation of responsibility. About trust. About subsidiarity.
It's a small-scale mirror of so many of our every day problems."
Anyway, my publication pages are updated regularly ... after a fashion. The rest not so much. If your time is short, I recommend to browse my 'Selected publications' and then move to the less frequently updated 'research interests' page. Much of my research is distilled into limericks at the bottom of this page. You are, of course, welcome to visit all the pages but there is always the possibility of getting lost in the overgrown meandering paths.
Often, when I finish developing a model or a theory, I write a limerick that summarises it. The weight and rythm are not always great, but they do describe, as accurately as possible in a limerick form, the modelling ideas and the relevant physics. Below are the limericks I wrote so far, each with a link to a paper that the work produced, if it has.