Generalizing the Archimedes’ Principle

Roberto Piazza talks to Russell Johnson about re-examining the Archimedes’ principle

Roberto Piazza1. You recently published a paper re-examining the Archimedes’ principle. What motivated you to take a second look at this?
We were actually trying to understand some rather weird instability phenomena we had observed in the settling of binary colloid mixtures (see J. Phys.: Cond Matt. 24, 284109, 2012 ). Of course, to perform a numerical simulating of sedimentation, one has to give a value for the buoyant force acting on a particle. Yet, if we consider a particle of type 1 settling in a suspension of particles of type 2, this is less trivial than expected: using the standard Archimedes’ principle, what value  for the density  of the “displaced fluid” should we use, that of the bare solvent or that of the  suspension of the host particles? Looking back at the literature, both attitudes can be found. It turns out that neither of the two is correct. 
(Ref. Soft Matter, 2012, DOI 10.1039/C2SM26120K)

2. Can you explain the main points you discovered in this paper?
The crucial point is precisely the expression “displaced fluid”. When a particle in inserted in a complex fluid, or in general in a liquid displaying strong correlations, it does not “displace” just its own volume. Because of its interactions with the solvent, the particle perturbs the local density of the surrounding too. In the example of binary hard-sphere colloids, for instance, a region “depleted” of the host particles forms around particle 1, followed at larger distances, if particles 2 are sufficiently concentrated, by a region where the latter display an oscillating concentration profile. The interesting point is that a the effect of these perturbations on buoyancy can be quantitatively evaluated using a very general expression obtained by extending a simple mechanical equilibrium argument used in elementary physics courses to derive the Archimedes’ principle. This corrected form of the buoyant form, which we call “Generalize Archimedes’ Principle” (GAP), may lead to curious and counterintuitive effects that we tested experimentally.

3. What challenges did you have to overcome in this project?
Trying to test the GAP with settling experiments by looking at the kinetics of settling is not easy,  for colloid sedimentation dynamics is very complex because of the presence of long-ranged hydrodynamic interactions between the particles. Besides, it is very hard to set apart concentration effects on buoyancy from those on transport coefficients. Hence, we resolved to study equilibrium sedimentation, by looking at the level where a very dilute suspension of particles 1 “float” in a “sea” of particles 2 at settling equilibrium – the so-called “isopycnic point”. This is very similar to what is done in Density Gradient Ultracentrifugation (DGU), a method extensively used in biology. Our results clearly show that the standard expression for the buoyant force is incorrect, while they quantitatively confirm the prediction of the GAP. Besides, they highlight the actual occurrence of apparently weird situations, also predicted by the GAP, where very dense (in our case, gold) particles float atop a suspension of different particles with a density comparable to water.

4. What other research topics are you working on at the moment?
Currently, my group is mostly concentrated on non-equilibrium effects in colloidal suspensions. One of the main topic, strictly related to what I just discussed, is restructuring and aging induced by the gravitational stress in colloidal gels and glasses. A second topic we have thoroughly investigated in the past few years is thermophoresis, namely, particle or macromolecule transport driven by thermal gradient, which is one of the simplest, but also most puzzling thermal non-equilibrium effect I am aware of. Presently, we are trying to exploit thermophoretic effects in microfluidics. 

5.  As science becomes increasingly interdisciplinary, how do you see the future of science, and soft matter science in particular?
Predicting the future of science is of course very daring, even over a short time span. But I tend to  agree with a claim by Dave Weitz, from Harvard, who once told me that “The physics of the 21st century will be the physics of the 19th century”. By this, he meant that there are still many open questions in classical physics that the quantum revolution of the last century swept somehow under the carpet. In particular, our understanding on non equilibrium effects is depressingly low – just think of  turbulence that, according to Feynman, not even God may possibly understand. Soft matter, allowing to devise simple model systems to tackle general statistical mechanics and condensed matter problems, will surely play a leading role in this challenging enterprise.

6. Collaborations form a large part of modern scientific research. Which scientist, past or present, would you really like to work with?
Of course, modern scientific research requires a strongly interconnected community. But let me say that, paradoxically, collaboration works when you don’t really need to collaborate. A prerequisite for a balanced cooperation is indeed that both parties are fully able to stand alone, profiting from complementary expertise. This work on buoyancy, where we join forces with Alberto Parola, a very competent and careful theoretician who rigorously justified our simple model using density functional theory, is to me an example of a sound collaboration.

But since this question is also about scientists of the past I would have liked to work with, let me use this occasion to try and beg my pardon to the superb scientist whose great intuition we dared to challenge: Archimedes, whose role in the development of modern science has been probably undervalued, as in general has been the incredible modernity of the scientists of the Hellenistic period. To appreciate how much we owe to these forgotten giants, I strongly recommend an amazing book by Lucio Russo, “The Forgotten Revolution: How Science Was Born in 300 BC and Why it Had to Be Reborn”, suggested to me by Daan Frenkel, that I am presently reading with great pleasure. Had Archimedes being aware of the wonders of the nanoword we play with, he would have surely spotted the need for updating his own principle much more easily than we did.

To keep up-to-date with all the latest research, sign up for the Soft Matter e-Alert or RSS feeds or follow Soft Matter on Twitter or Facebook.

Digg This
Reddit This
Stumble Now!
Share on Facebook
Bookmark this on Delicious
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)