Skip to content

Science1 publisher2 min readPublished

One calculation tracks PMMA's stiffness from terahertz vibrations down to millihertz lab rates

Working from an atomistic model of PMMA, the authors added a power-law memory kernel to non-affine lattice dynamics and matched molecular dynamics plus three kinds of measurement down to millihertz frequencies.

The Scientist · Science desk

Illustration accompanying One calculation tracks PMMA's stiffness from terahertz vibrations down to millihertz lab rates

What happened

  • A team modelling poly(methyl methacrylate) with 9,920 atoms calculated the polymer's shear modulus from its atomic vibrations, in work published in The Journal of Chemical Physics.
  • Instead of simulating slow deformation directly, they extracted the vibrational modes of the structure and how those modes couple to an imposed deformation.
  • The resulting curve runs from above the terahertz regime down to millihertz, a frequency range the authors describe as exceeding 20 orders of magnitude.

Compiled by The ScientistSomething wrong?How this is made

Why it matters

  • capability One calculation covers windows that four techniques have covered separately, so a stiffness at an untested deformation rate can be computed from atomic structure instead of interpolated between instruments.
  • contradiction Anyone quoting the headline span should know the endpoints given, above a terahertz and down to a millihertz, cover 15 decades while the account reports more than 20.
  • decision How the memory kernel's exponent was set decides whether an engineer can use the low-frequency end before measuring the material, or only after.
  • constraint The demonstration covers one polymer in one atomistic model, so a formulator weighing a filled or semicrystalline grade has no basis here for transferring the result.

Deform an ordered solid and you can picture every atom moving along with the imposed macroscopic strain. A polymer glass is structurally disordered, so its atoms do not all follow that deformation, and many of them make additional local rearrangements that collectively soften the material [9]. Non-affine lattice dynamics calculates that softening by connecting the vibrational modes of the atomic structure with the forces generated when the material is deformed, and the quantity it returns is the macroscopic shear modulus [10]. The author says he helped develop the framework with his longtime collaborator Dr. Tim Sirk at the U.S. Army Research Laboratory [8].

Vibrations alone do not reach laboratory timescales. In a simple friction model the resistance an atom feels depends only on what is happening at that instant, while a glassy polymer's response retains information about what happened before, which makes the dynamics non-Markovian [11]. The power-law memory kernel put into the equations for the non-affine motions is what carries the broad spectrum of relaxation processes [12]. How the exponent of that kernel was set goes unreported in the write-up, and that determines how much of the low-frequency curve is computed and how much is fitted [18].

A terahertz is 10^12 hertz and a millihertz is 10^-3 hertz, so the two frequencies quoted bracket 15 orders of magnitude, not the more than 20 the account reports [13][16]. Twenty decades above a millihertz floor would put the ceiling near 10^17 hertz [17]. The account describes the top of its range only as above the terahertz regime [13].

"The result surprised even us in terms of the range of scales that could be connected," wrote the researcher who described the work for phys.org [15]. The comparisons run down the frequency axis in order: direct molecular dynamics at the fastest scales, Brillouin light scattering in the gigahertz range, high-rate mechanical tests below that, and dynamic mechanical analysis at conventional laboratory frequencies [14]. All of those PMMA datasets existed before the calculation. Matching them tests the chain of reasoning.

The cost the approach avoids is simulation time. Molecular dynamics resolves the fastest atomic motions, and the shortness of the time step is what makes ordinary mechanical timescales so hard to reach [3]. Here the vibrational modes come out of one atomistic model and the frequency sweep happens in the theory [7]. Between those two ends, Brillouin scattering, ultrasonic measurement and high-strain-rate testing each fill portions of the gap [4].

What to watch

  • Whether the same calculation, with the kernel exponent fixed the same way, reproduces PMMA data at a second temperature.
  • Whether another group applies the method to a semicrystalline or filled polymer without retuning the memory kernel.
  • Whether the paper itself names the two frequencies it counts between, which would settle the 15-versus-20 decade question.
Loading claim ledger
Loading source directory links
Loading share composer
Loading topic controls
Loading related stories