Abstract
The force virial is a construct combining information about the current configuration of a mechanical system in motion with information about the acting forces; its long-time average turns out to be proportional to the long-time average of the system’s kinetic energy, that is, the virial theorem holds true; a version of this theorem is obtained when a connection between kinetic energy and temperature is established, as it happens in the kinetic theory of gases or in classical equilibrium statistical mechanics. In fact, a virial theorem holds in whatever mechanics. This paper is an exposition of its various formulations, from the simplest deterministic formulation for a single massy particle in Newtonian motion to the fairly more complicated statistical formulation, which makes the theorem a corollary of the equipartition theorem.
This is a preview of subscription content, log in via an institution to check access.
Notes
This remark is not meant as a criticism: the diameter of a Brownian particle is of the order of, say, 1 μm, way above atomic size.
“Sur la force complémentaire \(X\) nous savons qu’elle est indifféremment positive et négative, et sa grandeur est telle qu’elle maintient l’agitation de la particule que, sans elle, la résistance visqueuse finirait par arrêter.”
We see from this relation that time averaging eliminates the contribution to the virial of viscous/frictional force of Stokes type.
The case of discrete and continuous material systems with variable mass is studied in [8].
This formula was derived by Finger in 1897 (see [9], Sects. 216 and 219).
Taking all masses the same entails no substantial loss of generality in the developments to follow. Recall that taking the thermodynamic limit amounts to inspecting the system’s properties for \(N,V\rightarrow \infty\), keeping the \(V/N\) ratio fixed.
In other words, the box walls are modeled as perfectly reflecting.
Recall the well-known distributional relationships between the Heaviside and Dirac functions:
$$ -f(x_{0})= \int_{\mathbb{R}} \theta(x-x_{0})f^{\prime}(x)dx=- \int_{\mathbb{R}} D\theta(x-x_{0})f(x)dx\quad\text{for all test functions } f, $$(here \(D\) denotes distributional differentiation), whence
$$ D\theta(x-x_{0})=\delta(x-x_{0}). $$The fact that the notation we use for the microcanonical density keeps track of the dependence on \(E\) but not of those on \(N\) and \(V\), is not a good reason to overlook it.
It is often reported that, around phase changes, MD systems may become sluggish and correlation lengths may increase and even diverge.
By a spatial s-correlation function (see, e.g., [12], Sect. 79, and [10], Chap. 10) we mean a function giving the probability of finding \(s\) particles—say, those of Lagrangian coordinates \((q_{1},q_{2},\ldots,q_{s})\)—in a configurational box \(dq_{1}dq_{2}\ldots dq_{s}\) at a specified time.
Here a standard lemma of integration by parts is called upon, namely,
$$ \int_{A} \nabla\alpha d(vol_{A})= \int_{\partial A}\alpha \mathbf {n}_{\partial A}d(area_{A}), $$where \(\mathbf {n}_{\partial A}d(area_{A})\) is the oriented area measure over the boundary of the open bounded region \(A\) and where the scalar field \(\alpha\) is conveniently smooth, say, \(\alpha\in C^{0}(clo(A))\cap C^{1}(A)\).
We just saw that this result—which, we recall, was first derived and given its name by Clausius in 1870—is nothing but a corollary of the generalization (41) of the Equipartition Theorem, whose first formulation was given by Boltzmann only one year later, in 1871. We read in [9], Sect. 219, that the quantity \(\sum_{i} q_{i} f_{i}\) was introduced in 1837 by Möbius, Schweins, and Jacobi, in that order.
For a system confined in a box of finite volume, \(|q|\) stays finite at all times, and so does \(|p|\), because the total energy is conserved.
References
Marc, G., McMillan, W.G.: The virial theorem. Adv. Chem. Phys. 58, 209–361 (1985)
Podio-Guidugli, P.: Sparse notes in continuum and statistical thermodynamics. To appear, eventually
Maugin, G.A.: Material Inhomogeneities in Elasticity. Chapman and Hall, London (1993)
Gurtin, M.E.: Configurational Forces as Basic Concepts of Continuum Physics. Springer, New York (2000)
Ganghoffer, J.F.: On the virial theorem and Eshelby stress. Int. J. Solids Struct. 47(9), 1209–1220 (2010)
Langevin, P.: Sur la théorie du mouvement brownien. C. R. Acad. Sci. 146, 530 (1908)
Einstein, A.: Über die von der molekular-kinetischen Theorie der Wärme gefordete Bewegung von in ruhenden Flüssigkeiten suspendierten Teilchen. Ann. Phys. (Leipz.) 17, 549 (1905)
Ganghoffer, J.F., Rahouadj, R.: On the generalized virial theorem for systems with variable mass. Contin. Mech. Thermodyn. 28(1), 443–463 (2015)
Truesdell, C., Toupin, R.A.: The classical field theories. In: Handbuch der Physik III/1 (1960)
Sethna, J.P.: Entropy, Order Parameters, and Complexity. Oxford Master Series in Physics. Oxford University Press, London (2006)
Andersen, H.C.: Molecular dynamics simulations at constant pressure and/or temperature. J. Chem. Phys. 72(4), 2384–2393 (1980)
Landau, L.D., Lifshitz, E.M.: Course of Theoretical Physics. Vol. 5, Statistical Physics. Part 1, 3rd edn. Pergamon, Elmsford (1980)
Huang, K.: Statistical Mechanics, 2nd edn. Wiley, New York (1987)
Acknowledgements
I presented part of this material in two lectures given at OIST (Okinawa Institute of Science and Technology, Okinawa, Japan) during the Summer School on “Hierarchical multiscale methods using the Anderson-Parinello-Rahman formulation of molecular dynamics”, April 3–8, 2017. It is a pleasant duty for me to acknowledge OIST’s support and hospitality.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Podio-Guidugli, P. The Virial Theorem: A Pocket Primer. J Elast 137, 219–235 (2019). https://doi.org/10.1007/s10659-018-09716-6
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10659-018-09716-6