Refractive index and generalized polarizability
Krzysztof Pachucki, Mariusz Puchalski

TL;DR
This paper explores how retardation effects influence polarizability and refractive index, emphasizing the need for nonlocal dielectric models in interpreting precise optical measurements.
Contribution
It introduces the importance of nonlocal dielectric constant modifications due to retardation corrections in classical electromagnetic theory.
Findings
Retardation corrections significantly affect polarizability and refractive index.
Nonlocal dielectric models are necessary for accurate optical measurements.
Classical theory requires modifications to incorporate nonlocal effects.
Abstract
We investigate the role of retardation corrections to polarizability and to refractive index. We found that the classical electromagnetic theory of dielectrics requires corresponding modifications in terms of nonlocality of the dielectric constant. This nonlocality should be taken into account in the interpretation of accurate measurements of the optical refractivity.
| Polarizability | Expectation value |
|---|---|
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Refractive index and generalized polarizability
Krzysztof Pachucki
Faculty of Physics, University of Warsaw, Pasteura 5, 02-093 Warsaw, Poland
Mariusz Puchalski
Faculty of Chemistry, Adam Mickiewicz University, Umultowska 89b, 61-614 Poznań, Poland
Abstract
We investigate the role of retardation corrections to polarizability and to the refractive index. We found that the classical electromagnetic theory of dielectrics requires corresponding modifications in terms of the nonlocality of the dielectric constant. This nonlocality should be taken into account in the interpretation of accurate measurements of the optical refractivity.
I Introduction
A coherent monochromatic electromagnetic plane wave propagates in a disordered medium of polarizable particles, provided its wavelength is much larger than the average distance between particles. Then the multiple scattering has the only effect of changing the wavelength. The ratio of this medium wavelength to the vacuum one is the refractive index. The electromagnetic wave satisfies Maxwell’s equations on the macroscopic level, thus the refractive index is given by the dielectric and magnetic constants of the medium. If the wavelength is sufficiently short or a high-precision refractive index is intended, the wave-vector dependence of the dielectric constant cannot be neglected. High-precision measurements of the refractive index of a low-density helium gas are being performed in several laboratories egan:17 ; jousten:17 ; silvestri:18 .
Helium is the best choice as it is the simplest noble-gas atom, thus its properties can be calculated with the highest accuracy. The Clausius-Mosotti formula,
[TABLE]
relates the dielectric constant to the microscopic property, namely, to the electric dipole polarizability of an atom. For this reason, the helium electric dipole polarizability has been studied in a series of works summarized in Ref. puchalski:16 , resulting in a accuracy of about ppm, which comes mainly from the estimation of unknown higher-order corrections.
In this work we demonstrate that the electric dipole polarizability is not sufficient for a highly accurate description of the refractive index. On the fundamental level it should be obtained from the forward photon-scattering amplitude. Apart from the electric and magnetic dipole polarizabilities, it includes also the so-called retardation corrections in the form of a quadrupole and other types of polarizabilities. This leads to the dependence of the refractive index on the wave-vector . This dependence affects the relation between the and fields, which becomes nonlocal in the coordinate space [see Eq. (16)]. The wave-vector dependence of the dielectric constant has already been indicated for the model systems of finite-size spherical insertions in the electric dipole approximation ersfeld:98 . Here, we derive a complete set of retardation corrections and present their accurate numerical values for a gas of helium atoms. They affect the interpretation of the accurate measurements of the refractive index egan:17 ; jousten:17 ; silvestri:18 . As an additional result, we note that the dielectric tensor introduced in previous works on this topic ersfeld:98 ; forcella:17 can be expressed in terms of scalar dielectric and magnetic constants only.
II Photon propagator in the medium
We will obtain only the leading term in the small expansion of the refractive index, as it is sufficient for our purpose. For this we consider the photon propagator in the presence of the noble gas atoms. The photon wavelength is assumed to be much larger than the atomic size . Then, the electromagnetic interaction can be expanded in powers of a small factor and with inclusion of qubic terms takes the following form lwqed
[TABLE]
where denotes the derivative of with respect to , and the summation is over all the atomic electrons. The electromagnetic fields in the above are taken at the position of the nucleus. Since the electron-nucleus mass ratio is very small ( for He) we assume an infinite nuclear mass. Equation (2) demonstrates the existence of small nondipole couplings to the electromagnetic field and will be used below in the derivation of the previously neglected retardation correction to the refractive index of helium gas. We use theoretical units and throughout this work.
Consider now the photon propagation through the medium of noble-gas atoms. We assume, in accordance with planned measurements, that the photon frequency is much smaller than the excitation energy of the helium atom. Then the scattering amplitude is real and there is no photon absorption. Each scattering on an individual atom is averaged over the whole space assuming a homogeneous and isotropic density of atoms,
[TABLE]
which shows that photon propagation depends only the elastic forward off-shell amplitude for the scattering off the single atom, which is of the general form olmos_01
[TABLE]
where is the generalized electric polarizability, and is the magnetic one. The above Eq. (4) is the definition of these polarizabilities, and in the nonrelativistic limit , coincides with the standard electric, magnetic dipole polarizability. Moreover, we note that in the nuclear (particle) physics literature olmos_01 , the magnetic polarizability is denoted by .
Let us now define , , and to be the corresponding densities. The free-photon propagator in the gauge is landau
[TABLE]
Straightforward derivation of the photon propagator in the medium, which does not exclude different atoms being at the same positions or multiple interaction with the same atom, leads to
[TABLE]
This assumption that different atoms do not occupy the same position affects the second term in the expansion of in powers of and leads to the Clausius-Mossotti formula. So our formalism is correct only up the the first term in , the density of atoms. An identity
[TABLE]
with
[TABLE]
gives
[TABLE]
We note that and are completely independent quantities here.
III Refractive index
is the photon propagator in the medium and thus contains all the information about the electromagnetic field. In particular, the relation between the photon frequency and its wave-vector is obtained from the condition
[TABLE]
which corresponds to the pole in of the photon propagator at the fixed . The refractive index, up to the terms linear in the density of atoms is
[TABLE]
Since polarizabilities and depend on the frequency and the wave-vector , the above is a transcendental equation for . However, in the leading order of the atomic density , polarizabilities for the propagation of light in the medium can be taken at , namely, and . Accordingly, the forward-scattering amplitude for on-shell photons simplifies to
[TABLE]
We have not been able to derive the analog of the Clausius-Mossotti formula using the above propagator formalism, so we limit ourselves only to terms which are linear in the atomic density, although according to Ref. ersfeld:98 such a generalization is possible.
For the general electromagnetic field the notion of the refractive index is less appealing, because the wave equation no longer holds, as it is modified by the square of d’Alembertian. For example, the static Coulomb field is modified according to which is different from the modification of the light propagation.
IV Electrodynamics in matter
The effective action for the electromagnetic field including medium in terms of a scattering amplitude from Eq. (4) is
[TABLE]
where
[TABLE]
In general, the dependence of on is to be interpreted as a nonlocal relation between and fields forcella:17 , namely,
[TABLE]
similarly to the relation in the time domain. This nonlocal relation is obvious considering that atoms have a finite size. Indeed, the dependence of and goes with , which is assumed to be a small factor. However, when high precision is intended, as in Refs. egan:17 ; jousten:17 ; silvestri:18 , this nonlocal relation between and might play a role. For this one needs to investigate the dependence of generalized atomic polarizabilities on and the consequences on the Maxwell equations in the medium, for example, on boundary conditions close to metallic walls. Here, we aim to estimate the magnitude of the dependence on and perform calculations for the case of the helium atom.
V Retardation corrections to the generalized polarizability of the helium atom
Assuming an infinite nuclear mass and a vanishing overall electron spin and overall angular momentum, the interaction of the helium atom with an electromagnetic field is given by Eq. (2). The scattering amplitude obtained from this interaction is given by three contributions,
[TABLE]
is the scattering amplitude due to the nonrelativistic electric dipole polarizability landau ,
[TABLE]
where is defined below in Eq. (21). is due to the relativistic correction to the electric dipole polarizability and due to diamagnetic coupling,
[TABLE]
where is a correction coming from the Breit-Pauli Hamiltonian (see Refs. pachucki:01 ; puchalski:16 ). is a retardation correction,
[TABLE]
where
[TABLE]
where is the quadrupole moment operator. Only the is a new term, not considered yet in the context of the refractive index puchalski:16 . The corresponding corrections to the generalized polarizabilities are
[TABLE]
where is the fine structure constant. For small (real) photon frequencies and in atomic units
[TABLE]
where the last equation is a definition of the dimensionless coefficient .
VI Results and conclusions
Numerical calculations for He (see Tab. I) give . With m and nm egan:17 ; jousten:17 , the retardation correction to the refractive index is equal to
[TABLE]
which is larger than the previously calculated term in the expansion of the relativistic polarizability in the small , puchalski:16 . We note that the retardation correction to the refractive index significantly depends on the wavelength, and for the planned measurements at LNE silvestri:18 with nm this correction amounts to which is larger than the anticipated accuracy of these measurements.
Acknowledgements.
The authors wish to thank Karol Makuch and Bogumił Jeziorski for interesting discussions. This work was supported by the National Science Center (Poland) Grant No. 2017/27/B/ST2/02459.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1(1) P. F. Egan , J. A. Stone, J. E. Ricker, J. H. Hendricks, and G. F. Strouse, Optics Lett. 42 , 2944 (2017).
- 2(2) K. Jousten et al. , Metrologia 54 , S 146 (2017).
- 3(3) Z. Silvestri, F. Boineau, P. Otal and J. Wallerand, Helium-Based Refractometry for Pressure Measurements in the Range 1-100 k Pa, in 2018 Conference on Precision Electromagnetic Measurements (CPEM 2018) , Paris (IEEE, New York, 2018), pp. 1 and 2.
- 4(4) M. Puchalski, K. Piszczatowski, J. Komasa, B. Jeziorski, and K. Szalewicz, Phys. Rev. A 93 , 032515 (2016).
- 5(5) B. Ersfeld and B. U. Felderhof, Phys. Rev. E 57 , 1118 (1998).
- 6(6) D. Forcella, C. Prada, and R. Carminati, Phys. Rev. Lett. 118 , 134301 (2017).
- 7(7) K. Pachucki, Phys. Rev. A 69 , 052502 (2004).
- 8(8) V. Olmos de León et al. , Eur. Phys. J. A 10 , 207 (2001).
