Refractive indices of Y 2 SiO 5 in the near-infrared Journal: opticajournal Publication type: Research Article

Shijun Zhang 1, Jérôme Debray 1, Benoît Boulanger 1, Pierre Lemonde 1, Thierry Chanelière 1,* · 2026

Read Refractive indices of Y 2 SiO 5 in the near-infrared Journal: opticajournal Publication type: Research Article by Shijun Zhang 1, Jérôme Debray 1, Benoît Boulanger 1, Pierre Lemonde 1, Thierry Chanelière 1,* (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-26971.txt.

Introduction

Shijun Zhang, Jérôme Debray, Benoît Boulanger, Pierre Lemonde, and Thierry Chanelière measure how a key optical crystal bends near-infrared light. The crystal, yttrium orthosilicate, is a reference material for quantum optics, often doped with rare-earth atoms. Its refractive indices were well known only in visible light. Telecom-wavelength uses needed numbers that did not exist.

The authors first review historical measurements and their old models. Interferometric measurements in the telecom range then determined the principal refractive indices directly. A Sellmeier model, the standard equation fitting index against wavelength, is fitted across a broad range with explicit coefficients. The achievable accuracy is discussed openly. As a demonstration, white-light interference measured the thickness of a ten-micrometre crystal film precisely.

Catalog record

CREATORShijun Zhang 1, Jérôme Debray 1, Benoît Boulanger 1, Pierre Lemonde 1, Thierry Chanelière 1,*
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE31 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.26971

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Refractive indices of Y 2 SiO 5 in the near-infrared Journal: opticajournal Publication type: Research Article Shijun Zhang 1, Jérôme Debray 1, Benoît Boulanger 1, Pierre Lemonde 1, Thierry Chanelière 1,* Abstract Y 2 SiO 5 is a reference birefringent material for optical quantum technologies. The refractive index is primarily known in the visible spectrum, whereas this crystal is also used in the near-infrared. We begin by analysing historical measurements, as well as the modelling proposed at the time. The absence of refractive index tabulated values in the near- and mid-infrared ranges motivated us to carry out independent measurements. Using interferometric techniques in the telecom wavelength range, we demonstrate the ability to determine the principal refractive indices and propose a model spanning a broad spectral range based on Sellmeier equations for which we explicitly give the coefficients and discuss the achievable accuracy. Conversely, as an illustration, white-light interference enables us to precisely determine the thickness of a Y 2 SiO 5 thin film on the order of ten micrometers. 1 Introduction Rare-earth doped Y 2 SiO 5 has emerged as a promising materials for quantum photonics. There are numerous optical studies with different dopants covering the visible, the near-infrared (telecom) range. These are generally supplemented by studies of nuclear or electronic spin transitions, which provide particularly detailed knowledge of the materials. Surprisingly, the principal values of the refractive index in the dielectric frame (called b, D 1 , D 2 where b is the monoclinic crystal axis) are only approximately known because there is only a single measurement performed in the 430-650 nm range with a heavily doped Nd 3+ :Y 2 SiO 5 sample [ 2 ] (10% Nd 3+ concentration). Given the interest in ytterbium and erbium at 979 nm and 1536 nm respectively, it seems important to know the refractive indices in the infrared, especially since extrapolating Beach et al. measurements is not scientifically sound. We shall therefore begin by discussing in section 2 the limitations of the historical model and the alternative that we consider most appropriate. In the telecom range, we employ a spectral interferometric technique and use the crystal itself as a Fabry-Perot etalon. As coherent source are now widely available, interferometric methods, first in the spatial domain, have been rapidly identified as an precise alternative to the historical techniques based on refractometry [ 17 , 9 ] . Indeed, spatial fringes produced by the sample under study can be analysed to evaluate the index of refraction. Measurements in the spectral domain have also appeared early as extremely precise, with the advantage that white light interferometry (using broadband source as opposed to lasers) allows to acquire immediately the dispersion curve [ 14 , 6 , 1 ] . In that case, a standard interferometer (Michelson or Mach-Zehnder for example) is assembled and used as a measurement test-bed in which different samples can be inserted. Just to be perfectly clear, our aim is not to perform high-precision refractometry of Y 2 SiO 5 , even if the measurement techniques have been refined and specifically improved for optical crystals. The most commonly used method is the minimum deflection angle, whose accuracy meets the requirements of nonlinear optics for calculating phase-matching wavelength ranges, up to a 10^{-6} precision. It requires high-quality, well-polished and large-size crystal samples (typically ¿10 mm in size) to achieve reliable measurements. Reference [ 16 , and references therein] is a recent review and a good entry point for the literature. The technical challenges are compounded in the case of birefringent materials, for which several orientations must be investigated using different samples of equivalent qualities. In any case, the monoclinic nature of Y 2 SiO 5 imposes limitations on the precision of the index measurements. We will take the time in section 2 to discuss the accuracy of the measurements and their limitations in our case. Here, we directly use the millimeter-size crystal as a Fabry-Perot interferometer to complete in the infrared the historical index measurements. The relatively high index of Y 2 SiO 5 ( \sim 1.8 [ 2 , 4 ] ) is sufficient to observe well-contrasted fringes in the transmitted spectrum because of the facets reflection (Fresnel reflection). This technique doesn’t require any external interferometic setup but the parallelism of the opposite faces should be ensured during the polishing process. This approach has been used naturally in the mid-infrared (large wavelength) [ 18 , 20 ] or thin film samples [ 3 , 13 ] for which a moderate spectrometer resolution is sufficient. The method is actually well-adapted to the telecom range because commercially available high resolution spectrometers ( < 0.01nm) allow to observe fringes though millimetres of propagation. In turn, millimeter-size crystals can be shape as cuboids to probe three directions with the same sample as opposed to three independent thin samples (with the risk of relative disorientation of the crystalline axis between samples). In the C-band, an erbium-doped fiber amplifier (EDFA) offers an ideal broadband source with high brightness. Superluminescent diodes can alternatively be used and cover also the near-infrared with very similar properties. The experimental setup and measurements will be described in section 3 . As we shall see in this section, it is important to remember that measurements in the spectral domain, where the fringe spacing is simply the free spectral range of the etalon, give the optical path length with the group index and not the refractive index. We will see in section 4 how to use these measurements to extrapolate index values in the near-infrared. To illustrate the interest of our accurate index measurement in section 5 , we will use it to evaluate the thickness of a thin film of Y 2 SiO 5 by collecting the transmission with a commercial moderate-resolution spectrometer. 2 Extrapolating previous measurements It is important to begin by analysing the historical measurements that are still used as a reference, carried out by Beach et al. in 1990 [ 2 ] . They cover six wavelengths between 436nm and 644nm using spectral lamps (Hg, Cd, Na) (Fig. 1 ) and enable the dependencies of the principal indices to be obtained using a Sellmeier formula whose validity we will discuss a little later. We recall it here and will use it throughout the rest of this paper. n_{i}(\lambda)^{2}=A_{i}+\frac{B_{i}}{\lambda^{2}+C_{i}}+D_{i}\lambda^{2} (1) \lambda is the wavelength expressed in \mu m and A_{i} , B_{i} , C_{i} , D_{i} , the Sellmeier coefficients along the axis i (where i= b , D 1 or D 2 ), as tabulated below: axis i A_{i} B_{i} C_{i} D_{i} n_{b} 3.0895 0.0334 0.0043 0.0199 n_{D_{1}} 3.1173 0.0283 -0.0133 0.00 n_{D_{2}} 3.1871 0.03022 -0.0138 0.00 Table 1: Sellmeier coefficients along the axes b , D 1 , D 2 from Beach et al. [ 2 ] . Regardless of how accurately the parameters are fitted, it is important to discuss the physical significance of the equation 1 . This is the two-pole Sellmeier dispersion formula [ 7 ] , with an ultraviolet pole (coefficients B_{i} and C_{i} ) and an mid-infrared pole (coefficient D_{i} , assuming \lambda far from resonance). It is not necessary to extend to three poles given the desired precision, which we will discuss at the

Read the complete text