Complex Refractive Index Determination via Microspectroscopy Through Magnifying Optics: Challenges and Opportunities

arXiv:2607.03077 · physics.app-ph · Submitted 2026-07-03 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.

Kai: We're looking at the paper: "Complex Refractive Index Determination via Microspectroscopy Through Magnifying Optics".

Mira: Here's the gist of the paper. Complex refractive index determination via microspectroscopy presents a fast and non-destructive approach for extracting refractive indices from reflectance and transmittance spectra of samples with lateral dimensions down to the micrometer scale <ref:2607.03077#pg2> The authors present a numerical formalism explicitly accounting for the influence of the numerical aperture, allowing for precise retrieval of the refractive index without resorting to dispersion models <ref:2607.03077#pg5>.

Kai: Microspectroscopy Methodology Microspectroscopy enables the assessment of the optical response of multilayer stacks using high-magnification optics with correspondingly large numerical apertures <ref:2607.03077#pg2> This technique couples a spectrometer with a glass fiber to a standard optical microscope, enabling the assessment of reflectance and/or transmittance spectra with a spot size down to 1 µm to 2 µm <ref:2607.03077#pg3>.

Mira: The measurement setup involves an Olympus BX53-MTRF microscope coupled to a Horiba iHR 320 spectrometer, utilizing various objective lenses with different numerical apertures (NAs) <ref:2607.03077#pg2> Linearly polarized light is achieved by inserting a continuously rotatable linear polarizer in the illumination path <ref:2607.03077#pg3>.

Kai: Modeling and Data Processing The numerical procedure involves the inversion of the modeling process for determining layer thickness or crystal axis orientation when refractive indices are known <ref:2607.03077#pg2> The influence of the NA is considered by an objective lens dependent weighting function WTot, which comprises a geometric component based on the objective lens’s NA and a lens-dependent Gaussian shaped component defined by the parameter ζ <ref:2607.03077#pg3>.

Paper summary: Kai: So, looking at the paper "Complex Refractive Index Determination via Microspectroscopy Through Magnifying Optics: Challenges and Opportunities" and the authors Julian Schwarz, Johannes Bauer, Mathias Rommel, and Andreas Hutzler, what's your take on what they actually built here?

Mira: Well, it’s about getting those complex refractive indices from reflectance and transmittance data using microspectroscopy. They show that this is a fast way to do it for samples down to the micrometer scale.

Lev: It seems like they are tackling some real problems with traditional methods, especially when you look at materials like van der Waals crystals or oriented graphite flakes.

Kai: Right, and their approach explicitly accounts for the numerical aperture, which is a big deal because that NA actually changes how much light you capture during the measurement.

Mira: They use a specific numerical formalism to get the refractive index without needing those dispersion models they usually rely on. That’s pretty neat for experimental setups where you don't want to model everything from scratch.

Kai: And the caveat is that for those graphite flakes, the method struggles with probing the out-of-plane refractive indices because they aren't very sensitive to them. That’s a limitation you have to keep in mind for any real hardware application.

Lev: For us, it means if you’re building a device that relies on these nanoscale materials, you need to consider how this method handles those thickness variations and polarization effects. That leads us into how we actually implement this kind of measurement in the lab.

Conclusion: Kai: So, we're wrapping up our look at "Complex Refractive Index Determination via Microspectroscopy Through Magnifying Optics: Challenges and Opportunities" by Julian Schwarz, Johannes Bauer, Mathias Rommel, and Andreas Hutzler.

Mira: This whole paper is about moving the determination of complex refractive indices away from those heavy dispersion models that spectroscopic ellipsometry usually forces you into.

Lev: It focuses on using microspectroscopy to get those indices quickly from reflectance and transmittance spectra measured at the micrometer scale.

Kai: The authors show this numerical formalism explicitly accounts for how much light you capture based on the numerical aperture, or NA, which is a big part of the setup.

Mira: They demonstrate this works for things like transparent glass and absorbing silicon carbide, but they also test it on those tricky flakes of highly oriented pyrolytic graphite and molybdenum trioxide.

Lev: And the caveat there is that for those graphite flakes, the method struggles with probing the out-of-plane refractive indices because they aren't very sensitive to them. That’s a limitation you have to keep in mind for any real hardware application.

Kai: So, it’s promising as an alternative to spectroscopic ellipsometry because of how simple the setup is, but you still get those ambiguities with multiple solution branches if you don't know your material beforehand.

Mira: Exactly, and they discuss ways forward by suggesting we might assume the out-of-plane component gets less important as the flake gets thinner. That’s a key idea for future work.

Lev: For us, it means if you’re building a device that relies on these nanoscale materials, you need to consider how this method handles those thickness variations and polarization effects. That leads us into how we actually implement this kind of measurement in the lab.

Julian Schwarz, Johannes Bauer, Mathias Rommel, Andreas Hutzler

Electron Devices, Friedrich-Alexander-Universität Erlangen-Nürnberg · Fraunhofer Institute for Integrated Systems and Device Technology IISB · Forschungszentrum Jülich GmbH, Helmholtz Institute Erlangen-Nürnberg for Renewable Energy (IET-2)

physics.app-ph

Submitted: 2026-07-03

Updated: 2026-07-03

DOI: 10.1002/smtd.71167

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 79/100

The gist: Complex refractive index determination via microspectroscopy presents a fast and non-destructive approach for extracting refractive indices from reflectance and transmittance spectra of samples with…

Key concepts

Microspectroscopy
This technique combines a standard optical microscope with a spectrometer to measure the optical response of samples at very small spots (1-2 µm). It allows researchers to assess reflectance or transmittance spectra on samples with lateral dimensions down to the micrometer scale, providing high-magnification optical data.
Numerical Formalism Accounting for NA
The method uses a numerical procedure that specifically incorporates the influence of the numerical aperture (NA) of the objective lens. This is done through a weighting function that combines geometric and Gaussian components, enabling accurate retrieval of refractive indices directly from measurements without relying on pre-existing dispersion models.
Transfer Matrix Method (TMM)
The TMM modeling is used to simulate how light interacts with multilayer stacks. It employs Fresnel-based equations to accurately reproduce the detected multiply reflected light, which is crucial for correctly extracting the complex refractive indices from the measured spectra.

Terminology

Summary

Complex refractive index determination via microspectroscopy presents a fast and non-destructive approach for extracting refractive indices from reflectance and transmittance spectra of samples with lateral dimensions down to the micrometer scale <ref:2607. p. 0

Where the paper says it

Complex Refractive Index Determination via Microspectroscopy Through Magnifying Optics: Challenges and Opportunities Julian Schwarz ∗1 , Johannes Bauer 1 , Mathias Rommel ∗2 , and Andreas Hutzler 3 1Electron Devices, Friedrich-Alexander-Universität Erlangen-Nürnberg, Cauerstraße 6, 91058 Erlangen, Germany 2Fraunhofer Institute for Integrated Systems and Device Technology IISB, Schottkystraße 10, 91058 Erlangen, Germany 3Forschungszentrum Jülich GmbH, Helmholtz Institute Erlangen-Nürnberg for Renewable Energy (IET-2), Cauerstraße 10, 91058 Erlangen, Germany Abstract For the design and optimization of optoelectronic devices, accurate knowledge of the complex refractive indices of the constituent materials is essential. Herein, we present a fast and non-destructive approach for the extraction of the refractive indices from reflectance and transmittance spectra of samples with lateral dimensions down to the micrometer scale.

03077#pg2> The authors present a numerical formalism explicitly accounting for the influence of the numerical aperture, allowing for precise retrieval of the refractive index without resorting to dispersion models <ref:2607. p. 0
Where the paper says it

Microspectroscopy, based on the combination of a standard optical microscope and a spectrometer, enables the assessment of the optical response of multilayer stacks using high-magnification optics with correspondingly large numerical apertures. Employing a numerical formalism explicitly accounting for the influence of the numerical aperture, allows for precise retrieval of the refractive index without resorting to dispersion models.

03077#pg5>.

Microspectroscopy Methodology

Microspectroscopy enables the assessment of the optical response of multilayer stacks using high-magnification optics with correspondingly large numerical apertures <ref:2607.03077#pg2> This technique couples a spectrometer with a glass fiber to a standard optical microscope, enabling the assessment of reflectance and/or transmittance spectra with a spot size down to 1 µm to 2 µm <ref:2607. p. 1

Where the paper says it

An established technique enabling the optical characterization of samples on the micrometer scale is microspectroscopy [26–30], which can be easily implemented by coupling a spectrometer with a glass fiber to a standard optical microscope [31]. In this way the reflectance and/or transmittance spectra are assesed with a spot size down to 1 µm to 2 µm [28, 32].

03077#pg3>. The measurement setup involves an Olympus BX53-MTRF microscope coupled to a Horiba iHR 320 spectrometer, utilizing various objective lenses with different numerical apertures (NAs) <ref:2607. p. 2
Where the paper says it

1 Measurement setup and modus operandi The schematic of the employed measurement setup is shown in Fig. Reflectance and transmittance were measured with an Olympus BX53-MTRF microscope coupled to a Horiba iHR 320 spectrometer including a 1200 grooves/mm diffraction grating.

03077#pg2> Linearly polarized light is achieved by inserting a continuously rotatable linear polarizer in the illumination path <ref:2607. p. 2
Where the paper says it

Due to the characteristics of the halogen light sources and the Horiba Synapse CCD detector, the available spectral range is limited from 440 nm to 950 nm. Linearly polarized light was achieved by the insertion of a continuously rotatable linear polarizer in the illumination path.

03077#pg3>.

Modeling and Data Processing

The numerical procedure involves the inversion of the modeling process for determining layer thickness or crystal axis orientation when refractive indices are known <ref:2607.03077#pg2> The influence of the NA is considered by an objective lens dependent weighting function WTot, which comprises a geometric component based on the objective lens’s NA and a lens-dependent Gaussian shaped component defined by the parameter ζ <ref:2607. p. 3

Where the paper says it

these AOIs are weighted as per the definition of WTot. WTot comprises a geometric component based on the objective lens’s NA and a lens-dependent Gaussian shaped component, defined by the parameter ζ, accounting for potential inhomogeneities of the intensity between the individual AOIs.

03077#pg3>. For thick transparent substrates, a developed approach denoted partial backside reflectance - micro(spectro)scopy (PBRM) is applied to account for the partial detection of multiply reflected light <ref:2607. p. 3
Where the paper says it

Besides the incoherent behavior without interference fringes for thick substrates, multiply reflected light is either completely, partially, or not at all detected. In order to account for this, we apply our developed approach, denoted partial backside reflectance - micro(spectro)scopy (PBRM) [32], to both reflectance and transmittance.

03077#pg2>. The 4×4 transfer matrix method (TMM) modeling is employed, with Fresnel-based equations used to simulate individual R(αn) and T(αn) to accurately reproduce the detected multiply reflected light <ref:2607. p. 3
Where the paper says it

5 %, the error for SrTiO3 amounts to ca. As a consequence, Fresnel-based equations were used to simulate the individual R(αn) and T(αn) to accurately reproduce the detected multiply reflected light, since this approach allows the different orders of reflected and transmitted light to be separated in the calculations [47].

03077#pg3>.

Material Characterization Results

The method was demonstrated for several sample types, including large-area, homogeneous, optically incoherent samples such as transparent glasses and absorbing 4H-SiC <ref:2607.03077#pg2>. Precise refractive indices matching reference dispersions were extracted for nonabsorbing materials like BF33 and Al2O3 for the 10x and 100x objective lenses <ref:2607. p. 5

Where the paper says it

S8 demonstrate the feasibility of the proposed method. So, precise refractive indices matching the reference dispersions were extracted for these nonabsorbing materials for the 10x and 100x objective lenses.

03077#pg5>. For thick absorbing samples, reflectance and transmittance are required to unambiguously determine n and k, with simulations accurately replicating measurements for a c-plane oriented 4H-SiC sample <ref:2607. p. 5
Where the paper says it

On the contrary, for thick absorbing samples, reflectance and transmittance are required to unambiguously determine n and k. As an example, a c-plane oriented 4H-SiC sample was investigated, whereby the 4 ◦ off-axis cut was neglected.

03077#pg4>.

Anisotropy and Limitations

For flakes of highly oriented pyrolytic graphite (HOPG) and MoO3, anisotropic refractive indices were chosen from literature, and the isotropic approximation was applied for HOPG due to low NA objective lenses <ref:2607.03077#pg5>. The ambiguity of multiple solution branches for n and k in these flakes is a fundamental limitation of the method when no a priori knowledge is available <ref:2607. p. 6

Where the paper says it

The branch with the higher n, denoted as branch A, corresponds to the smaller k, and vice versa. Such ambiguity constitutes a fundamental limitation of the method.

03077#pg9>. The introduction of linearly polarized light enables the extraction of refractive indices along the in-plane crystal axes of biaxial materials like MoO3 <ref:2607. p. 7
Where the paper says it

Finally, the introduction of linearly polarized light enables the extraction of the refractive indices along the in-plane crystal axes of biaxial materials. Since linearly polarized light parallel to the in-plane axes effectively reduces the biaxial behavior to that of a uniaxial material like HOPG, the determined material properties are again, in principle, isotropic approximations showing thickness-dependent impact of the out-of-plane component.

03077#pg10>. The paper discusses possible extensions towards retrieving the full anisotropic tensor of the refractive index <ref:2607. p. 0
Where the paper says it

While the results prove excellent agreement with values reported in literature, the case of graphite highlights the limitation for probing the out-of-plane refractive indices due to reduced sensitivity. Finally, we discuss possible extensions towards retrieving the full anisotropic tensor of the refractive index, establishing the proposed approach as a methodologically sound alternative to spectroscopic ellipsometry.

03077#pg2>.

Conclusion and Future Directions

The proposed method is presented as a promising alternative to state-of-the-art methods for determining material dispersion, with a comparatively simple setup <ref:2607.03077#pg2>. The approach successfully resolves ambiguities in the SixNy thin film by superimposing solution spaces for lenses with different NAs <ref:2607. p. 9

Where the paper says it

Moreover, the intricate extinction coefficients of a thick absorbing 4H-SiC substrate were retrieved with high accuracy. The ambiguous solutions for the dispersion of a SixNy film, due to the interference pattern and the numerical nature of the approach, could be resolved by superimposing the solution spaces for lenses with different NAs.

03077#pg10>. The method shows excellent agreement with literature values, and the limitations regarding low sensitivity to the out-of-plane component are discussed <ref:2607. p. 0
Where the paper says it

We demonstrate the applicability of the proposed method for large-area, homogeneous, optically incoherent samples such as transparent glasses and absorbing 4H-SiC, for a SixNy thin film on glass substrate, and for mechanically exfoliated flakes of highly oriented pyrolytic graphite and MoO3 , as representatives of uniaxial and biaxial optical anisotropy. While the results prove excellent agreement with values reported in literature, the case of graphite highlights the limitation for probing the out-of-plane refractive indices due to reduced sensitivity.

03077#pg2>. The study concludes by proposing approaches such as assuming the influence of the out-of-plane component diminishes with decreasing flake thickness <ref:2607. p. 9
Where the paper says it

In principle, several approaches can be considered for the simultaneous extraction of the in-plane and out-of-plane complex refractive indices from microspectroscopic measurements. One possibility is based on the assumption that the influence of the out-of-plane component diminishes with decreasing flake thickness.

03077#pg2>.

--- Page 1 ---

Complex Refractive Index Determination via Microspectroscopy Through Magnifying Optics: Challenges and Opportunities <ref:2607.03077#pg2>

Through Magnifying Optics: Challenges and Opportunities <ref:2607.03077#pg2>

Julian Schwarz, Johannes Bauer, Mathias Rommel, and Andreas Hutzler are the authors of this paper <ref:2607.03077#pg2>.

Abstract For the design and optimization of optoelectronic devices, accurate knowledge of the complex refractive indices of the constituent materials is essential <ref:2607. p. 0

Where the paper says it

Complex Refractive Index Determination via Microspectroscopy Through Magnifying Optics: Challenges and Opportunities Julian Schwarz ∗1 , Johannes Bauer 1 , Mathias Rommel ∗2 , and Andreas Hutzler 3 1Electron Devices, Friedrich-Alexander-Universität Erlangen-Nürnberg, Cauerstraße 6, 91058 Erlangen, Germany 2Fraunhofer Institute for Integrated Systems and Device Technology IISB, Schottkystraße 10, 91058 Erlangen, Germany 3Forschungszentrum Jülich GmbH, Helmholtz Institute Erlangen-Nürnberg for Renewable Energy (IET-2), Cauerstraße 10, 91058 Erlangen, Germany Abstract For the design and optimization of optoelectronic devices, accurate knowledge of the complex refractive indices of the constituent materials is essential. Herein, we present a fast and non-destructive approach for the extraction of the refractive indices from reflectance and transmittance spectra of samples with lateral dimensions down to the micrometer scale.

03077#pg2>. Herein, we present a fast and non-destructive approach for the extraction of the refractive indices from reflectance and transmittance spectra of samples with lateral dimensions down to the micrometer scale <ref:2607. p. 0
Where the paper says it

Complex Refractive Index Determination via Microspectroscopy Through Magnifying Optics: Challenges and Opportunities Julian Schwarz ∗1 , Johannes Bauer 1 , Mathias Rommel ∗2 , and Andreas Hutzler 3 1Electron Devices, Friedrich-Alexander-Universität Erlangen-Nürnberg, Cauerstraße 6, 91058 Erlangen, Germany 2Fraunhofer Institute for Integrated Systems and Device Technology IISB, Schottkystraße 10, 91058 Erlangen, Germany 3Forschungszentrum Jülich GmbH, Helmholtz Institute Erlangen-Nürnberg for Renewable Energy (IET-2), Cauerstraße 10, 91058 Erlangen, Germany Abstract For the design and optimization of optoelectronic devices, accurate knowledge of the complex refractive indices of the constituent materials is essential. Herein, we present a fast and non-destructive approach for the extraction of the refractive indices from reflectance and transmittance spectra of samples with lateral dimensions down to the micrometer scale.

03077#pg2>. Microspectroscopy, based on the combination of a standard optical microscope and a spectrometer, enables the assessment of the optical response of multilayer stacks using high-magnification optics with correspondingly large numerical apertures <ref:2607. p. 0
Where the paper says it

Herein, we present a fast and non-destructive approach for the extraction of the refractive indices from reflectance and transmittance spectra of samples with lateral dimensions down to the micrometer scale. Microspectroscopy, based on the combination of a standard optical microscope and a spectrometer, enables the assessment of the optical response of multilayer stacks using high-magnification optics with correspondingly large numerical apertures.

03077#pg2>. Employing a numerical formalism explicitly accounting for the influence of the numerical aperture, allows for precise retrieval of the refractive index without resorting to dispersion models <ref:2607. p. 0
Where the paper says it

Microspectroscopy, based on the combination of a standard optical microscope and a spectrometer, enables the assessment of the optical response of multilayer stacks using high-magnification optics with correspondingly large numerical apertures. Employing a numerical formalism explicitly accounting for the influence of the numerical aperture, allows for precise retrieval of the refractive index without resorting to dispersion models.

03077#pg2>. We demonstrate the applicability of the proposed method for large-area, homogeneous, optically incoherent samples such as transparent glasses and absorbing 4H-SiC, for a SixNy thin film on glass substrate, and for mechanically exfoliated flakes of highly oriented pyrolytic graphite and MoO3 <ref:2607. p. 0
Where the paper says it

Employing a numerical formalism explicitly accounting for the influence of the numerical aperture, allows for precise retrieval of the refractive index without resorting to dispersion models. We demonstrate the applicability of the proposed method for large-area, homogeneous, optically incoherent samples such as transparent glasses and absorbing 4H-SiC, for a SixNy thin film on glass substrate, and for mechanically exfoliated flakes of highly oriented pyrolytic graphite and MoO3 , as representatives of uniaxial and biaxial optical anisotropy.

03077#pg2>. While the results prove excellent agreement with values reported in literature, the case of graphite highlights the limitation for probing the out-of-plane refractive indices due to reduced sensitivity <ref:2607. p. 0
Where the paper says it

We demonstrate the applicability of the proposed method for large-area, homogeneous, optically incoherent samples such as transparent glasses and absorbing 4H-SiC, for a SixNy thin film on glass substrate, and for mechanically exfoliated flakes of highly oriented pyrolytic graphite and MoO3 , as representatives of uniaxial and biaxial optical anisotropy. While the results prove excellent agreement with values reported in literature, the case of graphite highlights the limitation for probing the out-of-plane refractive indices due to reduced sensitivity.

03077#pg2>. Finally, we discuss possible extensions towards retrieving the full anisotropic tensor of the refractive index, establishing the proposed approach as a methodologically sound alternative to spectroscopic ellipsometry <ref:2607. p. 0
Where the paper says it

While the results prove excellent agreement with values reported in literature, the case of graphite highlights the limitation for probing the out-of-plane refractive indices due to reduced sensitivity. Finally, we discuss possible extensions towards retrieving the full anisotropic tensor of the refractive index, establishing the proposed approach as a methodologically sound alternative to spectroscopic ellipsometry.

03077#pg2>.

Keywords Microspectroscopy · High numerical aperture · Van der waals materials · Refractive index · Material properties · Transmittance Reflectance

--- Page 2 ---

The precise knowledge of the material properties is vital for the design and optimization of devices build from next-generation materials like van der Waals crystals, exploiting their unique optical and electronic features like tunable bandgaps [1], high carrier mobilities [2], and inherent anisotropy [3–5] for polarization-sensitive devices [6, 7] <ref:2607.03077#pg3>. The complex refractive index, consisting of the real part, the refractive index n, and the imaginary part, the extinction coefficient k, is commonly determined by spectroscopic ellipsometry (SE) [3–12–14], enabling the simultaneous determination of both quantities as well as the thickness [15] <ref:2607. p. 0

Where the paper says it

For tailoring the characteristics in specific applications employing van der Waals materials like field-effect transistors [2, 6], photodetectors [8, 9] and waveplates [10, 11], the (complex) refractive index is a crucial parameter for selecting appropriate materials as it shapes the light-matter interaction in (opto)electronic devices. The complex refractive index, consisting of the real part, the refractive index n, and the imaginary part, the extinction coefficient k, is commonly determined by spectroscopic ellipsometry (SE) [3, 12–14], enabling the simultaneous determination of both quantities as well as the thickness [15].

03077#pg3>. However, the application of classical SE is limited to large samples with lateral dimensions of several hundred micrometers or more [14, 16] <ref:2607. p. 0
Where the paper says it

The complex refractive index, consisting of the real part, the refractive index n, and the imaginary part, the extinction coefficient k, is commonly determined by spectroscopic ellipsometry (SE) [3, 12–14], enabling the simultaneous determination of both quantities as well as the thickness [15]. However, the application of classical SE is limited to large samples with lateral dimensions of several hundred micrometers or more [14, 16].

03077#pg3>.

∗Corresponding authors: julian. p. 0

Where the paper says it

However, the application of classical SE is limited to large samples with lateral dimensions of several hundred micrometers or more [14, 16]. ∗Corresponding authors: julian.

schwarz@fau. p. 0
Where the paper says it

∗Corresponding authors: julian. schwarz@fau.

de, mathias. p. 0
Where the paper says it

Complex Refractive Index Determination via Microspectroscopy Through Magnifying Optics: Challenges and Opportunities Julian Schwarz ∗1 , Johannes Bauer 1 , Mathias Rommel ∗2 , and Andreas Hutzler 3 1Electron Devices, Friedrich-Alexander-Universität Erlangen-Nürnberg, Cauerstraße 6, 91058 Erlangen, Germany 2Fraunhofer Institute for Integrated Systems and Device Technology IISB, Schottkystraße 10, 91058 Erlangen, Germany 3Forschungszentrum Jülich GmbH, Helmholtz Institute Erlangen-Nürnberg for Renewable Energy (IET-2), Cauerstraße 10, 91058 Erlangen, Germany Abstract For the design and optimization of optoelectronic devices, accurate knowledge of the complex refractive indices of the constituent materials is essential. Herein, we present a fast and non-destructive approach for the extraction of the refractive indices from reflectance and transmittance spectra of samples with lateral dimensions down to the micrometer scale.

rommel@iisb. p. 0
Where the paper says it

Complex Refractive Index Determination via Microspectroscopy Through Magnifying Optics: Challenges and Opportunities Julian Schwarz ∗1 , Johannes Bauer 1 , Mathias Rommel ∗2 , and Andreas Hutzler 3 1Electron Devices, Friedrich-Alexander-Universität Erlangen-Nürnberg, Cauerstraße 6, 91058 Erlangen, Germany 2Fraunhofer Institute for Integrated Systems and Device Technology IISB, Schottkystraße 10, 91058 Erlangen, Germany 3Forschungszentrum Jülich GmbH, Helmholtz Institute Erlangen-Nürnberg for Renewable Energy (IET-2), Cauerstraße 10, 91058 Erlangen, Germany Abstract For the design and optimization of optoelectronic devices, accurate knowledge of the complex refractive indices of the constituent materials is essential. Herein, we present a fast and non-destructive approach for the extraction of the refractive indices from reflectance and transmittance spectra of samples with lateral dimensions down to the micrometer scale.

fraunhofer. p. 0
Where the paper says it

rommel@iisb. fraunhofer.

de, a.hutzler@fz-juelich. p. 0
Where the paper says it

fraunhofer. hutzler@fz-juelich.

de <ref:2607. p. 0
Where the paper says it

hutzler@fz-juelich. de 1 arXiv:2607. p. 0

Where the paper says it

hutzler@fz-juelich. de 1 arXiv:2607.

03077#pg2>.

arXiv:2607.03077v1 [physics. p. 0

Where the paper says it

de 1 arXiv:2607. 03077v1 [physics.

app-ph] 3 Jul 2026 <ref:2607. p. 0
Where the paper says it

03077v1 [physics. app-ph] 3 Jul 2026

03077#pg2>.

--- Page 3 ---

This requirement is often not met for flakes of van der Waals materials produced by mechanical exfoliation, whose typical lateral dimensions range from a few to a few tens of micrometers <ref:2607.03077#pg3>. Hence, more specialized setups with smaller probe sizes are required for the characterization of such flakes, which can be achieved by imaging SE [17, 18], microspot SE [19, 20], and Fourier plane micro-ellipsometry [21, 22], among others <ref:2607. p. 1

Where the paper says it

This requirement is often not met for flakes of van der Waals materials produced by mechanical exfoliation, whose typical lateral dimensions range from a few to a few tens of micrometers. Hence, more specialized setups with smaller probe sizes are required for the characterization of such flakes, which can be achieved by imaging SE [17, 18], microspot SE [19, 20], and Fourier plane micro-ellipsometry [21, 22], among others.

03077#pg4>. But, these advanced methods entail setups with increasing complexity, prolonged measurement times, or complex data analysis and calibration <ref:2607. p. 1
Where the paper says it

Hence, more specialized setups with smaller probe sizes are required for the characterization of such flakes, which can be achieved by imaging SE [17, 18], microspot SE [19, 20], and Fourier plane micro-ellipsometry [21, 22], among others. But, these advanced methods entail setups with increasing complexity, prolonged measurement times, or complex data analysis and calibration.

03077#pg4>. Moreover, SE typically relies on optical dispersion models, such as the Drude–Lorentz [23, 24], Tauc–Lorentz [3, 14], and Sellmeier [25] formalism, especially for the simultaneous determination of the complex refractive indices and thicknesses of the materials under investigation <ref:2607. p. 1
Where the paper says it

But, these advanced methods entail setups with increasing complexity, prolonged measurement times, or complex data analysis and calibration. Moreover, SE typically relies on optical dispersion models, such as the Drude–Lorentz [23, 24], Tauc–Lorentz [3, 14], and Sellmeier [25] formalism, especially for the simultaneous determination of the complex refractive indices and thicknesses of the materials under investigation.

03077#pg3>.

An established technique enabling the optical characterization of samples on the micrometer scale is microspectroscopy [26–30], which can be easily implemented by coupling a spectrometer with a glass fiber to a standard optical microscope [31] <ref:2607. p. 1

Where the paper says it

Moreover, SE typically relies on optical dispersion models, such as the Drude–Lorentz [23, 24], Tauc–Lorentz [3, 14], and Sellmeier [25] formalism, especially for the simultaneous determination of the complex refractive indices and thicknesses of the materials under investigation. An established technique enabling the optical characterization of samples on the micrometer scale is microspectroscopy [26–30], which can be easily implemented by coupling a spectrometer with a glass fiber to a standard optical microscope [31].

03077#pg3>. In this way the reflectance and/or transmittance spectra are assesed with a spot size down to 1 µm to 2 µm [28, 32] <ref:2607. p. 1
Where the paper says it

An established technique enabling the optical characterization of samples on the micrometer scale is microspectroscopy [26–30], which can be easily implemented by coupling a spectrometer with a glass fiber to a standard optical microscope [31]. In this way the reflectance and/or transmittance spectra are assesed with a spot size down to 1 µm to 2 µm [28, 32].

03077#pg2>. From such measurements, parameters like the thickness [31, 33, 34] or the complex refractive indices [35–38] can be deduced by comparison with comprehensive modeling <ref:2607. p. 1
Where the paper says it

In this way the reflectance and/or transmittance spectra are assesed with a spot size down to 1 µm to 2 µm [28, 32]. From such measurements, parameters like the thickness [31, 33, 34] or the complex refractive indices [35–38] can be deduced by comparison with comprehensive modeling.

03077#pg3>. If linearly polarized light is applied, even the anisotropic optical properties [4, 10, 11, 35, 39] and the orientation of the crystal axes can be determined <ref:2607. p. 1
Where the paper says it

From such measurements, parameters like the thickness [31, 33, 34] or the complex refractive indices [35–38] can be deduced by comparison with comprehensive modeling. If linearly polarized light is applied, even the anisotropic optical properties [4, 10, 11, 35, 39] and the orientation of the crystal axes can be determined.

03077#pg3>. However, the modeling is often based on the assumption of normal incidence, which may lead to significant errors in the extracted parameters for objective lenses with high numerical apertures (NAs) and the accompanying large range of angles of incidence (AOIs) <ref:2607. p. 1
Where the paper says it

If linearly polarized light is applied, even the anisotropic optical properties [4, 10, 11, 35, 39] and the orientation of the crystal axes can be determined. However, the modeling is often based on the assumption of normal incidence, which may lead to significant errors in the extracted parameters for objective lenses with high numerical apertures (NAs) and the accompanying large range of angles of incidence (AOIs).

03077#pg3>. Additionally, the utilization of transmittance measurements is underrepresented in literature <ref:2607. p. 1
Where the paper says it

However, the modeling is often based on the assumption of normal incidence, which may lead to significant errors in the extracted parameters for objective lenses with high numerical apertures (NAs) and the accompanying large range of angles of incidence (AOIs). Additionally, the utilization of transmittance measurements is underrepresented in literature, whereby especially a lack of detailed description of the measurement process for such measurements is evident.

03077#pg2>.

For this reason, we present an in-depth study on the purely numerical extraction of the (complex) refractive index from reflectance and/or transmittance spectra obtained by microspectroscopy under arbitrary NA <ref:2607. p. 1

Where the paper says it

Additionally, the utilization of transmittance measurements is underrepresented in literature, whereby especially a lack of detailed description of the measurement process for such measurements is evident. For this reason, we present an in-depth study on the purely numerical extraction of the (complex) refractive index from reflectance and/or transmittance spectra obtained by microspectroscopy under arbitrary NA.

03077#pg2>. Therefore, the extensive modular modeling framework from our previous works [31, 32, 40–42] is employed and refined to account for transmitted light <ref:2607. p. 1
Where the paper says it

For this reason, we present an in-depth study on the purely numerical extraction of the (complex) refractive index from reflectance and/or transmittance spectra obtained by microspectroscopy under arbitrary NA. Therefore, the extensive modular modeling framework from our previous works [31, 32, 40–42] is employed and refined to account for transmitted light.

03077#pg2>. In the course of this work, our objective lens dependent modeling for the influence of the NA [31] is transferred from reflectance to transmittance measurements <ref:2607. p. 1
Where the paper says it

Therefore, the extensive modular modeling framework from our previous works [31, 32, 40–42] is employed and refined to account for transmitted light. In the course of this work, our objective lens dependent modeling for the influence of the NA [31] is transferred from reflectance to transmittance measurements, as well as the developed approach accounting for the partial detection of multiply reflected light in thick transparent substrates with incoherent behavior [32].

03077#pg2>. By incorporating linearly polarized light in the measurements and implementing it into modeling [42], we demonstrate the determination of the refractive indices along the different in-plane crystal axes of anisotropic materials <ref:2607. p. 1
Where the paper says it

In the course of this work, our objective lens dependent modeling for the influence of the NA [31] is transferred from reflectance to transmittance measurements, as well as the developed approach accounting for the partial detection of multiply reflected light in thick transparent substrates with incoherent behavior [32]. On top of that, by incorporating linearly polarized light in the measurements and implementing it into modeling [42], we demonstrate the determination of the refractive indices along the different in-plane crystal axes of anisotropic materials.

03077#pg2>. Lastly, we discuss the chances and limitations of microspectroscopy for the accurate extraction of the full anisotropic tensor of refractive indices <ref:2607. p. 1
Where the paper says it

On top of that, by incorporating linearly polarized light in the measurements and implementing it into modeling [42], we demonstrate the determination of the refractive indices along the different in-plane crystal axes of anisotropic materials. Lastly, the chances and limitations of microspectroscopy for the accurate extraction of the full anisotropic tensor of refractive indices are discussed.

03077#pg2>. We start with the measurement and modeling procedure for reflectance and, specificially, transmittance spectra <ref:2607. p. 1
Where the paper says it

Lastly, the chances and limitations of microspectroscopy for the accurate extraction of the full anisotropic tensor of refractive indices are discussed. We start with the measurement and modeling procedure for reflectance and, specificially, transmittance spectra.

03077#pg2>. Building on this, we present extracted complex refractive indices for several thick incoherent substrates, both non-absorbing and absorbing, as well as for a SixNy thin film on a glass substrate, and flakes of the uniaxial van der Waals material highly oriented pyrolytic graphite (HOPG) and the biaxial representative MoO3 <ref:2607. p. 1
Where the paper says it

We start with the measurement and modeling procedure for reflectance and, specificially, transmittance spectra. Building on this, we present extracted complex refractive indices for several thick incoherent substrates, both non-absorbing and absorbing, as well as for a SixNy thin film on a glass substrate, and flakes of the uniaxial van der Waals material highly oriented pyrolytic graphite (HOPG) and the biaxial representative MoO3 .

03077#pg2>.

--- Page 4 ---

In this work, we investigated several optically thick substrates <ref:2607.03077#pg3>. We assumed purely specular reflection and transmission, as all samples were double side polished <ref:2607. p. 1

Where the paper says it

1 Sample preparation and reference characterization In this work, we investigated several optically thick substrates. We assumed purely specular reflection and transmission, as all samples were double side polished.

03077#pg3>. The samples were 500 µm thick BF33 glass wafers from Schott AG [43], 100 µm AF 32eco thin wafers from Schott AG [44] <ref:2607. p. 1
Where the paper says it

We assumed purely specular reflection and transmission, as all samples were double side polished. The samples were 500 µm thick BF33 glass wafers from Schott AG [43], 100 µm AF 32eco thin wafers from Schott AG [44], 500 µm thick Al2O3 windows from S.

03077#pg3>, 500 µm thick Al2O3 windows from S. p. 1
Where the paper says it

We assumed purely specular reflection and transmission, as all samples were double side polished. The samples were 500 µm thick BF33 glass wafers from Schott AG [43], 100 µm AF 32eco thin wafers from Schott AG [44], 500 µm thick Al2O3 windows from S.

A.F.I.R., and 500 µm thick SrTiO3 substrates from CRYSTAL GmbH <ref:2607. p. 1
Where the paper says it

The samples were 500 µm thick BF33 glass wafers from Schott AG [43], 100 µm AF 32eco thin wafers from Schott AG [44], 500 µm thick Al2O3 windows from S. , and 500 µm thick SrTiO3 substrates from CRYSTAL GmbH.

03077#pg3>. A SixNy thin film was deposited on a BOROFLOAT 33 glass (BF33) wafer as a single-wafer process using plasma-enhanced chemical vapor deposition (PECVD) at a deposition temperature of 130 ◦C with NH3 and SiH4 as precursors <ref:2607. p. 1
Where the paper says it

, and 500 µm thick SrTiO3 substrates from CRYSTAL GmbH. A SixNy thin film was deposited on a BOROFLOAT 33 glass (BF33) wafer as a single-wafer process using plasma-enhanced chemical vapor deposition (PECVD) at a deposition temperature of 130 ◦C with NH3 and SiH4 as precursors.

03077#pg3>. The ca. 500 nm thick layer was applied in two consecutive 250 nm deposition procedures to increase the homogeneity of the layer thickness <ref:2607. p. 1
Where the paper says it

A SixNy thin film was deposited on a BOROFLOAT 33 glass (BF33) wafer as a single-wafer process using plasma-enhanced chemical vapor deposition (PECVD) at a deposition temperature of 130 ◦C with NH3 and SiH4 as precursors. 500 nm thick layer was applied in two consecutive 250 nm deposition procedures to increase the homogeneity of the layer thickness.

03077#pg3>. These samples were already employed in a previous study [32], where extensive reference characterization of focused ion beam (FIB) prepared lamellae was performed by high-angle annular dark-field (HAADF)-scanning transmission electron microscopy (STEM) and energy dispersive X-ray spectroscopy (EDX) and the thickness of the SixNy film was determined to be 536. p. 1
Where the paper says it

The measurements took place as-deposited without any further heat treatment. These samples were already employed in a previous study [32], where extensive reference characterization of focused ion beam (FIB) prepared lamellae was performed by high-angle annular dark-field (HAADF)-scanning transmission electron microscopy (STEM) and energy dispersive X-ray spectroscopy (EDX) and the thickness of the SixNy film was determined to be 536.

4 nm <ref:2607. p. 0
Where the paper says it

hutzler@fz-juelich. de 1 arXiv:2607.

03077#pg3>. The thickness of an n-doped 4H-SiC wafer with the typical 4◦ off-axis cut normal to the c-plane was determined to be 351. p. 1
Where the paper says it

These samples were already employed in a previous study [32], where extensive reference characterization of focused ion beam (FIB) prepared lamellae was performed by high-angle annular dark-field (HAADF)-scanning transmission electron microscopy (STEM) and energy dispersive X-ray spectroscopy (EDX) and the thickness of the SixNy film was determined to be 536. The thickness of an n-doped 4H-SiC wafer with the typical 4 ◦ off-axis cut normal to the c-plane was determined to 351.

5 µm as the average of 5 thickness measurements at different positions on the wafer obtained by a Logitech CG10 contact measurement gauge <ref:2607. p. 1
Where the paper says it

The thickness of an n-doped 4H-SiC wafer with the typical 4 ◦ off-axis cut normal to the c-plane was determined to 351. 5 µm as the average of 5 thickness measurements at different positions on the wafer obtained by a Logitech CG10 contact measurement gauge.

03077#pg3>. Flakes of HOPG and MoO3 were mechanically exfoliated in atmosphere and transferred onto 500 µm thick BF33 glass substrates <ref:2607. p. 1
Where the paper says it

5 µm as the average of 5 thickness measurements at different positions on the wafer obtained by a Logitech CG10 contact measurement gauge. Flakes of HOPG and MoO3 were mechanically exfoliated in atmosphere and transferred onto 500 µm thick BF33 glass substrates.

03077#pg3>. The thickness of the flakes was determined by atomic force microscopy (AFM) measurements using a Bruker Dimension ICON in intermittent-contact mode <ref:2607. p. 1
Where the paper says it

Flakes of HOPG and MoO3 were mechanically exfoliated in atmosphere and transferred onto 500 µm thick BF33 glass substrates. The thickness of the flakes was determined by atomic force microscopy (AFM) measurements using a Bruker Dimension ICON in intermittent-contact mode.

03077#pg3>.

--- Page 5 ---

The schematic of the employed measurement setup is shown in Fig. 1a <ref:2607. p. 0

Where the paper says it

hutzler@fz-juelich. de 1 arXiv:2607.

03077#pg2>. Reflectance and transmittance were measured with an Olympus BX53-MTRF microscope coupled to a Horiba iHR 320 spectrometer including a 1200 grooves/mm diffraction grating <ref:2607. p. 2
Where the paper says it

1 Measurement setup and modus operandi The schematic of the employed measurement setup is shown in Fig. Reflectance and transmittance were measured with an Olympus BX53-MTRF microscope coupled to a Horiba iHR 320 spectrometer including a 1200 grooves/mm diffraction grating.

03077#pg2>. The signal is collected with the center fiber of a fiber bundle, which has a diameter of 200 µm <ref:2607. p. 2
Where the paper says it

Reflectance and transmittance were measured with an Olympus BX53-MTRF microscope coupled to a Horiba iHR 320 spectrometer including a 1200 grooves/mm diffraction grating. The signal is collected with the center fiber of a fiber bundle, which has a diameter of 200 µm.

03077#pg2>. Due to the characteristics of the halogen light sources and the Horiba Synapse CCD detector, the available spectral range is limited from 440 nm to 950 nm <ref:2607. p. 2
Where the paper says it

The signal is collected with the center fiber of a fiber bundle, which has a diameter of 200 µm. Due to the characteristics of the halogen light sources and the Horiba Synapse CCD detector, the available spectral range is limited from 440 nm to 950 nm.

03077#pg2>. Linearly polarized light was achieved by the insertion of a continuously rotatable linear polarizer in the illumination path <ref:2607. p. 2
Where the paper says it

Due to the characteristics of the halogen light sources and the Horiba Synapse CCD detector, the available spectral range is limited from 440 nm to 950 nm. Linearly polarized light was achieved by the insertion of a continuously rotatable linear polarizer in the illumination path.

03077#pg2>. For transmittance measurements, an achromatic and aplanatic Olympus U-AAC universal condenser was used <ref:2607. p. 2
Where the paper says it

Linearly polarized light was achieved by the insertion of a continuously rotatable linear polarizer in the illumination path. For transillumination measurements, an achromatic and aplanatic Olympus U-AAC universal condenser was used.

03077#pg2>. All utilized objective lenses are listed in Tab. p. 2
Where the paper says it

For transillumination measurements, an achromatic and aplanatic Olympus U-AAC universal condenser was used. All utilized objective lenses are listed in Tab.

1 together with their NAs and the corresponding diameters of the measurement spot on the sample, defined by the respective magnification and the fiber diameter <ref:2607. p. 2
Where the paper says it

All utilized objective lenses are listed in Tab. 1 together with their NAs and the corresponding diameters of the measurement spot on the sample, defined by the respective magnification and the fiber diameter.

03077#pg2>. In general, the field stop was set to the smallest size to reduce stray light <ref:2607. p. 2
Where the paper says it

Objective lens 10x 50xUL1 100xUL1 100x NA 0. 90 Spot diameter/µm 20 4 2 2 In general, the field stop was set to the smallest size to reduce stray light [26, 32].

03077#pg2>. To account for background noise in reflectance, the spectrum of an ultra-low reflecting flock sheet from Musou Black was measured <ref:2607. p. 2
Where the paper says it

90 Spot diameter/µm 20 4 2 2 In general, the field stop was set to the smallest size to reduce stray light [26, 32]. To account for background noise in reflectance, the spectrum of an ultra-low reflecting flock sheet from Musou Black was measured [42].

03077#pg2>. In transmittance, the background signal was isolated by measuring the spectrum with the light path to the detector blocked <ref:2607. p. 2
Where the paper says it

To account for background noise in reflectance, the spectrum of an ultra-low reflecting flock sheet from Musou Black was measured [42]. In transmittance, the background signal was isolated by measuring the spectrum with the light path to the detector blocked.

03077#pg2>. All measurements were averaged over 10 acquisitions <ref:2607. p. 2
Where the paper says it

In contrast, a 100 % transmittance reference was used by aligning condenser and objective lens relative to each other without a sample in between. All measurements were averaged over 10 acquisitions.

03077#pg2>. For focusing on the upper surface of the sample in both illumination modes, the objective lens was focused by moving the stage until the image of the field diaphragm was brought into sharp focus <ref:2607. p. 2
Where the paper says it

All measurements were averaged over 10 acquisitions. For focusing on the upper surface of the sample in both illumination modes, the objective lens was focused by moving the stage until the image of the field diaphragm was brought into sharp focus.

03077#pg2>. In transmission mode, the condenser was subsequently adjusted so that the image of its field diaphragm was also sharply visible, after the light source in the epi-illumination path had been disabled <ref:2607. p. 2
Where the paper says it

For focusing on the upper surface of the sample in both illumination modes, the objective lens was focused by moving the stage until the image of the field diaphragm was brought into sharp focus. In transmission mode, the condenser was subsequently adjusted so that the image of its field diaphragm was also sharply visible, after the light source in the epi-illumination path had been disabled.

03077#pg2>. In general, the aperture diaphragm was fully opened in epi-illumination and in transillumination the condenser aperture was adjusted to match the NA of the objective lens to ensure the same range of AOIs in both illumination modes <ref:2607. p. 2
Where the paper says it

In transmission mode, the condenser was subsequently adjusted so that the image of its field diaphragm was also sharply visible, after the light source in the epi-illumination path had been disabled. In general, the aperture diaphragm was fully opened in epi-illumination and in transillumination the condenser aperture was adjusted to match the NA of the objective lens to ensure the same range of AOIs in both illumination modes.

03077#pg2>. However, this phenomenon is especially pronounced for the 100x objective lens with the highest NA of 0. p. 2
Where the paper says it

However, the illumination through a thick substrate by the condenser causes spherical aberrations due to the refractive index mismatch between the substrate and the surrounding air [45, 46]. This phenomenon is especially pronounced for the 100x objective lens with the highest NA of 0.

90 and the ocurring high AOIs, which is why the condenser aperture was reduced to an effective NA of 0. p. 2
Where the paper says it

This phenomenon is especially pronounced for the 100x objective lens with the highest NA of 0. 90 and the ocurring high AOIs, which is why the condenser aperture was reduced to an effective NA of 0.

60 for the transmittance measurements with this objective lens to mitigate the influence of spherical aberrations <ref:2607. p. 2
Where the paper says it

90 and the ocurring high AOIs, which is why the condenser aperture was reduced to an effective NA of 0. 60 for the transmittance measurements with this objective lens to mitigate the influence of spherical aberrations (cf.

03077#pg2>.

--- Page 6 ---

Fig. 1: a) Schematic of the employed measurement setup. p. 2

Where the paper says it

2 Microspectroscopic measurements and modeling 2. 1 Measurement setup and modus operandi The schematic of the employed measurement setup is shown in Fig.

b) Flowchart of the numerical procedure for extracting the (complex) refractive indices from measured reflectance and/or transmittance spectra <ref:2607. p. 2
Where the paper says it

2 Modeling framework Fig. 1b shows the flowchart of the numerical procedure for extracting the (complex) refractive indices from measured reflectance and transmittance spectra.

p. 2
Where the paper says it

2 Modeling framework Fig. 1b shows the flowchart of the numerical procedure for extracting the (complex) refractive indices from measured reflectance and transmittance spectra.

03077#pg2>.

Flowchart of the numerical procedure for extracting the (complex) refractive indices from measured reflectance and/or transmittance spectra <ref:2607.03077#pg2>.

--- Page 8 ---

Fig. 2: (a) Measured reflectance (R) and transmittance (T) spectra of a 351. p. 5

Where the paper says it

The simulations accurately replicate the measurements. 2: (a) Measured reflectance (R) and transmittance (T) spectra of a 351.

5 µm thick 4H-SiC sample for the 10x/NA 0. p. 5
Where the paper says it

2: (a) Measured reflectance (R) and transmittance (T) spectra of a 351. 5 µm thick 4H-SiC sample for the 10x/NA 0.

25 objective lens and corresponding modeled spectra, calculated using isotropic optical constants (n, k) from Khadivianazar et al. p. 5
Where the paper says it

5 µm thick 4H-SiC sample for the 10x/NA 0. 25 objective lens and corresponding modeled spectra, calculated using isotropic optical constants (n, k) from Khadivianazar et al.

[56] as reference <ref:2607. p. 0
Where the paper says it

hutzler@fz-juelich. de 1 arXiv:2607.

03077#pg4>. Extracted solution space, smoothed mean o… p. 5
Where the paper says it

[56] as reference. Extracted solution space, smoothed mean of the solution space, analytical solution based on the equations of Nichelatti [58], and the reference values by Khadivianazar et al.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Automated extraction of complex refractive indices from microspectroscopy data can be performed without relying on dispersion models like Drude-Lorentz or Tauc-Lorentz by using a numerical formalism that explicitly accounts for the influence of the numerical aperture (NA).

  2. The improved system can accurately determine the thickness and complex refractive indices of multilayer stacks and thick transparent substrates, such as a SixNy thin film on glass, by transferring objective lens dependent modeling from reflectance to transmittance measurements.

  3. The AI system can resolve ambiguities in extracting refractive indices for flakes of van der Waals materials like HOPG and MoO3 by superimposing the solution spaces obtained from different objective lenses or analyzing samples with varying thicknesses.

  4. The system can determine refractive indices along the in-plane crystal axes of biaxial materials like MoO3 by incorporating linearly polarized light measurements into the modeling framework, enabling the determination of material properties that are at least uniaxial approximations.

  5. The AI system can estimate isotropic refractive indices for anisotropic samples like HOPG by utilizing solution branch A, which corresponds to the smaller extinction coefficient k, and comparing results with literature data where available.

Abstract

For the design and optimization of optoelectronic devices, accurate knowledge of the complex refractive indices of the constituent materials is essential. Herein, we present a fast and non-destructive approach for the extraction of the refractive indices from reflectance and transmittance spectra of samples with lateral dimensions down to the micrometer scale. Microspectroscopy, based on the combination of a standard optical microscope and a spectrometer, enables the assessment of the optical response of multilayer stacks using high-magnification optics with correspondingly large numerical apertures. Employing a numerical formalism explicitly accounting for the influence of the numerical aperture, allows for precise retrieval of the refractive index without resorting to dispersion models. We demonstrate the applicability of the proposed method for large-area, homogeneous, optically incoherent samples such as transparent glasses and absorbing 4H-SiC, for a SixNy thin film on glass substrate, and for mechanically exfoliated flakes of highly oriented pyrolytic graphite and MoO3, as representatives of uniaxial and biaxial optical anisotropy. While the results prove excellent agreement with values reported in literature, the case of graphite highlights the limitation for probing the out-of-plane refractive indices due to reduced sensitivity. Finally, we discuss possible extensions towards retrieving the full anisotropic tensor of the refractive index, establishing the proposed approach as a methodologically sound alternative to spectroscopic ellipsometry.

Related papers