🔬 materials science

Determination of fundamental properties of nitrogen from first principles. III. Temperature and frequency dependence of the molecular polarizability and magnetic susceptibility

This paper presents first-principles calculations of the temperature and frequency dependence of nitrogen's molecular polarizability and magnetic susceptibility, which, while less accurate than high-precision experiments on their own, are combined with experimental data to generate highly accurate semi-empirical estimates for key reference temperatures and to address discrepancies in magnetic susceptibility measurements.

Original authors: Jakub Lang, Giovanni Garberoglio, Michał Przybytek, Michał Lesiuk

Published 2026-07-24

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Jakub Lang, Giovanni Garberoglio, Michał Przybytek, Michał Lesiuk

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to measure the temperature of a gas, not with a thermometer that touches it, but by shining a laser through it and watching how the light bends. This is the heart of a cutting-edge field called gas thermometry. To make this work, scientists need to know the "personality" of the gas molecules inside. Specifically, they need to know how easily the molecule's electron cloud can be squished by an electric field (called polarizability) and how it reacts to a magnetic field (called magnetic susceptibility). Think of polarizability like a spring: some molecules are stiff springs that are hard to squish, while others are loose springs that stretch easily. The problem is, these "springs" don't behave the same way all the time. Just like a rubber band gets looser when it's hot and stiffer when it's cold, these molecules change their squishiness depending on the temperature and the color (frequency) of the light hitting them. For a long time, scientists could only measure these properties at specific, frozen moments, leaving huge gaps in our knowledge of how they behave across the entire temperature range.

This paper is the third chapter in a story about the nitrogen molecule ( $N_2$ ), the most common gas in our atmosphere. The authors, a team of theoretical chemists, decided to build a perfect, virtual model of nitrogen from the ground up—using only the fundamental laws of physics, without relying on any experimental measurements to guide them. They wanted to map out exactly how nitrogen's "squishiness" and magnetic reaction change as the gas heats up from a chilly 50 Kelvin to a scorching 2000 Kelvin, and how it reacts to different colors of light. They didn't just guess; they used two different, super-complex mathematical methods to simulate the molecule's behavior, essentially running a double-check to ensure their virtual nitrogen was acting exactly like the real thing.

The team's main discovery is a detailed, high-resolution map of nitrogen's behavior. They calculated how the molecule's polarizability changes with temperature and frequency, producing a set of numbers called Cauchy coefficients that act like a recipe for predicting the molecule's reaction at any condition. They found that while their pure-theory numbers are incredibly detailed, they are still slightly less precise than the best real-world experiments currently available. However, this doesn't mean the theory failed. Instead, the authors realized that the real magic happens when you mix their theoretical map with existing experimental data. By combining the two, they were able to create "semi-empirical" estimates—super-accurate guesses for values that have never been measured directly.

For instance, they determined that at a standard room temperature of 303 Kelvin, the static polarizability of nitrogen is 11.735962 a.u., and at the freezing point of water (273.16 K), it is 11.735585 a.u. These numbers are so precise they could serve as new reference points for future science. They also tackled the molecule's magnetic susceptibility, finding that a specific "paramagnetic" contribution (where the molecule acts like a tiny magnet) is crucial, even though it is tiny compared to the "diamagnetic" part. Interestingly, they found a significant mismatch between their theoretical predictions for this magnetic value and some existing experimental data, suggesting that the old measurements might need a second look.

The authors were careful to note that while their simulations are powerful, they are not a replacement for direct measurement yet. The theoretical results have an uncertainty that is about two orders of magnitude larger than the best experimental data. However, the paper explicitly rules out the idea that the magnetic susceptibility changes significantly with frequency (color of light) in the range used for these experiments, suggesting that effect is likely smaller than 1%. They also proved that the motion of the molecule's center of mass doesn't mess up the magnetic measurements in a way that matters for their accuracy goals. Ultimately, this work doesn't just give us numbers; it gives us a reliable way to fill in the blanks of the nitrogen story, allowing scientists to predict how this gas will behave in conditions we haven't even tested yet, bridging the gap between what we can measure and what we need to know.

Technical Summary: Determination of Fundamental Properties of Nitrogen from First Principles. III. Temperature and Frequency Dependence of the Molecular Polarizability and Magnetic Susceptibility

Problem Statement
This work addresses the theoretical determination of the temperature and frequency dependence of the molecular polarizability ( $\alpha$ ) and magnetic susceptibility ( $\chi$ ) of the nitrogen molecule ( $N_2$ ). These properties are critical for modern gas thermometry, specifically Refractive Index Gas Thermometry (RIGT), where the Lorentz–Lorenz equation links macroscopic refractive index measurements to microscopic molecular properties. While previous studies in this series established the interaction potential and electronic ground state of $N_2$ , a systematic theoretical treatment of the temperature and frequency dependence of $\alpha$ and $\chi$ was lacking. Experimental measurement of these quantities across the full range of admissible temperatures and frequencies is often time-consuming or impossible due to experimental limitations. Consequently, a robust theoretical framework is required to supplement experimental data, particularly for generating semi-empirical estimates of quantities that have not yet been measured with high precision.

Methodology
The authors employ a composite ab initio approach combined with two independent methods for temperature averaging to ensure result validation.

Electronic Structure Calculations:
- Static Properties: The purely electronic contributions to the static polarizability and Cauchy coefficients (up to sixth order, $\alpha_0$ through $\alpha_6$ ) and the isotropic magnetic susceptibility are computed as a function of internuclear distance ( $R$ ) within the range $R \in [1.75, 2.60]$ bohr.
- Composite Scheme: A high-level coupled-cluster (CC) hierarchy is used, including CCSD, CC3, CCSDT, and CCSDTQ levels of theory. The scheme incorporates frozen-core and all-electron corrections, as well as one-electron relativistic corrections (via the Douglas–Kroll–Hess method).
- Frequency Dependence: Instead of calculating dynamic properties at specific frequencies, the dynamic polarizability is expressed as a power series in frequency (Cauchy expansion), $\alpha(\omega, R) = \sum \alpha_{2k}(R)\omega^{2k}$ . This avoids the need for interpolation between frequency grids.
- Magnetic Susceptibility: Calculated using the Van Vleck theory, separating diamagnetic and paramagnetic contributions. The paramagnetic term is evaluated using coupled-cluster response theory to bypass explicit summation over excited states. A correction for the coupling of the center-of-mass (COM) motion to the magnetic field is derived and found to be negligible.
Temperature Dependence:
Two independent methods are used to average the $R$ -dependent properties over the canonical ensemble to obtain temperature-dependent values ( $T = 50$ K to $2000$ K):
- Rovibrational Averaging: The properties are averaged over a set of 345 rovibrational levels ( $\nu, J$ ) using weights derived from the canonical ensemble. The integration over $R$ is truncated to the interval where the probability amplitude is non-negligible.
- Path Integral Monte Carlo (PIMC): A quantum-classical mapping is used to simulate the partition function. This method accounts for quantum effects beyond the classical picture and serves as an independent validation of the rovibrational averaging results.
Analytic Representation:
The calculated data points are fitted using short-range analytic functions (exponential-polynomial forms) to facilitate direct use in thermometry applications. Uncertainties are propagated using functional differentiation and covariance analysis.

Key Contributions and Results

High-Precision Electronic Data: The authors provide high-accuracy values for the static polarizability and Cauchy coefficients ( $\alpha_0, \alpha_2, \alpha_4, \alpha_6$ ) and the magnetic susceptibility ( $\chi_0$ ) as functions of internuclear distance.
Temperature Dependence: The study establishes the temperature dependence of these properties from 50 K to 2000 K. The results from rovibrational averaging and PIMC show near-perfect agreement, validating the methodology.
Magnetic Susceptibility: The paper reports theoretical values for $\chi_0$ , highlighting the importance of the paramagnetic contribution. It identifies a significant discrepancy between the theoretical $\chi_0$ and existing experimental data, suggesting the experimental values may require re-evaluation.
Semi-Empirical Estimates: By combining the theoretical temperature dependence and Cauchy coefficients with recent high-precision experimental thermometry data (Egan and Yang), the authors generate semi-empirical estimates for the static polarizability at key reference temperatures:
- $\alpha_0(303 \text{ K}) = 11.735962$ a.u.
- $\alpha_0(273.16 \text{ K}) = 11.735585$ a.u.
  These estimates are claimed to have an accuracy of approximately 10 ppm.

Significance and Claims
The authors explicitly state that the accuracy of their purely theoretical polarizability values is currently lower (by roughly two orders of magnitude) than the most precise experimental measurements available. Therefore, they do not claim that theory can replace direct measurement at present.

Instead, the primary significance of this work lies in its utility as a hybrid tool. The theoretical data provides the necessary frequency and temperature dependence to extrapolate or interpolate available experimental data to conditions where measurements are unavailable. This capability allows for the generation of highly accurate semi-empirical estimates for quantities that have never been measured directly with comparable precision, such as the static polarizability at specific reference temperatures. The work fills a critical gap in the systematic theoretical understanding of nitrogen's properties, providing a foundation for future metrological applications and resolving discrepancies in magnetic susceptibility data.

More like this