Source-linked AI summary

SpectraPlot.com: Integrated Spectroscopic Modeling of Atomic and Molecular Gases

Christopher S. Goldenstein, Victor A. Miller, R. Mitchell Spearrin, Christopher L. Strand

arXiv:1709.06220v1physics.chem-ph

TL;DR

Spectroscopic databases can be difficult for nonexperts to use, while simulation results depend on the capabilities and limits of their models. This paper describes SpectraPlot’s modular simulation tools and evaluates the error from its far-wing cutoff; the reported errors are small for tested H2O and CO2 conditions.

  • Problem

    Spectroscopic databases can be difficult for nonexperts to use, motivating accessible tools whose simulation capabilities and limitations users can understand.

    Source: §1.1 · pp. 1–2
  • Method

    The paper describes SpectraPlot’s modular architecture, spectroscopic models, and assumptions, including its far-wing cutoff approximation.

    Source: §1.2 · p. 2, §1.1 · p. 2, §3 · p. 13
  • Results

    At tested conditions, far-wing cutoff errors were <0.1% of peak-band absorbance for H2O and <10^-4% for CO2.

    Source: §2.4.1 · pp. 10–11, §3 · p. 13
  • Takeaways & Limitations

    The cutoff limits included lines and the wavelength range simulated to improve computational efficiency.

    Source: §3 · p. 13

    For NIST transitions lacking a listed A value, SpectraPlot assigns 10^7 s^-1 by default, so corresponding spectra require extreme caution.

    Source: §2.2.2 · p. 8

Abstract

from arXiv · show

SpectraPlot is a web-based application for simulating spectra of atomic and molecular gases. At the time this manuscript was written, SpectraPlot consisted of four primary tools for calculating: 1) atomic and molecular absorption spectra, 2) atomic and molecular emission spectra, 3) transition linestrengths, and 4) blackbody emission spectra. These tools currently employ the NIST ASD, HITRAN2012, and HITEMP2010 databases to perform line-by-line simulations of spectra. SpectraPlot employs a modular, integrated architecture, enabling multiple simulations across multiple databases and/or thermodynamic conditions to be visualized in a single interactive plot window. The primary objective of this paper is to describe the architecture and spectroscopic models employed by SpectraPlot in order to provide its users with the knowledge required to understand the capabilities and limitations of simulations performed using SpectraPlot. Further, this manuscript discusses the accuracy of several underlying approximations used to decrease computational time, namely, the use of far-wing cutoff criteria.

1. Introduction

SpectraPlot aims to make atomic and molecular gas simulations accessible while clarifying the capabilities and limits of its models. Its modular tools draw on multiple spectroscopic databases and present simulations and transition data through interactive visualizations.

1.1. Background and Motivation

  • Because spectroscopic databases can be difficult for nonexperts, SpectraPlot aims to offer accessible web-based atomic and molecular spectrum simulations and explain their limits.Source: §1.1 · pp. 1–2

1.2. Architecture

  • Each data type has its own processor and database, while a shared interface plots results for comparisons across databases; the design is database-agnostic.Source: §1.2 · p. 2, Figure 1 · p. 2

1.3. Tools and User Interface

  • The four tools calculate atomic and molecular absorption, emission, transition linestrengths, and blackbody emission spectra; the paper focuses on the first three.Source: §1.3 · p. 3

1.3.1. Contents

  • SpectraPlot draws on NIST ASD, HITEMP2010, and HITRAN2012, covering 78 atomic and 37 molecular species, though the databases are not comprehensive.Source: §1.3.1 · p. 3

1.3.2. Absorption and Emission Tools

  • Absorption and emission tools simulate up to three species independently using selected gas and spectral conditions, then return interactive spectra with zoom, cursor readouts, and export.Source: §1.3.2 · pp. 3–5

1.3.3. Linestrength Survey

  • The linestrength survey displays individual transitions in an interactive stick plot; temperature, pressure, and mole fraction adjust strengths or linecenters, and hovering reveals transition parameters.Source: §1.3.3 · p. 5

2. Spectroscopic Models

This section describes the most pertinent details required to understand the spectroscopic models employed by SpectraPlot.

  • The section presents the spectroscopic-model details needed to understand SpectraPlot’s models.Source: §2 · p. 5

2.1. Absorption and Emission Spectroscopy

SpectraPlot calculates absorption and emission spectra for uniform gas samples with collimated light. Absorbance is related to gas properties, while emission is modeled relative to blackbody radiance.

2.1. Absorption and Emission Spectroscopy

  • Calculations assume a uniform gas sample and collimated light.The assumed orientation of light and gas sample is illustrated for absorbance and emission calculations.Source: §2.1 · p. 6, Figure 4 · p. 6

2.1.1. Calculation of Absorbance Spectra

  • Beer’s Law defines spectral absorbance, which is related to gas properties for a uniform sample containing one radiating species.The relation uses linestrength, pressure, number density, species mole fraction, line shape, and gas path length.Source: §2.1.1 · p. 6

2.1.2. Calculation of Emission Spectra

  • For a gas in thermal equilibrium, emissivity equals absorptivity, and emitted radiance is calculated relative to blackbody radiance.The blackbody radiance is calculated using Planck’s Law.Source: §2.1.2 · pp. 6–7
  • An optically thick gas (α >> 1) emits a spectrum equal to that of a blackbody.The spectral radiance is expressed in units of power per solid angle per unit area per unit frequency or wavelength.Source: §2.1.2 · p. 7

2.2. Calculation of Linestrengths

SpectraPlot calculates transition linestrengths using parameters from HITRAN, HITEMP, and NIST ASD, accounting for stimulated emission. The database-specific calculations use temperature-dependent corrections or population estimates.

2.2. Calculation of Linestrengths

  • Linestrength describes a transition’s propensity to absorb and emit light, and SpectraPlot accounts for stimulated emission in its calculations.The calculations use parameters from HITEMP2010, HITRAN2012, and NIST ASD.Source: §2.2 · p. 7

2.2.1. HITRAN and HITEMP

  • HITRAN and HITEMP list linestrengths at a reference temperature of 296 K, which SpectraPlot adjusts for the specified temperature.The temperature adjustment uses the partition function and lower-state energy; simulations include database isotopologues and scale reference linestrengths by natural abundance.Source: §2.2.1 · p. 7
  • The linestrength survey tool outputs HITRAN and HITEMP linestrengths in cm−2-atm−1, convertible to cm−1/molecule-cm−2 for an ideal gas.Source: §2.2.1 · p. 7

2.2.2. NIST ASD

  • For NIST ASD transitions, SpectraPlot calculates linestrength using the tabulated Einstein-A coefficient, energies, and state degeneracies, with electronic temperature as an additional input.For a gas in equilibrium, electronic temperature equals kinetic temperature.Source: §2.2.2 · pp. 7–8
  • When NIST ASD lacks a transition’s Einstein-A value, SpectraPlot assigns a default of 107 s−1; such transitions are included in spectra but excluded from line-survey results.The paper advises extreme caution when simulating spectra for transitions assigned this default.Source: §2.2.2 · p. 8

2.3. Calculation of Partition Function

SpectraPlot obtains partition functions through database-specific approaches: interpolation or extrapolation of HITRAN Global Data for HITRAN/HITEMP, and a generated lookup table for NIST ASD.

2.3.1. HITRAN and HITEMP

  • For HITRAN and HITEMP, SpectraPlot calculates partition functions at the specified temperature by interpolating or, if needed, extrapolating HITRAN Global Data.The data provide total internal partition sums for species and isotopologues from 70 to 3000 K.Source: §2.3.1 · p. 8

2.3.2. NIST ASD

  • Because NIST ASD lacks pre-calculated partition sums, SpectraPlot generated an electronic-partition-function lookup table for atomic species and ions up to 800 eV.Values at the specified temperature are obtained by interpolation within the table or, if necessary, extrapolation.Source: §2.3.2 · p. 8

2.4. Calculation of Lineshapes

SpectraPlot selects Gaussian, Lorentzian, or Voigt profiles according to the Lorentzian-to-Doppler ratio and models collisional broadening using database parameters. Its pressure-based far-wing cutoff limits simulation scope while introducing small residual errors in the tested H2O and CO2 spectra.

2.4. Calculation of Lineshapes

  • SpectraPlot uses Gaussian profiles when the Lorentzian-to-Doppler ratio is <0.001, Lorentzian profiles when it is >1000, and Voigt profiles otherwise.The Voigt profile uses the algorithm developed by Mclean et al. and Martin and Puerta.Source: §2.4 · p. 8
  • HITRAN and HITEMP collisional widths use self- and air-broadening coefficients, with n_self = 0.75 estimated because the databases omit self-broadening temperature exponents.The coefficients are adjusted with a power-law temperature model.Source: §2.4 · pp. 8–9
  • For NIST ASD atomic spectra, users must provide a collisional-broadening coefficient and temperature exponent for the transition and gas mixture.The example Na-in-N2 coefficient is 0.25 cm−1-atm−1, described as representative of these transitions in dilute N2.Source: §2.4 · pp. 9–10, Figure 6 · p. 10

2.4.1. Far-Wing Cutoff

  • The cutoff is 200 cm−1 below 10 atm, 20 cm−1/atm × P between 10 and 100 atm, and 2000 cm−1 otherwise; it limits included lines and each simulated lineshape’s frequency span.Transitions are selected by linecenter within the user’s frequency bounds extended by the cutoff, and each lineshape is set to zero outside its cutoff span.Source: §2.4.1 · p. 10
  • At 0.1 and 100 atm and 296 and 2500 K, cutoff errors were <0.1% for H2O and <10^-4% for CO2, as percent of peak-band absorbance.The comparisons used HITRAN2012; larger local errors may occur beyond the cutoff from strong band edges, and HITEMP2010 residuals are expected to increase modestly.Source: §2.4.1 · pp. 10–11, Figure 7 · p. 11

2.5. Calculation of Transition Linecenter

SpectraPlot takes vacuum linecenter frequencies from HITRAN2012, HITEMP2010, and NIST ASD. It accounts for pressure shifting with HITRAN and HITEMP parameters, but not for NIST ASD simulations.

2.5. Calculation of Transition Linecenter

  • Vacuum linecenter frequencies come from HITRAN2012, HITEMP2010, and NIST ASD.Source: §2.5 · p. 11
  • HITRAN and HITEMP simulations account for pressure shifting using the air pressure-shift coefficient tabulated at 296 K.The temperature dependence uses a power law with an estimated exponent of 0.96 for all transitions, following.Source: §2.5 · p. 12
  • NIST ASD simulations omit pressure shifting because the database does not provide pressure-shift coefficients.Source: §2.5 · p. 12

3. Conclusions

SpectraPlot combines four spectroscopic tools and three databases in a modular application for integrated visualization. The authors report that their far-wing cutoff introduces negligible error for H2O and CO2 across the tested range of temperatures and pressures.

3. Conclusions

  • SpectraPlot provides tools for atomic and molecular absorption, emission, transition linestrengths, and blackbody emission calculations.Source: §3 · pp. 12–13
  • Its modular architecture uses NIST ASD, HITEMP2010, and HITRAN2012 to visualize results across databases, species, and thermodynamic conditions in one interactive interface.Source: §3 · p. 13
  • The far-wing cutoff approximation introduced <0.1% of peak-band absorbance error for H2O across a wide range of tested temperatures and pressures.The cutoff was developed by simulating H2O and CO2 spectra over varied conditions.Source: §3 · p. 13
Loading 1709.06220v1…