The temperature dependence (203–293 K) of the absorption cross sections of O3 in the 230–850 nm region measured by Fourier-transform spectroscopy

https://doi.org/10.1016/S1010-6030(01)00480-4Get rights and content

Abstract

Absolute absorption cross sections of O3 were measured in the 230–850 nm (11765–43478 cm−1) region at five different temperatures (203–293 K) using a Fourier-transform spectrometer, at a spectral resolution of 5.0 cm−1 (corresponding to about 0.027 nm at 230 nm and to about 0.36 nm at 850 nm). The spectral accuracy of the data is better than 0.1 cm−1 — about 0.5 pm at 230 nm and about 7.2 pm at 850 nm — validated by recording of I2 absorption spectra in the visible using the same experimental set-up. O3 absorption spectra at different concentrations were recorded at five different sample temperatures in the range 203–293 K, and at each temperature at two total pressures (100 and 1000 mbar) using O2/N2 mixtures as buffer gas. Within the limits of experimental uncertainties, no influence of total pressure on the O3 spectrum was observed in the entire spectral region, as expected from the short lifetimes of the upper electronic states of O3. The temperature dependence of the O3 absorption cross sections is particularly strong in the Huggins bands between 310 and 380 nm, as observed in previous studies. An empirical formula is used to model the temperature dependence of the O3 absorption cross sections between 236 and 362 nm, a spectral region that is particularly important for atmospheric remote-sensing and for photochemical modelling.

Introduction

Ozone, O3, is an important minor constituent of the Earth’s atmosphere [1], [2]. It plays a significant role in atmospheric radiative transfer and photochemistry [3], [4]. For many years, atmospheric O3 concentrations have been determined using UV–VIS spectroscopy [5], [6]. For this purpose, a number of laboratory measurements of the absorption spectrum of O3 have been reported in the past [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43]. In spite of the number of studies, significant uncertainties in the UV–VIS absorption cross sections of O3 still exist. In order to monitor small changes in the atmospheric O3 concentrations, it is essential to know the absorption strength of this molecule to high accuracy.

In the last decade, a new generation of powerful satellite-borne UV–VIS–NIR spectrometers for atmospheric remote-sensing (e.g. GOME [44] and SCIAMACHY [45]) has been developed, that will be followed by many other similar instruments on different meteorological satellites. These instruments observe backscattered, reflected and transmitted light from the Earth’s atmosphere, and measure simultaneously the entire spectral region between about 250–800 nm at medium or high spectral resolution (0.2–0.4 nm). Such experiments require precise reference spectra of O3 and of many other atmospheric absorbers in this spectral region.

In order to provide these reference spectra, new O3 absorption spectra were recorded using the GOME Flight-Model (GOME-FM) spectrometer in 1994–1995, in the entire region 230–800 nm at five different atmospheric temperatures in the range 202–293 K, as reported previously [43]. This data set has immediately been used in many remote-sensing and photochemical studies, although the spectral calibration of these cross sections is only accurate to about 0.03 nm, the limitations arising from the experimental set-up.

While the accuracy of the spectral wavelength calibration is not critical for many scientific applications, it is essential for the retrieval of atmospheric concentrations of O3 and other constituents, such as NO2, NO3, OClO, BrO, SO2, CH2O, and ClO, from remote-sensing measurements in the UV and visible. During the computer analysis of atmospheric spectra by the differential optical absorption spectroscopy (DOAS) technique [46], a non-linear wavelength fitting procedure (“shift and squeeze”) is often applied in order to take into account systematic errors in the wavelength calibration. However, it is well established that additional fitting parameters lead to correlations between the retrieved atmospheric concentrations and to larger statistical uncertainties.

To resolve this issue, an impressive number of new laboratory spectra of NO2 [47], [48], [49], [50], O2/O4 [51], [52], [53], [54], SO2 [55], [56], H2O [57], [58], [59], BrO [60], [61], and OClO [62] have been recorded in the past years using UV–VIS Fourier-transform spectroscopy (FTS). This experimental technique combines the advantages of high spectral resolution, of a well-known instrumental line shape, and of a linear wave number scale [63]. However, up to the present study, no temperature-dependent absorption spectra of the most important trace gas O3 have been recorded in the UV–VIS spectral region using FTS. In order to address this deficit, we have performed laboratory measurements of O3 spectra at five different temperatures in the range 203–293 K, with a Fourier-transform spectrometer suitable for use in the ultraviolet and visible spectral regions.

Section snippets

Experimental set-up

Absorption spectra of O3 at different temperatures were recorded using the high-resolution Fourier-transform spectrometer (Bruker IFS-120 HR) at the University of Bremen, Germany. The instrument has a maximum spectral resolution of 0.004 cm−1 [64] and operates in the spectral range between 600 and 45,000 cm−1, using different combinations of broad-band light sources, beamsplitters/combiners, and detectors. For all spectra used in the present study, the same interferometer configuration was used

Results and discussion

Within the limits of experimental accuracy, no pressure dependence of the O3 absorption cross sections in the entire region 230–850 nm was observed; i.e. the only differences are due to noise and baseline changes. This result is in agreement with expectations from the upper state’s lifetime, because all excited electronic states of O3 leading to the absorption spectrum between 230 and 850 nm are repulsive or highly predissociated. The only region where a possible pressure dependence could have

Conclusion

The new absorption cross sections at different temperatures are an important contribution to our knowledge of the UV–VIS–NIR spectrum of O3, in particular, because of the high wavelength accuracy and spectral resolution over the entire 230–850 nm region which has — to the best of our knowledge — not been achieved before. Comparisons with previous studies show general agreement concerning the temperature dependence of the O3 cross sections, while a detailed quantitative analysis is currently in

Acknowledgements

This study was partly supported by the European Space Agency ESA-ESTEC under contract no. 11340/95/NL/CN, by the German Space Agency DARA/DLR under contract no. 50/EP/9207, and by the University of Bremen. The authors wish to thank J.W. Brault (Boulder, USA), A.P. Thorne and R.P. Learner (London, UK) for their suggestions about FT–UV spectroscopy, and M. Birk and G. Wagner (Oberpfaffenhofen, Germany) for a ray-tracing software of the Bruker IFS-120 HR spectrometer optics. We are much indebted

References (73)

  • D.E. Freeman et al.

    Planet Space Sci.

    (1984)
  • D. Daumont et al.

    Planet Space Sci.

    (1983)
  • J. Brion et al.

    Chem. Phys. Lett.

    (1993)
  • J.P. Burrows et al.

    J. Quant. Spectrosc. Rad. Transfer

    (1999)
  • A.C. Vandaele et al.

    J. Quant. Spectrosc. Radiat. Transfer

    (1998)
  • P. Bernath et al.

    Chem. Phys. Lett.

    (1998)
  • R. Schermaul et al.

    J. Quant. Spectrosc. Radiat. Transfer

    (1999)
  • R. Schermaul

    J. Quant. Spectrosc. Radiat. Transfer

    (1999)
  • S. Voigt et al.

    J. Mol. Spectrosc.

    (1996)
  • R.P. Wayne, Chemistry of Atmospheres, 3rd Edition, Oxford University Press, Oxford,...
  • World Meteorological Organisation, Scientific Assessment of Ozone Depletion: 1998, WMO Global Ozone Research and...
  • G. Brasseur, S. Solomon, Aeronomy of the Middle Atmosphere, Reidel, Dordrecht, 1986 (reprinted in...
  • B.J. Finlayson-Pitts, J.N. Pitts, Jr., Atmospheric Chemistry, Wiley, New York,...
  • W.B. Grant (Ed.), Ozone Measuring Instruments for the Stratosphere, Vol. 1, Collected Works in Optics, Optical Society...
  • R.E. Huffmann, Atmospheric Ultraviolet Remote-Sensing, Vol. 52, International Geophysics Series, Academic Press, San...
  • J. Chappuis

    C. R. Acad. Sci.

    (1880)
  • J. Chappuis

    C. R. Acad. Sci.

    (1882)
  • G.L. Humphrey et al.

    J. Phys. Chem.

    (1947)
  • A. Vassy et al.

    J. Chem. Phys.

    (1948)
  • L. Lefebvre

    C. R. Acad. Sci.

    (1953)
  • E. Vigroux

    Ann. Phys. Paris

    (1953)
  • E.C.Y. Inn et al.

    J. Opt. Soc. Am.

    (1953)
  • Y. Tanaka et al.

    J. Chem. Phys.

    (1953)
  • A.G. Hearn

    Proc. Phys. Soc. London

    (1961)
  • W.B. DeMore et al.

    J. Phys. Chem.

    (1964)
  • E. Vigroux

    Ann. Phys.

    (1967)
  • M. Griggs

    J. Chem. Phys.

    (1968)
  • R.D. Hudson

    Rev. Geophys.

    (1971)
  • R.D. Hudson

    Can. J. Chem.

    (1974)
  • J.C.D. Brand et al.

    Can. J. Phys.

    (1977)
  • C.M. Penney, NASA Contract Report 158977,...
  • K.F. Klenk

    Appl. Opt.

    (1980)
  • J. Brion et al.

    J. Phys. Lett. Orsay Fr.

    (1984)
  • A.M. Bass, R.J. Paur, in: C. Zerefos, A. Ghazi, D. Reidel (Eds.), Proceedings of the Quadrennial Ozone Symposium,...
  • R.J. Paur, A.M. Bass, in: C. Zerefos, A. Ghazi (Eds.), Proceedings of the Quadrennial Ozone Symposium, Halkidiki,...
  • L.T. Molina et al.

    J. Geophys. Res. D

    (1986)
  • Cited by (183)

    • Heterogeneity of volatile sources along the Halmahera arc, Indonesia

      2021, Journal of Volcanology and Geothermal Research
    • Vertically increased NO<inf>3</inf> radical in the nocturnal boundary layer

      2021, Science of the Total Environment
      Citation Excerpt :

      NO2 and O3 were measured in the near UV spectral region and analyzed in the fitting range of 349–374 nm and 309–341 nm, respectively, while NO3 was analyzed in the wavelength range of 617–672 nm. The absorption cross-sections of NO2 (Voigt et al. 2002), O3 (Voigt et al. 2001), NO3 (Yokelson et al. 1994), H2O (from The HITRAN Database) and solar spectra (Kurucz et al. 1985) were considered in the spectra fitting procedure. The purpose of adding the solar spectra is to eliminate the interference of the Fraunhofer line.

    View all citing articles on Scopus

    Dedicated to the memory of Richard C.M. Learner.

    View full text