Adaptation of a Fourier transform spectrometer as a reference instrumentfor solar UV irradiance measurementsPeter Meindl, Christian Monte, and Martin Wähmer Citation: AIP Conf. Proc. 1531, 829 (2013); doi: 10.1063/1.4804898 View online: http://dx.doi.org/10.1063/1.4804898 View Table of Contents: http://proceedings.aip.org/dbt/dbt.jsp?KEY=APCPCS&Volume=1531&Issue=1 Published by the AIP Publishing LLC. Additional information on AIP Conf. Proc.Journal Homepage: http://proceedings.aip.org/ Journal Information: http://proceedings.aip.org/about/about_the_proceedings Top downloads: http://proceedings.aip.org/dbt/most_downloaded.jsp?KEY=APCPCS Information for Authors: http://proceedings.aip.org/authors/information_for_authors

Downloaded 21 Aug 2013 to 193.5.60.125. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

Adaptation of a Fourier Transform Spectrometer as a Reference Instrument for Solar UV Irradiance

Measurements

Peter Meindl, Christian Monte and Martin Wähmer

Physikalisch-Technische Bundesanstalt (PTB), Abbestraße 2-12, D-10587, Berlin, Germany

Abstract. Within the European Metrology Research Project Traceability for surface spectral solar ultraviolet radiation a commercially available Fourier transform spectrometer (FTS) will be adapted for spectral solar UV irradiance measurements to demonstrate the feasibility of using this type of device as an alternative reference spectroradiometer. The investigated Fourier transform spectrometer has a comparatively small wavelength uncertainty between 2 pm to 5 pm in the wavelength range from 250 nm to 500 nm, and the wavelength scale of this type of instrument is inherently traceable to the SI units. The investigation of the FTS has shown that the dynamic range is limited to about three to four orders of magnitude. However, improvements of the dynamic range seem to be possible mainly by reducing the bandwidth of the instrument.

Keywords: Solar UV irradiance, Dissemination, Radiometric units, Fourier transform spectroradiometer. PACS: 92.70.Qr

INTRODUCTION

Several types of primary standards are in use at national metrology institutes for the realization of radiometric units. These primary standards realize the units with lowest possible uncertainty. However, primary standards like e.g. black body radiators [1] or cryogenic radiometers [2] are not suitable for the dissemination of radiometric units to remote measuring sites that are used for solar irradiance measurements. For this purpose, portable reference instruments have to be used as transfer standards. Up to now, scanning spectroradiometers are in use as reference instruments [3]. This type of radiometer has a couple of disadvantages. For example, the solar spectrum is scanned sequentially which needs minutes of time for each spectrum and limits the temporal resolution of the radiometer. This is a big disadvantage considering varying atmospheric conditions. The standard uncertainty of these instruments for solar irradiance measurements is around 2.3% to 4.4% depending on wavelength and solar zenith angle [3].

It is the goal of the European Metrology Research Project "Traceability for surface spectral solar ultraviolet radiation" (EMRP ENV03) to enhance the reliability of spectral solar UV radiation measurements at the Earth’s surface by developing new techniques and devices that enable a traceability of solar UV irradiance measurements of better than 2%. Wavelength accuracies of better than 50 pm are required to reach nominal uncertainties of 1% to 2% due to the steep decrease of solar UV irradiation below 330 nm over many orders of magnitude. For the same reason, a high dynamic range over at least five orders of magnitude is necessary to cover the wavelength range between 280 nm and 400 nm. Furthermore, the rapid temporal variation of solar UV radiation due to atmospheric conditions (e.g. fast moving clouds) requires fast spectroradiometers. It should be noted that traceability of solar UV irradiance measurements to SI units concerns the irradiance itself but also the wavelength scale of the measured spectra.

Fourier transform spectroradiometers are in use for solar irradiance measurements with high wavelength resolution [4]. In contrast to these facilities, the goal of this project is to evaluate the usability of a commercially available Fourier transform spectrometer as a portable reference spectroradiometer. The usage of Fourier transform spectroradiometers may improve the dissemination of absolute irradiance scales due to the specific advantages of these instruments [5, 6]. Fourier transform spectrometers have a high throughput which is caused by the circular aperture of these instruments (Jacquinot or throughput advantage). There are no diffraction losses to higher-order spectra as it is the case in grating spectrometers. All wavenumbers are measured simultaneously which allows fast measurements (multiplex advantage). Fourier transform spectrometers have a wide free spectral range and therefore cover broad spectral ranges with high resolution and high wavenumber accuracy. Modern FT spectrometers often

Radiation Processes in the Atmosphere and Ocean (IRS2012)AIP Conf. Proc. 1531, 829-832 (2013); doi: 10.1063/1.4804898

© 2013 AIP Publishing LLC 978-0-7354-1155-5/$30.00

829

Downloaded 21 Aug 2013 to 193.5.60.125. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

use built in HeNe lasers for the measurement of the position of the moving mirror. This laser can be used for the wavenumber calibration of the FTS.

The concept of this work is to use a commercially available FTS (Bruker Vertex80v) which will be fitted with a global entrance optics to be able to perform measurements of the solar UV irradiance. The irradiance calibration of this spectroradiometer will be performed by using the calculable radiation of a black body radiator [7, 8]. The calibration will therefore be conducted in the same way as it is done with scanning spectroradiometers [9]. This work is currently under preparation.

WAVENUMBER TRACEABILITY OF FOURIER TRANSFORM SPECTROMETERS

The knowledge of the position of the moving FTS interferometer mirror is necessary to calculate the absolute wavenumber scale of the measured spectra. The Fourier transform spectrometer Bruker Vertex 80v uses the radiation of a HeNe laser at 633 nm to determine these scan positions of the interferogram. The HeNe laser beam is coupled into the interferometer coaxial to the radiation under investigation. Following a recommendation of the International Committee for Weights and Measures (CIPM), the frequency (respectively the wavelength) of unstabilized HeNe lasers has been added to the list of standard frequencies of the mise en pratique of the definition of the meter [10, 11]. For this reason, the meter can be realized by using the wavelength of an unstabilized HeNe laser as a primary standard. Consequently, the wavenumber scale of the FTS can be assumed to be traceable to the SI as it is. However, a few conditions have to be fulfilled in order to use the built-in HeNe laser as a wavelength primary standard:

• The laser has to be a non-tuneable helium–neon laser operating on the 3s2 → 2p4 transition (633 nm) without contamination by radiation from other transitions (e.g. 640 nm) [11].

• The built-in laser has to be perfectly in line with the beam under test because any misalignment leads to a wavenumber error.

Both conditions may be checked by performing an FTS measurement of the wavenumber of an external HeNe laser whose properties are well known. If the HeNe laser operates at the 3s2 → 2p4 transition, the CIPM recommended value of the vacuum wavelength of this laser is 632.9908 nm with a relative standard uncertainty of 1.5 × 10-6. The corresponding vacuum wavenumber is (15798.018 ± 0.024) cm-1.

Several measurements concerning the wavenumber calibration and the spectral resolution of the FTS have been performed. In particular, the wavelength uncertainty has been measured by using the radiation of a well known external HeNe laser and of a mercury pencil lamp (Fig. 1). It has been found that the measured HeNe wavenumber agrees with the SI value within the standard uncertainty of the measurement which was about 0.03 cm-1. However, the measurements of the mercury spectral lines show a slight systematic deviation from the literature values [12, 13]. These deviations are probably caused by a marginal misalignment of the internal HeNe laser. The deviations can be corrected or otherwise simply added to the wavenumber uncertainty of the FTS.

FIGURE 1. Deviation of the measured HeNe wavenumber from the SI value [11] and the deviations of the measured mercury

peak wavenumbers from the literature data (upper line: wavenumbers from [12], lower line: wavenumbers from [13]). The uncertainty bars represent standard uncertainties and include the uncertainty of the peak wavenumber determination and the

uncertainty of the literature data.

-­‐0.50

-­‐0.25

0.00

0.25

0.50

15000 20000 25000 30000 35000 40000

Wavenumber  Deviation  

(Vertex  80v  -­‐Literature)  /  cm-­‐1

Wavenumber  /  cm-­‐1

measurements  of  external  HeNe  laser  at  15798.018  cm-­‐1

measurements  of  mercury  spectral  lines

830

Downloaded 21 Aug 2013 to 193.5.60.125. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

A further uncertainty contribution originates from misaligned parts of the beam under investigation caused by large entrance apertures (off-axis beams). As a result the combined wavenumber uncertainty for measurements of spectral lines in the UV can be assumed to be around 2 pm at 250 nm and 5 pm at 500 nm. These wavelength uncertainties are an order of magnitude below the demanded 50 pm which are required to reach irradiance uncertainties of 1% to 2%. As a result, the wavelength scale of the FTS is inherently traceable to the SI with an uncertainty that will not limit the irradiance uncertainty at solar UV measurements. This is a clear advantage of this type of instrument compared to scanning spectroradiometers.

DYNAMIC RANGE OF THE FOURIER TRANSFORM SPECTROMETER

A dynamic range over at least five orders of magnitude is demanded for solar UV irradiance measurements to cover the wavelength range between 280 nm and 400 nm. However, it has been found that the dynamic range of the FTS is limited by ghost spectra (Fig. 2). This artefact can be explained by periodic position errors of the scanning mirror of the FTS.

According to the Nyquist–Shannon sampling theorem, the maximum resolvable frequency is half of the sampling frequency [14]. Consequently, the interferogram has to be sampled many times at each HeNe laser interferogram period to resolve the wavenumbers in the UV. Actually, the interferogram is sampled at a frequency which is eight times the frequency of the HeNe laser interferogram. In this way, the minimum resolvable wavelength is 158 nm which covers the range of solar UV. The sampling positions are calculated by a signal processor that interpolates the interferogram of the HeNe laser radiation [15, 16]. The ghost spectra can be explained by a slight periodic error of the interpolation of the sample positions. The amplitude of the ghost spectra is about three to four orders of magnitude below the amplitude of the original spectrum. For this reason, the expected dynamic range to be covered by the spectrometer is about three to four orders of magnitude.

The periodic position-error ghosts can be reduced by a number of possibilities. First of all, slowing down the speed of the moving mirror can reduce the ghost amplitudes by a factor of two or three (when sampling only with 2.5 kHz instead of 10 kHz HeNe laser interferogram frequency). The reduction of the signal bandwidth by using detectors with limited spectral range (e.g. GaP detector) or optical filters is another way to minimize ghost spectra. Due to the systematic and periodic behaviour of the position errors an at least partial mathematical correction seems to be possible. Eventually, the position-error ghosts could be reduced with an improved FTS design adapted to measurements in the UV spectral range. It seems possible to improve the dynamic range to at least five orders of magnitude in a limited spectral range of 280 nm to 400 nm. However, further investigations are necessary to prove this.

FIGURE 2. Xenon lamp spectrum recorded with a silicon detector. The xenon spectral lines around 11000 cm-1 occur at several other wavenumbers as ghost lines.

831

Downloaded 21 Aug 2013 to 193.5.60.125. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

SUMMARY

The goal of this project is to evaluate the usability of a commercially available Fourier transform spectrometer as a portable reference spectroradiometer for solar UV irradiance measurements. The usage of Fourier transform spectroradiometers may improve the dissemination of absolute irradiance scales to remote measuring sites due to the specific advantages of this type of instrument.

The Fourier transform spectrometer has a comparatively small wavelength uncertainty between 2 pm to 5 pm in the wavelength range from 250 nm to 500 nm. This is well below the 50 pm which are demanded to obtain an uncertainty of irradiance measurements of less than 2%. The wavelength uncertainty will therefore not limit the uncertainty of solar UV irradiance measurements. Furthermore, the wavelength scale can be assumed to be traceable to the SI via the built-in HeNe laser. This property is a big advantage compared with scanning spectroradiometers. The dynamic range of the unmodified commercially available FTS is limited by systematic distortions of the spectrum to about three to four orders of magnitude. Few improvements are possible, and it seems to be possible to resolve a dynamic range of at least five orders of magnitude by reducing the bandwidth of the instrument.

Future investigations will concern the calibration of the spectral irradiance responsivity including the uncertainty of the calibration. This will also include the investigation of the stability of the irradiance responsivity which is of importance in order to achieve a stable irradiance traceability to the SI and the ability to use the FT spectroradiometer as a reference instrument also under field conditions. Finally, a comparison of the calibrated Fourier transform spectroradiometer with the portable scanning spectroradiometer QASUME [3] will be performed.

ACKNOWLEDGMENTS

The work leading to this study was partly funded by the EMRP ENV03 Project "Traceability for surface spectral solar ultraviolet radiation". The EMRP is jointly funded by the EMRP participating countries within EURAMET and the European Union.

REFERENCES

1. J. Hartmann, Physics Reports 469, 205-269 (2009). 2. J. E. Martin, N. P. Fox and P. J. Key, Metrologia 21, 147-155 (1985). 3. J. Gröbner, J. Schreder, S. Kazadzis, A. F. Bais, M. Blumthaler, P. Görts, R. Tax, T. Koskela, G. Seckmeyer, A. R. Webb and

D. Rembges: Appl. Opt. 44 (25), 5321-5331 (2005). 4. Solar spectral atlases obtained with the Fourier Transform Spectrometer at the McMath/Pierce Solar Telescope situated on

Kitt Peak, Arizona, USA, operated by the National Solar Observatory, NOAO - http://nsokp.nso.edu/dataarch.html. 5. S. P. Davis, M. C. Abrams and J. W. Brault: "Fourier Transform Spectrometry", Waltham, MA: Academic Press, 2001. 6. W. Herres and J. Gronholz, Comp. Appl. Lab. 2, p.216 (1984). 7. J. Hollandt, R. Friedrich, B. Gutschwager, D. R. Taubert and J. Hartmann: High Temperatures - High Pressures 35/36, 379-

415 (2003/2004). 8. J. Hartmann, K. Anhalt, R. D. Taubert and J. Hollandt, Int. J. Thermophys. 32, 1707-1718 (2011). 9. J. Gröbner and P. Sperfeld, Metrologia 42, 134–139 (2005). 10. 96th Meeting of the CIPM (International Committee for Weights and Measures), Recommendation 2 (CI-2007): “On the

value and uncertainty of unstabilised He-Ne lasers”. 11. J. A. Stone, J. E. Decker, P. Gill, P. Juncar, A. Lewis, G. D. Rovera and M. Viliesid, Metrologia 46, 11–18 (2009). 12. C. J. Sansonetti, M. L. Salit and J. Reader, Appl. Opt. 35 (1), 74-77 (1996). 13. C. J. Sansonetti and J. Reader, Phys. Scripta 63, 219-242 (2001). 14. C. E. Shannon, Proc. IRE 37, 10–21 (1949). 15. J. W. Brault, Appl. Opt. 35, 2891-2896 (1996). 16. P. R. Griffiths and J. A. de Haseth, "Fourier Transform Infrared Spectrometry", Hoboken, NJ: Wiley, 2007.

832

Downloaded 21 Aug 2013 to 193.5.60.125. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions