Document not found! Please try again

Terahertz atmospheric attenuation and continuum

4 downloads 0 Views 831KB Size Report
From these data, the absorption coefficient as a function of ... sensing, can cause a 30dB difference between calculated and observed attenuation. ... mp and sample .... Also shown in the Figure 3 are the effects of the instrumentation distortion.
Terahertz atmospheric attenuation and continuum effects

David M. Slocum,*a Thomas M. Goyette,a Elizabeth J. Slingerland,†a Robert H. Giles,a and William E. Nixonb a

Submillimeter-Wave Technology Laboratory, University of Massachusetts Lowell, Lowell, MA 01854, United States; bUnited States Army, National Ground Intelligence Center, Charlottesville, VA 22911, United States ABSTRACT Remote sensing over long path lengths has become of greater interest in the terahertz frequency region. Applications such as pollution monitoring and detection of energetic chemicals are of particular interest. Although there has been much attention to atmospheric effects over narrow frequency windows, accurate measurements across a wide spectrum is lacking. The water vapor continuum absorption spectrum was investigated using Fourier Transform Spectroscopy. The continuum effect gives rise to an excess absorption that is unaccounted for in just a resonant line spectrum simulation. The transmission of broadband terahertz radiation from 0.300THz - 1.5THz through air with varying relative humidity levels was recorded for multiple path lengths. From these data, the absorption coefficient as a function of frequency was determined and compared with model calculations. The intensity and location of the strong absorption lines were in good agreement with spectral databases such as the 2008 HITRAN database and the JPL database. However, a noticeable continuum effect was observed particularly in the atmospheric transmission windows. A small discrepancy still remained even after accounting for continuum absorption using the best available data from the literature. This discrepancy, when projected over a one kilometer path length, typical of distances used in remote sensing, can cause a 30dB difference between calculated and observed attenuation. From the experimental and resonant line simulation spectra the air-broadening continuum parameter was calculated and compared with values available in the literature. Keywords: water vapor; atmospheric absorption; continuum; terahertz; spectroscopy

1 INTRODUCTION The propagation of electromagnetic radiation, in particular at terahertz frequencies, through the Earth’s atmosphere is an important field of study. Applications like remote sensing, imaging, detection of concealed items, astronomical observations, and more are used in industries like defense, medical, environmental, and academia and are more commonly employing terahertz frequencies. Water vapor has thousands of rotational and vibrational absorption lines positioned throughout the whole electromagnetic spectrum and is a main contributor to atmospheric attenuation in the terahertz regime. Propagation of radiation over extended distances is not feasible at many frequencies. Measurements to be performed in this frequency region require significant attention to the specific environment. The absorption peaks of water vapor have been extensively studied1-3 however the window regions between the strong absorption lines are more difficult to characterize. Current models4-8 require the use of an empirical continuum contribution to the spectrum for an accurate comparison to experimental spectra. In the microwave region there is much evidence that the continuum contribution scales as the square of the frequency with a negative dependence on temperature.8-10 However, at higher frequencies in the terahertz region only a few studies have seen the quadratic dependence on frequency.11-13 Further complicating the issue, there is uncertainty as to the cause of the continuum as well as the values of the continuum parameters. Since the continuum contribution is empirical, it depends on both experimental data as well as model predictions. The choice of line parameters, line shape function, and number of * Corresponding author. Email: [email protected] phone: 978-934-1300 www.stl.uml.edu † Current address: Metron Inc., 1818 Library St Ste 600, Reston, VA 20190, United States

Terahertz Physics, Devices, and Systems VII: Advanced Applications in Industry and Defense, edited by Mehdi F. Anwar, Thomas W. Crowe, Tariq Manzur, Proc. of SPIE Vol. 8716, 871607 © 2013 SPIE · CCC code: 0277-786X/13/$18 · doi: 10.1117/12.2015471 Proc. of SPIE Vol. 8716 871607-1 Downloaded From: http://proceedings.spiedigitallibrary.org/ on 08/02/2013 Terms of Use: http://spiedl.org/terms

absorption lin nes used in thee model can sig gnificantly affeect the calculatiion of the conttinuum parameeters. Also a lim mited number of ex xperimental stu udies performed at terahertz frequencies f resstricts a compreehensive and aaccurate determ mination of the continu uum parameterrs. Only a few w broadband stu udies have beenn performed inn the terahertz rregion, most of which were perform med in a laboratory environment using eitheer pure nitrogenn or oxygen gaas as the foreignn broadening ggas.12-15 Control over the experimen ntal conditions is gained throu ugh the sacrificce of the broaddening gas, which is of a diffeerent composition from Earth’s atmosphere. a Other O works hav ve studied atmoospheric air,11,16-18 however tthese studies doo not have control over many of the t experimenttal parameters.. The present study s used dry air as the foreiign broadening g gas, which haas the same com mposition of aatmospheric airr, while maintaining control c over th he experimental parameters. Broadband B trannsmission dataa was taken bettween 0.3-1.5T THz over multiple path h lengths and humidity h levels to determine the t absorption coefficient of atmospheric aiir. These resullts were combined wiith simulated ab bsorption coeffficient data callculated using line parameterrs from the 20008 HITRAN Database19 (H HITRAN) to fiind the air-broaadened continu uum coefficientt. When the coontinuum contrribution is incorporated into the absorp ption coefficien nt calculationss, the resulting spectrum can bbe used to idenntify the best frequency ran nge for which an experiment is to be perforrmed.

2 PR ROCEDURE E mental setup 2.1 Experim Fourier Transform Spectrosscopy (FTS) was w used to colllect transmissioon data of terahhertz radiationn through air wiith varying relative humidity leevels and path lengths. A Pik ke Technologiees variable pathh length gas cell was used to contain the sample an nd allow for multiple m path len ngth measurem ments. Figure 1 shows a scheematic of the exxperimental settup along with a diagram of thee variable path length gas celll. The cell wass mounted intoo the sample chhamber of a Brruker Vertex 80V Fourier F Transfo form Infrared (F FTIR) Spectrom meter with a m mercury arc lam mp source alignned to the opticcal path of the spectro ometer. The trransmitted radiiation was colleected using a liiquid helium cooled bolometter from IR labs with a vacuum pum mp attached to the t liquid heliu um chamber to lower the operrating temperatture to 1.6K.

(b)

'......-..-..

(a)

HL imidity

Se!nsor

,

AI\

I

-.C-

.