Raman efficiencies of natural rocks and minerals: Performance of a remote Raman system for planetary exploration at a distance of 10 meters

https://doi.org/10.1016/j.saa.2005.02.030Get rights and content

Abstract

Raman spectroscopy is a powerful technique for materials analysis, and we are developing and analyzing a remote Raman system for use on a planetary lander or rover. We have acquired data at a distance of 10 m from a variety of geologic materials using different instrument designs. We have employed a pulsed laser with both an ungated detector and a gated detector. A gated detector can reduce long-lived fluorescence while still collecting all Raman signal. In order to design a flight instrument, we need to quantify how natural surfaces will respond to laser stimulus. We define remote Raman efficiency of natural surfaces as the ratio of radiant exitance leaving a natural surface to the irradiance of the incident laser. The radiant exitance of a natural surface is the product of the sample radiance, the projected solid angle, and the full-width-half-maximum of the Raman signal. We have determined the remote Raman efficiency for a variety of rocks and minerals. The best efficiencies are achieved for large, clear, single crystals that produce the most radiant exitance, while darker fine-grained mineral mixtures produce lower efficiencies. By implementing a pulsed laser, gated detector system we have improved the signal detection and have generally decreased the integration time necessary to detect Raman signal from natural surfaces.

Introduction

In situ measurements and observations are an integral part of planetary exploration. The instruments carried by landers and rovers can be highly specialized to perform detailed chemical and mineralogical analyses. Future landers and rovers can carry new generations of instruments to Mars as they build upon the legacy of the two 2004 Mars Exploration Rovers (MERs). Remote sensing instruments are a priority for lander and rover design as they can provide independent investigations of the surface as well as support sample collection. Sample selection is necessary to identify important rocks and minerals for further analysis either by on-board instruments or for a sample return and more thorough analysis in terrestrial laboratories. By developing a quick and efficient way of selecting the most important samples for analysis, rover time and power can be conserved. Remote Raman spectroscopy is a capable technology for such functions.

Raman spectroscopy relies on a change in induced polarization to make mineralogical identifications, versus passive spectroscopy, which requires a change in a permanent dipole for detection. Viking [1] and Mars Pathfinder [2] used multispectral imaging to select samples while the 2004 MERs use thermal IR spectroscopy combined with multispectral images [3]. Both of these techniques are passive; however, because Raman spectroscopy is an active technique, it provides capabilities beyond those of multispectral and IR spectroscopy. Emission and reflectance spectra (VIS and IR) are prone to broad, overlapping spectral features. On the other hand, Raman scattering produces sharp, narrow lines. These Raman lines can be used to identify a wide range of materials including minerals, water, ice, glass (water and glass are strong IR absorbers), biomarkers, and biomolecules (e.g. [4], [5], [6], [7]). For many of these materials precise compositions can be obtained because the positions of the Raman features depend upon molecular bonds. Inelastic scattering of light is the fundamental principle of Raman spectroscopy, and therefore, Raman spectroscopy is based on a different mechanism than VIS/IR spectroscopy and does not require changes in permanent dipolar molecules. Raman spectroscopy will contribute greatly to the current knowledge of Mars by complementing current passive vibrational spectroscopy datasets. In situ Raman spectroscopy was proposed on planetary missions for mineral analyses [8]. Here we continue our development of the remote Raman system at the University of Hawaii, previously described [9], [10] for use on a planetary rover or lander. We show that many materials of planetary interest can be identified at 10 m.

Stand-off hard-target Raman spectroscopy has been demonstrated as a portable remote system of moderate size that can acquire spectra up to a distance of 20 m [11]. We emphasize the planetary applications of such an instrument. In order to design a flight-ready instrument, its performance on another planet must be predicted. On a spacecraft, resources (mass, time, electric power) are limited, and the amount of laser power required for a given number of Raman spectra must be known. Raman scattering is weak when compared to other remote sensing techniques, and approximately only one in 108 photons induces Raman shift in a molecule [12]. By optimizing the efficiency of our system and reducing noise and background, we can increase the quality of spectra and the number of low-Raman cross-section minerals that can be detected by remote Raman spectroscopy on the Martian surface with a limited supply of power.

Section snippets

Equipment

Our excitation source is a 532 nm, pulsed ULTRA CFR Big Sky Nd:YAG laser with 8 ns pulse width, repetition rate of 20 Hz, and a 35 mJ/pulse power. A 12.5 cm (5 in.) diameter f/15 Meade ETX-125 Maksutov-Cassegrain telescope collects the Raman scattered light. The light is transmitted through a holographic notch filter mounted near the back of the telescope. This filter prevents unshifted laser light from entering the spectrometer. We used a HoloSpec f/2.2 spectrometer coupled to a cooled, intensified

Raman spectra and interpretation

We examined a number of rock and mineral Raman spectra collected at a distance of 10 m for characteristic Raman peaks. In the CW mode with coaxial FO configuration of the remote Raman system (Fig. 1), the Raman signal from darker rocks, which are typically weak Raman scatters, was not detectable above the background. The reduction in Raman signal may possibly be due to the strong absorption of 532 nm laser radiation by the sample and the presence of low-intensity background radiation in our

Radiance and efficiency

The radiance and calculated remote Raman efficiencies for natural surfaces in the FO-coupled, coaxial CW detection mode are summarized in Table 2. The variation within each sample depends somewhat on orientation. In Table 2, data from different sampling runs is averaged together. The highest efficiencies in CW (coaxial, directly coupled) mode are derived from quartz, fibrous gypsum, gypsum selenite (#95), calcite (crystal), magnesite, and dolomite marble: all having sample efficiencies ∼10−8.

Application

Other instruments and analytical techniques are available for planetary missions, but each has limitations. Mars Pathfinder utilized Alpha Proton X-ray spectrometry [31], which requires close contact with the sample surface. X-ray fluorescence as implemented by the Viking landers (e.g. [32]), as well as mass spectrometry [33], require sample acquisition and preparation. And Raman spectroscopy is capable of detecting and identifying many more minerals than are possible with Fe-Mössbauer

Summary

Remote Raman spectroscopy is a viable technique for material analysis on future planetary missions. We have observed characteristic Raman shift for a variety of natural rocks and minerals at a distance of 10 m. After testing several system designs, we achieve the best signal detection in the oblique, directly coupled gated mode. We will not abandon the coaxial, fiber-optic-coupled system as this design is very useful for integration onto planetary landers and rovers. We will continue to

References (40)

  • D.D. Wynn-Williams et al.

    Icarus

    (2000)
  • S.K. Sharma et al.

    Spectrochim. Acta: Part A

    (2003)
  • A.K. Baird et al.

    J. Geophys. Res.

    (1977)
  • M.P. Golombek et al.

    J. Geophys. Res.

    (1999)
  • S.W. Squyres et al.

    Science

    (2004)
  • D.L. Dickensheets et al.

    J. Raman Spectrosc.

    (2000)
  • M.C. Sparrow et al.

    J. Appl. Spectrosc.

    (2001)
  • M.C. Storrie-Lombardi et al.

    Rev. Sci. Instrum.

    (2001)
  • A. Wang et al.

    J. Geophys. Res.

    (1995)
  • P.G. Lucey et al.

    Lunar Planet. Sci. Conf.

    (1998)
  • S.M. Angel et al.

    Appl. Spectrosc.

    (1992)
  • T.J. Wdowiak et al.

    Lunar Planet. Sci. Conf.

    (1995)
  • A.K. Misra, S.K. Sharma, C.H. Chio, P.G. Lucey, B. Lienert, Spectrochim. Acta: Part A (2005), this...
  • F.R. Dollish et al.

    Characteristic Raman Frequencies of Organic Compounds

    (1974)
  • J. Urmos et al.

    Am. Mineral.

    (1991)
  • F. Pauer et al.

    Geophys. Res. Lett.

    (1996)
  • H.H. Eysel et al.

    J. Raman Spectrosc.

    (1988)
  • R.L. McCreery
  • J.D. Stopar et al.

    Lunar Planet. Sci. Conf.

    (2003)
  • J.D. Stopar et al.
  • Cited by (46)

    View all citing articles on Scopus
    View full text