Time-resolved Raman spectroscopy for in situ planetary ...authors.library.caltech.edu/20171/1/Blacksberg2010p11444Appl... · Time-resolved Raman spectroscopy for in situ planetary

Time-resolved Raman spectroscopy forin situ planetary mineralogy

Jordana Blacksberg,1,* George R. Rossman,2 and Anthony Gleckler3

1Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California 91109, USA2California Institute of Technology, Division of Geological and Planetary Sciences, Pasadena, California 91125, USA

3GEOST, Inc., Tucson, Arizona 85741, USA

*Corresponding author: [email protected]

Received 12 April 2010; revised 20 July 2010; accepted 12 August 2010;posted 13 August 2010 (Doc. ID 126635); published 8 September 2010

Planetary mineralogy can be revealed through a variety of remote sensing and in situ investigations thatprecede any plans for eventual sample return. We briefly review those techniques and focus on the cap-abilities for on-surface in situ examination of Mars, Venus, the Moon, asteroids, and other bodies. Overthe past decade, Raman spectroscopy has continued to develop as a prime candidate for the next gen-eration of in situ planetary instruments, as it provides definitive structural and compositional informa-tion of minerals in their natural geological context. Traditional continuous-wave Raman spectroscopyusing a green laser suffers from fluorescence interference, which can be large (sometimes saturatingthe detector), particularly in altered minerals, which are of the greatest geophysical interest. Takingadvantage of the fact that fluorescence occurs at a later time than the instantaneous Raman signal,we have developed a time-resolved Raman spectrometer that uses a streak camera and pulsed miniaturemicrochip laser to provide picosecond time resolution. Our ability to observe the complete time evolutionof Raman and fluorescence spectra in minerals makes this technique ideal for exploration of diverse pla-netary environments, some of which are expected to contain strong, if not overwhelming, fluorescencesignatures. We discuss performance capability and present time-resolved pulsed Raman spectra collectedfrom several highly fluorescent and Mars-relevant minerals. In particular, we have found that conven-tional Raman spectra from fine grained clays, sulfates, and phosphates exhibited large fluorescent sig-natures, but high quality spectra could be obtained using our time-resolved approach. © 2010 OpticalSociety of AmericaOCIS codes: 300.6500, 300.6450, 300.6280, 300.6190.

1. Introduction

Since its discovery more than 80 years ago, and ele-vation 30 years later with the invention of the laser,Raman spectroscopy has evolved into the techniqueof choice for the in situ exploration of planetarybodies because it addresses a primary goal of miner-alogical analysis: determination of structure andcomposition. With their high spectral and spatial re-solution, laser Raman spectrometers for both surfaceand subsurface analysis are currently under develop-

ment for a diverse set of planetary targets, includingMars and its moons Phobos and Deimos, Venus,Earth’s moon, and asteroids. Localized analyses ofplanetary surfaces using laser Raman spectroscopycomplement global satellite remote sensing usingvisible-infrared imaging from orbit and are the primeconsideration for preselection of rock samples priorto caching for potential sample return missions.

A. Comparison of Techniques

Several techniques can be used for mineralogicalanalysis of planetary bodies. From orbit, spectacularresults have been achieved using visible-IR imagingspectroscopy of Mars and the Moon. Two prime

0003-6935/10/264951-12$15.00/0© 2010 Optical Society of America

10 September 2010 / Vol. 49, No. 26 / APPLIED OPTICS 4951

examples are the Compact Reconnaissance ImagingSpectrometer for Mars (CRISM) on the Mars Recon-naissance Observer (MRO) and theMoonMineralogyMapper (M3) on the Lunar Reconnaissance Orbiter(LRO). The resulting global mineralogy maps can beplaced in detailed context when combined with highresolution imaging provided by instruments such asHiRISE on MRO. Additionally, gamma ray mappingof hydrogen (e.g., on Mars Odyssey) has led to theconclusion that huge amounts of water exist over alarge percentage of Mars.

Guided by orbital data, a host of in situ techniquessummarized in Table 1 can provide mineralogicalinformation locally within a well-defined geologicalcontext. Of these techniques, only x-ray diffraction(XRD) and Raman spectroscopy can definitively iden-tify nearly all crystallineminerals based upon uniquenarrowband spectroscopic features. For XRD, samplecollection is achieved through the use of a robotic armthat delivers a rock fragment that is typically pow-dered for analysis. Laser Raman spectroscopy offersthe important advantage that it is nondestructiveto the sample, can be operated meters away from thecontaminated area of the spacecraft (e.g., on a roverarm), and is directed to a specific target samplewithinthe context of its natural mineral setting. It is the di-rectional, intense, coherent nature of the laser thatprovides the important ability to focus down to a verysmall spot size comparable to mineralogical grains.This provides the foundation for two-dimensionalRaman mapping.

Although a highly successful method for mineralo-gical analysis from orbit, visible to near-IR spectro-scopy is inherently low-resolution because it relieson the solar reflectance spectra of solid-phasemateri-als. Hence, much of the subtlety of this technique isfound in painstaking analysis of low-resolution, poorcontrast features. Laser Raman spectroscopy [1],however, is inherently a high spectral resolution tech-nique, because the returned Raman spectra showextremelynarrowvibrational features. For a gas, theycan be considered infinitely narrow (<0:03 cm−1),but even for a solid, the Raman spectral featuresare considered extremely sharp (∼3–10 cm−1). Theuse of a moderately high resolution spectrometer cantherefore provide unique Raman fingerprints for un-ambiguousmineral identification.LaserRamanspec-troscopy fromamineral sampleunder laser excitationrelies on the measurement of scattered radiation,which differs in wavelength from the incident beamas a result of interaction with phonons. Because eachband in a Raman spectrum represents interaction ofthe incident light with a vibrational mode in the crys-tal, it is highly specific for a given mineral and can beused for identification and structural characteriza-tion of unknown samples. An added benefit is thatthe observed Raman shifts (defined as the shifts ofRaman peaks in wavenumber from the laser line)are nearly independent of the excitation wavelength,greatly simplifying the development of Raman data-

bases, which can draw from data taken using anyexcitation wavelength.

B. Mineralogy of Mars and Venus

As a result of the impressive orbital and in situ dataadded over the past decade or so, we have come to un-derstand Mars as a diverse body thought to have hadtwo distinct geological histories. The early historymaybe characterizedbyawater-rich environment en-abling the formation of minerals, such as clays. Thisneutral-to-alkaline environment is thought to haveevolved to a sulfurous acidic environment as aqueousactivity diminished and volcanic activity increased.Most of the aqueous alteration is likely to have oc-curred in the early Martian history with the forma-tion of minerals such as goethite; jarosite; Fe-, Mg-,and Ca- sulfates; hydrated sulfates (e.g., kieserite);phyllosilicates (e.g., montmorillonite clays); and Fe-,Mg-, and Ca- carbonates [2]. It is this depiction ofearlier Martian conditions that also leads us to theexciting prospect of discovering evidence of past lifeon Mars.

A number of richly successful orbital missions havelaid the foundation for in situ exploration onMars [3].Our current understanding of Martian planetarymineralogy is in large part inferred through the anal-ysis of data from the Thermal Emission Spectrometer(TES) on the Mars Global Surveyor (launched in1996), the Thermal Emission Imaging System(THEMIS) and Gamma Ray Spectrometer (GRS) onthe Mars Odyssey (launched in 2001), Visible andInfrared Mineralogical Mapping Spectrometer(OMEGA) on the Mars Express (launched 2003),and,most recently, theCompactReconnaissance Ima-ging Spectrometer for Mars (CRISM) [4] on the MarsReconnaissance Orbiter (MRO) (launched 2005) andits complementary HiRISE camera [5]. It is throughthese missions that we have evidence for the over-arching basaltic nature of Mars, and the discoveryof altered regions containing minerals such assulfates, phyllosilicates, iron oxides, zeolites, andwater ice.

Although Mars is primarily basaltic with typicallyonly a weak degree of alteration, much of our interestlies in the altered regions. It is these regions that weaim for with landers carrying in situ instruments.Evidence for past liquid water on Mars comes, inpart, from evidence of sulfate-rich sediments thatcontain minerals such as jarosite, calcium and mag-nesium sulfates [6], and hematite spheres, indenti-fied spectroscopically with Mini-TES on the MarsExploration Rover (MER) [7], that led to the realiza-tion that some of the sandstones were derived froman evaporitic source [8]. The presence of goethite atthe Columbia Hills site is of great importance be-cause, unlike hematite, this mineral can only formin the presence of water. However, with the exceptionof Fe-bearing sulfates, hematite, and goethite identi-fied by the MER Mössbauer instrument, there havebeen no alteration phases unambiguously identifiedby in situ instruments to date [2].

4952 APPLIED OPTICS / Vol. 49, No. 26 / 10 September 2010

Tab

le1.

Summaryofin

situ

Tec

hniques

forPlanetaryMineralogya

InSituTechn

ique

sforPlane

tary

Mineralog

y

Technique

Cap

abilities

App

roach

Exa

mples

ofus

ein

insitu

instru

men

ts—laun

chda

te

Las

er-ind

uced

brea

kdow

nsp

ectroscopy

(LIB

S)

Elemen

talcompo

sition

ofrock

andsoil

Highpo

wer

shortlaserpu

lseinitiatesaplas

mafrom

which

iscollectedtheligh

tem

ission

from

excited

stateelem

ents

andions

.

Che

mCam

onMarsScien

ceLab

oratory(M

SL)—

2011

;Ram

an/LIB

Son

prop

osed

Exo

MarsRov

er—20

18

X-ray

diffraction(X

RD)*

Struc

ture

and

compo

sition

Rockor

soilsa

mples

arepr

ocessedinto

apo

wde

ran

dan

alyz

edon

boardtheland

er.X

rays

aredirected

throug

hthesa

mplean

dadiffractionpa

tternis

collected

usingan

energy

discriminatingde

tector.

Che

Min

onMSL—20

11

X-ray

fluo

rescen

ce(X

RF)

Elemen

talcompo

sition

ofrock

andsoil(m

ost

sensitive

tohe

avier

elem

ents)

Rockor

soilsa

mples

arepr

ocessedinto

apo

wde

ran

dan

alyz

edon

boardtheland

er.Anx-raysour

cege

neratesfluo

rescen

txrays

inthesa

mplethat

arede

tected

byan

energy

discriminatingde

tector.

XRFon

Viking1&

2MarsLan

ders—19

75;

XRFon

Ven

era13

&14

Ven

usLan

ders—19

81;

Che

Min

onMSL—20

11

Neu

tron

spectroscopy

Elemen

talcompo

sition

ofrock

andsoil;pr

esen

ceof

water

inminerals

Neu

tron

sarege

neratedan

dpe

netratetheplan

etarysu

rface.

Scattered

neutrons

arede

tected

;region

swithwater

crea

temorelow

energy

thermal

neutrons

.

Dyn

amic

albe

doof

neutrons

(DAN)on

MSL—20

11;Neu

tron

-Activated

Gam

ma

Ray

Spe

ctrometer

(NAGRS)forthepr

oposed

SAGE-Ven

us—20

16mission

;NSHEND

prop

osed

forPho

bos-Gru

nt—20

11Alpha-pa

rticle

x-ray

spectroscopy

(APXS)

Elemen

talcompo

sition

ofrock

andsoil

Alpha

particles,

proton

s,an

dxrays

gene

ratedby

aradioa

ctive

sour

ceim

ping

eon

thesu

rface.

Backs

cattered

alph

apa

rticles

andge

neratedpr

oton

sor

xrays

canallbe

detected

for

elem

entalan

alysis.

APXSon

MarsPathfinde

r-199

6;APXSon

MarsExp

loration

Rov

er(M

ER)-200

3;APXS

onMSL—20

11

Mössb

auer

spectroscopy

*Com

position

ofiron

-con

tainingminerals

Gam

marays

gene

ratedby

aradioa

ctivesour

ceim

ping

eon

thesu

rface.

Reson

antab

sorp

tion

prov

ides

inform

ationon

thech

emical

environm

entof

iron

.

MiniaturizedMössb

auer

Spe

ctrometers

(MIM

OSII)on

MER—20

03

Las

erRam

ansp

ectroscopy

*Structure

and

compo

sition

ofminerals

intheirnaturalcontex

t

Las

erligh

tscatteredfrom

thesu

rfaceis

detected

.The

wav

elen

gthof

theRam

anscatteredligh

tpe

aks

arehigh

lymineral-spe

cific.

Ram

an/LIB

Son

prop

osed

Exo

MarsRov

er—20

18;

Propo

sedinstru

men

tsforDeimos

andVen

us

a Ofthesetech

niqu

es,thede

finitive

metho

dsformineralog

ical

iden

tification

deno

tedby

*arelimited

tox-raydiffraction,

Mössb

auer

spectroscopy,an

dRam

ansp

ectroscopy.


The next planned in situmission toMars, the MarsScience Laboratory [(MSL) to be launched in 2011] isexpected to change this fact [9]. It will carry theChemCam and CheMin instruments for both ele-mental andmineralogical analysis, respectively, withthe Sample Analysis at Mars (SAM) suite providingdetailed isotopic analysis in pyrolysis and combus-tion experiments. Because of the desire to studyan altered region with the MSL, the selection ofthe landing site is crucial. The orbital data has there-fore been essential in identifying the four potentiallanding sites at Mawrth Vallis, Gale Crater, Ebers-walde Crater, and Holden Crater. The MSL willuse the high resolution MastCam and the laser in-duced breakdown spectroscopy (LIBS) or ChemCamto rapidly survey the geology in the vicinity of therover, and the neutron spectrometer to survey forwater-containing minerals. The samples of greatestinterest will be processed into a powder and deliv-ered to CheMin (XRD/XRF) and SAM. The data fromCheMin will be the first truly definitive mineralogi-cal data outside of that obtained on iron-containingminerals using Mössbauer.

In situ Mars exploration beyond MSL is expectedto undertake the challenging goal of sample return.Because the quantity of rock and soil that could bereturned is limited (likely od the order of severalhundreds of grams, total), this would require precisein situ instruments that could definitively identifyinteresting samples for caching and possible return.Raman spectroscopy is the technique of choice be-cause, unlike XRD, it does not require sample pre-paration, and Raman spectra can be acquired onmultiphase rock and soil samples as they are foundin nature, preserving rock fabrics and textures. Theraw spectra provide unique fingerprints that canoften be easily interpreted without recourse to com-plex spectral deconvolution. The power of Ramanspectroscopy to unambiguously identify minerals,for example in a mixed phase Mars meteorite, hasbeen well-demonstrated [10]. Raman spectroscopicmapping in conjunction with complementary tech-niques such as imaging, elemental analysis (e.g.,LIBS), and ultraviolet fluorescence (for organic de-tection) [11] would provide a strong comprehensiveset of measurements for successful identification ofminerals and trace components.

Relative to Mars, very little is known about thesurface of Venus. The planet is surrounded by adense carbon dioxide atmosphere with a pressure90 times that of Earth. In addition, the surface tem-perature is around 450 °C, making it very difficult fora lander or probe to survive more than a few hours.The prospect of a rover spending months or years onthe surface, as has been the model for Mars, is highlyunlikely. If we are to plan for mineralogical analysison the Venus surface, rapid acquisition of spectrawould be crucial. The XRD experiment (CheMin)on theMars Science Laboratory will typically requirehours for sample preparation and integration timesof about10 × h for a single measurement. It is there-

fore prohibitively long for Venus without furthermodification. On the other hand, Raman spectro-scopy is a prime candidate for rapid mapping. Byeliminating the need for extensive sample prepara-tion, we could hope to surveymany rock and soil sam-ples during the short course of a landed mission,either for in situ analysis or potential sample returnselection.

Our current knowledge of the surface of Venus [12]is derived from a series of Russian Venera and Vegaprobes from 1970 to 1985, which returned imagesand high resolution radar maps, as well as x-rayfluorescence (XRF) spectra of rock and soil samples.The Venera–Vega landing sites were comprised ofsoil and finely bedded rock, with evidence suggestinga primarily mafic surface composed of tholeiiticbasalts and alkaline basalts. The Magellan missionin 1990 obtained radar images of the Venera–Vegalanding sites, piecing together after the fact thatthese missions explored what are described as plainswith wrinkle ridges. Magellan has also led to theidentification of the tessera terrain as the first can-didate for future in situ analysis and/or sample re-turn. Tessera, on which no spacecraft has landedto date, is morphologically diverse with ridges andgrooves, and it is expected to contain the oldestmaterial recognized on Venus. Our knowledge ofVenus mineralogy is based primarily on modelingof the XRF data, indicating high concentrations of si-licon, aluminum, iron, magnesium, and calciumoxide. Unfortunately, XRF does not provide informa-tion on lighter elements, and therefore, hydrogencontent, and thus hydration in minerals, is unknown.Modeling of minerals under Venus conditions sug-gests the stability of hydrated minerals such as hy-drous silicates, as well as OH−, containing nominallyanhydrous minerals such as pyroxenes [13]. Becausethis subject of mineralogical alteration and weather-ing in the Venus environment is of great interest, itperhaps suggests the need to get below the surface toaccess unweathered materials for analysis.

C. Mineralogy of the Terrestrial Moons and Asteroids

The Earth’s moon and the moons of Mars (Phobosand Deimos) are the only satellites of terrestrial pla-nets in our solar system. The origin, composition, andevolution of the Martian moons are still very much inquestion with various theories, including the captureof asteroids into the Martian orbit [14]. In addition,many asteroids are thought to be surviving protopla-nets and planetesimals, and can act as windows intovarious stages of the formation of the solar system. Aplanned sample return mission to the Mars moonPhobos (Phobos-Grunt) in 2011 by the Russian SpaceAgency [15] aims to clarify some of the questions ofits nature and origin, providing a more expansive un-derstanding of terrestrial moons, asteroids, and theformation of the planets. In addition to returningsamples to Earth, that mission includes an in situlander with a robotic arm that will scoop samples,with elemental/mineralogical analysis relying on


techniques such as alpha particle x-ray spectro-scopy (APXS), neutron spectrometry, and mass spec-trometric analysis of pyrolyzed samples. For thesecond Mars moon, Deimos, NASA is consideringproposals to land on the surface with laser Ramanspectroscopy playing a key role in mineralogicalassessment.

Our understanding of the geology of the Earth’smoon comes from a long history of Earth-based tele-scopic observations, orbiter observations, and directsampling. Lunar sample return from the Earth’smoon has been successfully accomplished by a seriesofmanned (Apollo, 1969–1972) and unmanned auton-omous (Soviet Luna, 1970–1976) missions. Two dis-tinct regions have been identified: the highlands,which are anorthositic in composition, and the darkmaria,whichare basaltic innature.Thegeology of nu-merous other regions is not yet well-understood. Anexample is the South Pole Aitken basin, the oldest,largest, and deepest impact feature on the Moon. Itcontains mineralogy that is expected to vary greatlyfrom what has been found at the Apollo landing sites.This was confirmed by spectroscopic data acquired byGalileo, Clementine, and Lunar Prospector missions.This region is therefore of great interest for futurein situ and potential sample return missions.

Raman spectroscopy has already been proven as asuccessful technique for determining the mineralogy,structure, and composition of lunar rocks, soils, andglasses returned to Earth by the Apollo missions[16]. The primary constituents in lunar samples,such as feldspar, pyroxene, and olivine, have been ea-sily identified using Raman spectroscopy. Interest-ingly, fluorescence interference was observed insome of the lunar samples, pointing to the potentialvalue of a time-resolved approach. Potential futurein situ missions to the Earth’s moon and the moonsof Mars could take advantage of Raman spectroscopy,particularly for studying permanently shadowedareas, such as those found in the lunar South Pole.In such regions, passive visible-IR reflectance spec-trometry could not be used, and an active source(laser) would likely be necessary to obtain usefulspectroscopic data from either enhanced visible-IRor from laser Raman interrogation.

Small and primitive planetary bodies like aster-oids are thought to be representative of the earlystages of solar system formation. In situ mineralogi-cal analysis of these bodies, while of great interest,would present additional challenges associated withlanding in a very low gravity environment of un-known topography. This underscores the need forpreliminary gravity and topographic mapping tohelp lower the risk of approaching and touchingdown onto these low-g bodies. Furthermore, if laserRaman instrumentation (e.g., focusing optical head)must be brought to surface samples using a roboticarm, a sampling strategy must be developed forlow-g implementation, which could mean challen-ging ground-testing and verification before launch.

D. Next Generation Laser Raman Spectrometers

There are two types of Laboratory Raman spectro-meters in widespread use: dispersive Raman andFourier Transfer (FT)-Raman. Dispersive Ramanspectrometers typically operate with visible excita-tion sources, fixed gratings, no moving parts, andlarge-format detectors such as CCDs. FT-Ramanspectrometers typically operate with infrared lasersources and movable mirrors, which allow for a vari-able path length and the generation of an interfero-gram collected on a single element detector such asGe or InGaAs. Of the dispersive Raman spectrom-eters, multichannel dispersive systems have sur-passed single channel systems due to technologicaladvancements, particularly in low-noise large-formatCCD detectors. Multichannel spectrometers offer fas-ter collection speeds and higher sensitivity, and areoften shot-noise limited, providing thehighest achiev-able signal-to-noise ratio (SNR). CW dispersive greenRaman spectroscopy is commonly used in the lab-oratory for identification and characterization ofminerals. For example, the RRUFF project [17] wasestablished to create a complete set of high qualityspectral data from well-characterized mineral sam-ples to be used as standards for Raman fingerprint-ing. Much of the data was acquired using CW greenRaman spectroscopy . However, for many minerals,Raman spectra are difficult or impossible to acquireusing a CW green source due to a large interferingfluorescence signature.The fluorescence problemwasaddressed in 1986 by the introduction of the FT-Ramanspectrometerwithan infrared laser excitationsource (commonly 1064nm) [18,19]. A primary ad-vantage of FT-Raman is that it greatly reduces thenumber of cases in which fluorescence interferenceis problematic, since infrared typically does not pro-vide enough energy to excite fluorescence. It hasgained popularity as a laboratory technique andhas proven particularly useful for high resolutioncharacterization of organic and biological materials,which often contain large fluorescence signatureswhen excited in the visible [20,21]. This advantageis often enough to justify using 1064nm excitation,despite the significantly reduced signal level thatcomes with using long wavelength excitation (ac-cording to the 1=λ4 dependence of the Raman return).Another recent development designed to combatfluorescence has been the near-IR dispersive spectro-meter [22]. When combined with low noise detectors,dispersive spectrometers offer inherently largerSNRs than FT-spectrometers. However, the sensi-tivity of CCD detectors is too low at 1064nm to becompetitive with FT-Raman. An approach usingdispersive spectrometers and shorter wavelengthnear-IR excitation (700–800nm) works to reducefluorescence in many cases, although not as signifi-cantly as 1064nm. Because each method offers its in-herent advantages, the choice of spectrometer designwill be sample dependent. For minerals, the use of IRexcitation has yielded mixed results. In many cases,fluorescence interference from rare earth elements


has been observed using a 1064nm excitation leadingto ambiguous results [23–25]. In addition, the fluores-cence return is expected to be enhanced at low tem-perature, an important consideration for Mars andother cold bodies.

For several reasons, we have chosen a time-resolved approach for planetary mineralogy, ratherthan an infrared Raman approach to overcome fluor-escence. First, the use of a green pulsed laser allowsus tobroaden the scope of our instrument to obtainnotonly Raman, but also complementary fluorescence,and potentially LIBS information. Second, the higherRaman return for green compared to infrared allowsfor faster collection times and lower laser power. Andlast, although Fourier transform IR (FTIR) systemshave been designed for planetary missions (e.g., themini-TES on the MER), dispersive systems offerthe advantage of containing no moving parts.

Fluorescence interference is likely to be proble-matic on Mars, as evidenced by Raman measure-ments of Mars meteorites. For example, nearly halfof the 362 Raman spectra taken on the Zagami Marsmeteorite at 532nm yielded no Raman informationdue to fluorescence [26]. Many minerals found onMars are known to fluoresce strongly on Earth—forexample, clays and sulfates. Because so little is stillknown aboutVenusmineralogy, it is difficult to assessthe extent to which fluorescence interference wouldbe a concern for Raman spectroscopy on Venus. Asone example,modeling suggests thepossible presenceof hydrous minerals, such as tremolite on Venus [27],and these minerals have been known to fluoresce onEarth. Because of the risks associated with unknownmineralogy, it would be prudent to plan for an envir-onment that includes fluorescence.

Raman scattering occurs instantaneously in time,while fluorescence occurs on longer time scales.Minerals can contain as many as 25 luminescencecenters, each with different decay times from nsecto msec [28,29]. We use the distinct time scales ofRaman and fluorescence to separate the spectra. Indoing so, we can collect fluorescence-free Ramanspectra using traditional visible laser excitation(532nm) and simultaneously measure time-resolvedfluorescence spectra for the detection of trace ioniccomponents. Through the combined use of these tech-niques, the probability of successful mineral identifi-cation is greatly increased.

Time-resolved pulsed Raman spectroscopy is idealfor fluorescence rejection, since the Raman signal iscollected simultaneously with the laser pulse, andphotons emitted after the laser pulse can be rejected.Laser pulse widths of the order of 10 ns are typicallycombined with an intensified charge-coupleddevice (ICCD) based gated detection system [29].By gating so as to detect only photons emitted duringthe laser pulse, the majority of the fluorescence sig-nal is rejected. Rather than just reject the longerduration fluorescence with a time-gated cutoff, itwould be preferable to collect and record the full timehistory of the Raman and fluorescence signals, and

this can be captured with high temporal resolution(∼1psecs) streak cameras [30–32]. Spectra collectedwith a streak camera are swept across a CCD usinghigh speed voltage ramps in a streak tube. The resultis a three-dimensional plot of intensity versus fre-quency and time. By using the entire time evolutionrather than a fixed gate, it becomes possible to viewRaman and fluorescence simultaneously. Other ef-fects, such as plasma and LIBS related effects, canbe seen, as well, if the energy density at the sampleis high. The streak camera is therefore an invaluabletool for identifying and characterizing complexmineral samples as they are found in nature, con-taining both Raman and fluorescence centers.

E. Challenges of Developing On-SurfacePlanetary Spectrometers

As successful Apollo lunar missions illustrated, theelemental and mineralogical analysis of samplesreturned from planetary bodies can be made with ex-tremely high accuracy and precision using state-of-the-art spectrometers of various types (spectrometersbased onmass,Raman, neutron, x-ray, alpha-particle,Mossbauer, etc.) housed in laboratories throughoutthe world. These instruments are, by definition, verylarge, consume high power, and often take up one orseveral rooms in size. Also, despite significant instru-ment automation, they usually require several staffmembersand students to operateandhaveassociatedlarge computing facilities for data processing andanalysis.

In situ planetary instruments face challenges thatgreatly exceed even the challenges of developingin-space Earth orbiting instruments. First, the basicinstrument design must be applicable to a diverse setof target bodies, some relatively nearby, such asMars or Venus, and some a long way out, such asSaturn’s moon Titan, or distant asteroids. Individualbodies may be associated with extreme pressures ortemperatures—absolute and diurnal or seasonalchanges. Because of restricted payload masses forthese ambitious missions, planetary in situ instru-mentsmust be very small (a fewkg, atmost), consumelow power (20W or less), take up low volume (a fewliters, at most), and be highly automated from instru-ment turn-on to data transmission. Flight instrumentpractices by bothNASAand other international agen-cies require environmental and operational testingthat exceeds the expected planetary environment.The instrument must be designed to be readily testa-ble (environmental and vibration), easily calibrated,and, in most cases, be able to withstand pyrotechnicshock, launch loads, planetary protection require-ments, and radiation levels that strongly depend ontarget body and mission duration.

To develop a small, robust instrument that cansurvive all of the above described demands whileachieving measurement capability comparable tostate-of-the-art laboratory instruments will not bepossible. Furthermore, long integration times toenhance signal-to-noise ratios may not be possible


(e.g., touch-and-go analysis of asteroids or Venus sur-face measurements). However, there is an importantrole for laser Raman spectrometers, not only in on-surface mineralogical analysis, but also rock prese-lection prior to sample return, and various effortshave produced instruments that are feasible for pla-netarymissions. Key subsystems of potential risk arethe laser source itself and signal detector, and theirflight qualification, with the receiving optics, opticaltrain, and electronics considered more technicallymature for flight. Estimates by JPL engineers for atime-gated green laser Raman spectrometer fora future Mars mission, for example, suggest that acomplete instrument with a 5 μJ=pulse laser andsolid-state detector could weigh only 3kg total mass,consume 15W of power, and take up only 1L in vol-ume. This compares well with existing in situ instru-ments for Mars, including those described in Table 1.

2. Experiment

A. Experimental Setup

A schematic of our time-resolved Raman instrumentis shown in Fig. 1. A 532nm passively Q-switchedpulsed microchip laser (Arctic Photonics) delivers a5 μJ pulse over ∼800ps at a repetition rate of

1kHz. A small portion (<1%) of the beam is sampledand delivered to a trigger photodiode, which triggersthe streak camera to sweep. The main beam goesthrough an optical delay line timed to ensure thatthe Raman return is collected and synchronized withthe camera. A dichroic edge filter reflects the laserlight to an objective lens, which focuses the beamonto the mineral sample. The Raman/fluorescencereturn passes back through the dichroic and is fo-cused onto the input slit of a modified Kaiser OpticalHolospec spectrometer. Inside the spectrometer, aholographic notch filter cuts out the laser light againby ∼6 orders of magnitude. The light is focused ontoa vertical slit prior to passing through a holographicgrating, which spectrally disperses the light horizon-tally onto a 10mm wide, 75 μm high slit that definesthe entrance to a custom Axis Photonique streakcamera using a Photonis P925 streak tube.

The streak camera collects data in synchroscanmode in which the laser is synchronized to the sweepelectronics. The pulser electronics sweeps the spec-trum vertically over a chosen time base, which canbe varied from 4 ns to 500 μs. The streak camera re-peatedly streaks small pulses across the phosphorscreen, and theCCD integrates these pulses. The out-put on theCCD is a three-dimensional imagewith thespectrumon one axis, time on another axis, and inten-sity on the third axis. The benefit of this approach isthat a small and inexpensive low pulse energy laser(microchip laser) can beused, and thatwedonot incurCCD read noise with every pulse, which significantlyimproves SNR. The front-end gain of the streak cam-era coupled with the low dark current and read noiseof theCCDresults inanextremely low totalmeasurednoise of <2 photons per minute. Because only the re-turn is synchronized with the laser, background lightis eliminated (allowing for daylight operation) alongwith unwanted fluorescence. As a result, a pulsed sys-tem operating at the same average laser power as aCW system can achieve a given Raman peak SNRin a shorter collection time.

In our setup, an objective lens is used to focus thebeam onto the mineral sample. The Raman return iscollected in the 180° configuration back through theobjective lens. A set of objective lenses of 4×, 10×, and40× were used. The data presented here were ac-quired with the 40× objective, which has the highestNA, but also the shortest working distance.

Frequency calibration was performed using cyclo-hexane and acetonitrile standards. Radiometricanalysis using cylcohexane as a standard [33] con-firmed that the Raman return detected at the CCDis >60% of the expected Raman return from the sam-ple within the acceptance angle of the spectrometer.This is reasonable, considering the losses in theoptical chain after the sample.

3. Results and Discussion

A typical streak camera image of calcite is shown inFig. 2(a). The Raman peaks are simultaneous withthe laser pulse and therefore appear as 800 ps

Fig. 1. (Color online) Schematic diagram of the time-resolvedRaman instrument.


streaks in time. This spectrum was collected in syn-chroscan mode with a total integration time of 30 s.In Fig. 2(b), a broadband peak is present after the la-ser pulse. This peak is likely emission from a plasmacreated near the sample surface. Figs. 2(a) and 2(b)show two different calcite samples, with one showingmore plasma effects. One noticeable difference is thatthis calcite sample was flatter, smoother, and moretransparent. The plasma effects are particularly no-ticeable in an opaque, smooth sample, such as silicon,where the entire laser pulse is absorbed in a verysmall sample volume, leading to high energy density.

Two-dimensional plots of intensity versus Ramanshift shown as insets of Fig. 2 are created from thestreak images by summing the counts over the timescale of the laser pulse width. This selects the Ramanreturn and rejects the fluorescence. In cases in whichthe plasma emission is significant during the laserpulse, a background is still observed, although it is

not fluorescence related. This is illustrated in the in-set of Fig. 2(b), where the background is higher. Itshould be noted that without time resolution, thisbackground would be indistinguishable from fluores-cence, and important information relating to poten-tial sample damage or alteration would be lost.

A. Fluorescence Rejection

We demonstrate fluorescence rejection using a vari-ety of fluorescent minerals, some of which are highlyrelevant to the Martian environment. Montmorillo-nite, brushite, and magnesium sulfate are alteredminerals that have been inferred on Mars, and allhave exhibited strong fluorescence, leading to theinability to measure Raman spectra using a CWRaman system. These samples are representativeof the most challenging fluorescence conditions thatwe can expect to find on Mars. In order to demon-strate the fluorescence rejection capabilities of ourtime-resolved Raman system, we have also perfor-med measurements on the same samples using a Re-nishaw M1000 Micro Raman Spectrometer Systemwith an Ar ion laser, 514:5nm through an opticalmicroscope. Both measurements were compared tospectra in the RRUFF database, as well as reportsin the literature. The RRUFF samples are typicallychosen with minimal fluorescence centers in order togenerate the best Raman spectra for the database,and are not necessarily indicative of the expectedfluorescence from a given mineral. Alternatively,many RRUFF spectra are taken using infrared exci-tation to avoid fluorescence in highly fluorescentsamples.

B. Willemite

Willemite is a trigonal zinc silicate mineral(Zn2SiO4) and aminor ore of zinc.Willemite is knownfor its strong green fluorescence peaked at 535nmowing to the presence of Mn2þ [28]. The sampleCIT1487 originated in Franklin, New Jersey. Thismineral is highly fluorescent and provides a usefuldemonstration of the power of pulsed Raman underthe most extreme fluorescence conditions. Figure 3shows the time-resolved spectrum of willemite. Asshown, the Raman spectrum from this sample wasnot measurable in the CW Raman system, even withonly 1% of the laser power, due to saturation of thedetector by fluorescence at all wavelengths. Thetime-resolved spectrum matches well with theRRUFF spectrum taken on an unoriented samplewith a 785nm excitation wavelength.

C. Spodumene

Spodumene is a monoclinic pyroxene mineral con-sisting of LiAlðSiO3Þ2. The origin of the sample2919 is San Pedro Mine, Pala, California. The pinkcolor of this sample is connected with Mn impuritiesresponsible for strong fluorescence in the region ofinterest, saturating the detector at 1080 cm−1. Thetime-resolved spectrum in Fig. 4 matches well with

Fig. 2. (Color online) Streak camera images of calcite a) showingstrong Raman return and b) showing Raman return and plasmaeffects. The horizontal scale covers 4 ns, and the vertical scale cov-ers 235–2600 cm−1. Insets show spectra averaged over time for thelength of the laser pulse. Photos of each calcite sample show dif-ferences in opacity of the samples. Note that the oscillating natureof the plasma peak is an artifact of the streak camera sweep cir-cuitry, causing small variations in sweep speed with time.


the RRUFF spectrum taken on an unoriented samplewith a 785nm excitation wavelength.

D. Montmorillonite

Montmorillonite is a clay mineral of the smectite fa-mily and is a common alteration product of manytypes of rock. Spectroscopic data (e.g., MER APXS)supports the presence of clay minerals on Mars.Definitive data on these minerals would provide amodel for the alteration of the martian surface sup-porting neutral- or slightly alkaline pH. The Inde-pendence Class rocks of the Columbia Hills haveundergone substantial alteration. The evidence sug-gests that they are dominated by montmorillonite or

its compositional equivalent [34]. Obtaining cleanRaman spectra from clay minerals is highly challen-ging. They tend to be fine grained and powdery, withweak Raman bands owing to the small polarizabilityof Si-O bonds. They tend to be highly fluorescent, inpart due to the presence of iron(III) hydroxide andorganic matter [35]. We obtained Raman spectrafrom the Clay Mineral Standard H-19. The over-whelming fluorescence under CW excitation satu-rated the detector over the entire wavelengthrange. The time-resolved spectrum exhibited weakRaman peaks attributed to montmorillonite and aminor phase of calcite, as shown in Fig. 5. The mostextensive Raman study of montmorillonite was per-formed using FT Raman with a 1064nm laser tominimize fluorescence [36]. The strongest bandswere reported in the 278–285 cm−1 range and the708–725 cm−1 range. Variations are due to composi-tional differences between montmorillonite samples.

E. Brushite

Brushite is a phosphate mineral with a chemical for-mulaCaHPO4 · 2H2O that can be formed at lowpHbyreaction of phosphate-rich solutions with calcite andclay. It has been inferred onMars in the Gusev Crateroutcrops, rocks, and soils using MER APXS data [2].We have observed strong fluorescence under CW ex-citation, saturating theCCDat 760 cm−1. The fluores-cence background from the natural sample fromMoorbaCave inWesternAustraliawas toohigh todis-tinguish Raman peaks over the entire wavelengthrange. Time-resolved Raman revealed a Raman sig-nature representative of brushite, as shown in Fig. 6.In addition to brushite, the presence of the anhydrousphase CaHPO4 is also hypothesized, based on the twostrong peaks associated with the P-O stretching

Fig. 3. (Color online) Raman spectra of willemite (pictured inthe inset) obtained using pulsed and CW Raman. Under CW illu-mination, the fluorescence saturates the CCD detector. TheRRUFF spectrum acquired using a 785nm source is shown forreference.

Fig. 4. (Color online) Raman spectra of spodumene (picturedin the inset) obtained using pulsed and CW Raman. TheRRUFF spectrum acquired using a 785nm source is shown forreference.

Fig. 5. (Color online) Raman spectra of montmorillonite (picturedin the inset) obtained using pulsed and CW Raman. Under CWillumination, the fluorescence saturates the CCD detector. Apulsed Raman spectrum of calcite is shown, as well, revealing thata minor phase of calcite is likely present.


modes at 988 and 998 cm−1 for the anhydrous andhydrated forms, respectively [37].

F. Magnesium Sulfate

Magnesium sulfate (MgSO4) is hygroscopic and canexist in a range of hydration states, depending on itsenvironmental history. Mg-sulfates are inferred onMars, for example, in the Columbia Hills and Meri-diani Planum outcrops [34]. Raman spectroscopy canprovide a means to determine the degree of hydra-tion, where Raman peaks shift to a higher wavenum-

ber with decreasing hydration [38]. The origin ofthe sample 2222 is Stassfurt, Germany. CW Ramanmeasurements of this kieserite sample revealedoverwhelming fluorescence with no ability to detectRaman. Pulsed Raman revealed a fully hydratedsample, with peak positions coincident with thoseof MgSO4 in aqueous solution, as shown in Fig. 7.

4. Conclusions

We have reviewed techniques for in situ mineralogyon planetary surfaces and presented time-resolvedlaser Raman spectroscopy as the next generationin situ instrument for definitive mineralogy of multi-phase rock and soil samples preserving geologicalcontext. We have shown that time-resolved Ramanspectroscopy is an excellent approach to identifyingminerals when a high fluorescence background ispresent. This technique would have broad applica-tion to planetary mineralogy—for example, on Marsand its moons, Phobos and Deimos; Venus; Earth’smoon; and asteroids. In particular, it is recommendedfor an in situ instrument on Mars, where fluores-cence is likely to be of concern. For Mars, the objec-tives for a proposed 2018 mission point to a potentiallanding site with access to outcrops of diverse miner-alogy where ancient habitability is hypothesized,and a possibility exists for detecting organics and bio-signatures [39,40]. This suggests a heavily alteredregion with secondary minerals, many of which aresimilar to those chosen for this work (e.g., clays, sul-fates, and phosphates). Our demonstration of strongfluorescence interference in these samples is motiva-tion for selecting a time-resolved Raman instrumentthat operates regardless of fluorescence background.The comprehensive data required for potentialsample return preselection would be acquired bymapping the mineralogy of a sample using time-resolved Raman spectroscopy. Mapping of the miner-alogy is crucial if we can hope to make sense ofcomplementary data on trace ions and organics with-in the context of the host rock. Such data would beobtained by other instruments on the same payload,such as time-resolved fluorescence spectroscopy foridentifying trace ions, and ultraviolet fluorescencespectroscopy for mapping of organics. With the em-phasis in planetary exploration shifting toward pos-sible sample return fromMars, the Moon, Venus, andasteroids, Raman spectroscopy is expected to becomean essential tool for in situ mineralogical analysis. Itwould be important in preselection for sample cach-ing and potential return to Earth, as well as nondes-tructive high resolution measurement on samplesafter return to Earth.

We acknowledge invaluable discussions on Mars2018 with Sabrina Feldman at the Jet PropulsionLaboratory (JPL). The research described in thispublication was carried out at the Jet PropulsionLaboratory, California Institute of Technology, undera contract with the National Aeronautics and SpaceAdministration (NASA). Continuous-wave Ramanmeasurements were performed at the Mineral

Fig. 6. (Color online) Raman spectrum of Brushite (loose powderpictured in the inset) obtained using pulsed and CW Raman.The time-resolved spectrum has been background subtracted.The RRUFF spectrum is also shown for reference. The peaklocations of the P-O stretching modes for the anhydrous and hy-drated phases from [37] are noted and reveal the presence of bothphases.

Fig. 7. (Color online) Raman spectrum of kieserite (loose whitepowder pictured in the inset) obtained using pulsed Raman. UnderCW illumination, the fluorescence saturates the CCD detector.Data from [38] are plotted, as well, and show that the peaks shiftto a higher wavenumber with decreasing hydration state. Oursample spectrum matches well with the completely hydratedsample.


Spectroscopy Laboratory in the Department of Geo-logical and Planetary Sciences at the CaliforniaInstitute of Technology, and time-resolved experi-ments at the JPL.

References

1. R. L. McCreery, Raman Spectroscopy for Chemical Analysis(Wiley, 2000).

2. D. W. Ming, R. V. Morris, and B. C. Clark, “Aqueous alterationon Mars,” in The Martian Surface: Composition, Mineralogyand Physical Properties, J. F. Bell III, ed. (Cambridge Univer-sity Press, 2008).

3. L. A. Soderblom and J. F. Bell III, “Exploration of the Martiansurface: 1992–2007,” in The Martian Surface: Composition,Mineralogy and Physical Properties, J. F. Bell III, ed.(Cambridge University Press, 2008).

4. S. Murchie, R. Arvidson, P. Bedini, K. Beisser, J.-P. Bibring,J. Bishop, J. Boldt, P. Cavender, T. Choo, R. T. Clancy,E. H. Darlington, D. Des Marais, R. Espiritu, D. Fort,R. Green, E. Guinness, J. Hayes, C. Hash, K. Heffernan,J. Hemmler, G. Heyler, D. Humm, J. Hutcheson, N. Izenberg,R. Lee, J. Lees, D. Lohr, E. Malaret, T. Martin, J. A. McGovern,P. McGuire, R. Morris, J. Mustard, S. Pelkey, E. Rhodes,M. Robinson, T. Roush, E. Schaefer, G. Seagrave, F. Seelos,P. Silverglate, S. Slavney, M. Smith, W.-J. Shyong,K. Strohbehn, H. Taylor, P. Thompson, B. Tossman,M. Wirzburger, and M. Wolff, “Compact reconnaissance imag-ing spectrometer for Mars (CRISM) on Mars reconnaissanceorbiter (MRO),” J. Geophys. Res. [Planets] 112(E5), E05S03(2007).

5. A. S. McEwen, E. M. Eliason, J. W. Bergstrom, N. T. Bridges,C. J. Hansen, W. A. Delamere, J. A. Grant, V. C. Gulick,K. E. Herkenhoff, L. Keszthelyi, R. L. Kirk, M. T. Mellon,S. W. Squyres, N. Thomas, and C. M. Weitz, “Mars re-connaissance orbiter’s high resolution imaging scienceexperiment (HiRISE),” J. Geophys. Res. [Planets] 112(E5),E05S02 (2007).

6. P. R. Christensen, M. B. Wyatt, T. D. Glotch, A. D. Rogers, S.Anwar, R. E. Arvidson, J. L. Bandfield, D. L. Blaney, C.Budney, W. M. Calvin, A. Fallacaro, R. L. Fergason, N.Gorelick, T. G. Graff, V. E. Hamilton, A. G. Hayes, J. R.Johnson, A. T. Knudson, H. Y. McSween, G. L. Mehall,L. K. Mehall, J. E. Moersch, R. V. Morris, M. D. Smith,S. W. Squyres, S. W. Ruff, and M. J. Wolff, “Mineralogy at Mer-idiani Planum from the Mini-TES experiment on the Oppor-tunity Rover,” Science 306, 1733–1739 (2004).

7. P. R. Christensen, J. L. Bandfield, R. N. Clark, K. S. Edgett,V. E. Hamilton, T. Hoefen, H. H. Kieffer, R. O. Kuzmin, M. D.Lane, M. C. Malin, R. V. Morris, J. C. Pearl, R. Pearson, T. L.Roush, S. W. Ruff, and M. D. Smith, “Detection of crystallinehematite mineralization on Mars by the Thermal EmissionSpectrometer: evidence for near-surface water,” J. Geophys.Res. [Planets] 105, 9623–9642 (2000).

8. J. P. Grotzinger, R. E. Arvidson, J. F. Bell, W. Calvin, B. C.Clark, D. A. Fike, M. Golombek, R. Greeley, A. Haldemann,K. E. Herkenhoff, B. L. Jolliff, A. H. Knoll, M. Malin,S. M.McLennan, T. Parker, L. Soderblom, J. N. Sohl-Dickstein,S. W. Squyres, N. J. Tosca, and W. A. Watters, “Stratigraphyand sedimentology of a dry to wet eolian depositional system,Burns formation,Meridiani Planum,Mars,”Earth Planet. Sci.Lett. 240, 11–72 (2005).

9. P. Mahaffy, “Sample analysis at Mars: developing analyticaltools to search for a habitable environment on the red planet,”Geochem. News 141 (2009).

10. A. Wang, B. L. Jolliff, and L. A. Haskin, “Investigating surfacemineralogy, alteration processes, and biomarkers on Mars

using laser Raman spectroscopy,” in Sixth International Con-ference on Mars, abstract no. 3270 (2003).

11. M. C. Storrie-Lombardi, J-P. Muller, M. R. Fisk, C. Cousins, B.Sattler, A. D. Griffiths, and A. J. Coates, “Laser-inducedfluorescence emission (L.I.F.E.): searching for Mars organicswith a UV-enhanced PanCam,” Astrobiology 9(10), 953–964(2009).

12. A. T. Basilevsky, M. A. Ivanov, J. W. Head, M. Aittola, and J.Raitala, “Landing on Venus: past and present,” Planet. SpaceSci. 55, 2097–2112 (2007).

13. M. Y. Zolotov, B. Fegley, Jr., and K. Lodders, “Hydrous silicatesand water on Venus,” Icarus 130, 475–494 (1997).

14. A. J. Ball, M. E. Price, R. J. Walker, G. C. Dando, N. S. Wells,and J. C. Zarnecki, “Mars Phobos and Deimos survey(M-PADS)—A Martian Moons orbiter and Phobos lander,”Adv. Space Res. 43, 120–127 (2009).

15. E. M. Galimov, “Phobos sample return mission: scientific sub-stantiation,” Sol. Syst. Res. 44(1), 5–14 (2010).

16. A. Wang, B. L. Jolliff, and L. A. Haskin, “Raman spectros-copy as a method for mineral identification on lunarrobotic exploration missions,” J. Geophys. Res. 100(E10),21189–21199 (1995).

17. RRUFF Project: http://rruff.info/.18. T. Hirschfeld and B. Chase, “FT-Raman spectroscopy: develop-

ment and justification,” Appl. Spectrosc. 40 (2), 133–137(1986).

19. B. Chase, “Fourier transform Raman spectroscopy,” J. Am.Chem. Soc. 108, 7485–7488 (1986).

20. L. Burgio and R. J. H. Clark, “Library of FT-Raman spectra ofpigments, minerals, pigment media and varnishes, and sup-plement to existing library of Raman spectra of pigmentswith visible excitation,” Spectrochim. Acta Part A 57,1491–1521 (2001).

21. E. B. Hanlon, R.Manoharan, T-W. Koo, K. E. Shafer, J. T. Motz,M. Fitzmaurice, J. R. Kramer, I. Itzkan, R. R. Dasari, andM. S.Feld, “Prospects for in vivo Raman spectroscopy,” Phys. Med.Biol. 45, R1–R59 (2000).

22. Y. Wang and R. L. McCreery, “Evaluation of a diode laser/charge coupled device spectrometer for near-infrared Ramanspectroscopy,” Anal. Chem. 61, 2647–265 (1989).

23. H. G. M. Edwards, S. E. Jorge Villar, J. Jehlicka, andT. Munshi, “FT–Raman spectroscopic study of calcium-richand magnesium-rich carbonate minerals,” Spectrochim. ActaPart A 61, 2273–2280 (2005).

24. P. Makreski and G. Jovanovski, “Minerals from Macedonia.XXII. Laser-induced fluorescence bands in the FT-Ramanspectrum of almandine mineral,” J. Raman Spectrosc. 39,1210–1213 (2008).

25. A. Aminzadeh, “Fluorescence bands in the FT-Raman spectraof some calcium minerals,” Spectrochim. Acta Part A 53693–697 (1997).

26. T. Frosch, N. Tarcea, M. Schmitt, H. Thiele, F. Langenhorst,and J. Poppet, “UV Raman imaging: a promising tool forastrobiology: comparative Raman studies with differentexcitation wavelengths on SNC Martian meteorites,” Anal.Chem. 79, 1101–1108 (2007).

27. N. M. Johnson and B. Fegley, “Longevity of fluorine-bearingtremolite on Venus,” Icarus 165(2), 340–348 (2003).

28. M. Gaft, R. Reisfeld, and G. Panczer, Modern LuminescenceSpectroscopy of Minerals and Materials (Springer-Verlag, 2005).

29. M. Gaft and L. Nagli, “Gated Raman spectroscopy: potentialfor fundamental and applied mineralogy,” Eur. J. Mineral. 21,33–42 (2009).

30. T. Tahara and H. Hamaguchi, “Picosecond Raman spectro-scopy using a streak camera,” Appl. Spectrosc. 47(4), 391–398(1993).


http://rruff.info/

http://rruff.info/

31. N. Everall, T. Hahn, P. Matousek, A. W. Parker, andM. Towrie,“Picosecond time-resolved Raman spectroscopy of solids: cap-abilities and limitations for fluorescence rejection and the in-fluence of diffuse reflectance,” Appl. Spectrosc. 55, 1701–1708(2001).

32. H. Hamaguchi and T. L. Gustafson, “Ultrafast time-resolvedspontaneous and coherent Raman spectroscopy: the structureand dynamics of photogenerated transient species,” Annu.Rev. Phys. Chem. 45, 593–622 (1994).

33. R. L. McCreery, “Photometric standards for Raman spectro-scopy,” in Handbook of Vibrational Spectroscopy, J. M.Chalmers and P. R. Griffiths, eds. (Wiley, 2002).

34. S. W. Ruff, P. R. Christensen, T. D. Glotch, D. L. Blaney, J. E.Moerch, and M. B. Wyatt, “The mineralogy of Gusev craterand Meridiani Planum derived from the Miniature ThermalEmission Spectrometers on the Spirit and Opportunity ro-vers,” in The Martian Surface: Composition, Mineralogy andPhysical Properties, J. F. Bell III, ed. (Cambridge UniversityPress, 2008).

35. J. M. Alia, H. G. M. Edwards, F. J. Garcia-Navarro,J. Parras-Armenteros, and C. J. Sanchez-Jimenez, “Applica-

tion of FT-Raman spectroscopy to quality control in brick claysfiring process,” Talanta 50, 291–298 (1999).

36. R. L. Frost and L. Rintoul, “Lattice vibrations of montmoril-lonite: an FT Raman and x-ray diffraction study,” Appl. ClaySci. 11, 171–183 (1996).

37. J. Xu, I. S. Butler, and D. F. R. Gilson, “FT-Raman and high-pressure infrared spectroscopic studies of dicalcium phos-phate dihydrate (CaHPO4 · 2H2O) and anhydrous dicalciumphosphate (CaHPO4),” Spectrochim. Acta Part A 55,2801–2809 (1999).

38. A. Wang, J. F. Freeman, B. L. Jolliff, and I-M Chou, “Sulfateson mars, a systematic Raman spectroscopic study of hydrationstates of magnesium sulfates,” Geochim. Cosmochim. Acta7024, 6118–6135 (2006).

39. P. Christensen, “Scientific overview of Mars sample return,”presented at the 22ndMEPAGMeeting, Monrovia, California,17–18 March 2010.

40. C. G. Salvo and A. Elfving, “Proposed Mars astrobiologyexplorer–Cacher (MAX-C) & ExoMars 2018 (MXM-2018)Mission Formulation Status,” presented at the 22nd MEPAGMeeting, Monrovia, California, 17–18 March 2010.


Documents

Time-resolved Raman spectroscopy for in situ planetary ...authors.library.caltech.edu/20171/1/Blacksberg2010p11444Appl... · Time-resolved Raman spectroscopy for in situ planetary