Spectrochimica Acta Part B 101 (2014) 335–341

Contents lists available at ScienceDirect

Spectrochimica Acta Part B

j ourna l homepage: www.e lsev ie r .com/ locate /sab

Stand-off laser-induced breakdown spectroscopy of aluminum andgeochemical reference materials at pressure below 1 torr☆

Kang-Jae Lee, Soo-Jin Choi, Jack J. Yoh ⁎Department of Mechanical and Aerospace Engineering, Seoul National University, 1 Gwanakro, Gwanakgu, Seoul 151-742, Korea

☆ Selected paper from the 7th Euro-MediterraneanBreakdown Spectroscopy (EMSLIBS 2013), Bari, Italy, 16–⁎ Corresponding author. Tel.: +82 2 880 9334; fax: +8

E-mail address: jjyoh@snu.ac.kr (J.J. Yoh).

http://dx.doi.org/10.1016/j.sab.2014.06.0090584-8547/© 2014 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 27 December 2013Accepted 12 June 2014Available online 25 June 2014

Keywords:Laser-Induced Breakdown SpectroscopyStand-offLow pressure

Laser-induced breakdown spectroscopy (LIBS) is an atomic emission spectroscopy that utilizes a highly irradiat-ed pulse laser focused on the target surface to produce plasma. We obtain spectroscopic information from themicroplasma and determine the chemical composition of the sample based on its elemental and molecularemission peaks. We develop a stand-off LIBS system to analyze the effect of the remote sensing of aluminumand various geochemical referencematerials at pressures below 1 torr. Using a commercial 4 inch refracting tele-scope, our stand-off LIBS system is configured at a distance of 7.2m from the four United States Geological Survey(USGS) geochemical samples that include granodiorite, quartz latite, shale-cody, and diabase, which are selectedfor planetary exploration. Prepared samples were mixed with a paraffin binder containing only hydrogen andcarbon, and were pelletized for experimental convenience. The aluminum plate sample is considered as a refer-ence prior to using the geochemical samples in order to understand the influence of a low pressure condition onthe resulting LIBS signal. A Q-switched Nd:YAG laser operating at 1064 nm and pulsed at 10 Hz with 21.7 to48.5 mJ/pulse was used to obtain signals, which showed that the geochemical samples were successfully detect-ed by the present stand-off detection scheme. A low pressure condition generally results in a decrease of thesignal intensity, while the signal to noise ratio can vary according to the samples and elements of varioustypes. We successfully identified the signals at below 1 torr with stand-off detection by a tightly focused lightdetection and by using a relatively larger aperture telescope. The stand-off LIBS detection at low pressure ispromising for potential detection of the minor elements at pressures below 1 torr.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

The stand-off LIBS has attracted growing interest in recent years.Since it requires only a small sample portion to sufficiently identifythe atomic emission spectrum without sample preparation, LIBS is be-coming more recognized as a reliable remote detection technique. Forexample, the National Aeronautics and Space Administration (NASA)launched a rover to Mars in Nov 2011 and performed analyses of Marssoil using ChemCam, an on-board LIBS system. The rover successfullylanded on the surface of Mars in Aug 2012 and began its mission withthe stand-off LIBS equipment [1,2]. In 2018, the European Space Agency(ESA) also plans to land ExoMars, which is equipped with a combinedRaman-LIBS system [3]. Sallé et al. used geological samples for lens tosample distances of 3 ~ 12 m at 7 torr to simulate the Mars atmosphere[1,2]. They conducted quantitative analysis of in-situ and stand-off con-ditions by calculating the calibration curves of various elements such asmagnesium and silicon. Thompson et al. carried out quantitative

Symposium on Laser Induced20 September 2013.2 2 887 2662.

analysis and classification of Mars meteorite samples for a lens tosample distance of 5.4 m at 7 torr [4]. The weight percent of oxidesfrom the samples was analyzed in order to categorize the samples.Dyar et al. performed quantitative analysis using multivariate methodswith Principal Component Analysis (PCA) and Partial Least Squares(PLS) for discrimination of the geological samples [5,6]. The experimentwas also carried out in aMars atmospheric pressure condition at 7 ~ 9m.Several strategies were considered to detect sulfides and sulfatesthrough the chemometrics. Cousin et al. attempted pyroxene samplesat 3mdistance inside a 7 torr chamber in order to build aMars databasefor the Mars mission [7]. A meaningful comparison was made with theNational Institute of Standards and Technology (NIST) database whichis obtained at atmospheric pressure.

Further works of the stand-off LIBS were carried out at standardearth atmosphere. Palanco et al. used aluminum and titanium standardsamples for lens to sample distances of 30 ~ 100 m at 760 torr ambientpressure to construct a field-deployable LIBS system [8]. An open-trusstype telescope with a 16 inch aperture was introduced and applied incases such as the detection of aerosols and explosive contaminated fin-gerprints and the studies of beam propagation [9–11]. Gottfried et al.used geomaterials and explosive samples for lens to sample distancesof 20 ~ 30 m at 760 torr and carried out multivariate analyses such as

336 K.-J. Lee et al. / Spectrochimica Acta Part B 101 (2014) 335–341

PCA and Partial Least Squares Discriminant Analysis (PLS-DA) tocategorize each sample [12–15].

In this study, we configured a stand-off LIBS system with a vacuumchamber to study the low pressure effects on the remote sensing of alu-minum and various geochemical reference materials. It is commonlyunderstood that signal detection in a lowpressure condition is consider-ablymore difficult than in an atmospheric condition.We considered thelower pressure conditions reaching 0−2 torr and provide first timestand-off LIBS results that consist of strong signals of several ions thathave not been observed at the higher pressure conditions.

2. Experimental Setup

Fig. 1 shows the schematics of in-situ and stand-off detection. Asillustrated in Fig. 1a), a Q-switched Nd:YAG laser (Surelite I, Continu-um) operating at 1064 nm and pulsed at 10 Hz with the energy of48.5 mJ/pulse was used for the in-situ system. The laser beam passesthrough Lens 1 (planoconvex, focal length 350 mm, fused silica) and isfocused on the target surface. After the plasma generation, detectorslocated behind Lens 2 (planoconvex, focal length 300 mm, fused silica)and Lens 3 (same as Lens 2) receive plasma light into an opticalfiber that is connected to the spectrometer. An intensified-CCD(ICCD)-coupled echelle grating spectrometer (Mechelle and iStar,Andor) covering a wavelength range of 200 nm to 975 nm then obtainsthe spectrum image from the plasma light and eventually transfers thisto the personal computer. Both the spectrometer and the Nd:YAG laserare connected to the pulse generator for synchronization.

The time delay of the spectrometer, as used for a pressure above10 torr, was 1 μs, while it was 0.7 μs for tests below 1 torr. The gatewidth was 20 μs. The 200 μm fiber optic cable was connected to thespectrometer to collect the plasma emission. Samples were placed onthe sample holder inside the vacuum chamber. The laser was focusedon the sample surface and 5 shots were fired at the same location.

Fig. 1b) shows a stand-off LIBS system. We used the same laser asthat for the in-situ system, but with the energy of 21.7 ~ 48.5 mJ/pulsefor a stand-off scheme. The laser beam is expanded by the 4 x beamexpander for a higher beam collimation. Lens 4 (planoconvex, focallength 5000 mm, fused silica) focuses the laser beam on the samples.As a stand-off detector of the emitting plasma light, we used a commer-cial refracting telescope (SE102, Kenko) with a 4 inch aperture and thefocal length of 500mm. The telescope is located behind the laser system,but is set at a higher position so the laser system does not interrupt theability of the telescope to detect plasma light. The samples were placedat a distance of 5.7m from the laser and 7.2m from the detector at roomtemperature for a stand-off system.

The pressure inside the chamber varied from10−2 torr to 760 torr. Asample holder is placed inside the chamber, of which the volumetriccapacity is 180 mm3. The vacuum chamber is fabricated of anodized

Fig. 1. Experimental setup (a) In-situ d

aluminum and designed to depressurize to as low as 10−4 torr usingthe rotary and turbo pumps in only a few minutes.

Four powdered samples from the USGS were selected as the geo-chemical standard materials that include granodiorite, quartz latite,shale-cody, and diabase, which are shown in Table 1. The contents ofeach sample are shown in Table 2. Each sample was mixed with a par-affin binder at a proportion of 90% concentration and was pelletizedwith 10 tons of pressure, 2.5 minutes of dwell time, and 1.5 minutesof release time. The paraffin binder contains only hydrogen and carbon(CnH2n + 2) to ensure a straightforward concentration adjustment andbonding of the powdered sample. In addition, an aluminum platesample was prepared to further understand the influence of the lowpressure condition on the LIBS signal intensity, since aluminum isknown to retain strong spectra without any interference with otherelements.

3. Results and Discussion

3.1. In-situ measurement with different detector locations

The location of the detector is critical in the low pressure condition,as illustrated in Fig. 2. Optical fiber 2 shown in Fig. 1a) is placed on theside of the plasma light, which is parallel to the sample surface, and op-tical fiber 1 is placed in the central direction of the plasma light, which isperpendicular to the surface. Both results are normalized in order tocompare the effect of the detector location.When the detector is placedon the side, the signal intensity of the aluminum plate at atmosphericpressure showed amaximumand decreased significantlywith the pres-sure drop. It is almost impossible to detect the aluminum signals fromthe side at a pressure below 1 torr, although Al I peaks (394.40 nmand 396.15 nm) are easily detectable for LIBS in general. However, thesignal intensity is sufficiently high to be detected when the detector isplaced in the central direction of the plasma, even in the low pressureregion.

At low pressure, plasma expands quickly, whichmakes it difficult tocapture all of the plasma from the side. Most of the plasma volumequickly moves towards the laser along the beam axis after ablation;thus, the detector location must be changed toward the farther pointfrom the sample surface. However, it is much more unlikely to expectan exact location of the plasma volume since it differs according to thepressure variations. The plasma volume is also very sensitive to thedelay time because the plasma moves outside the capture range. How-ever, in the case of a detector located in the central direction of the plas-ma along the laser beam, plasma detection is possible regardless ofwhether the chamber pressure is high or low. Although the plasma ex-pansion is very fast at low pressure, most of the plasma volume remainsalong the axis of the laser beam. In this case,most of the plasma volume,except someof the expanded plasma outside the detector's range, is stilldetectable even at a pressure below 1 torr, as illustrated in Fig. 2. From

etection, (b) Stand-off detection.

Table 1List of geochemical samples.

Category SRM Description

Geochemical reference materials GSP-2 GranodioriteQLO-1 Quartz LatiteSCo-1 Shale-CodyW-2 Diabase

Table 2Concentration of major elements in geochemical samples.(Concentrations are in mg/kg, while noted by a single asterisk in %).

SRM Mg Si Ca Ti Fe Al K Na Mn

GSP-2 0.58* 31.30* 1.50* 0.40* 3.43* 7.88* 4.48* 2.06* 320QLO-1 0.60* 30.66* 2.26* 0.37* 3.04* 8.57* 2.99* 3.12*SCo-1 1.64* 29.35* 1.87* 0.38* 3.58* 7.25* 2.30* 0.67* 410W-2 3.84* 24.62* 7.76* 0.64* 7.56* 8.18* 0.52* 1.63* 130

337K.-J. Lee et al. / Spectrochimica Acta Part B 101 (2014) 335–341

this result, one can expect that detection in the central direction ispreferred (or required) for the proper detection of the LIBS signal atlow pressure. All our stand-off detections were performed with thisalignment near the beam axis in all low pressure experiments.

3.2. In-situ vs. stand-off detection at low pressure region

The 394.40 nmand 396.15 nmaluminumpeaks are analyzed in bothin-situ and stand-off conditions, as shown in Fig. 3. In both peaks, max-imum intensity is obtained between 1 ~ 10 torr where the stand-offmeasurement showed a higher signal intensity at 1 torr. It is interestingthat signal intensities at above 10 torr aswell as below 1 torr are dimin-ished compared to those of the 1 ~ 10 torr zone. This result directlyrelates to the electron number density and themean free path of ablatedatoms.

The linewidth in the plasmapeaks obtained by the Stark broadeningcan be used to estimate the electron number density. The spectra of Al I,namely the 394.40 nm and 396.15 nm peaks from in-situ and stand-offwith different pressure conditions, are indicated in Fig. 4, where thelaser energy used is 48.5 mJ/pulse. Fig. 4 shows that Stark broadeningoccurs for each pressure. The line widths of the Al I 396.15 nm peaksare acquired in full width at half maximum (FWHM) through Lorentzfitting of the measured spectra, as shown in Table 3. Stark broadeningis expressed by the following equation,

Δλ1=2 ¼ 2wNe

Nr

� �;

Fig. 2. In-situ signal intensity comparison between different

where Δλ1/2 is the line width in FWHM, w is the electron impact widthparameter, Ne is the plasma electron number density, and Nr is thereference electron density [16–18]. The electron impact broadeningwidth parameter w for the Al I 396.15 nm line is given by Griem withthe reference electron density, Nr = 1018 cm−3 [19]. The parameter wis temperature dependent, and was interpolated for each appropriatetemperature calculated for each pressure case. The observed linewidth needs to be corrected by subtracting the instrumental width in-duced by themonochromator itself [20].Wemeasured the instrumentalwidth with the Hg lines from a mercury lamp. The instrumental widthof 0.06 nm in our case is subtracted from the observed line width inorder to calculate the true value of the linewidth. The resulting electronnumber density for each pressure is shown in Table 3. It is reported thatthe plasma behavior differs between ambient pressures higher than10 mbar and lower than 1 mbar [17]. This tendency concurs with thatin our experiment and we can understand this in terms of the meanfree paths of each pressure and the corresponding plasma persistencetime.

The mean free path is expressed by the following form,

l ¼ kBTffiffiffi2

pπd2p

;

where l is the mean free path in meters, kB is the Boltzmann constant inJ/K, T is the temperature in K, d is the diameter of gas particles inmeters,and p is pressure in Pascals. The mean free path is mainly dependenton temperature and pressure, since other parameters are constant.Pressure is more dominant in determining the mean free path in thecase of plasma, because the temperature of the plasma is in the rangeof several thousand Kelvin, whereas the pressure change for each caseis in the order of 10s [21]. When pressure is lower than 1 torr, themean free path becomes longer, such that the chance of atomic collisionis sufficiently low to avoid the atoms becoming excited or ionized. Thisphenomenon is intensified as the pressure decreases, which in turn cre-ates a lower signal intensity. On the other hand, the mean free path isshort at a pressure higher than 10 torr, which renders it difficult to ob-tain enough velocity for atomic collision. In this case, the collision isweak or it seldom occurs and the signal intensity is low [22]. Self-absorption can also cause lower signal intensity at the high pressure re-gion [20]. At high pressure, breakdown can be initiated by inversebremsstrahlung and the plasmas absorb the laser when the pressureof ambient gas increases [23]. The plasma shielding effect also increasesdue to the high density of the surrounding air.When plasma shields thelaser beam, the laser energy is trapped inside the plasma plume andcannot be delivered to the target, causing a decrease of the ablatedmass. As a result of this laser plasma interaction, the line width of the

detector location (a) Al I 394.40 nm, (b) Al I 396.15 nm.

Fig. 3. Signal intensities of in-situ vs. stand-off detection (a) Al I 394.40 nm, (b) Al I 396.15 nm.

338 K.-J. Lee et al. / Spectrochimica Acta Part B 101 (2014) 335–341

emission lines increases and the signal intensity decreases at a pressurehigher than 10 torr.

The pressure dependence of the emission intensity of LIBS is coveredin detail by Yalcin et al. [24]. The authors compared Al I (396.15 nm)peaks at an atmospheric pressure to those at 4 torr. Significant signalenhancement is noticed at 4 torr and the signal to noise ratio (SNR)also increased. A Mg I (383.83 nm) peak is easily detected at 4 torr,while it is not detected at atmospheric pressure. The authors explainthat this result is due to a long persistence time of plasma at 4 torr. Asimilar experimental result was also reported by Choi et al. [22], where-by the plasma persistence time at 1 torr is peculiarly high, indicatingthat an optimal detecting pressure range may exist near the 1 torrregion. Thus, it is reasonable to assume that the signal intensities ofeach case in our experiment are maximized at 1 torr.

In all tested pressure ranges, the stand-off measurements remainedstrong enough for detection, even at below 1 torr. Moreover, a positiveaspect of the low pressure condition is that the spectra widths arenarrower. In the case of wide spectra, many peaks interfere with eachother and we cannot recognize the exact peak widths. This is a seriousproblem when analyzing samples with complex elements, especiallythose containing minor elements with relatively low signal intensities,if further elemental peaks can interfere. However, a low pressurizedenvironment provides clearly observable peaks that are not obstructedby adjacent peaks due to the narrow line width. This is shown in Fig. 4with the spectrum itself and in Table 3 with FWHM according to each

Fig. 4. Spectra of Al I at different pressure (a)

pressure, whereby the width of the spectra is narrower when pressuredecreases.

The SNR is calculated for the 394.40 nm and 396.15 nm peaks asshown in Fig. 5. SNR increases as the pressure drops in most cases,since noise level decreases significantly. It can be inferred that the signalintensities at 1 ~ 10 torr are the highest, so that SNR also increaseswhenthe pressure reaches 1 ~ 10 torr. Moreover, at high pressure, there is noevidence of saturation of the signal, and the noise here can still be high,rendering SNR lower in the high pressure region. However, it is also no-ticed that the SNRs at 10−1 torr and 10−2 torr are higher than those at100 torr and 760 torr, respectively. This means that the noise level ismuch lower than that of the high pressure case, since the signal intensi-ty is low at 10−2 torr. Due to a low noise level, the signal detection istherefore much easier in the low pressure region.

Aluminum peaks are measured with laser energies varied from 21.7~ 48.5 mJ/pulse, as indicated in Fig. 6. Signals at 1 torr consistentlyshowed the highest value amongst other pressure conditions, regard-less of the laser energy. Noticeably, even at 10−2 torr, the Al I peaksare still detectable. When in stand-off condition, we only consider theplasma effect at low pressure inside the chamber. Therefore, the air out-side the chamber may cause errors or a fluctuation of signals by affect-ing the laser beam propagation [10]. Secondly, the focused region iselongated when using a focusing lens with a long focal length. Thismay cause variety in the focusing point, which is regarded as a reasonfor different plasma intensities, particularly at the 1 ~ 10 torr region

In-situ detection, (b) Stand-off detection.

Table 3Line width (FWHM) and electron number density of Al I.

Pressure [torr] Width (FWHM) [nm] Ne [1018 cm−3]

0.01 0.339 0.3050.1 0.328 0.4031 0.333 0.39810 0.367 0.558100 0.485 0.696760 0.693 0.832

339K.-J. Lee et al. / Spectrochimica Acta Part B 101 (2014) 335–341

where the error bar is the largest. However, signals at 10−2 torr can bedetected at a low laser energy such as at 21.7 mJ/pulse.

The ionized aluminum peak (624.34 nm) shown in Fig. 7 is analyzedby using varying laser energies and pressures. The Al II peak(624.34 nm) begins to appear below 1 torr. The reasonwhy the ionizedatoms can be more easily detected at low pressure is related to theatomic reaction [23]. When the atom is excited by the laser ablation,electrons are released and the atom is ionized. After the ionizationprocess, the electrons are attached through the collision and interactionbetween atoms. However, the lowpressure condition delays this atomiccollision and interaction process because of the low electron numberdensity, which provides easier detection of ionized peaks at lowpressure. Also, the plasma shielding effect diminishes when the pres-sure is low. The plasma shielding effect is almost negligible when inlow pressure condition because the plasma expands quickly at lowpressure. For both reasons, the ionized peaks are easier to detect. It is

Fig. 5. Signal to noise ratio of Al I peaks (a) In

Fig. 6. Stand-off signal intensities with different lase

very encouraging that the Al II peak is shownwith a noticeable intensityat 10−2 torr, suggesting that the stand-off LIBS detection in the lunarcondition is feasible.

3.3. Stand-off detection of geochemical samples at low pressure region

Geochemical referencematerials are introduced in order to examinevarious elements other than aluminum. They are originally in a pow-dered condition and pelletized for experimental convenience. This pow-dered condition offers uniformity of sample components and generallyof its surface, but it also leads to a difficulty in the detection of signals asthe powders are scattered during the ablation process. This scatteringphenomenon in our experiment worsens at low pressure condition,with the particles becoming attached to the chamberwindow. Althoughwe used 7 tons of pressure in pelletizing, the samples are not as rigid asthe solid rocks. Thus, the overall signal intensities of the geochemicalsamples are much lower than those of the rigid samples such as analuminum sample.

The highest signal intensity is observed at 1 torr in both samples(GSP-2 and SCo-1), as shown in Fig. 8. The signals decreased significant-ly at pressures lower than 1 torr. The sodiumpeak (589.59 nm) is low inthe SCo-1 sample, which contains the least amount of sodiumcomparedto the other geochemical samples, as indicated in Table 2. It is promisinghowever, that signals from these pelletized geochemical samples can bedetected at such low pressure (10−2 torr) in spite of this unfavorablecondition.

-situ detection, (b) Stand-off detection.

r energy (a) Al I 394.40 nm, (b) Al I 396.15 nm.

Fig. 7. Stand-off signal intensities of Al II 624.34 nm peak with different laser energy. Fig. 9. Stand-off signal to noise ratio of GSP-2 with various elements.

340 K.-J. Lee et al. / Spectrochimica Acta Part B 101 (2014) 335–341

The SNR is calculated for each element in GSP-2, and the results areshown in Fig. 9. The SNRs of Na I and Al I at 10−1 torr are 4 times higherthan those of the results at 760 torr. When the pressure is high, thenoise level is higher compared to that of the lowpressure result. The sig-nal intensity is lower at 10−1 torr than that at 1 torr, but the noise levelis much lower, resulting in such a high SNR. From Fig. 9 we can estimatethe low degree of noise level at 10−2 torr, since the signal intensity isthe smallest amongst all pressure regions. Therefore, a pressure below1 torr is preferred for a stand-off LIBS for the various samples of interest.

The calibration curves are generated from the four geochemicalsamples and are shown in Fig. 10. The signal intensity of each sampleis compared with its original concentration provided by the USGS. Thisunivariate method of quantitative analysis is inaccurate compared tothe multivariate methods such as PCA or PLS-DA. However, it does pro-vide an initial intuitive comparison. The signal intensities for calculatingthese calibration curves are normalized to an internal standard. The Rsquare of potassium at 100 torr acquired from the experiment is 0.91and that of potassium at 10−2 torr is 0.99. The results are reasonableconsidering that various geochemical samples with different composi-tions are used to build the calibration curves. As a result, the combinedstand-off and low pressure detection offered a certain accuracy and canbe further optimized for minor element detection. PCA or PLS-DAcould be an appropriate step towards accurate classification of variousgeochemical samples.

Fig. 8. Stand-off signal intensities of geochemical reference

4. Conclusion

In this research, we performed a stand-off LIBS analysis at lowpressure conditions below 1 torr. We then compared the obtainedsignals to those of the in-situ measurements. Al peaks at a stand-offdistance reached a maximum at 1 torr, while the signals at 10−1 torrand 10−2 torr were still identifiable. Ionized Al II peaks at pressuresbelow 10−1 torr were successfully detected with a low laser energy of21.7 mJ/pulse. Stand-off signals of geochemical samples were alsostrong at 1 torr, and the SNR was higher at pressures below 1 torr,indicating that such low pressure detection is possible. The obtainedcalibration curves for low pressure conditions seemed reasonable fordetermining the concentration of the unknown sample. Future researchis aimed at developing the lunar rock sample analysis at a pressuresignificantly below 1 torr. Our results suggest an optimal stand-offLIBS scheme conducted at a low pressure condition for the minorelement detection in combination with the PCA or PLS-DA data analysistechnique.

Acknowledgements

The authors are grateful for the financial support from the KoreaNational Research Foundation under theNational Space Laboratory Pro-gram 2009 (NRF-2009-0092017) through the IAAT at Seoul NationalUniversity.

materials with various elements (a) GSP-2, (b) SCo-1.

Fig. 10. Stand-off calibration curves of K from geochemical reference materials (a) 100 torr, (b) 0.1 torr.

341K.-J. Lee et al. / Spectrochimica Acta Part B 101 (2014) 335–341

References

[1] B. Salle, J. Lacour, E. Vors, P. Fichet, S. Maurice, D.A. Cremers, R.C. Wiens, Laser-induced breakdown spectroscopy for Mars surface analysis: capabilities at stand-off distances and detection of chlorine and sulfur elements, Spectrochim. Acta B59 (2004) 1413–1422.

[2] B. Salle, D.A. Cremers, S. Maurice, R.C.Wiens, P. Fichet, Evaluation of a compact spec-trograph for in-situ and stand-off Laser-Induced Breakdown Spectroscopy analysesof geological samples on Mars missions, Spectrochim. Acta B 60 (2005) 805–815.

[3] G.B. Courreges-Lacoste, B. Ahlers, F.R. Perez, Combined Raman spectrometer/laser-induced breakdown spectrometer for the next ESA mission to Mars, Spectrochim.Acta A 68 (2007) 1023–1028.

[4] J.R. Thompson, R.C. Wiens, J.E. Barefield, D.T. Vaniman, H.E. Newsom, S.M. Clegg,Remote laser-induced breakdown spectroscopy analyses of Dar al Gani 476 andZagami Martian meteorites, J. Geophys. Res. 111 (5) (2006) 1–9.

[5] M.D. Dyar, J.M. Tucker, S. Humphries, S.M. Clegg, R.C. Wiens, M.D. Lane, Strategiesfor Mars remote Laser-Induced Breakdown Spectroscopy analysis of sulfur ingeological samples, Spectrochim. Acta B 66 (2011) 39–56.

[6] M.D. Dyar, M.L. Carmosino, J.M. Tucker, E.A. Brown, S.M. Clegg, R.C. Wiens, J.E.Barefield, J.S. Delaney, G.M. Ashley, S.G. Driese, Remote laser-induced breakdownspectroscopy analysis of East African Rift sedimentary samples under Marsconditions, Chem. Geol. 294 (2012) 135–151.

[7] A. Cousin, O. Forni, S. Maurice, O. Gasnault, C. Fabre, V. Sautter, R.C. Wiens, J.Mazoyer, Laser induced breakdown spectroscopy library for the Martian environ-ment, Spectrochim. Acta B 66 (2011) 805–814.

[8] S. Palanco, C. Lopez-Moreno, J.J. Laserna, Design, construction and assessment of afield-deployable laser-induced breakdown spectrometer for remote elementalsensing, Spectrochim. Acta B 61 (2006) 88–95.

[9] L.A. Alvarez-Trujillo, A. Ferrero, J.J. Laserna, Preliminary studies on stand-off laserinduced breakdown spectroscopy detection of aerosols, J. Anal. At. Spectrom. 23(2008) 885–888.

[10] J.J. Laserna, R.F. Reyes, R. Gonzalez, L. Tobaria, P. Lucena, Study on the effect ofbeam propagation through atmospheric turbulence on standoff nanosecond laserinduced breakdown spectroscopy measurements, Opt. Express 17 (12) (2009)10265–10276.

[11] P. Lucena, I. Gaona, J. Moros, J.J. Laserna, Location and detection of explosive-contaminated human fingerprints on distant targets using standoff laser-inducedbreakdown spectroscopy, Spectrochim. Acta B 85 (2013) 71–77.

[12] J.L. Gottfried, F.C. De Lucia Jr., C.A. Munson, A.W. Miziolek, Double-pulse standofflaser-induced breakdown spectroscopy for versatile hazardous materials detection,Spectrochim. Acta B 62 (2007) 1405–1411.

[13] J.L. Gottfried, R.S. Harmon, F.C. De Lucia Jr., A.W. Miziolek, Multivariate analysis oflaser-induced breakdown spectroscopy chemical signatures for geomaterial classifi-cation, Spectrochim. Acta B 64 (2009) 1009–1019.

[14] F.C. De Lucia Jr., J.L. Gottfried, Classification of explosive residues on organicsubstrates using laser induced breakdown spectroscopy, Appl. Opt. 51 (7) (2012)B83–B92.

[15] J.L. Gottfried, Influence of metal substrates on the detection of explosive residueswith laser-induced breakdown spectroscopy, Appl. Opt. 52 (4) (2013) B10–B19.

[16] M. Milan, J.J. Laserna, Diagnostics of silicon plasmas produced by visible nanosecondlaser ablation, Spectrochim. Atca B 56 (2001) 275–288.

[17] J.S. Cowpe, R.D. Pilkington, J.S. Astin, A.E. Hill, The effect of ambient pressure onlaser-induced silicon plasma temperature, density and morphology, J. Phys. D.Appl. Phys. 42 (2009) 165202–165210.

[18] A.M. El Sherbini, A.A.S. Al Amer, A.T. Hassan, T.M. El Sherbini, Measurements ofplasma electron temperature utilizing magnesium lines appeared in laser producedaluminum plasma in air, Opt. Photonics J. 2 (2012) 278–285.

[19] H.R. Griem, Plasma Spectroscopy, McGraw-Hill, New York, 1964.[20] M. Sabsabi, P. Cielo, Quantitative analysis of aluminum alloys by laser-induced

breakdown spectroscopy and plasma characterization, Appl. Spectrosc. 49 (4)(1995) 499–507.

[21] D.A. Cremers, L.J. Radziemski, Handbook of Laser-induced Breakdown Spectroscopy,John Wiley & Sons Ltd, Chichester, 2006.

[22] S. Choi, J.J. Yoh, Laser-induced plasma peculiarity at low pressures from theelemental lifetime perspective, Opt. Express 19 (23) (2011) 23097–23103.

[23] A. Miziolek, V. Palleschi, I. Schechter, Laser-induced Breakdown Spectroscopy,Cambridge University Press, New York, 2006.

[24] S. Yalcin, Y.Y. Tsui, R. Fedosejevs, Pressure dependence of emission intensity infemtosecond laser-induced breakdown spectroscopy, J. Anal. At. Spectrom. 19(2004) 1295–1301.