cmrs UNIVERSITY DE i'Al J Fl DFS PAYS f]lf- I 'AiJfX IK Femtosecond laser ablation Femtosecond Laser ablation/ICPMS for trace element and isotope ratio measurement: application to biomineral, petroleum industry and forensic science... And poetry! Christophe Pecheyran Laboratoire de Chimie Analytique Bio-lnorganique et Environnement (LCABIE), UMR CNRS-UPPA 5254 IPREM, Helioparc Pau Pyrenees, 64053 Pau, France christophe.pecheyran@univ-pau.fr Laser ablation for direct solid sampling Linear scan Direct analysis Micro sampling (5-200 Jim) Macro sampling (>200 |im) Minimizing invasive approach Screening Bulk analysis Geosciences Environment Protéomic Petrochimistry Forensic Archeology Surface analysis 1 : r. Depth profiling 2 Laser basic principle Spontaneous emission Stimulated emission Spontaneous emission: the energy accumulated by an atom is released via the emission of a photon. Stimulated emission (discovered by A. Eintein) induced by a photon on a excited atom, this phenomenon leads to the emission of a photon of the same energy, same direction and phase (coherent light). When a population inversion is present the rate of stimulated emission exceeds that of absorption, and a net optical amplification can be achieved This property is used to generate laser light... Excited state Ground state 3 Laser basic principle Miror PumpinglL Active medium The population inversion is triggered by an external source of energy (pumping: light flash, electric source, or even lasers) Combined with a resonator (2 mirors and an active medium), the laser light can be accumulated in the axis of the resonator. Resonator The active medium might be a crystal, a gas, or a liquid. When the accumulated energy is high enough, and before destroying the resonator the end mirror is made transparent to the light (using a Pockels cell) 4 Laser ablation-ICPMS • Limits of Detection from sub ppb to ppb range • Dynamic range from LOD to up to few precents concentration • Isotopic information • Easy set up 5 Laser ablation-ICPMS set up Wet versus Dry plasma: In some conditions (i.e when the oxydes level is not a problem), the LA/ICPMS coupling can be used in wet plasma configuration with some advantages: In-torch aerosol mixing! Dry aerosol (laser ablation) • Wet aerosol (nebuliser) 0 Easier daily tuning (mass calib, lens detector calibration, etc..) 0 Plasma stability monitoring 0 Internal normalisation 0 Better plasma robustness A The ICPMS family... Laser ablation-ICPMS set up Various laser types... Type Longueur d'onde Energie/ impulsion Longueur d'une impulsion Diametre faisceau primaire Frequence de tir nm J ns mm Hz C02 10600 0.3 100-1000 2-5 100-1000 Rubis 694 1 10-100 10-25 0.1 Nd :YAG fondamental 1064 0.5 -4 - 100 5-10 1-20 Nd :YAG harmonique 4 266 0.003-0.009 a u a Nd :YAG harmonique 5 213 > 0.001 ti a ti KrF 248 0.002 10-100 20x10 1-100 ArF 193 0.2 10-100 20x10 1-100 Femtoseconde 196-265-8 00-257 -515- 1030 0.0005 0.0001 - 0.0004 4 1-10000 •UV nanosecond lasers (266, 213 and 193 nm) are the most used in analytical chemistry. * Femtosecond lasers, more expansive, though having better performance are less used. 9 Some definitions Gaussian shape: the beam energy distribution of a laser is naturally gaussian: when drilling a hole, the resulting crater shape is gaussian (or cor\\ca\ in a first approximation) Flat Top : some optical arrangements allow homogenising the energy profile : the resulting crater is now a cylinder. Flat top Fluence (J/cm2) : energy per pulse/ laser beam size at the surface of the sample. Power density (W/cm2) : Fluence/pulse duration Gaussian shape 10 LA/ICPMS since 1985 in 3D... lOHz lOOOHz Repetition rate Advances in LA/ICPMS since 2003 in 3D Ablation processes LASER 3 main processes take place during the ablation: 1st step; the laser energy is absorbed, converted into heat, when the heat is not dissipated then melting and sublimation can occur. This occurs at low power density (<106W/ cm2). In contrast, when the power density is high enough (>109W/cm2), ie high energy or/and short pulse duration, "spontaneous" sublimation occurs that will generate explosion and ejection of particles by fragmentation. • 2d step; free electron occuring in the vapor phase and those formed by thermal isonisation lead to the formation of a micro plasma at the surface of the sample. • 3rd step; the plasma expands and generate a shock wave. 13 laser ablation time scale Electrons absorb Electron emission Plasma formation Particles ejection photons from ^e surface Nanoseconds Microseconds Femtoseconds Picoseconds Time(seconds) 14 Adapted from R. Russo - Winter conference 2004 Short pulse versus Ultra Short pulse Short pulse few nanoseconds (10-9 s) Ejected molten MATERIAL Long pulse laser beam Surface debris RECAST LAYER DAMAGE CAUSED TO ADJACENT STRUCTURES heat transfer o surrounding material ©1999 Clsrk-IVKR, Inc. Thermal effects accuracy => in-depth resolution hundreds femtosecond es (10-15s) u lt rarest laser pulses NO recast layer NO surface debris NO damage caused TO adjacent structures Plasma pl un e NO MELT ZONE NO M ICR O CRACKS NO SHOCK WAVE NO HEAT TRANSFER TO SURROUNDING MATERIAL OT, DENSE K> IfELECT ROM SOUP [I.E. Plasnai ©1999 Clark-MXR. Inc. Very limited thermal effects => better accuracy (due to small particles) => better in-depth resolution Ablation processes At low power density (106 W/cm2), the vaporisation is the dominant process and the ablated material is made of vapor, aerosol, melted particles. These conditions should be avoid as much as possible (see below). Beyond 109 W/cm2, the process becomes explosive and thermal effects are reduced. In addition, the micro plasma formation can interact with the sample and generate additional melting and vaporisation. The interaction between the laser light and the sample is a dynamic process governed by the laser properties AND the sample characteristics : LASER Laser source Wavelength, energy, pulse duration, repetition rate laser beam Size, focus SAMPLE Optical absorption, reflectance Thermal conductivity, specific heat of vaporisation Mechanic 16 Typical craters Glass samples Nanosecond IR (1064 nm) laser: evidence of high thermal effects (melting) composition... "Micro- and macro-scale investigation of fractionation and matrix effects in LA-ICP-MS at 1064 nm and 266 nm on glassyi7 materials", M. Motelica-Heino, P. Le Coustumer, and O. F. X. Donard, J. Anal. At. Spectrom., 2001, 16, 542-550 Typical craters Carrier gas Sorrier gas influence: the analytical performances are better when Ar is replaced by He : => better cooling effect (then smaller particles), and particles are more easily extracted from the bottom of the crater. The particles are however better transported in an Ar stream, that's why Ar is added post ablation cell. The sensitivity improvement is about 2-3 when using He in the ablation cell. Ar Carrier Gas -3.8 x 10_5cal/sec Tern He Carrier Gas ~3.2 x 10 cal/sec °C cm UM« DON Pi Liu jo 7: Gas Effects: Crater Characteristics in NEST 612 Glass (images courtesy of USGS Denver CO, USA) CetacLSX 200-266 nm In addition, the spectral background being reduced when adding He in the ICPMS, LOD are finally improved by a factor 10 (this phenomenon^ is more pronounced at 193 nm compared to 266 nm). He 1.2 l/min Ar 1.0 l/min Vers spectrometre D. Gunther et al, JAAS, 1999,14,1363-1368 19 Sample preparation LA/ICPMS requires very limited sample preparation. However flat sample surface generally facilitates laser focussing (better visualisation). Powders (eg: clays, sands, aerosols, sediments, etc) must be fixed: compaction (binder addition (ex : paraffin), drying and palletisation with high pressure hydraulic press). Lithium borate fusion (Li2B407). Small samples must be embedded into a resin Binders and Li borate might induce contamination. It is well indicated to pre-clean the sample surface by ablating the desired zone with low energy (close to the ablatio threshold) laser conditions; only few shots are required. 20 Quantification The reason why you will need, sometimes, to spend a lot of money for high performance laser systems (from 60 k€ to >400 k€...) Quantification The quantification is sometimes challenging properties. the Achilles heel of laser ablation. Quantification is because the ablation processes depends on the sample The ablation yield varies for different sample matrices (e.g. the ablation rate is not the same for copper than for glass or polymers... The ablated mass depends on the laser energy but not linearly : f luence (=> f ocalisation), pulse duration, etc.. The detected elemental composition of the ablated mass might depends on the laser energy (elemental fractionation) We will see later how to identify the elemental fractionation, how to calculate it and what are the physical processes behind it... And eventually how to counterbalance it or at least limit it... 2?ng 30% Cu, 70% Zn Sample: Brass 50% Cu, 50% Zn 70% Cu, 30% Zn Sample: Brass 50% Cu, 50% Zn 50% Cu. 50% Zn Sample: Brass 50% Cu, 50% Zn Adapted from R. Russo - Winter conference 2004 Quantification: External calibration with internal standard The quantification of a sample is generally carried out by external calibration using standards of similar matrix. For synthetic materials (glass, etc.), numerous standards are commercially available. Bias resulting from laser ablation processes (laser energy drift, focussing problems, sample transport, etc..) and the ICPMS can be corrected using an Internal Standard (a given element that we know the concentration in the standard AND the sample). The IS must be homogenously distributed (generally part of the matrix) as it is used for normalisation. The IS in the sample can be obtained by XRF measurement for instance. A With Astdi: NET (blank corrected) signal intensity oftheAnalyte for standard i ISstdj: NET (blank corrected) signal intensity of the Internal Standard for the standard i Asamp: NET (blank corrected) signal intensity oftheAnalyte for the sample ISstdj: NET (blank corrected) signal intensity of the Internal Standard for the sample 4 £q -—> I ] Concentration Internal standard normalised calibration curve Quantification: External calibration with internal standard Then, knowing the concentration of IS in the sample, the concentration of the analyte in the sample is expressed as follow : This equation WORKS ONLY IF the internal standard behaves similarly as the analyte in terms of laser/material interaction AND particle atomisation in the ICPMS. For natural samples, it is not always true due to the wide range of matrices and the lack of Certified Reference Materials (CRM). This makes the quantification more challenging... Quantification: Evidence of Elemental Fractionation Ideally, when ablating a homogenous sample, all the elements should have the same signal profile. In fig A, a crater was drilled in CRM NIST610 (glass). It is clear that Ca and U exhibit quite similar profiles (the progressive signal extinction is due to laser fluence drop as the crater becomes deeper. One can also imagine that particles are less efficiently removed from the bottom of the crater). However, Pb profile departs from Ca and U which indicates that all the element don't behave similarly. This calls into question the use of an internal standard to correct for ablation processes drift, especially when the sample matrix differs from the standards one... The quantification is directly affected as illustrated in figB, even after internal normalisation (here Ca) as the calibration slopes differ from glass to CaC03... THIS IS THE ELEMENTAL FRACTIONATION.... D. Guntheretal, Spectr. Chim. Acta, 1999, 54, 381-409 A. Barats etal, PhD thesis - 2006 Quantification: Origin of the elemental fractionation Two main processes can explain the elemental fractionation 1/ Selective evaporation : It appears as a result of the laser/material interaction where the vapour phase is enriched with some elements with low sublimation enthalpy : this is then a non stoichiometric sampling where the aerosol composition (particles + vapor) does not reflects the sample composition... It is governed by the ablation conditions (temperature, the time duration of the interaction, induced plasma, the surface of interaction, etc..) and the physical properties of the sample. Melting and melted material deposited around the crater enhance this phenomenon. This fractionation affects mostly siderophile elements (Zn, Pb, Au, Tl) whereas refractory elements (Ca, U, Th, rare earth) are less affected. Over the course of the ablation, the sample is more and more heated (as the induced plasma interacts with the crater wall) which generated the vapor enrichment with "volatile" elements : this fractionation can be evidenced by monitoring the ratio of a "volatile element" versus a refractory element (Pb/U) : the ratio starts from a "true" value that progressively increases. By using shorter laser wavelength (deep UV), these thermal effects (and then the fractionation indexes) are much less pronounced which explain the success of 193 nm, 213 nm or 266nm lasers. Quantification: Origin of the elemental fractionation 2/ Particle size: Incomplete atomisation of the laser induced particles, into the ICP, has been proposed more recently. Here the fractionation does not occur at the ablation site, but in the plasma of the ICPMS, the biggest particles being less efficiently atomised (>150 nm) than the smallest. In these hot plasma conditions (7500 K), the most volatile elements contained in the particle diffuse more rapidly than the most refractory which generates a bias if the particle is not completely atomised : the remaining particles are then highly enriched in refractory element that will never be ionised (and then detected). This is illustrated here with U (heat of vaporisation=417 kJ.mol1) and Th (heat of vaporisation=514 kJ.moh1) This phenomenon is also indirectly governed by the laser wavelength since the particles size distribution is shifted towards smaller particles when using short wavelengths (see below). 27 Quantification: Origin of the elemental fractionation During the ablation, the particle size distribution changes; big particles (up to 2|j.m) being produced during the first pulses whereas smaller particles are gradually produced. The U/Th ratio is then biased at the beginning and tends towards true value at the end of the ablation. Toward true CO 1 HI 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0,4 0.2 O.O 9 o © © © © o Vapour phase condensation-Coagulation slow energy removal Helium Argon high thermal conductivity low thermal conductivity Adapte de Horn et al Applied Surf. Sei, 2003 The key is the particle formation... 2 - Hydrodynamic sputtering (spherical particle>100 nm) More info in Hergenroeder J. Anal. At. Spectrom., 2006, 21, 517-524 | 517 The thickness of the melting layer depends : * the wavelength * Pulse duration (very thin layer with femtosecond pulses) Droplet ejection due to Kelvin-Helmholtz The key is the particle formation... Melted material 266 nm UV laser Transparent glass sample (NIST 610) 193 nm UV laser ICP dependent critical particle size for incomplete vaporisation d 193 nm, He H. Kuhn etAI, abc, 2005,383,434-441 The aerosols is consituted by agglomerated primary particles and spherical particles resulting from melting. At 266 nm more spherical particles (and also bigger) than at 193 nm, Particle size Fi**t 12 Schematic diagram of influence of ICP conditions on incom plete particle vaporisation and the contribution of the latter to elemental fractionation. M. Guiflong etAI, J A AS, 2002,17,831-837 32 How to quantify the elemental fractionation... For a given analyte, the fractionation index (FI) is defined with respect to another element taken as reference (generally Ca or Si). It reports the variation of the ratio over the course of the ablation taking into account 2 zones (Pb) / {Ca) riPb!Ca ~ / (Pb) 2 zone! (Ca) zone! CO Q. o Ü CO 10,000,000 ' 1, Zone 1 Zone 2 .A. 100 150 SECONDS 33 250 Fractionation Index Fractionation Index W S a © ^ = s s > 3 =L r * • 3 S a « ^ & ^ IJ ■ - £ o -■ S _ ö PUS 2 3 " 2 a 5 ,J es = = I ■_I E3 Ol» a B ■ i & e s h - — & 3 I 5 3 o ES a c W 3 V w 3 V w 3 3 op 3 fD fD QJ 3 to O QJ fD 00 CO fD Element Ratio 3 i p 00 LI7 -B11 -Na23 -Mg24-AI27 Si29 -P31 -Ca42 -Ti49 -V51 -Cr53 -Mn55 -Fe57 -Co59 -Ni60 -Cu65 -Zn66 -As75 -Sr88 -Zr90 -Mo98 -Ag107 -Sn118 Sb121 Cs133 Ba137 La 139 CeHO Pr141 EU153 Tb159 Ho165 Tm169 Lu175 Hf178 Ta181 W182 Au197 T1205 Pb208 BJ209 Th232 U238 ro ro ^ t£> Ol w w Volume distribution dV/dR / [fLm3 urn'1) J.-i-\ .MM-1-1-1 I I J J I I 9 W M -1 OS _k to a U tri 3 3 3 3 3 3 SU O ft O 0) 0) o fD 213 nm vs 266 nm lasers Accuracy of 266nm vs 213nrn No Difference Eetter by 213 nm Insects Minerals Fish ears Soils Tree rings U/Pb dating Gl Fluid Inclusions Skins Metals Teeth - Plastics Shells Cöurt*?y öf Tjt T. Jě&í&s, ITtfural Tfatroy Ttfusfranm v»6 1e*4 Ca 42 NIST610 - 195 ***** 213 nm 100 t2t) Time/wc Precision of 266nm vs 213nm Better ty 26£ nm No Difference Better by 213 im Insects Soils Fish ears Minerals Tree rings U/Pb dating Fluid inclusions Glasses Skins Metals Plastics Teeth I Shells C'jurtefy of It. T. Tdfrra: Natural Hüflroy THu£*HLim. Solutions for elemental fractionation... ELEMENTAL FRACTIONATION Non stoichiometric detection Solution? Matrix Matching using certified reference material similar fractionation in the sample and the standard UNIVfKSlll Using less fractionating lasers => Deep UV, => Fluence => Power density, => Ultra short pulses (fs), Etc 37 Tuning the LA/ICPMS Tuning of the LAICPMS coupling : •According to "normal" ICPMS recommendation in liquid configuration • Efficient particles atomization Use a NIST 612 glass sample and set the ICP parameters in order to get U/Th =1 ±0,05 (theoretical value U/Th=l). The U/Th ratios tend to be higher than 1 due to incomplete particles atomization. Robust ICP conditions are mandatory to reach accuracy. They are very often obtained while sacrificing signal sensitivity, and vary from one instrument to the other : •High power (>1200W) • Reduced carrier gas flow (compared to the optimal signal sensitivity conditions) •Reduced auxiliary gas Isotope fractionation nsLA/MC-ICPMS of chalcopyrite 0.4460 £ 0.4450 O O co to "B ín 0.4440 CO 0.4430 0.6070 □2 - 0.6060 C N (O (O 0.6050 50 100 150 Time (S) 200 250 Like elemental fractionation, isotope fractionation occur also. It is mainly attributed to incomplete particles atomisation and faster diffusion of light isotopes... Fig. 1 Time resolved 65Cu/63Cu ratios, corrected for mass bias, for ablation of chalcopyrite using Ar (bold) and He (light) as ablation gases (ablation commenced at 20 s). The 65Cu/63Cu ratios gradually increased towards the value determined by solution-MC-ICP-MS analysis for the same chalcopyrite sample. The thin line shows the raw ratios for an Ar supported semi-dry Zn aerosol, produced by a desolvating nebulisation system, added to the carrier gas stream during the ablation of chalcopyrite in Ar. The constant 66Zn/64Zn ratio during ablation indicates that the increase in Cu isotope ratios was a laser-induced phenomenon. Data from Botfield (1999) and unpublished. S. Jackson and D. Guenther, 3. Anal. At. Spectrom., 2003, 18, 205-212 39 Plasma robustness... • Only particles < 100-150 nm can be efficiently atomised into the ICP • The ICP can tolerate a finite amount of particles (plasma loading) • Specific desorption of the species at different places into the plasma * Need for improving atomisation and ionisation efficiency * Need for improving the plasma robustness. Insights on the ablation cell => Aerosol transport efficiency affects sensitivity and accuracy => The finite size of the cell limits the size of the sample => Washout time : very important for keeping high spatial resolution (imaging). Insights on the ablation cell The volume of the cell affects: •spatial resolution in dynamic ablation (scar\s for instances) •extraction efficiency and then signal intensity •Washout time End of laser shot 1E+06 3 Cell A Cell B ® 1E+05 i o O CO >> 1E+04 i 1E+03 i 1E+02 120 130 ' 140 150 160 170 180 190 200 210 220 230 240 250 Time (s) 42 Insights on the ablation cell D. Bleiner, A. Bogaerts/Spectrochimica Acta Part B 62 (2007) 155-168 Towards mini cells for better washout time Towards mini cells for better washout time LFC, New Wave Research Laurin cell 2: Resonetics, Cetac New developements: ultrafast washout time : 30 ms! Wang et al, ana L chem. 2013 Supercell, New Wave Research The femtosecond world... ©1999 Clark-MXR. Inc. Ultra short pulses : femtosecond lasers Nanosecond pulses: Large heated zone laser/plasma interaction Large particle ejection due to melting Femtosecond pulses: The ablation process is confined in time and spatially No laser/plasma interaction Very thin particles are produced Less thermal effects 4 0000 3 0000 & E 2 0000 E □_ 10000 10 Tns=Anat • ns ■ fs 0.22 mj, 50 micron Fluence=11.2 J/cm' Distance: 0.6 mm T =A t fs fs 9^ 100 Time (ns) 1El9n 5E18- LU 1E18 10 ns fs 0.22 mj, 50 micron Fluence=11.2 mJ/cm Distance: 0.6 mm n =B t &,ns ns 50 100 Time(ns) 47 R. Russo - Winter conference 2004 Short pulses versus ultra short pulses Fs 800 nm Brass 25KU 1 0 h» m ■ £ mm 18mm ns 800 nm FIGURE 4. Fs- and iks-ablaiion of polished brass directly revealed different ablation mechanism. In both cases no beam shaping was used. Whereas the fs crater has a clean rim, die ns crater shows the typical pattern of inciting. (Modified from Elsevier, Spectrochim. Acta B 55(2000)1771). Glass Fs 266 nm ns 266 nm Poitrasson et al, Anal. Chem, 2003,75,6184-6190 Hergenroeder et al, spectrochim. Acta B, 55,2000,1771 48 Femtosecond ablation 266 nm « Nanosecond ablation 266 nm Short pulses versus ultra short pulses BÖO- n. + 0 4.0 ■ 30- m 20 in- 0.01 O. í ] panicle diameter [jun] 10 Ü.0 experimental valut- ---hulk value < 1.53) 0,01 0.] I particle diameter [\xm] to 800 nm nanosecond LA He, fluence 2.5 J.cm2 c N + O 3000 2006 1000 - 0 Particle size distributions 800 nm femtosecond LA He, fluence 2.5 J.cm2 4.0- C N O Zo- ro- TTT I ..Li 1 I experimental value ---bulk v;iliK-(KM) ■ n 11 f*i (Mil 0.1 I til 0.01 TT+F->l * l*nh| I ■ •> li" i 10 Particle diameter [pm] Particle diameter [pm] KocAi et a/, J - 4t, Spectrom., 2004, 19f 267-272 50 Short pulses versus ultra short pulses Cu65 samples 1COOOQO A i O o 100000 H co O 10000- 7 20 t—■—]—i—i—'—r 40 60 30 100 t—i—i—i—r 120 140 160 i—i—i 180 200 Timefs) Very limited elemental fractionation due to: • Very limited thermal effect (no preferentia evaporation) • production of thin particles easily atomised in the ICP • Augmentation du signal d'un facteur 10 Stability improvement Sensitivity improvement (mostly on metals) 5 4 2 E E 8 Oř DC £ TJ tu .DJ zA 4- 3 J 2- I A 0 ns lase ■—NIST 610 — NIST612 Monaztte Managotry — Monazite Moacyr zircon 31500 50 150 Time (S> fs laser Poitrasson et al, Anal. Chem, 2003,75,6184-6190 t 0 SD —i-1-r~ 1 DO 1S0 Time ťS^ ::::o 250 Is non matrix match calibration possible? 0 .1 ~\-1-1-1-1-t—i—i—t~ |-r-1-1-1-t—i—t—r-] 0-1 1 10 concentration ratio of Zn/Cu Fig- 3 The experimental 64Zn/63Cu ratio against the certified ratios for five brass, two aluminium and one glass sample measured with different fluences. The straight lines represent fits to the measured data (full line 42 J cm2; dotted line 34 J cm2; dashed-dotted line 23 J cm2; dashed line 12 J cm2; short dashed line 5.7 J cm2). The data for 42 J cm2 were recorded with a tighter focused beam while the fluence of the other measurements was varied by attenuation of the laser beam. Q. Bian, JAAS, 2006,21,187-191 Non-matrix matched calibration • High fluence for glass sample • Low fluence for high thermal conductivity and low melting point sample (Al) 52 Isotopic fractionation? A 12 10 „ 8 "5 > 6 >! ■*—f 4 0 o o B 16.85 O 16.84 CO Q 16.83 g 16.82 16.81 16.80 JM Puratronic (pure Fe-metal) • 5 Low energy (less than 2 mJ) > High repetition rate (<100 kHz) > pulse duration = 360 fs > Crystal: KGW doped with Yb, 1030 nm in the fundamental Diode pumped laser — Compact — Robust 75 cm <10 kHz E <0,1 mJ <100 kHz E<2mJ Amplitude Systemes, Pessac, France s-Pulse HP First High repetition rate femto-ICPMS Alfamet : developed by LCABIE - Amplitude - Novalase mplUu< Fs pulses Compact Mobile Fully integrated Automated High spatial resolution Precise energie control High analytical performances Low energy High repetition rate First High repetition rate femto-ICPMS..: Low energy high repetition rate (10 KHz) Fast laser beam movement r * * Surface analysis High spatial res Low sensitivity SENSITIVITY Virtual beam shaping Galvanometric mirrors Small crater l~ 10 urn) High repetition rate ( (A 1 1,4 o c 3 0,9 10,4 -0,1 88Sr R2 = 0,999991 0 20 40 60 80 Concentration (ppm) 100 Glass (NIST 614) [Co] = 0,73 ppm v> Iß a. o 80000 70000 60000 50000 Reproducibility : 5 % 0 150 „m •£ 40000 5 30000 20000 10000 0 0 50 100 Temps (s) 150 200 Max energy (here 0,5 uJ) @ 10 KHz 60 Virtual beam shaping for flash ablation : LOD Using a Thermo X7 quad ICPMS Femtosecond (360 fs) Nanosecond (8 ns) flash ablation UV 266 nm, 20 Hz, 2mJ Limites de detection (ng/g) (nist612) Femtoseconde Improvement factor compared to 266 nm ns laser 59Co 14,1 x25 65Cu 38,6 xl2 88Sr 3,3 x2i 107Ag 1,10 x8 mCd 197Au 0,73 xlO 205J| 0,42 xl3 208Pb 1,10 xll 232Th 0,15 x41 238U 0,13 x 16 Signal stability Femtosecond IR 1030 nm Energy : 75 |jJ/pulse Repetition rate : 500 Hz Ablated width : 100 |jm; Stage speed: 10 |jm/s Single line scanning ablation mode & 3,E+06 i f 2.E+06 -c I 1,E+06- (0 c (75 = 0,E+00 0 20 RSD = 3.1% 40 Time (s) 60 80 Ablation of a certified glass sample (NIST 612) Indium Nanosecond UV 266 nm Energy : 1.3 mJ uJ/pulse Repetition rate :20 Hz Ablated width : 100 um; Stage speed: 10 (jm/s 3.E+06 n a o £ 2.E+06 ^ C ~ 1,E+06 ^ .2* (75 0,E+00 0 RSD = 14% 20 40 60 Time (s) 80 100 62 Elemental fractionation CO X o u <0 ,5 r A ,3 ,2 ,1 0,9 O 0,8 0J 0,6 0,5 0* n Femto 5 kHz D Nano 266 nm 2 mJ 20 Hz o ■5.. 1 1 63 Elemental fractionation 1.5-j 1.4- 1.3- Li_ 1.2- LU 1.1- ~o CD M 1.0 1 0.9- o 0.8- 0.7- o CM 0.6- 0.5- The intrinsec properties of fs pulses are preserved even at high repetition rate... Glass CRM: Nist 610 0100 iim 10 kHz 5 kHz 1 kHz 0.3 kHz 0.1 kHz 1030 nm Fluence : 14 J cm-2 Scanner speed : 1.5 mm S"1 J A AS, 24 (7), 200% pp. 891-902 Al mau fraction [ing/kg] o LS-ICP-MS L4 ICP-MS _I_ LA ■JU *A r- CT > L13-GD-MS LlUCP-MS L19-LA-ICP-MS L2-ICP-ÖE5 L20 ICP-M5 As mass fraction fmg/kg] o vi H* k* L™ -j L20-ICP-MS l.il-5parlí-0ES L4-ICP-MS L3-ETV-ICP-ÜES LS-ICP-MS L17-ICP-MS 03 ö L7 DCnaft-OES 3 Q L19-LA-ICP-MS L14-INAA Ltí-ICP-MS L24CP-ÜE5 L13-GD-MS L16-ICP-MS LlMCP MS L10-INAA vi ♦ I 4» i i > t ._! 3 o O OJ (D 3 f? 3 ň 5* n CD rn TJ O n "O O In cn to "D o to LO Ag mass fraction [mg/Kgj LG-lCP-OtE> l9-ICP-m5 LW-iNM ĽMCP-OES Llí-lA-ICPM* L1V-ICP MÍ! 17-DC-arc-OK L13-GD-MS LHCPMS > ona Lia-icp-ws UQ-lNAA p i ď) o (D (D (D (D 3 (fi O O "U (D I TJ O c Q. 73 O G-3 Trace elements in copper - Round Robin Femtosecond laser ablation provides good precision and accuracy. 2.5 Note that due to the high thermal conductivity of Cu, laser ablation is not a priori the best indicated approach. Ě v\ 1/1 n E 1.5 i 43 3..S 2.S 2 -i LS 1 0.5 0 f 0.5 ^^;::::ž::::ž::::i:::^::4::::4:::" Cd i2 1 S/l 4 la LLl O ■ u ■ i Cu 3 s ■ I Laboratory code / Method t 5 3 ■ The laser beam is virtually at two places simultaneously => the laser ablates simultaneously the sample and an isotopically enriched sample the 2 aerosols are mixed into the cell > then the isotope dilution takes place into the cell enriched solid spike Direct iri-cell Isotope Dilution analysis Paramětres laser se avelenght Repetition rate nergiy canner speed (ablation) Scanner jump virtual beam shape Distance between 2 pellets ALFAMET 360 fs 1030 nm 1000 Hz 45|iJ 1 mm.s1 100 mm.s"1 150 \irr\ 1 mm Galvanometric mirrors Catalyseur Echantillon Etalon Direct In-cell isotope dilution In-cell ID/ICPMS < 6 min GBW-07405 200 i 150 -100 0 Sediment PACS-2 400 300 -\ - 200 ■ CD 100 0 Cu Conventional ID/ICPMS 6 Hours 800 600 i Ž 400 i 200 m—f 0 Sn 200 -150 - 4» 100 i 50 H 0 n □ In-cell / Fs-LA-ICP-IDMS Digestion / ICP-IDMS Certified values z n Cu ■rHT*1 Sn Pb APPLICATION TO SOILS AND SEDIMENTS ANALYSIS > in 6 min vs 6 hours limitations due to sample inhomogeneity Pb Fernandez et a I, Anal. Chem. 2008, 80, 6981-6994 Direct In-cell standard addition Sample Zr addO 20 Standard (catalyst) add 1 Add 3 60 Add 2 Add 1 AddO \ add 2 Constant nb of shots Identical ablation yield from one pellet to the other 80 Temps (s) add 3 100 120 140 60 er 8 40 £ c o E o C O u 0 ■ XRF ou ICP-AES B LA-ICPMS (Ajouts doses) LA-ICPMS (Et. externe matrix match) LA-ICPMS (Et. externe verres) Validation 20 ■JkTl inn Ba Zr Ce MgO AI203 Si02 E GL Q. (LI 3 O" "Ü5 tu E c o •4= CO u C o 6 4 2 0 i XRF ou ICP-AES H LA-ICPMS (Ajouts doses) ■ LA-ICPMS (Et. externe matrix match) ' LA-ICPMS (Et. externe verres) Pt Pd Rh La Nd Hf Application of the virtual beam shaping to biocarbonate analysis European eelAnguilla anguilla : biological cycle • Ubiquitous declining species with great local and european economical stakes Long live predator, high fat content, high resistance to physico-chemical conditions. Estuary Anthropogenic activity influence on the Adour ecosystem : eel as a model Markers of physico-chemical condition : from the molecular response level to the reconstruction of the individual history through the otolith (larvae period) otoliths as a tracer of migration High spatial resolution and sensitive analytical technique LA-ICPMS Trace element and isotope microchemistry RÉPUBLIQUE FRANCAISE CARTE CHDENTITÉ NATIONALE IM- ; 0Ä0. Familie : Salmoniden Nationalste Francabe ~_ Néle: 23 mars 2006 Genre Salrrvo A: Frayére de Espéce : Salär Departement: 64 Cařacténsitque: SaumondeIPadour Origine: Sauvage I0FRATAF<<<<<<<<««<«<«<«<13132A 08CK980498(K9 SAUMON << ATLA 1114 141M7 Ba, So Mg, Mn, Zn, Cd, Pb. Li. Ca... 87Sr/86Sr Virtual beam shaping Increasing the ablation rate to improve the signal to noise ratio... Crater (20 |jm) Mn Continuous scan (width 20 urn) i Mn nucleus 0,5 Flash ablation mm Trace element detection Flash ablation to understand a given life-time history Otolith border = sampling place Flash ablation to understand a given life-time history atal zone of interest Flash ablation of the post natal growing zone for high sensitivity... Ablation rateflash = 20 x Ablation rate scan 2500 2000 <1500 o u 1000 500 0 0 17 ppb (LOD = 0.6 ppb} 208Pb flash 208Pb 100 nm scan 50 100 150 200 _Time ts)_ 250 300 Lrt D O u 1000000 900000 1 800000 700000 600000 500000 400000 300000 200000 100000 0 18 ppm 138Ba flash 138Ba 100 pun scan 50 100 150 200 250 300 Time (s) Simulating a corona-shape laser beam Perrier et al, Can. J. Fish. Aquat. Sci. 68:977-987 (2011) Fig. 2. Pictures of (a) juvenile and (b) aduit Atlantic salmon otoliths showing the area ablated by the laser prior to inductively coupled plasma mass spectrometry analysis. The annulus ablated is ZOO-GOO pm, is centered on the primordium, and corresponds to the juvenile period of growth only. Simulating a corona-shape laser beam Fig. 3. Representative otolith image with an annular ablation trajectory along the first feeding mark (15 jim away from the primordium) - Eel. Sim a semi corona- laser beam í Trace metal analysis Ba, Sr, Mg, Mn, Zn, Cd, Pb, Li, Ca... Isotope distribution 87Sr/86Sr,... Simulating a sharp blade laser beam Scanning with a virtual sharp blade 10 |im 10 pirn bjO a3 £ o 1000 Hz Large spot vs sharp blade ablation : -> better spatial resolution while keeping high signal sensitivity o B CP CO O Time Sharp blade ablation scan, growing axis, life history Quantitation and quality control taw u»w t>u V I ___: t i 5 u tM* 11 m 4*3 m IN u m !-1 HI Ml M 3n M *-» * It tm J IV cum *j; id mat »111! 1 ■ * 71 * *■ 4 » ETALON WAGE 1 [m*«nj 9 in* 'i D.MOQO ftflOQCQ ■ CMDQQ "3" 0 ttX'J* * emtotecond law oWow Curl k M wtttm' [irUftt* CI I Ml etftift* *u * ftftftM* We have developed our own VBA Macro working with Excel (FOCAL series) for data processing and quantification. FOCAL series allows rapid quantification with quality control tools for the various laser ablation approaches : ablation lines (continuous scan spot for across a sample, ablation, rasters imaging). When images are required, the processed data are imported with ImageJ for easy handling Otolith imaging The sharp-blade shape allows real square pixels and better resolution. Otolith imaging Otolith imaging as a decision tool to select the best ablation scan direction for routine analysis ^Uiix/zu pixels; ib-Dit; Z-BMB PPotof r028Bad! - 010414 - ^H? 100« vlBal 38-2 $$5^4x0.00 pixels (1028x253E~B=BfiT2:56k 0.0015 0.0010 u 03 5 m a 0.0005 CO 0.0000 0 100 200 300 400 Distance (pixels] 500 600 700 BOO List Save... | Copy..*. LlV€ Live IX-15.7, Y-9-22E-4 FsLA- 2D-SF-ICPMS : Simultaneous isotopic and trace element information LAMBDA 3 360 Fs lHz-100 000 Hz 257 nm Using the same laser trace Element XR (Jet Interface) High sensitivity (HRICPMS) Nu Plasma HR High isotopic precision (MCICPMS) FsLA- 2D-SF-ICPMS : Simultaneous isotopic and trace element information IR-Vis-Uv^ Fs Laser gas flow rate tuning split adjustment Plasma viscosity + and other parameters Stability Sensitivity Accuracy Jet-HR-ICPMS (wet plasma) FsLA- 2D-SF-ICPMS: Simultaneous isotopic and trace element information Isotopic data reduction for fast transient signal using an innovative Linear Regression Slope approach 0,98 - 0,78 0,58 0,38 H 0,02 63 0,7600 0,7500 0,7400 0,7300 0,7200 0,7100 0,7000 73 83 Temps (s) 93 □ □ □ □ Very easy Natural weigthing of the statistic precision improved by x 10 (minimum) mass bias correction using 88Sr/86Sr Ttetzke et al.(2008), J AAS, Sr by LA/MCICPMS •Epov et al.(2010), Anal. Chem: Hg by GC/MCICPMS ^Sanabria et al (2012), Anal.Chem; Pb by GC/MCICPMS \~LCABIE •Resano et al (2012), JAAS, Cu by LA/MCICPMS 87/86 0,8 5 £ 0,4 - 0,0048 ,99999 1.0 1 CO 0,5 H 0,0 86/88 y = 0.115472X + 1,096915 R2 = 0,999999 ■10,0 -8,0 -6,0 -4,0 -2,0 88 {V) 0,0 Precision on fast transient signal using the Linear Regression Slope approach External Precision (2xRSD) n=10 : 200 ppm i.e. (0,02 %) Internal precision (2xRSE): 600 ppm, (i.e. 0,06%) 0,7098 0,7096 0,7094 H 0,7092 0,7090 0,7088 0,7086 0,7084 Pelletized Nies22 otolith powder i 2 3 4 5 ■ 87/86 measured 87/86 ref value 6 t 7 8 9 10 FsLA- 2D-SF-ICPMS : Simultaneous isotopic and trace element information 1600 Downstream 0,2 0,4 0,6 0,8 0,712 0,711 ' 0,71 0,709 3 0,708 0,707 0.706 Typical LODs with fsLA/Jet-HRICPMS Sr =40 ppb i.e. 1,2femtograms Ba = 6 ppb i.e. 180 attograms Cd = 4 ppb i.e. 120 attograms Versatile approach with complementary information. Relevant when some geochemical signatures are not significantly pronounced (here 87/86Sr) Application of the virtual beam shaping to crude oil Direct anaysis of trace metals in petroleum product Better understanding in petroleum reservoir => upstream (Exploration). A. Ant id inn I trap B* Fallit trap Oil-Oil correlation Oil-source rock correlation Deposits conditions Oil maturity Oil migration Degrees of biodegradation Geochemistry Analytical challenge Organic product analysis still not easy in ICPMS : • high sample dilution (crude oil) • 02 feeding • Mineralisation Trace element analysis in crude oil by FsLA- ICPMS : Laser Ablation • Direct analysis • Very limited sample preparation No sample dilution • Low organic sample introduction Hg/min) => better plasma tolerance => No need for oxygen feeding (~20 High repetition rate FsLA/ ICPMS coupling • Fast analysis (< 2 min compared to 20 min in conventional nebulisation) Trace element analysis in crude oil by FsLA- ICPMS : He Galvanometric mirrors 0 2mm max Neb • Wavelength : 1 030 nm, • Pulse duration : 360 fs, • 17 u.m laser beam • Repetition rate : up to 10 000 Hz L |_|jg^ sensitivity • Fast 2D laser ablation trajectories (up to 280 mm.s"1) Trace element analysis in crude oil by FsLA- ICPMS : droplets) just after ejection => Limit contamination Trace element analysis in crude oil by FsLA- ICPMS : Normalisation 115 In 200 ppb Pb 300000 n 250000 - Q. U 200000 - 150000 00 o 100000 - 50000 - Huile CRM 1634c Pb208 208/115 t 3 + 2,5 + 2 t 1,5 + 1 + 0,5 a. oo o transport efficiency ~ 30 % Sample preparation: • Indium (internal standard) + Xylene (dilution 2 to 5) LOD (Hg/g) 51V 0.005 60 Ni 0.008 59 Co 0.001 75 As 0.007 98 Mo 0.002 208 Pb 0.0006 Evaporation under Nitrogen stream Trace element analysis in crude oil by FsLA- ICPMS External calibration with NIST 1634c => Fast (2 min) and accurate 3 100 3 ition 10 ® 1 Cone 0,1 0,01 Venezuelan oil 1 v LA/1 CP MS Digestion ICPMS Co Ni Mo In press in ABC Pb CUD 1000 100 10 o VP H3 U c O O 0,1 0,01 Venezuelan oil 2 □ LA/ICPMS ■ Digestion ICPMS Co Ni Mo Pb Forensic applications Fighting against Wine counterfeiting • Is this bottle counterfeited? • Where is it from (country)? • Who did it? Trace element analysis in glass FsLA- ICPMS : 257 nm Femtosecond laser LAMBDA 3 Galvanometnc mirrors ICPMS Ablation : 100 x 500 urn Calibration with NIST 612 (40 ppm) Trace element analysis in glass FsLA- ICPMS : 0 5 10 15 20 25 30 35 40 45 50 55 60 m/z BO0Q Trace element analysis in glass FsLA- ICPMS : Laulan — Sauterne std 612 — A 150 m/z 155 160 165 170 175 180 2000 | I two 1000 soo 0 ISO Femtawjcond tog* obtahon Laulan A Sauterne -std 612 1S5 190 195 200 205 210 m/z 215 220 225 230 235 240 75 elements analysed by UVfs/HRICPMS 4 bottles of de Pauillac + 4 suspect bottles (Asia) + 1 copy of Pomerol (Asia) Li7 Ti47 Rb85 Sn118 Tb159 Be9 V51 Sr88 SM21 Dy163 B11 Cr52 Y89 Te125 Ho165 Na23 Mn55 Zr901127 Er166 Mg24 Fe56 Cs133 in 93 Pt195 Au197 Hg202 Tm169 TI205 AI27 Co59 Mo95 Ba138 Yb172 Pb208 Si28 Ni60 Mo98 La139 Lu175 Bi209 P31 Cu63 Ru101 Ce140 Hf178 Th232 S32 Zn66 Rh103 Pr141 Hf179 U238 CI 35 Ga69 Pd105 Nd146 Ta181 Pu239 K39 Ge72 Ag107 Sm147 W182 Ca44 Sc45 As75 Se78 Cd111 ln11S Eu153 Gd157 Re 185 Os189 Very discriminating Discriminating Poorly discriminating LOD: 0,06 ng/g (Pr) - 2 ng/g (Pb) - 20 ng/g (Sn) all analyzed in medium resolution (so with only 10% transmission) => 20 discriminating elements with 9 highly discriminating Unambiguous autheratification using PCA Le millesime 2006 suspect est en fait un vrai Pauillac Forensic sciences at the particle size particles of few nanograms => detection of few pico, femto and even attograms ORIGIN La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er 100,00 j-1-1-1-1-1-1-1-1-!-!— _£ 0,10 0,01 USE U naturel U civil (exemple) {< 5% 235U) U-238 99.3% U-238 96% U militaire (exemple > 20% 235U) ^ Uranium isotopic composition of micrometric particles f n the framework of the Non Proliferation Treaty (NPT, 1968), whose aim^ is to avoid spreading of nuclear weapons, the International Atomic Energy Agency (IAEA) is in charge of controlTing the nuclear installations worldwide in the countries that established IAEA safeguard agreements with the IAEA (183 countries have signed the treaty, 50 operated nuclear facilities). IRAK, 1991: due to the discover of a clandestine nuclear enrichment program (not detected by regular inspections), the IAEA became aware of the limitations of the controls previously carried out, only on nuclear materials declared and accessible to the inspectors in declared facilities. ► Undeclared facilities => no control ► Undeclared nuclear material => not collected by inspectors Urgent need to reinforce legal protocols and analytical methods ^ Uranium isotopic composition of micrometric particles V The challenge is to detect and characterize nuclear materials (uranium, plutonium) present in a nuclear facility, without access to these materials Physico-chemical processes applied to a material, or even the displacement itself of a material release some very thin particles {< u.m to few u.m) that will spread in the nuclear facilities or around. Declared activities Undeclared zone Cotton cloth (swipe samples), used for collecting particles on surfaces (walls, tubings, etc.), and in the vicinity of the nuclear facilities. Uranium isotopic composition of micrometric particles ► Aim: determine the use (military, civilian, natural,....) and, if possible the associated treatment (irradiation, purification, enrichment process,...) and the origin of the nuclear material. ► Isotope ratios to discriminate between civilian and military use of the nuclear material : 2351 I 7238 U/238^ 234^238^ 236y/238U 240 Pu/239Pu (241Pu/239Pu, 242Pu/239Pu) U-235 0.72% Natural U U-235 4« Exemple of civilian U (< 5% 235U) Example of military U (> 20% 235U) U-235 60% U-234 0.0055% U-238 99.3% U-238 96% U-234 0.6% U-238 39% U-236 0.4% Uranium isotopic composition of micrometric particles ► The widespread method of "bulk analysis" (dissolution, chemical purification and isotopic analysis of the cotton cloth as a whole sample) provides an average isotopic composition of the particles and may allow detecting undeclared materials. ► However, this method is not efficient when the sample contain only one particle enriched in fissile 235U isotope (or a very limited number of these particles) among a vast majority of particles coming from declared materials (for instance natural uranium or uranium depleted in 235U). • * Declared activity A single particle holding evidence of an undeclared activity ► Towards analysing isotopic signature of individual particles... ^ Uranium isotopic composition of particles (CEA, DAM) 1 - Particles extraction and deposition on disks 3 - Isotopic analysis of U isotopes Particles are directly analyzed 2 - Localization and selection of the most pertinent particles "AW 30 0 kV E • mm ii >Ü - 4 ■ 4 Scanning Electron Microscopy Detection and localization of the "largest" particles (0 > 1 u m Fission tracks i detection and localization of the high 235U content-particles TIMS 41- Micro-sampling of the particles, deposition on a Re filament and introduction in the TIMS source LA-ICPMS ■a. Particles are directly analysed © Not selective for23Hj enriched-particies © Selective for 235U enriched-particles p) 4 analysis per hour ® 1 analyses per hour 4X Potential?? Jk Uranium isotopic composition of micrometric particles ►Particles coordinates known at ± 5-lOu.m ►Particles analysed : < 3 urn diameter 1 kHz 26 J/cm- M-f. V fipol Magii Pel WD Txp |-1 JO 0m 30 0 KV S 3 SE 130 l CnllQOOllr 2 kHz fsLA/ICPMS Crater of 40 \im diameter to make sure to ablate the particles of interest 24 J/cm 5 kHz 20 J/cm 5(1 V Spnl Mflgn Hot w 0 Fxp I-1 W lint II Acc V Spot Magn Dct WO Txp l.fi 1 Ci il^OOII? N^H 3fl 0 fcV 5 3 75ft* SE 13 6 1 20 Mm \ Uranium isotopic composition of micrometric particles ► Particles heterogeneity => size, shape, volume, density,... 10 15 20 Time (s) 15 20 Time (s) ► Very short transient signal ► Large signal intensity 25 2 - S1.5 en 1 0,5 0 ..............gwJ 10 15 20 Time (s) 25 2 25 2.5 2 H 3-5 i O) i/5 1 H 0,5 0 10 'íime (i) 0 25 £1,5 <55 1 - 0,5 0 10 2,5 Si,5 en 1 0,5 - 0 15 20 Time (s) 10 15 20 Time (s) 3 25 2,5 2 -I 1 I 0,5 0 10 25 2,5 2 gl .5 "I 1 - 0,5 i 0 10 15 20 Time (s) i— 15 20 Time (s) 25 25 ► Particles of Natural Isotope ■ ■» composition 0 3 mnUO 238U: 150 pg 235U: 1 pg 234U: 8fg ^ Uranium isotopic composition of micrometric particles FsLA/ Quad ICPMS Natural Isotope composition 1.0E-02 r 9.0E-03 8.0E-03 .0E-03 6.0E-03 5.0E-03 50 ms 300 Hz 100 Hz I 20 ms 0 5 10 15 20 25 30 35 40 Typical within run precision (la) = 7% to 15% for 235U/238U ► No reliable information on 234U nor 236U [V Uranium isotopic composition of micrometric particles ► Sector Field => higher ion transmission, => higher sensitivity when using the IC 238U, 235U, 234u, 236u MC-ICPMS -Nu Plasma HR Uranium isotopic composition of micrometric particles ► Effect of the detector... 0,008 r 00 0,006 m rsi $0,005 Aw NIVtKSI III i fAV t r dlsI 0,007 t 0,004 h 0,003 238u => Faraday 235u => Ion counter 238u => Faraday 235u => Faraday 0 _ ^_______#, # • • Theoretica Value No internal mass bias correction No external mass bias correction SIGNAL RANGE (at peak height) (Transient signal peak width 10 s) Faraday 238U Faraday 235U Ion counter 234U cps 8V 0,058 V 27702 cps 0,25 V 0,002 V 803 cps ■........................■ 5 10 15 20 25 Uranium isotopic composition of micrometric particles Comparing the analytical techniques... Aw NIVtKSI III I PAV t r dlsI „¥SPl i AIXXH J 235U/238U 234U/238U fsLA/Q-ICPMS Average Internal precision Mean values (n=9) Bias from theoretical value 10% 0.00736 ±0,00027 (3,7%) 1,5% < 100 particles /day fsLA/MC-ICPMS Average Internal precision Mean values (n=16) Bias from theoretical value Fara day/Faraday 0,68% 0.007178 ± 0.000084 (1,2%) -1,04% Ion counter/Faraday 2.4% <100 0.0000514 ± 0.0000014(3.2%) part'c,es /day -7.2% FT-TIMS Average Internal precision Mean values (n=5) Bias from theoretical value 1.5% 0.00740 ± 0.00004 ( 5.2%) 2% <6 particles /day SIMS Average Internal precision Mean values (n=6) Bias from theoretical value 0.2% 2% < 24 0.007246 ± 0.000058 (0.8%) 0.0000540 ± 0.0000011 (2%) partlcles /day -0.1% -2.5% geanx proto lithe An artistic installation merging science and poetry.... 2 GEANT Mitftfe b onranosnwt eta cjeul Tos bs pcrerts nt des gecrts dn gecrts or bin tffaK Tm provito* gcarti bMiAidBttiQi*toi(ta bi tjurtft