Laser Ablation - Inductively Coupled Plasma - Mass...

Karel de Grote-Hogeschool

Katholieke Hogeschool Antwerpen Departement Industriële Wetenschappen en Technologie

Campus Hoboken

Laser Ablation - Inductively Coupled

Plasma - Mass Spectroscopy

(LA-ICP-MS)

door Bart Van den Broeck

Promotor: Mirja Piispanen, Oulun Yliopisto, Kemian laitos

Promotor: Andoni Michelena, Campus Hoboken

Proefschrift tot het

behalen van de graad van

Industrieel Ingenieur in de

Chemie optie Chemie

Hoboken, juni 2004

Laser Ablation – Inductively Coupled Plasma – Mass Spectrometry (LA-ICP-MS)

Bart Van den Broeck I

Table of content

Table of content ................................................................................ I Table of Figures ............................................................................... II

Preface ........................................................................................... V Literature Part................................................................................1

1. Introduction ................................................................................1 2. Laser Ablation .............................................................................2

2.1. Introduction ..........................................................................2 2.2. The general laser principal ......................................................3

2.2.1. Absorption and Spontaneous Emission ................................3 2.2.2. Stimulated Emission..........................................................4

2.2.3. Creating a Population Inversion..........................................6 2.3. Technical realisation ...............................................................8

2.3.1. Amplification of light .........................................................8

2.3.2. Energizing the amplifying medium ......................................8 2.3.3. Laser oscillator ...............................................................10

2.3.4. Summary ......................................................................12 2.4. Properties of Laser light ........................................................13

2.4.1. Coherence .....................................................................13 2.4.2. Monochromatic...............................................................13

2.4.3. Parallel light (collimated) .................................................13 2.4.4. Brightness .....................................................................14

2.5. Different types of lasers and their properties ...........................14 2.5.1. Nd:YAG laser .................................................................15

2.5.1.1. General introduction..................................................15 2.5.1.2. Q-switching ..............................................................16

2.5.1.3. Theory of Harmonic Generation ..................................16 2.5.1.4. Optical attenuator .....................................................17

3. Inductively Coupled Plasma Mass Spectrometry (ICP-MS)...............19

3.1. Introduction ........................................................................19 3.2. Vacuum system ...................................................................20

3.2.1. Mean free path...............................................................20 3.2.2. Vacuum pumps ..............................................................21

3.2.2.1. Rotary mechanical oil pumps ......................................21 3.2.2.2. Turbo Molecular pumps..............................................22

3.2.2.3. Other pumps ............................................................22 3.3. ICP.....................................................................................24

3.3.1. Instrumentation .............................................................24 3.3.2. Formation of ICP discharge ..............................................24

3.3.3. Ionization of the sample ..................................................25 3.4. Interface.............................................................................27

3.5. Ion focussing system............................................................30 3.5.1. Role of the ion optics ......................................................30

3.5.2. Dynamics of the ion flow .................................................31

3.5.3. Instrumentation .............................................................33 3.6. Collision and reaction cells.....................................................36

3.6.1. Spectroscopic interferences in ICP-MS...............................36 3.6.2. Previous methods of interference reduction........................36


Bart Van den Broeck II

3.6.3. Basic processes in collision and reaction cells .....................38

3.7. Mass analyzer......................................................................41

3.7.1. Quadrupole mass filter ....................................................41 3.7.1.1. Basic mechanism ......................................................41

3.7.2. Quadrupole performance criteria ......................................44 3.7.2.1. Resolution................................................................44

3.7.2.2. Abundance sensitivity ................................................48 3.8. Detector..............................................................................50

3.8.1. Instrumentation .............................................................50 3.8.1.1. Discrete dynode electron multiplier .............................50

3.8.2. Extending the dynamic range ...........................................51 3.8.2.1. Filtering the ion beam................................................51

3.8.2.2. Using two detectors...................................................52 3.8.2.3. Using two scans with one detector. .............................52

3.8.2.4. Using one scan with one detector................................53 Experimental part ........................................................................56

1. Introduction ..............................................................................56

2. Sample preparation....................................................................57 2.1. LA-ICP-MS samples ..............................................................57

2.2. Fusion samples ....................................................................57 2.2.1. Na2CO3 and Na202 as flux.................................................57

2.2.2. Li2B4O7 as flux................................................................58 3. Configuration of LA-ICP-MS.........................................................59

3.1. Laser parameters .................................................................59 3.2. ICP-MS parameters ..............................................................60

4. Measurements with LA-ICP-MS....................................................61 4.1. Detection of 34S without CCT or internal standard.....................61

4.2. Measurements using CCT and internal standard .......................64 4.3. Comparison different bulk analysis .........................................69

5. Conclusions...............................................................................71 References .....................................................................................72

Table of Figures Fig. 1 Schematic of a laser ablation system [8] .....................................3

Fig. 2 Ground state [5] ......................................................................4 Fig. 3 Absorption [5]..........................................................................4

Fig. 4 Excited state [5] ......................................................................4 Fig. 5 Emission [5] ............................................................................4

Fig. 6 Ground state [5] ......................................................................5 Fig. 7 Absorption [5]..........................................................................5

Fig. 8 Excited state [5] ......................................................................5 Fig. 9 Stimulated emission [5] ............................................................5

Fig. 10 Populaton inversion [5] ...........................................................6

Fig. 11 Population inversion [5]...........................................................7 Fig. 12 Spontaneous emission [5] .......................................................7

Fig. 13 Stimulated emission and spontaneous emission [5] ....................7 Fig. 14 Cascade process [5]................................................................7


Bart Van den Broeck III

Fig. 15 Four level system [36] ............................................................7

Fig. 16 Amplification of light [5] ..........................................................8

Fig. 17 Energy input by pumping [5] ...................................................9 Fig. 18 Pumping with flashtube [5] ......................................................9

Fig. 19 Gaseous amplifying medium [5] .............................................10 Fig. 20 Laser oscillation [5] ..............................................................10

Fig. 21 Summary laser action [37] ....................................................12 Fig. 22 Units [25] ............................................................................14

Fig. 23 Different lasers and their properties [37] .................................15 Fig. 24 Half wave plate [23] .............................................................17

Fig. 25 Mass spectrometer [31].........................................................19 Fig. 26 Mean free path distance as a function of the system pressure [14]

...............................................................................................20 Fig. 27 Plasma torch and RF coil relative to the MS interface [30]..........24

Fig. 28 Formation of inductively coupled plasma [30] ..........................25 Fig. 29 Different temperature zones in the plasma [30] .......................26

Fig. 30 Mechanism of conversion of a droplet to a positive ion in the ICP

[30] .........................................................................................26 Fig. 31 Detailed view of the interface region [31] ................................27

Fig. 32 Pinch effect or secondary discharge [31] .................................28 Fig. 33 Electrons diffuse [33] ............................................................32

Fig. 34 Higher mass-to-charge ratio will dominate the centre or the ion beam [33] ................................................................................32

Fig. 35 A commercially available multi-component lens system [33] ......33 Fig. 36 Cylindrical ion lens, combined with a grounded stop [33]...........34

Fig. 37 The optimum lens voltage is placed on every mass in a multi-element run [33] .......................................................................34

Fig. 38 Schematic diagram of ICP-MS with reaction-collision cell [32] ....38 Fig. 39 Types of ion-molecule reaction in a collision reaction cell [18] ....39

Fig. 40 Principle of quadrupole mass filter [34] ...................................42 Fig. 41 Data acquisition system [34]..................................................43

Fig. 42 Mathieu stability diagram [34]................................................45

Fig. 43 Intensity loss due to high resolution [34].................................45 Fig. 44 Sensitivity comparison of two copper isotopes, 63Cu and 65Cu, at

resolution settings of 0,70 and 0,50 amu. [34]..............................46 Fig. 45 Plot of the "exact mass" divided by the "integer mass" vs. the

"integer mass" [14] ...................................................................47 Fig. 46 Spectral peak drop off [34] ....................................................48

Fig. 47 Example of the importance of abundance sensitivity [34] ..........49 Fig. 48 Discrete dynode electron multiplier [35] ..................................51

Fig. 49 Dual-stage discrete dynode detector [35] ................................53 Fig. 50 Cross calibration [35]............................................................54

Fig. 51 Normalized cross calibration [35]............................................54 Fig. 52 NIST-2691 and NIST-1633b...................................................56

Fig. 53 Laser parameters for blank measurements ..............................60 Fig. 54 ICP-MS configuration for detection of 34S without CCT or internal

standard...................................................................................61

Fig. 55 Acquisition parameters for 34S................................................61 Fig. 56 Results 34S in weight % .........................................................62


Bart Van den Broeck IV

Fig. 57 Calibration line ajo 43 std A 34S..............................................63

Fig. 58 ICP-MS configuration using CCT .............................................64

Fig. 59 Acquisition parameters using CCT ...........................................64 Fig. 60 Detection of 44Ca and 56Fe with 40Ca as internal standard...........65

Fig. 61 Detection of 56Fe with 40Ca as internal standard........................65 Fig. 62 Detection of 56Fe with 40Ca as internal standard........................66

Fig. 63 Detection of 24Mg, 27Al, 48Ti and 56Fe with 40Ca as internal standard...............................................................................................66

Fig. 64 Detection of 28Si, 31P, 34S and 44Ca with 40Ca as internal standard...............................................................................................66

Fig. 65 Calibration line ajo 19 std A 24Mg............................................67 Fig. 66 Calibration line CCTajo 6 std A 24Mg........................................67

Fig. 67 Calibration line CCTajo 7 std A 34S ..........................................68 Fig. 68 Comparison several bulk analysis on std A detecting Mg, S, P and

Ti.............................................................................................69 Fig. 69 Comparison several bulk analysis on std A detecting Al, Si, Ca and

Fe............................................................................................70


Bart Van den Broeck V

Preface

Since the University of Oulu bought last year a LA-ICP-MS to expand their

ways of research concerning corrosion and agglomeration problems in fluidized bed boilers. The objective of this thesis was to give a review

about LA-ICP-MS, what are its capabilities, how to calibrate, what are the start up procedures and what kind of measurements can we make? The

thesis will discus every step of the LA-ICP-MS; ablation, ionisation and detection applied on the LA-ICP-MS of the University of Oulu. After the

literature part we try to improve our measurement strategies with the LA-

ICP-MS, in order to get better results out of it.

I would like to thank the University of Oulu for giving me the opportunity to write my thesis here. While writing my thesis I had a lot of help from

Risto Laitinen, Minna Tiainen, Harri Kola, Heikki Ollila, Mirja Piispanen and Susanna Arvilommi, I would like to thank them for that. Thanks go also

to Andoni Michelena and Bert De Smet for reading my thesis through. Finally I would like to thank my parents for their financial and mental

support.


Bart Van den Broeck 1

Literature Part

1. Introduction

Since many ‘real samples’ are solid and often require cumbersome digestion procedures to prepare them for analysis by standard ICP-based

procedures, the development of techniques to introduce solid samples into

analytical instruments has been a long-term goal of analytical methodology research. Indeed, solid sample introduction is part of the

dream of analytical and process chemists for a procedure in which solid samples enter a one end of a black box and precise and accurate numbers

come out of the other end. Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS) is a technique which comes pretty close

to this dream. Laser ablation uses a focused, pulsed laser. A significant fraction of the ablated material is carried into a continuous flow of argon

to the ICP where it is converted to ions for quantification in a mass spectrometer. This technique enjoys the benefits of solid sampling

capability and the extraordinary detection power of modern ICP-MS instrumentation. Another big advantage of the LA-ICP-MS is the ability to

focus and direct a laser precisely, thereby creating a microprobe technique with the capability of spatially resolved sampling [8].

The LA-ICP-MS will be discussed in two main chapters:

• Laser Ablation

• Inductively Coupled Plasma Mass Spectrometry



2. Laser Ablation

2.1. Introduction The word ‘ablate’ has a Latin derivation and means to carry away, with no

implications about the mechanism by which the material is transported. While there is an extensive literature on the nature of ablation event, the

processes involved are still poorly understood. The word ablation thus describes the act of sampling by a laser.

The LASER (Light Amplification by Stimulated Emission of Radiation) is the

realisation of the magic ‘ray gun’ of science fiction. It was developed in the 1960’s and its potential sampling capabilities rapidly attracted the

attention of the analytical community, which whished to analyse solid samples without a prior digestion. Initially, the laser was used not only as

the sampling source, but also as a source of excitation and ionization

using the laser-induced plasma for optical emission, atomic absorption, and mass spectrometric analysis [19]. However, as the realisation

dawned that a laser ablated aerosol could be transported with surprising efficiency to the superior excitation and ionizing tool of the ICP, laser

sampling-ICP techniques began to evolve. The combination of laser ablation with an ICP is a powerful entity which produces what is often

referred to as either a LA-ICP (Laser Ablation – Inductively Coupled Plasma) or a LAM(P)-ICP (Laser Ablation Microprobe – Inductively Coupled

Plasma) depending upon the goal of the analyst, either bulk sampling or microanalysis [8].

The first paper using laser ablation as a sample introduction technique for

ICP-MS was by Alan Gray in 1985 [6] in which limits of detection of less than 1 µg/g were demonstrated using relatively high laser powers tot

generate large pits. It demonstrated the potential of laser ablation to be a

powerful tool in the analytical sciences. While Gray’s study produced large pits of 0,5-0,7 mm in diameter, earlier reports of laser ablation had

already demonstrated the ability of lasers to make pits that approached the diffraction limit of focus, which is typically a few micrometers. Since

Gray’s work, there has been a developing trend towards small spot size, micro-analytical developments and applications of laser ablation ICP-MS,

and exciting data from very small pits have been demonstrated in a variety of sample materials, including geological minerals. [8, 24, 15, 10,

9]

The hardware of a laser ablation microprobe is relatively simple (Fig. 1). In its simplest form, a pulsed laser beam of suitable controlled energy is

detected and focused onto a sample located in a gas tight cell. At the focal point, a plasma is produced and a sample is removed, a portion of

which is transported in a continuous flow of a suitable carrier gas, typically

argon, to the ICP. Some form of visual imaging, usually a TV Camera and Monitor, is employed for location sampling sites and monitoring ablation.



The use of TV technology avoids potential eye damage if direct

observation was used. [8]

Fig. 1 Schematic of a laser ablation system [8]

2.2. The general laser principal

The principle of lasers, the Light Amplification by Stimulated Emission of Radiation, is best explained when looking at the process of photon

emission. It is important to know how light can interact with individual atoms within an amplifying medium (‘atoms’ will be used to include

molecules and ions). Atoms consist of a positively charged core (nucleus) which is surrounded by negatively charged electrons. According to the

quantum mechanical description of an atom, the energy of an atomic electron can have only certain values and these are represented by energy

levels. The electrons can be thought of as orbiting the nucleus, those with the largest energy orbiting at greater distances from the nuclear core.

There are many energy levels that an electron within an atom can occupy, but here we will consider only two. Also, we will consider only the

electrons in the outer orbits of the atom as these can most easily be

raised to higher unfilled energy states. [5]

2.2.1. Absorption and Spontaneous Emission A photon of light is absorbed by an atom in which one of the outer

electrons is initially in a low energy state denoted by 0.



Fig. 2 Ground state [5]

Fig. 3 Absorption [5]

Fig. 4 Excited state [5]

Fig. 5 Emission [5]

The energy of the atom is raised to the upper energy level, 1, (Fig. 3) and

remains in this excited state for a period of time that is typically less than 10-6 second (Fig. 4). It then spontaneously returns to the lower state, 0,

with the emission of a photon of light (Fig. 5). Absorption is referred to as a resonant process because the energy of the absorbed photon must be

equal to the difference in energy between the levels 0 and 1. This means that only photons of a particular frequency (or wavelength) will be

absorbed. Similarly, the photon emitted will have energy equal to the difference in energy between the two energy levels. The emitted photon

all have the same energy but have the following less properties:

• Photons go to random directions • There is no relation with time, photons are emitted at random points

in time, there is no phase coherence

• Most of the time they are not monochromatic because several transitions are possible

These common processes of absorption and spontaneous emission cannot

give rise to the amplification of light. The best that can be achieved is that for every photon absorbed, another is emitted. [5, 36, 37]

2.2.2. Stimulated Emission Above it was stated that an atom in a high energy, or excited, state can

return to the lower state spontaneously. However, if a photon of light interacts with the excited atom, it can stimulate a return to the lower

state.



Fig. 6 Ground state [5]

Fig. 7 Absorption [5]

Fig. 8 Excited state [5]

Fig. 9 Stimulated emission [5]

One photon interacting with an excited atom results in two photons being

emitted (Fig. 9). Furthermore, the two emitted photons are said to be in phase, i.e. thinking of them as waves, the crest of the wave associated

with one photon occurs at the same time as on the wave associated with

the other. This feature ensures that there is a fixed phase relationship between light radiated from different atoms in the amplifying medium and

results in the laser beam produced having the property of coherence. Thus the photon released under stimulated emission has the following

good properties:

• The photon has the same frequency as the incoming photon • The photon is in the same phase, it is coherent with the incoming

photon • The photon goes in the same direction as the incoming photon

Stimulated emission is the process that can give rise to the amplification

of light. As with absorption, it is a resonant process; the energy of the incoming photon of light must match the difference in energy between the

two energy levels. Furthermore, if we consider a photon of light

interacting with a single atom, stimulated emission is just as likely as absorption; which process occurs depends upon whether the atom is

initially in the lower or the upper energy level. However, under most conditions, stimulated emission does not occur to a significant extent.

The reason is that, under most conditions, that is, under conditions of thermal equilibrium, there will be far more atoms in the lower energy

level, 0, than in the upper level, 1, so that absorption will be much more common than stimulated emission. If stimulated emission is to

predominate, we must have more atoms in the higher energy state than in the lower one. This unusual condition is referred to as a population



inversion and it is necessary to create a population inversion for laser

action to occur. [5, 36, 37]

2.2.3. Creating a Population Inversion

Finding substances in which a population inversion can be set up is central to the development of new kinds of laser. The first material used was

synthetic ruby. Ruby is crystalline alumina (Al2O3) in which a small

fraction of the Al3+ ions have been replaced by chromium ions, Cr3+. It is the chromium ions that give rise to the characteristic pink or red colour of

ruby and it is in these ions that a population inversion is set up in a ruby laser.

Fig. 10 Populaton inversion [5]

In a ruby laser, a rod of ruby is irradiated with the intense flash of light from xenon-filled flashtubes. Light in the green and blue regions of the

spectrum is absorbed by chromium ions, raising the energy of electrons of the ions from the ground state level to the broad F bands of levels.

Electrons in the F bands rapidly undergo non-radiative transitions to the two meta-stable E levels. A non-radiative transition does not result in the

emission of light; the energy released in the transition is dissipated as

heat in the ruby crystal. The meta-stable levels are unusual in that they have a relatively long lifetime of about 4 milliseconds (4 x 10-3 s), the

major decay process being a transition from the lower level to the ground state. This long lifetime allows a high proportion (more than a half) of the

chromium ions to build up in the meta-stable levels so that a population inversion is set up between these levels and the ground state level. This

population inversion is the condition required for stimulated emission to overcome absorption and so give rise to the amplification of light (Fig.

10). In an assembly of chromium ions in which a population inversion has been set up, some will decay spontaneously to the ground state level

emitting red light of wavelength 694,3 nm in the process. This light can then interact with other chromium ions that are in the meta-stable levels

causing them to emit light of the same wavelength by stimulated



emission. As each stimulating photon leads to the emission of two

photons, the intensity of the light emitted will build up quickly. This

cascade process in which photons emitted from excited chromium ions cause stimulated emission from other excited ions is indicated below:

Fig. 11 Population inversion [5]

Fig. 12 Spontaneous emission [5]

Fig. 13 Stimulated emission and

spontaneous emission [5]

Fig. 14 Cascade process [5]

The ruby laser is often referred to as an example of a three-level system. More than three energy levels are actually involved but they can be put

into three categories. These are: the lower level form which pumping takes place; the F levels into which the chromium ions are pumped; and

the meta-stable levels from which stimulated emission occurs. Other types of laser operate on a four level system. (Fig. 15)

Fig. 15 Four level system [36]

Usually the fourth level is then situated between the ground state and the

meta-stable level. If a photon in this situation is emitted, the electron drops from the meta-stable level to the extra fourth level, and from there



it drops back in a non-radiative manner to the ground state. The

advantage is that unwanted absorptions from the fourth level to the meta-

stable level is avoided because there are almost no electrons on the fourth level since they all go back to the ground state in a fast non-radiative

process, the result is that more stimulated emission can occur. In general, the mechanism of amplification differs for different lasing

materials. However, in all cases, it is necessary to set up a population inversion so that stimulated emission occurs more often than absorption.

[5, 36, 37]

2.3. Technical realisation

2.3.1. Amplification of light All lasers contain an energized substance that can increase the intensity of

light passing through it. This substance is called the amplifying medium or, sometimes, the gain medium, and it can be solid, liquid or gas.

Whatever its physical form, the amplifying medium must contain atoms, molecules or ions, a high proportion of which can store energy that is

subsequently released as light (see 2.2).

Fig. 16 Amplification of light [5]

In a neodymium YAG (Nd:YAG) laser, the amplifying medium is a rod of yttrium aluminium garnet single crystal (YAG) containing ions of the

lanthanide metal neodymium (Nd). In a dye laser, it is a solution of a fluorescent dye in a solvent such as methanol. In a helium-neon laser, it

is a mixture of the gases helium and neon. And in a laser diode, it is a thin layer of semiconductor material sandwiched between other

semiconductor layers. The factor by which the intensity of the light is increased by the amplifying medium is known as the gain. The gain is not

a constant for a particular type of medium. Its magnitude depends upon the wavelength of the incoming light, the intensity of the incoming light,

the length of the amplifying medium and also upon the extent to which the amplifying medium has been energized. [5, 37]

2.3.2. Energizing the amplifying medium Increasing the intensity of a light beam that passes through an amplifying

medium amounts to putting additional energy into the beam. This energy



comes from the amplifying medium which must in turn have energy fed

into it in some way. In laser terminology, the process of energizing the

amplifying medium is known as "pumping" (Fig. 17).

Fig. 17 Energy input by pumping [5]

There are several ways of pumping an amplifying medium. When it is a

solid, pumping is usually achieved by irradiating it with intense light. This

light is absorbed by atoms or ions within the medium raising them into higher energy states. Xenon-filled flashtubes positioned as shown below

are used as a simple source of pumping light (Fig. 18). Passing a high voltage electric discharge through the flashtubes causes them to emit an

intense flash of white light, some of which is absorbed by the amplifying medium. The assembly of flashtubes is enclosed within a polished metal

reflector (not shown in the diagram below) to concentrate as much light as possible on the amplifying medium. A laser that is pumped in this way

will have a pulsed output.

Fig. 18 Pumping with flashtube [5]

Pumping an amplifying medium by irradiating it with intense light is referred to as optical pumping. The source of pumping light can be

another laser. Some types of laser that were originally pumped using xenon-filled flashtubes are now pumped by laser diodes.

Gaseous amplifying media have to be contained in some form of enclosure

or tube and are often pumped by passing an electric discharge through the medium itself (Fig. 19). The mechanism by which this elevates atoms

or molecules in the gas to higher energy states depends upon the gas that

is being excited and is often complex. In many gas lasers, the end windows of the laser tube are inclined at an angle and they are referred to

as ‘brewster windows’. Brewster windows are able to transmit a beam



that is polarized in the plane of the diagram without losses due to

reflection. Such a laser would have an output beam that is polarized. [3]

Fig. 19 Gaseous amplifying medium [5]

The diagram illustrates pumping by passing a discharge longitudinally

through the gaseous amplifying medium but, in some cases, the discharge takes place transversely from one side of the medium to the other. Many

lasers that are pumped by an electric discharge can produce either a pulsed output or a continuous output depending upon whether the

discharge is pulsed or continuous.

Various other methods of pumping the amplifying medium in a laser are used. For example, laser diodes are pumped by passing an electric current

across the junction where the two types of semiconductors within the

diode come together. [5, 37]

2.3.3. Laser oscillator Pumped amplifying media as described in 2.3.2 could be used to increase

the intensity of light at particular wavelengths and such amplifiers are

often incorporated into laser systems. However, except in a few exceptional cases, light amplifiers would not be regarded as lasers. A

laser consists of a pumped amplifying medium positioned between two mirrors as indicated below (Fig. 20).

Fig. 20 Laser oscillation [5]

The purpose of the mirrors is to provide what is described as 'positive feedback'. This means simply that some of the light that emerges from

the amplifying medium is reflected back into it for further amplification.



Laser mirrors usually do not reflect all wavelengths or colours of light

equally well - their reflectivity is matched to the wavelength or colour at

which the laser operates. In appearance, they do not look like ordinary mirrors and are transparent at some wavelengths. An amplifier with

positive feedback is known as an oscillator.

The space between the two mirrors is known as the laser cavity. The beam within the cavity undergoes multiple reflections between the mirrors

and is amplified each time it passes through the amplifying medium. One of the mirrors reflects almost all of the light that falls upon it (total

reflector in Fig. 20). The other mirror reflects between 20% and 98% of the incident light depending upon the type of laser, the light that is not

reflected is being transmitted through the mirror. This transmitted portion constitutes the output beam of the laser.

The laser cavity has several important functions:

• Increasing the light intensity by multiple passes through the amplifying medium

• Ensuring that divergence of the beam is small • Improving the spectral purity of the beam

Following pumping, spontaneous emission of light initiates the emission of

low intensity light into the laser cavity. This light is increased in intensity by multiple passes through the amplifying medium so that it rapidly builds

up into an intense beam. When the number of photons produced by stimulated emission exceeds the number of photons produced randomly:

the system gain exceeds the losses and laser action can proceed. In the absence of cavity mirrors, this self-starting process, or oscillation, would

not occur.

The cavity ensures that the divergence of the beam is small. Only light

that travels in a direction closely parallel to the axis of the cavity can undergo multiple reflections at the mirrors and make multiple passes

through the amplifying medium. More divergent rays execute a ‘zig-zag path’ within the cavity and wander out of it.

The laser cavity also improves the spectral purity of the laser beam thanks

to the resonance effect in the cavity. The amplifying medium will amplify light within a narrow range of wavelengths. Especially in gas lasers atoms

move all the time so when they emit a photon, they become a moving source of waves. For the observing environment the frequencies are

shifted (= Doppler effect). Since all the atoms are moving in different directions with different velocities, the result is an observed frequency

band around the emitted frequency. However, within this narrow range, only photons which have the resonance frequency can undergo repeated

reflection up and down the cavity and have there amplitude increased

(resonance). The characteristics that a light beam within the cavity must possess in order to undergo repeated reflections define what is referred to



as a cavity mode, only those modes corresponding to multiples of half a

wavelength can be supported. Light which may still be amplified by the

amplifying medium but which does not belong to one of these special modes of oscillation is rapidly attenuated and will not be present in the

output beam. This behaviour is similar to that of a vibrating guitar string in that a particular string will only vibrate at certain frequencies. In a

similar way, an optical cavity will only sustain repeated reflections for particular well-defined wavelengths of light. [5, 36, 37]

2.3.4. Summary

Fig. 21 Summary laser action [37]

The basic requirements of any laser are similar, they all comprise: [37]

• An active medium with a suitable set of energy levels to support laser action

• A source of pumping energy in order to establish a population inversion

• An optical cavity to introduce optical feedback and so maintain the gain of the system above all losses



We should keep in mind that optimum laser design is more complex than

outlined here in this paper. For example in all our analyses so far we have assumed an optical cavity bounded by parallel reflectors, in practice many

different mirror arrangements involving combinations of curved and plane are adopted. However for LA-ICP-MS analysis it is enough to have a

feeling about what is involved.

2.4. Properties of Laser light

2.4.1. Coherence If we compare the phase of photons in a beam and we calculate the time

between two photons which are not in phase anymore, then this time τ is

a measure for the coherence of the beam. Usually τ is recalculated to the

coherence length (Lc = c . τ , with ‘c’ the velocity of light). The expression

‘laser light is coherent’ only means that this light is much more coherent

then visible light. [36]

• Normal visible light Lc = 1 µm

• Mercury lamp, green line Lc = 30 cm

• He-Ne-laser Lc = 30 cm • Mono-mode He-Ne Lc = 3 m

• Largest known coherence length Lc = 1000 km

2.4.2. Monochromatic

The Doppler-shift shows that a beam is not perfectly monochromatic. The mirror system can select one or a few frequencies, yet the result still has

a certain bandwidth. For the Hg-lamp the bandwidth is roughly 10-3 nm or 10ç Hz, which is comparable with the Doppler widening in the Ne-He-

laser. The laser suffers, after selection by the mirrors, of fluctuations by

Doppler-effect and by mechanical and thermal deforming of the tube. [36]

2.4.3. Parallel light (collimated) The light from a typical laser emerges in an extremely thin beam with

very little divergence. Another way of saying this is that the beam is

highly ‘collimated’. An ordinary laboratory helium-neon laser can be swept around the room and the red spot on the back wall seems about the same

size at that on a nearby wall. [21] For a round opening (= diameter, ‘d’ of the laser beam) applies that the

diffraction minimum has an opening-angle given by: λθ ⋅=⋅ 2,1sind



Typical values are: θ = 1 mrad, λ = 632,8 which results in d = 0,8 mm

after one meter. [36, 37]

2.4.4. Brightness

Because all the light is concentrated into a narrow spatial band, the light possesses high luminous intensity per unit area. The luminance is

expressed in 222

m

cd

msr

srcd

msr

lm=

⋅

⋅=

⋅

Quantity Unit

Luminous flux Lumen: srcdlm ⋅=

Solid angle Steradian: sr

Luminous intensity Candela: cd

Surface Square meter: ém

Fig. 22 Units [25]

A He-Ne-laser with power of mW1 has a brightness of 2

11102msr

lm

⋅⋅ . In a

pulsed regime it is possible to attain W910 , the brightness of this source is

2

2310msr

lm

⋅. [36]

2.5. Different types of lasers and their

properties

Many hundreds of different lasers are available now, but only a few types are in regular use in engineering. Fig. 23 gives a broad insight of the

operation and properties of such lasers. [37]

LASER & LED PARAMETERS

Type HeNe Argon Ruby Ruby Nd-YAG GaAlAs LED

Gas Gas

Free-

running

solid state

Q-

switched

solid state

Q-

switched

solid state

Semi-

conductor

Semi-

conductor

Power or

Energy 5 mW 1.5 W 1 J 50 mJ 250 mJ 10 mW 20 mW

Wavelength 632.8

nm

514.5

nm 694.3 nm 694.3 nm 1064 nm 820 nm 880 nm

Pulse

duration cw cw

350 µs

(FR)

30 ns

(QS)

10 ns

(QS) cw cw

Divergence

(full angle) 1 mrad 1 mrad 5 mrad 5 mrad 5 mrad 20° 40°

Linewidth 1.5

GHz 1 GHz 330 GHz 330 GHz 180 GHz 4 nm 50 nm



Spontaneous

lifetime 100 ns 3 ms 3 ms 550 µs 1 ns

Refractive

index 1 1 1.5 1.5 1.82 3.6

Beam diam 0.8 mm 1 mm 10 mm 5 mm 5 mm

active area

diam 200 µm 200 µm

Threshold

Current 80 mA

Forward

voltage drop 2 V 1.5 V

Max

Forward

current

200 mA 100 mA

Max power

dissipation 220 mW 200

Approx Cost £800 £25000 £35000 £15000 £35000 £200 £0.40

Fig. 23 Different lasers and their properties [37]

As stated above there are many of hundreds of lasers available now, it is beyond the purpose of this paper to consider all of them. However we will

discuss the Nd:YAG laser a bit further because that’s the laser that’s been used in our LA-ICP-MS analysis.

2.5.1. Nd:YAG laser

2.5.1.1. General introduction

The Nd:YAG laser uses traditional flash lamp pumping. Distilled water cools the lamp and gain medium, reducing thermal lensing. When

electrically pulsed, the lamp emits light that excites the laser gain medium, an Nd:YAG rod. The laser crystal is an yttrium aluminium garnet

single crystal, doped with neodymium ions (hence the term Nd:YAG)

Similar to a capacitor storing electrical energy, the Nd:YAG rod absorbs the flash lamp’s optical energy. Neodymium atoms that have been

excited to a higher electronic state (the lasing level) store this energy. These atoms remain excited for a fraction of a millisecond before

spontaneous emission starts.

In the absence of Q-switching, spontaneous emission (lasing, or light

amplification through stimulated emission) begins as soon as the cavity gain overcomes its losses. The duration of this spontaneous laser pulse is

almost as long as the driving lamp pulse. This non Q-switching pulse has high energy, but its peak power is low, because of its relatively long

width. The Q-switch improves performance by both increasing the



amount of energy stored in the rod and by preventing or delaying

spontaneous emission.

Whilst closed, a Q-switch in the laser cavity introduces an additional loss

and blocks spontaneous emission, allowing the number of excited atoms in the rod to build further. When instantaneously opened, it releases the

cavity’s stored energy in a shorter pulse with both higher average and peak power. [22]

2.5.1.2. Q-switching The idea of a Q-switched laser is that the resonator is prevented from

being effective until after the pumping pulse and most of the atoms are in the upper energy state (the population inversion in as complete as

possible). Its so-called Q is spoiled by disabling one of the mirrors. This can be accomplished mechanically by simply rotating the mirror or an

optical element like a prism between the mirror and the lasing medium, or electro-optically using something like a Pockel's cell in a similar location.

[4] In our LA-ICP-MS we used a Pockel's cell, this is a high speed electrically controlled optical shutter. In response to an electric field the

refraction index of the crystal changes, this change in index is linearly proportional to the electric field. [16] With the cavity not able to resonate

(mirror blocked, mirror at the wrong angle or wrong refraction index),

there can be no build up of stimulated radiation. There will still be the spontaneous emission but this is a small drain on the upper energy state.

At a point in time just after the pumping is complete (nearly total

population inversion), the Q is restored so that the resonator is once more intact – in our case a Marx bank (a number of capacitors which are

charged in parallel and discharged in series, this generates a high voltage impulse [17]) suddenly applies a high voltage to the crystal, this causes

the refraction index to change so that light can re-enter the laser rod. This light is now free to oscillate between the cavity and mirrors. During

these oscillations, the light increases in energy by extracting the energy stored in the gain medium. The resultant laser pulse is 3-5 nanoseconds

long, with high peak and total power. [22] Peak optical output power can be much greater than it would be without the Q-switch. Because of the

short pulse duration - measured in nanoseconds or picoseconds (or even

less), peak power of megawatts or gigawatts may be produced by even modest size lasers, this high peak power is generally considered to result

in better “coupling” with the sample. [4]

2.5.1.3. Theory of Harmonic Generation

The laser can produce laser light at other frequencies besides the natural, or fundamental frequency of its Nd:YAG gain medium. In some crystals, a

non-linear process known as harmonic generation produces additional



frequencies that are multiples (double, triple, quadruple etc) of the

fundamental. [22, 2]

The exact mechanisms behind harmonic generation are not considered

here, this would lead us to far from the subject. Sketched in broad outlines it comes down to two fundamental photons (1064 nm) combining

to give one first harmonic photon (532 nm). A second harmonic photon (355 nm) can be produced by a combination of a fundamental photon

(1064 nm) with a first harmonic photon (532 nm). For us it is important to know that besides from the laser’s fundamental (1064 nm), we can also

use its harmonics (532 nm, 355 nm, 266 nm, 213 nm). [22, 2] For some samples it is better to use a shorter wavelength, for example highly

transparent samples are better analysed with shorter wavelengths, this results in better ablation efficiency. [8] In our experiments we used

always a 213 nm beam.

2.5.1.4. Optical attenuator

The optical attenuator serves to control the laser energy without affecting the beam quality. The optical attenuator is designed to work on the 1064

nm beam, so it is placed directly after the IR head, before any harmonic generation crystals. The optical attenuator consists of a half wave plate

(sometimes also called retardation plates), followed by a polarizer.

The exact mechanisms of half wave plates are not considered here,

sketched in broad outlines the half wave plate can be used to rotate the polarization state of a plane polarized light as shown in Fig. 24.

Fig. 24 Half wave plate [23]

Suppose a plane-polarized wave is normally incident on a wave plate, and

the plane of polarization is at an angle θ with respect to the fast axis, as

shown. After passing through the plate, the original plane wave has been

rotated through an angle 2θ. A half wave plate is very handy in rotating

the plane of polarization from a polarized laser to any other desired plane

(especially if the laser device itself is too large to rotate). [23] The half



wave plate is secured to a motorized rotating mount. The servo motor

controlled angle is set by input from the control panel.

After the wave plate the laser beam enters a polarizer, a polarizer will

allow light polarized parallel to it to get through; light polarized perpendicular to the polarizer will be rejected to a beam dump. The

polarizer is permanently aligned to transmit vertically polarized light. So because the wave plate changes the angle of the polarized light which is

actually changing the magnitude of the laser beam component parallel to the polarizer, we have a device which controls the laser energy. [11]



3. Inductively Coupled Plasma Mass

Spectrometry (ICP-MS)

3.1. Introduction

In the previous chapter we discussed the sample introduction by laser

ablation (LA). In the next step the sample will be analyzed through an inductively coupled plasma mass spectrometer (ICP-MS). Mass

spectrometers come in various kinds and sizes, but also have a lot of similarities among them. It’s not the purpose of this thesis to discus

every mass spectrometer available on the market, however we will discuss the aspects which in most mass spectrometers are the same and which

are particularly used in our ICP-MS system.

We will discuss the ICP-MS in several stages from the ICP till the detector as shown in Fig. 25. We will discuss 7 stages: [14]

• Vacuum system

• ICP • Interface

• Ion optics • Collision and reaction cell

• Mass analyzer • Detector

Fig. 25 Mass spectrometer [31]



3.2. Vacuum system A vacuum is needed for the simple reason that ions will not get very far in

the presence of one atmosphere pressure of air. Without a mean path length of more than one metre, it is simply impossible to have much of a

mass spectrometer. It is not a trivial matter to obtain and maintain a suitable low pressure and to monitor this pressure. Usually the vacuum

system is the cause of the majority of the problems with instrument maintenance. [14]

3.2.1. Mean free path The fundamental parameter of importance in vacuum systems is the mean

free path which is the mean of average distance which an ion (or neutral) can move before it hits another particle. While clearly longer is better,

very low pressures are not cheaply obtained. In ICP systems it is especially difficult to obtain very low pressures as the “hole” in the

instrument is in the order of 1mm in diameter, through which a large volume of gas flows. Fig. 26 gives mean free path distances in metres as

a function of the system pressure expressed in units of Torr. These values are suitable for N2 or O2, which together compose 99% of air. (The

calculation is based on an ambient temperature of 25°C and a molecular

diameter of 370 pm)

Pressure (Torr)

Mean Free Path (m)

760 (1 atm) 10-7 1 5 . 10-5

0;05 10-3

10-5 5 10-6 50

10-7 500 10-8 5000

Fig. 26 Mean free path distance as a function of the system pressure [14]

At atmospheric pressure an ion does not go very far (0,1 µm) before a

collision occurs, clearly not a satisfactory condition. At a pressure of 10-5 Torr, a minimum operational pressure of a quadrupole ICP-MS

instruments, the mean free path is a reasonable 5 m in length. This is adequate for many studies since there is a low probability of an ion

colliding with another ion, molecule, or atom on its path through the instrument. On the other hand there are high precision applications

where even lower pressures and their accompanying longer mean free path are needed because even a small amount of scattering can result in

an ion being diverted. With relative errors of a few ppm, especially when measuring large isotope ratios, lower pressures are needed. Typical

quadrupole and TOF (time of flight) instruments are operated at a slightly



lower pressure, closer to 10-6 Torr. Pressures in “high precision” MS are

often 10-8 Torr and even lower. [14]

3.2.2. Vacuum pumps

The “pumps” remove the air from systems creating the vacuum. It is important to note that vacuum systems operate in equilibrium. The

ultimate pressure is an equilibrium between the pumps removing gas from

the system and gas coming into the system from “leaks” and from “out gassing” from system walls (especially water from metallic surfaces). All

systems leak and if the pump is closed off from the system pressure will rise. If the leak gets too high, then unsatisfactory operation of the

instrument can result, even though the pressure gauges may be indicating a satisfactory pressure. If the pressure rises linearly with time when the

pump is isolated from the system, this is an indication of a leak; otherwise out gassing is suggested. The pressure is only an indication that the

system leaks are “small enough” to allow good operation. Thus the concept of the “speed” of a pump is important, being not the lowest

ultimate pressure which the pump can obtain, but the speed of removal of gas from the system. The pump speed is often related to the pump size,

with larger pumps having a higher speed. In modern MS, the user will usually find only “rotary mechanical oil pump(s)”. Other pumps will be

briefly mentioned. [14]

3.2.2.1. Rotary mechanical oil pumps

Rotary mechanical oil pumps are the first line pumps, which are used to drop the pressure from atmospheric to less than one Torr. They are

ultimately capable of, or specified to have, ultimate pressures of 10-4 Torr,

a pressure which can realistically only be reached with a system of almost zero volume, as the pump speed of these pumps is not high. They are

based upon a simple mechanical system in which an eccentric rotor isolates a volume of the system gas, using oil to make gas tight seals

which then pushes this volume of gas out of the system to the ambient atmosphere. They are mechanical and noisy; the oil becomes

contaminated and for long pump life should be changed frequently (monthly). Older pumps had belt drives, with frequently belt failure, but

modern units have direct drive. Care is needed so that in case of a system failure that oil does not get “sucked” into the vacuum system,

although modern pumps contain much more robust protective devices than were available in older pumps. Systems often include automatic or

manual “Down To Air” valves which open to allow atmosphere to enter the pump during power outages. Rotary pumps are fundamental to system

operation as “high vacuum” pumps are never capable of operating

between low pressures and atmospheric, but do operate between low pressures and that is easily obtained using the rotary pump and hence the

common term “backing” pump in which the rotary “backs” the low



pressure pump. In all ICP-MS units the rotary pump directly pumps the

expansion region, in which a pressure of between 0,5 and 2 Torr is usual.

This same rotary pump or a separate rotary pump may be used to back the several turbo pumps. [14]

3.2.2.2. Turbo Molecular pumps

Turbo molecular pumps, usually simply called turbo pumps, are almost

universal in modern instrumentation. The technology has been around for some time, but the widespread acceptance of the pump has awaited the

“trivial” engineering problem of creating bearings which will run 24 hours a day at speeds of up to 100 000 RPM, which is really “screaming”. The

mode of failure is still the “simple” failure of the bearings resulting in the need for a replacement or rebuild unit. The pump is very clean as it

consists of 10 to 20 “fan” blades similar to the design of jet engine turbine blades. As residual gas molecules get close to a blade they are “hit” by

the blade and forced towards the next blade, etc. As mentioned the pump is “backed” by a mechanical pump, although in very simple situations the

pump is capable of operating without backing, but with higher ultimate pressures obtained. Progress has been made in the development of

bearings, with modern units often being magnetically suspended, and used without lubrication. The bearings are sometimes cooled by the

instrument water system. [14]

3.2.2.3. Other pumps

Other pumps are not commonly encountered in new instruments, but will be briefly mentioned for completeness and because they may be found on

older facilities.

Ion pumps are found in very low pressure systems; they are simple and

reliable. This device has two electrodes, both at ambient temperature. A “high” voltage of typically from one to three kilo volts is applied. When a

residual gas molecule gets close to the positive electrode an electron is removed to the electrode (field ionisation) and a positive gas phase ion

results. This positive ion is accelerated to the negative electrode, where it is “buried”. The opposite can occur in which negative ions form, but

positive ion formation is more probable. Magnetic fields are used to increase the path length so that multiplicative effects take place. These

devices are very clean and are found in the lowest pressure portion of MS systems. As a pressure measuring device, an error can be exhibited due

the significant ion currents which they create, however this current is a good indicator of system pressure. They are extremely reliable, as they

use only a moderate current, and a moderate high voltage supply.

Diffusion pumps were for decades the pump of choice. They are cheap to

make and can be “home made” either in metal of glass. The operation



principle is that oil or Hg is heated and the vapour passed through “jets”

at super sonic speeds, after which some kinetic energy is transferred from

the pumping fluid to the residual gas and forced to the other end of the pump to which is connected a rotary pump. The use of oil required

periodic change, however the most serious disadvantage of “diff” pumps was in the case of system failure when hot oil vapour was “sucked” into

the MS, which in some organic instruments made the system unusable. First generation ICP-MS, of which a few are still in use, were supplied with

“diff” pumps.

Cryo pumps which can be as simple as a cooled trap are not commonly used in ICP-MS, but still have a place in the specialised tool box of

vacuum systems. The first generation ICP-MS units from Sciex used a cryo pump which is simply a refrigeration system which used He as a

carrier gas and operated at 15 K, making a simple fast pump, however one which took hours to reach operational pressures and required a

weekly “defrosting” of the system. Various cooled absorbers can be

significantly enhanced by the use of molecular sieves, charcoal, etc. to enhance removal of gas phase molecules form the system. [14]



3.3. ICP

3.3.1. Instrumentation A plasma torch, a radio frequency (RF) coil, and RF power supply are the

basic components used to generate an inductively coupled plasma. Fig. 27

Fig. 27 Plasma torch and RF coil relative to the MS interface [30]

The plasma torch consists of three concentric tubes, which are usually made from quartz. In Fig. 27, these are shown as the outer tube, middle

tube, and sample injector. The torch can either be one-piece, with all three tubes connected, or it can be a demountable design in which the

tubes and the sample injector are separate. The gas (usually argon) used to form the plasma (plasma gas) is passed between the outer and middle

tubes at a flow rate of 12–17 l/min. A second gas flow, the auxiliary gas, passes between the middle tube and the sample injector at 1 l/min and is

used to change the position of the base of the plasma relative to the tube and the injector. A third gas flow, the nebulizer gas, also flowing at 1

l/min carries the sample, in the form of a fine-droplet aerosol, from the sample introduction system [29] and physically punches a channel

through the centre of the plasma. The sample injector is often made from

materials other than quartz, such as alumina, platinum, and sapphire, if highly corrosive materials need to be analyzed. It is worth mentioning

that although argon is the most suitable gas to use for all three flows, there are analytical benefits in using other gas mixtures, especially in the

nebulizer flow. The plasma torch is mounted horizontally and positioned centrally in the RF coil, approximately 10–20 mm from the interface. [30]

3.3.2. Formation of ICP discharge First, a tangential (spiral) flow of argon gas is directed between the outer

and middle tube of a quartz torch. A load coil, usually copper, surrounds the top end of the torch and is connected to a radio frequency generator.

When RF power (typically 750–1500 W, depending on the sample) is applied to the load coil, an alternating current oscillates within the coil at a

rate corresponding to the frequency of the generator. In most ICP generators this frequency is either 27 or 40 MHz. This RF oscillation of the

current in the coil causes an intense electromagnetic field to be created in the area at the top of the torch.



Fig. 28 Formation of inductively coupled plasma [30]

With argon gas flowing through the torch, a high-voltage spark is applied

to the gas, which causes some electrons to be stripped from their argon atoms. These electrons, which are caught up and accelerated in the

magnetic field, then collide with other argon atoms, stripping off still more

electrons. This collision induced ionization of the argon continues in a chain reaction, breaking down the gas into argon atoms, argon ions, and

electrons, forming what is known as an inductively coupled plasma discharge. The ICP discharge is then sustained within the torch and load

coil as RF energy is continually transferred to it through the inductive coupling process. The sample aerosol is then introduced into the plasma

through a third tube called the sample injector. This whole process is shown in Fig. 28. [30]

3.3.3. Ionization of the sample To better understand what happens to the sample on its journey through

the plasma source, it is important to understand the different heating zones within the discharge. Fig. 29 shows a cross-sectional

representation of the discharge along with the approximate temperatures for different regions of the plasma.



Fig. 29 Different temperature zones in the plasma [30]

As mentioned previously, the sample aerosol enters the injector via the spray chamber. When it exits the sample injector, it is moving at such a

velocity that it physically punches a hole through the centre of the plasma discharge. It then goes through a number of physical changes, starting at

the preheating zone and continuing through the radiation zone before it

eventually becomes a positively charged ion in the analytical zone. To explain this in a very simplistic way, let’s assume that the element exists

as a trace metal salt in solution. The first step that takes place is vaporization of the water in the droplet. With the water molecules

stripped away, it then becomes a very small solid particle. As the sample moves further into the plasma, the solid particle changes first into a

gaseous form and then into a ground state atom. The final process of conversion of an atom to an ion is achieved mainly by collisions of

energetic argon electrons (and to a lesser extent by argon ions) with the ground state atom. The ion then emerges from the plasma and is directed

into the interface of the mass spectrometer. [28] This process of conversion of droplets into ions is represented in Fig. 30. [30] Note that in

LA-ICP-MS we don’t have liquid droplets but already solid particles, so the spray chamber is not used here.

Fig. 30 Mechanism of conversion of a droplet to a positive ion in the ICP [30]

The next is probably the most crucial area of an ICP mass spectrometer — the interface region — where the ions generated in the atmospheric

plasma have to be sampled with consistency and electrical integrity by the mass spectrometer, which is under extremely high vacuum. [31]



3.4. Interface The ICP has long been recognised as a good source of ions. The problem

which awaited the mid 80’s was how to get ions efficiently, consistently, and with electrical integrity from the atmospheric pressure in the plasma

into a suitable low pressure vacuum system containing a mass spectrometer. The interface between the atmospheric pressure ICP and

the 10-6 Torr vacuum was solved by not trying to make the lowering of pressure of 8 orders of magnitude in one step. [14]

The interface consists of two metallic cones with very small orifices. After

the ions are generated in the plasma, they pass through the first cone, known as the sampler cone, which has an orifice diameter of 0,8 – 1,3

mm. The pressure drop across the sampler cone is from atmospheric pressure to between 0,5 – 2 Torr, the vacuum being maintained by a

mechanical rotary pump. The pressure is affected by the size (speed) of

the pump and the diameter of the orifice. Thus large volumes of gas are carried to the pump, which has both good and bad effects on the oil

contained in the pump. The high volume of gases tends to “clean” out the oil, thus the time needed between oil changes might be longer than

expected, however a monthly change contributes to the long live of the pump. The pressure in the region following the sampler (expansion

region) can affect the instrumental sensitivity. Gas flow through the sampler has been estimated at from 1 to 2 litres per minute

(approximately 100% of the sample carrier gas passes this orifice), and shortly after the sampler orifice, the gas reaches supersonic velocities,

giving the ions a forward velocity towards the analyzer portion of the instrument. [14, 31]

Fig. 31 Detailed view of the interface region [31]

From the sampler orifice the ions travel a short distance (~1 cm) to the

skimmer cone, which is generally sharper than the sampler cone and has a much smaller orifice diameter of 0,4 – 0,8 mm. This cone is called the

skimmer cone, because it “skims” the plasma, allowing through approximately 1% of the sample. The plasma at this point has expanded

significantly, and while the true temperature has dropped tremendously



(due to the expansion), the ions seems to have “frozen” in the more

rarefied atmosphere. The pressure behind the skimmer is typically around

10-4 Torr. The pressure in this region is, in all modern instruments, maintained by a turbo pump backed by a mechanical rotary pump.

Usually, but not always, a separate rotary oil pump is used to back the turbos and an other one is used to pump the expansion region. [14] [31]

Both cones are usually made of nickel, although aluminium, copper and

platina (for corrosive liquids) are also sometimes used. The requirement is that the metal has good thermal conductivity, so heat from the plasma

can be dissipated to a water cooled plate. Stainless steel for example, is not suitable due to its poor thermal conductivity. [14] [31]

Some distance beyond the skimmer another opening is placed. This

opening can be much larger than the sampler and skimmer orifices, as it separates the region operating at 10-4 Torr from a slightly lower pressure

region operating at 10-6 Torr or lower pressure. This is the analyzer and

detector region and is usually pumped by a second turbo pump, see [Fig. 25]. An additional vacuum region is possible in larger sector instruments,

especially important where very high precision isotope ratios are to be determined. [14]

This process sounds fairly straight-forward but proved very problematic

during the early development of ICP-MS because of an undesired electrostatic (capacitive) coupling between the load coil and the plasma

discharge, producing a potential difference of 100-200V. Although this potential is a physical characteristic of all inductively coupled plasma

discharges, it is particularly serious in an ICP mass spectrometer because the capacitive coupling creates an electrical discharge between the plasma

and sampler cone. This discharge, commonly called the pinch effect or secondary discharge shows itself as arcing in the region where the plasma

is in contact with the sampler cone. This process is shown very

simplistically in Fig. 32.

Fig. 32 Pinch effect or secondary discharge [31]

If not taken care of, this arcing can cause all kinds of problems (especially when using many different sample types, requiring different operation

parameters), including an increase in double charged interfering species, a



wide kinetic energy spread of sampled ions, formation of ions generated

from the sampler cone, and a decreased orifice lifetime. The problem was

first eliminated by grounding the induction coil at the centre, which had the effect of reducing the radio frequency (RF) potential to a few volts.

Originally, the grounding was implemented by attaching a physical grounding strap from the centre turn of the coil to the interface housing.

In today’s instrumentation the grounding is achieved in a number of different ways, depending on the design of the interface. Some of the

most popular designs include balancing the oscillator inside the circuitry of the RF generator; positioning a grounded shield or plate between the coil

and the plasma torch; or using two interlaced coils where the RF fields go in opposing directions. They all work differently but achieve a similar

result of reducing or eliminating the secondary discharge. [31]

A true test of the design of the interface occurs when plasma conditions need to be changed, when the sample matrix changes, or when a dry

sample aerosol is being introduced into the ICP-MS (for example when

using laser ablation or using cool plasma conditions). Analytical scenarios like these have the potential to induce a secondary discharge, change the

kinetic energy of the ions entering the mass spectrometer, and affect the tuning of the ion optics. It is therefore critical that the interface grounding

mechanism can handle these types of real-world applications. [31]



3.5. Ion focussing system

3.5.1. Role of the ion optics The ion focusing system is a crucial area of the ICP mass spectrometer

where the ion beam is focused before it enters the mass analyzer (or collision and reaction cell if used). Sometimes known as the ion optics, it

consists of one or more ion lens components, which electrostatically steer

the analyte ions from the interface region into the mass separation device (or collision and reaction cell if used). The strength of a well-designed ion

focusing system is its ability to produce low background levels, good detection limits, and stable signals in real-world sample matrices.

The ion optics are positioned between the skimmer cone and the mass

separation device, and consist of one or more electrostatic controlled lens components. They are not traditional optics that we associate with ICP

emission or atomic absorption but are made up of a series of metallic plates, barrels, or cylinders that have a voltage placed on them. The

function of the ion optic system is to take ions from the hostile environment of the plasma at atmospheric pressure via the interface

cones and steer them into the mass analyzer, which is under high vacuum. As mentioned before, the plasma discharge, interface region,

and ion optics have to be designed in concert with each other. It is

absolutely critical that the composition and electrical integrity of the ion beam is maintained as it enters the ion optics. For this reason it is

essential that the plasma is at zero potential to ensure that the magnitude and spread of ion energies are as low as possible.

A secondary but also very important role of the ion optic system is to stop

particulates, neutral species, and photons from getting through to the mass analyzer and the detector. These species cause signal instability

and contribute to background levels, which ultimately affect the performance of the system. For example, if photons or neutral species

reach the detector, they will elevate the background noise and therefore degrade detection capability. In addition, if particulates from the matrix

penetrate farther into the mass spectrometer region, they have the potential to deposit on lens components and, in extreme cases, get into

the mass analyzer. In the short term this will cause signal instability and,

in the long term, increase the frequency of cleaning and routine maintenance.

Basically two approaches will reduce the chances of these undesirable

species making it into the mass spectrometer. The first method is to place a grounded metal stop (disk) behind the skimmer cone. This stop

allows the ion beam to move around it but physically blocks the particulates, photons, and neutral species from travelling downstream.

The other approach is to set the ion lens or mass analyzer slightly off axis. The positively charged ions are then steered by the lens system into the

mass analyzer, while the photons and neutral and non-ionic species are



ejected out of the ion beam. It is also worth mentioning that some lens

systems incorporate an extraction lens after the skimmer cone to

electrostatically pull the ions from the interface region. This has the benefit of improving the transmission and detection limits of the low-mass

elements (which tend to be pushed out of the ion beam by the heavier elements), resulting in a more uniform response across the full mass

range of 0–300 amu. In an attempt to reduce these space charge effects, some older designs have used lens components to accelerate the ions

downstream. Unfortunately this can have the effect of degrading the resolving power and abundance sensitivity (ability to differentiate an

analyte peak from the wing of an interference) of the instrument because of the much higher kinetic energy of the accelerated ions as they enter

the mass analyzer. [33]

3.5.2. Dynamics of the ion flow

To fully understand the role of the ion optics in ICP-MS, it is important to have an appreciation of the dynamics of ion flow from the plasma through

the interface region into the mass spectrometer. When the ions generated in the plasma emerge from the skimmer cone, there is a rapid

expansion of the ion beam as the pressure is reduced from 760 Torr (atmospheric pressure) to approximately 10-3 to 10-4 Torr in the lens

chamber with a turbo molecular pump. The composition of the ion beam

immediately behind the cone is the same as the composition in front of the cone because the expansion at this stage is controlled by normal gas

dynamics and not by electrodynamics. One of the main reasons for this is that, in the ion sampling process, the Debye length (the distance over

which ions exert influence on each other) is small compared with the orifice diameter of the sampler or skimmer cone. Consequently there is

little electrical interaction between the ion beam and the cone and relatively little interaction between the individual ions in the beam. In this

way, compositional integrity of the ion beam is maintained throughout the interface region.



Fig. 33 Electrons diffuse [33]

With this rapid drop in pressure in the lens chamber, electrons diffuse out of the ion beam. Because of the small size of the electrons relative to the

positively charged ions, the electrons diffuse farther from the beam than the ions, resulting in an ion beam with a net positive charge. [Fig. 33]

Fig. 34 Higher mass-to-charge ratio will dominate the centre or the ion beam

[33]

The generation of a positively charged ion beam is the first stage in the charge separation process. Unfortunately, the net positive charge of the

ion beam means that there is now a natural tendency for the ions to repel each other. If nothing is done to compensate for this, ions with a higher

mass-to-charge ratio will dominate the centre of the ion beam and force the lighter ions to the outside. The degree of loss will depend on the

kinetic energy of the ions: those with high kinetic energy (high mass

elements) will be transmitted in preference to ions with medium (mid-mass elements) or low kinetic energy (low-mass elements). [Fig. 34]



The second stage of charge separation is therefore to electrostatically

steer the ions of interest back into the centre of the ion beam by placing voltages on one or more ion lens components. However, that this is

possible only if the interface is kept at zero potential, which ensures a neutral gas-dynamic flow through the interface and maintains the

compositional integrity of the ion beam. It also guarantees that the average ion energy and energy spread of each ion entering the lens

systems are at levels optimum for mass separation. If the interface region is not grounded correctly, stray capacitance will generate a

discharge between the plasma and sampler cone and increase the kinetic energy of the ion beam, making it very difficult to optimize the ion lens

system. [33]

3.5.3. Instrumentation

Over the years, there have been many different ion optic designs. Although they all have their own characteristics, they perform the same

basic function: they discriminate undesirable matrix- or solvent-based ions so that only the analyte ions are transmitted to the mass analyzer.

The most common ion optics design used today consists of several lens components, which all have a specific role to play in the transmission of

the analyte ions. With these multi-component lens systems, the voltage

can be optimized on every lens of the ion optics to achieve the desired ion specificity. Over the years this type of lens configuration has proven to be

very durable and has been shown to produce very low background levels, particularly when combined with an off-axis mass analyzer.

Fig. 35 A commercially available multi-component lens system [33]

However, because of the interactive nature of parameters that affect the signal response, the more complex the lens system the more variables

have to be optimized, particularly if many different sample types are being analyzed. This isn’t such a major problem because the lens voltages are

all computer controlled, and methods can be stored for every new sample



scenario. Fig. 35 is a commercially available multi-component lens

system, with an extraction lens and off-axis quadrupole mass analyzer,

showing how the ion beam is deflected into the mass analyzer, while the photons and neutral species travel in a straight line and strike a metal

plate.

Another, more novel approach is to use just one cylindrical ion lens, combined with a grounded stop positioned just inside the skimmer cone as

shown in Fig. 36. With this design, the voltage is dynamically ramped on-the-fly, in concert with the mass scan of the analyzer (typically a

quadrupole). The benefit of this approach is that the optimum lens voltage is placed on every mass in a multi-element run to allow the

maximum number of analyte ions through, while keeping the matrix ions to an absolute minimum. This is represented in Fig. 37, which shows a

lens voltage scan of six elements: lithium, cobalt, yttrium, indium, lead, and uranium, at 7, 59, 89, 115, 208, and 238 amu, respectively. We can

see that each element has its own optimum value, which is then used to

calibrate the system, so the lens can be ramp-scanned across the full mass range. This type of approach is typically used in conjunction with a

grounded stop to act as a physical barrier to reduce particulates, neutral species, and photons from reaching the mass analyzer and detector.

Fig. 36 Cylindrical ion lens, combined

with a grounded stop [33]

Fig. 37 The optimum lens voltage is

placed on every mass in a multi-

element run [33]

Although this design produces slightly higher background levels, it offers excellent long-term stability with real-world samples. It works well for

many sample types but is most effective when low mass elements are being determined in the presence of high-mass–matrix elements.

It is also worth emphasizing that a number of ICP-MS systems offer what

is known as a high-sensitivity interface. These all work slightly differently but share similar components. By using a combination of slightly different

cone geometry, higher vacuum at the interface, one or more extraction lenses, and slightly modified ion optic design, they offer as much as 10

times the sensitivity of a traditional interface. However, this increased

sensitivity is usually combined with inferior stability and an increase in



background levels, particularly for samples with a heavy matrix. To get

around this degradation in performance one must usually dilute the

samples before analysis, which limits the system’s applicability for real-world samples. However, they have found a use in non-liquid based

applications in which high sensitivity is crucial, for example in the analysis of small spots on the surface of a geological specimen using laser ablation

ICP-MS. For this application, the instrument must offer high sensitivity because a single laser pulse is used to ablate the sample and sweep a tiny

amount of the dry sample aerosol into the ICP-MS.

The role of the ion focusing system cannot be overestimated. It affects the background noise level of the instrument. It has a huge impact on

both long- and short-term signal stability, especially in real-world samples, and it also dictates the number of ions that find their way to the

mass analyzer. However, it must be emphasized that the ion optics are only as good as the ions that feed it, and for this reason it must be

designed in concert with both the plasma source and the interface region.

There is no question that this area is crucial to the design of the whole ICP mass spectrometer. [33]



3.6. Collision and reaction cells

3.6.1. Spectroscopic interferences in ICP-MS ICP-MS is a mass spectrometric technique that separates ions on the basis

of their mass to charge ratio (m/z). The mass resolution of most spectrometers is limited, and for a quadrupole, the type of analyzer most

commonly used in laser-ablation applications, only unit mass resolution is

achievable. [14] This means that at a given mass/charge ratio (m/z), ions of more than one element or polyatomic species may be measured

simultaneously. Spectroscopic interferences can thus prevent the accurate and/or precise determination of an element in a sample.

Several types of spectroscopic interferences are possible including: [13]

• Isobaric overlap between tow elements (naturally occurring) have

isotopes of nominally the same mass • Polyatomic ions, which are complexes based around the

recombination of two or more atomic species that are abundant in the plasma, e.g. ArO+.

• Refractory metal oxide species, which form due to incomplete breakdown of the sample or recombination of ions in the plasma,

e.g. BaO+.

• Double charged ions that may or may not be significant depending upon the second ionisation potential of the element

Spectrometric interferences are one of the main limiting factors in ICP-MS

analysis, leading to poorer detection limits for many elements and degraded precision due to the limitations of a fluctuating background

signal. Furthermore, the production of some spectroscopic interferences may be affected by the bulk composition of the sample. Such

enhancement or suppression effects are part of the problem commonly referred to as “matrix effects”. They may be compensated for by carefully

matching the major element chemistry of the samples and calibration standards used in ICP-MS analysis but is not easy to do this in laser-

ablation work due to a lack of readily available, homogeneous reference materials. The elimination or control of interferences can thus be critical

for accurate analysis by laser ablation ICP-MS. [18]

3.6.2. Previous methods of interference reduction

Several methods were already in existence, prior to the development of collision and reaction cells, for the reduction or elimination of interferences

including:

• Optimisation of plasma properties

• Use of mixed gas plasmas • Use of mass spectrometers of high resolving power



Interferences can be minimised in all ICP-MS systems through a careful

optimisation of instrument parameters such as sampling depth in the plasma (i.e. position of the sampling cone orifice relative to the load coil),

sampling cone orifice diameter, and geometry of the sampling and skimmer cones and the use of the correct carrier gas flow rates.

However, although optimal plasma properties give minimum interference levels, it is usually not possible to completely eliminate the problem. [18]

Other methods for controlling spectroscopic interferences include reducing

the plasma RF forward power to produce a cool or cold plasma and adding additional supplementary gases to the torch to form a mixed gas plasma.

[20] These techniques are very useful for some specific applications in solution nebulisation ICP-MS analysis but both have limitations when

applied to multi-element analysis and have not yet been widely used in laser-ablation work. Cool or cold plasmas have a lower ionisation

‘temperature’, usually produced by reducing the RF forward power. This

results in a reduction in polyatomic ion density, but also reduces sensitivity for elements with relatively high first ionisation potential (>8

eV), a less efficient break down of refractory oxide species and an increase in non-spectroscopic interferences i.e. matrix effects. A reduced

RF forward power is sometimes used in laser-ablation ICP-MS because less energy is required to break down the solid particles form a modern

UV laser-ablation system than is required to volatise the solvent from a nebulised liquid droplet, for which RF power is typically optimised.

However, interferences still persist under these operating conditions and levels of refractory oxides may be increased. Mixed gas plasmas (with the

addition of O2, N2, or H2) are not widely used, as they tend to introduce different polyatomic interferences and can reduce sensitivity for some

elements, depending upon the gas added to the plasma. Helium is often used as a carrier gas in laser ablation to give improved transport

efficiency of particles from ablation site to plasma. [7] The presence of

Helium in the plasma can lead to small changes in ionisation condition resulting in lower background count rates for some polyatomic ions. [18,

32]

The most thorough and appropriate method for assessing and avoiding interferences is to use a high-resolution double-focussing magnetic sector

mass spectrometer. [12] However since we used in our experiments a quadrupole mass spectrometer we will not discuss this further. And even

if we would use a double-focussing magnetic sector mass spectrometer this wouldn’t solve everything because it’s not able to resolve any

naturally overlapping elemental isobars or many metal oxide and doubly charged interferences. Other disadvantages are reduced sensitivity and

detection limits, and it scans slower than the quadrupole. [18]

The conclusion is that there remains a need to find an analytical technique

to reduce polyatomic interferences and to resolve elemental isobaric overlap whilst maintaining instrument detection limits and stability.



Collision and reaction cells have major advantage over the use of high

resolution mass spectrometer for interferences reduction because they

have the potential to do this for many elements without loss in sensitivity. This is achieved by inducing reactions with either the interfering or the

analyte ions respectively; a process called chemical resolution. Collision and reaction cells can be combined with modern fast-scanning quadrupole

mass spectrometers or even time-of-flight mass spectrometers, ideal for transient signal analysis. [18]

3.6.3. Basic processes in collision and reaction cells

Both physical and chemical processes affect the composition of the ion beam that passes through the collision and reaction cell. The cell is

positioned behind the skimmer cone and ion extraction lens and in front of

the quadrupole or other mass analyzer. See Fig. 38

Fig. 38 Schematic diagram of ICP-MS with reaction-collision cell [32]

With collision and reaction cell technology, ions enter the interface in the

normal manner, where they are extracted under vacuum into a collision and reaction cell that is positioned before the analyzer quadrupole. A

collision and reaction gas such as hydrogen or helium is then bled into the cell, which consists of a multipole (a quadrupole, hexapole, or octapole),

usually operated in the radio frequency (RF)-only mode. A multipole can be classified on the basis of the number of pairs of rods through which an

RF field is introduces, thus a quadrupole is second order with four rods and a hexapole is third order and so on. [18] The RF-only field does not

separate the masses like a traditional quadrupole, but instead has the effect of focusing the ions, which then collide and react with molecules of

the collision and reaction gas. By a number of different ion-molecule collision and reaction mechanisms, polyatomic interfering ions like 40Ar, 40Ar16O, and 38ArH, will either be converted to harmless non-interfering species, or the analyte will be converted to another ion which is not

interfered with.



The collision cell technology was originally designed for organic MS to

generate daughter species to confirm identification of the structure of the

parent molecule [38], but in ICP-MS they are used to stop the appearance of many argon based spectral interferences. There is an important

distinction between the techniques as the processes that take place in the two types of cells are fundamentally different from one another. In an

organic mass spectrometers molecules are accelerated into the collision cell and are broken down through a physical process called collision

induced dissociation (CID) whereas in the collision and reaction cells described here this type of fragmentation is not thought to be a major

process. It is important to remember that the multipole device itself does not directly lead to the solution of interference problems; rather it

provides a stability field within which gas phase ion-molecule reactions can occur. [18]

Reactions in a collision and reaction cell often occur in complex series and

a prediction of reaction schemes becomes difficult. Different types of

reactions are possible: Reaction type Reaction form Examples

Neutralisation I+ + R => I+ + R Ar+ + NH3 => Ar + NH3+

Ar+ + H2 Ar + H2+

Association I+ + R => IR+ H2O+ + H => H3O

+

Rh+ + NH3 Rh(NH3)+

Condensation I+ + R => IR1+ + R2 Ar+ + H2 => ArH+ + H

Sr+ + O2 SrO+ + O

Fragmentation I+ + R => I1+ + I2 + R

I1+ + I2+ + R

I1 + I2 + R+

Key: I-ion (interfering ion, analyte ion or product of preceding reactions), R-gas molecule

Fig. 39 Types of ion-molecule reaction in a collision reaction cell [18]

The previous example is a very simplistic explanation of how a collision and reaction cell works. In practice, complex secondary reactions and

collisions take place, which generate many undesirable interfering species.

If these species were not eliminated or rejected, they could potentially lead to additional spectral interferences. Basically two approaches are

used to reject the products of these unwanted interactions:

• Discrimination by kinetic energy (RF-only multipoles) • Discrimination by mass. (dynamic reaction cell quadrupole)

The major differences between the two approaches are in the types of

multipoles used and their basic mechanism for rejection of the interferences. Since we used in our experiments an RF-only hexapole as

collision and reaction cell we will only discuss discrimination by kinetic energy. [32]

In a typical ICP-MS instrument, ions gain kinetic energy as they are

extracted through the expansion chamber of the interface. Some of this

kinetic energy is lost during collisions with a buffer gas (collision gas) in



the collision and reaction cell. New ions produced inside the collision and

reaction cell during reaction have less kinetic energy. Thus a potential

barrier can be applied at the downstream end of the multipole device to discriminate between the two types of ions and hence remove the

unwanted products of ion-molecule reactions. [18]

Discrimination by kinetic energy is typically achieved by setting the collision and reaction cell bias slightly less positive than the mass filter

bias, this means that the collision-product ions, which have the same energy as the cell bias, are discriminated against and rejected, while the

analyte ions, which have a higher energy than the cell bias, are transmitted.

The inability to adequately control the secondary reactions meant that low

reactive gases like He, H2, and Xe were the only option. The result was that ion-molecule collisional fragmentation (and not reactions) was

thought to be the dominant mechanism of interference reduction. So

even though the ion transmission characteristics of a hexapole were considered very good (with respect to the range of energies and masses

transmitted), background levels were still relatively high because the interference rejection process was not very efficient. For this reason, its

detection capability (particularly for some of the more difficult elements, like Fe, K and Ca) offered little improvement over the cool plasma

approach. Recent modifications to the hexapole design have significantly improved its collision and reaction characteristics. In addition to offering

good transmission characteristics and kinetic energy discrimination, they now appear to offer basic mass-dependant discrimination. This means

that the kinetic energy discrimination barrier can be adjusted with analytical mass, which offers the capability of using small amounts of

highly reactive gases like ammonia. [32]



3.7. Mass analyzer The mass separation device, sometimes called the mass analyzer, is the

region of the ICP mass spectrometer which separates the ions according to their mass-to-charge ratio (m/z). This selection process is achieved in

a number of different ways, depending on the mass separation device, but they all have one common goal: to separate the ions of interest from all

the other non-analyte, matrix, solvent, and argon-based ions.

As shown in Fig. 25, the mass analyzer is positioned between the ion optics (or collision and reaction cell if used) and the detector and is

maintained at a vacuum of approximately 10-6 Torr with a second turbo molecular pump. Assuming the ions are emerging from the ion optics (or

CCT) at the optimum kinetic energy, they are ready to be separated according to their mass-to-charge ratio by the mass analyzer. There are

basically four kinds of commercially available mass analyzers: quadrupole

mass filters, double focusing magnetic sector, time-of-flight, and collision–reaction cell technology. They all have their own strengths and

weaknesses. Since we used for our experiments a quadrupole mass filter, we will discuss this in greater detail. The collision-reaction cell technology

was discussed in the previous chapter. [34]

3.7.1. Quadrupole mass filter

The quadrupole mass filter was developed in the early 1980’s, nowadays quadrupole-based systems represent approximately 90% of all ICP mass

spectrometers used. This design was the first to be commercialized; as a result, today’s quadrupole ICP-MS technology is considered a very

mature, routine, high throughput, trace-element technique. A quadrupole consists of four cylindrical or hyperbolic metallic rods of the same length

and diameter. They are typically made of stainless steel or molybdenum, and sometimes have a ceramic coating for corrosion resistance.

Quadrupoles used in ICP-MS are typically 15–20 cm in length and about 1 cm in diameter and operate at a frequency of 2–3 MHz.

3.7.1.1. Basic mechanism As ions enter the quadrupole mass filter the mass spectrometer does not

measure mass in the sense that a balance measures mass. A mass spectrometer measures mass-to-charge (m/e) and not mass itself, there

is a bit of confusion, but mainly plain disagreement of what to call the units of measurement. Dalton is the more formal unit of m/e, but remains

one seldom used, at least in friendly conversation. Atomic Mass Unit

(AMU or amu) is often used, without much formal recognition. “Mass” is clearly formally incorrect, and when multiple charged ions are detected

the meaning is especially confusing. However, “mass” remains the colloquial word of choice, in the laboratory and is used in many



publications. But it’s important to remember that m/e is meant.

Fortunately most ions have a single charge; only a few have double plus

charge, and higher charged ions are not seen. [14]

A thorough understanding requires an ability to visualise in three dimensions and an appreciation of second order differential equations,

however this is beyond the scope of this thesis. Quad means four, and in this analyzer there are four rods, originally about the diameter of a broom

stick, but now getting smaller. The four rods are ideally of hyperbolic shape rather then circular. Pairs of opposing rods are connected together

electrically, with the two connections brought out of the vacuum system to a combination RF (Radio Frequency), often of 1MHz frequency, and DC

(Direct Current) electrical supply. [14]

Ions move in response to an electric field. This force which moves them is proportional to the different voltage. Suppose that the rods in the x-plane

are supplied with a voltage +V0 and the rods in the y-plane are supplied

with an equal and opposite voltage –V0. If an ion enters the quadrupole field the ion experiences a force pushing it away from the left hand rod

(+) and attracting it to the top rod (-). As the ion moves towards the top rod the voltages reduce and reverse. So that by the time the ion is close

to the top rod, the rod potentials have swapped and now the ion is sent off towards the right rod. The oscillation of the ion in the changing

conditions leads to ions being repeatedly sent in different directions. Depending on the applied frequency and chosen voltage, only one ion at

any one time has the correct acceleration speed to cycle all the way through the mass filter. The same voltage will accelerate a light mass ion

to high speed, but a heavy mass to slow speed. If the speed on the acceleration of the ion does not coincide with the voltage changes, it will

hit the rods and is discharged. [73] Fig. 40 shows this in greater detail. [34] The more oscillations to which an ion is subject, the higher the

possible resolution.

Fig. 40 Principle of quadrupole mass filter [34]

In this simplified example, the analyte ion (black) and four other ions (colored) have arrived at the entrance to the four rods of the quadrupole.



When a particular RF-DC voltage is applied to the rods, the positive or

negative bias on the rods will electrostatically steer the analyte ion of

interest down the middle of the four rods to the end, where it will emerge and be converted to an electrical pulse by the detector. The other ions of

different mass-to-charge ratios will pass through the spaces between the rods and be ejected from the quadrupole. This scanning process is then

repeated for another analyte at a completely different mass-to-charge ratio until all the analytes in a multi-element analysis have been

measured. Since the quad passes only one mass for a given set of applied RF and DC, this analyzer operates only as a sequential analyzer.

The process for the detection of one particular mass in a multi-element

run is represented in Fig. 41. It shows a 63Cu ion emerging from the quadrupole and being converted to an electrical pulse by the detector. As

the RF-DC voltage of the quadrupole, corresponding to 63Cu, is repeatedly scanned, the ions as electrical pulses are stored and counted by a multi-

channel analyzer. This multi-channel data acquisition system typically has

20 channels per mass, and as the electrical pulses are counted in each channel, a profile of the mass is built up over the 20 channels,

corresponding to the spectral peak of 63Cu.

Fig. 41 Data acquisition system [34]

In a multi-element run, repeated scans are made over the entire suite of

analyte masses, as opposed to just one mass represented in this example. Quadrupole scan rates are typically on the order of 2500 atomic mass

units (amu) per second and can cover the entire mass range of 0–300 amu in about 0,1 s. However, real-world analysis speeds are much slower



than this, and in practice 25 elements can be determined in duplicate with

good precision in 1–2 min. [34]

3.7.2. Quadrupole performance criteria

Two very important performance specifications of a mass analyzer govern its ability to separate an analyte peak from a spectral interference.

The first is resolving power (R), which in traditional mass spectrometry is represented by the following equation: R = m/∆m, where m is the

nominal mass at which the peak occurs and ∆m is the mass difference between two resolved peaks. However, for quadrupole technology, the

term resolution is more commonly used, and is normally defined as the width of a peak at 10% of its height. The peak width can be measured at

any relative height, and can, depending upon the desires of a writer, be used to make the analyzer look better. It is normal practice to set up the

instrument so that resolution, defined this way, is constant. Typical values are between 0,7 an 1,0 mass units. [14]

The second specification is abundance sensitivity, which is the signal

contribution of the tail of an adjacent peak at one mass lower and one mass higher than the analyte peak. [34] It is thus a measure of the

“interference” of a mass peak on its two neighbours. Notably for quad

instruments there is a larger tailing on the low mass side compared to the high mass side. [14]

Even though they are somewhat related and both define the quality of a

quadrupole, the abundance sensitivity is probably the most critical. If a quadrupole has good resolution but poor abundance sensitivity, it will

often prohibit the measurement of an ultra-trace analyte peak next to a major interfering mass. [34]

3.7.2.1. Resolution The ability to separate different masses with a quadrupole is determined

by a combination of factors including:

• Shape, diameter, and length of the rods • Frequency of quadrupole power supply

• Operating vacuum • Applied RF-DC voltages

• Motion and kinetic energy of the ions entering and exiting the quadrupole

All these factors will have a direct impact on the stability of the ions as they travel down the middle of the rods and thus the quadrupole’s ability

to separate ions of differing mass-to-charge ratios. This is represented in



Fig. 42, which shows a simplified version of the Mathieu mass stability plot

[27] of two separate masses (A and B) entering the quadrupole at the

same time.

Fig. 42 Mathieu stability diagram [34]

Any of the RF-DC conditions shown under the light blue plot will allow only

mass A to pass through the quadrupole, while any combination of RF-DC voltages under the yellow plot will allow only mass B to pass through the

quadrupole. If the slope of the RF-DC scan rate is steep, represented by the light blue line (high resolution), the spectral peaks will be narrow, and

masses A and B will be well separated (equivalent to the distance between the two blue arrows). However, if the slope of the scan is shallow,

represented by the red line (low resolution), the spectral peaks will be

wide, and masses A and B will not be so well separated (equivalent to the distance between the two red arrows). On the other hand, if the slope of

the scan is too shallow, represented by the grey line (inadequate resolution), the peaks will overlap each other as shown by the green area

of the plot) and the masses will pass through the quadrupole without being separated. In theory, the resolution of a quadrupole mass filter can

be varied between 0,3 and 3,0 amu.

Fig. 43 Intensity loss due to high resolution [34]

However, improved resolution is always accompanied by a sacrifice in

sensitivity, as seen in Fig. 43, which shows a comparison of the same mass at a resolution of 3,0, 1,0, and 0,3 amu. We can see that the peak

height at 3,0 amu is much larger than the peak height at 0,3 amu but, as

expected, it is also much wider. This would prohibit using a resolution of 3,0 amu with spectrally complex samples. Conversely, the peak width at

0,3 amu is very narrow, but the sensitivity is low. For this reason, a compromise between peak width and sensitivity usually has to be



reached, depending on the application. This can clearly be seen in Fig.

44, which shows a spectral overlay of two copper isotopes — 63Cu and 65Cu — at resolution settings of 0,70 and 0,50 amu.

Fig. 44 Sensitivity comparison of two copper isotopes, 63Cu and 65Cu, at

resolution settings of 0,70 and 0,50 amu. [34]

In practice, the quadrupole is normally operated at a resolution of 0,7–1,0 amu for most applications. It is worth mentioning that most quadrupoles

are operated in the first stability region, where resolving power is typically; 400. If the quadrupole is operated in the second or third

stability regions, resolving powers of 4000 and 9000, respectively, have been achieved. However, improving resolution using this approach has

resulted in a significant loss of signal. Although there are ways of

improving sensitivity, other problems have been encountered, and as a result, to date there are no commercial quadrupole instruments available

based on this design. Some instruments can vary the peak width on-the-fly, which means that the resolution can be changed between 3,0 and 0,3

amu for every analyte in a multi-element run. For some challenging applications this can be beneficial, but in reality they are rare. So, even

though quadrupoles can be operated at higher resolution (in the first stability region), until now the slight improvement has not become a

practical benefit for most routine applications. [34]

What is the interest in having an instrument which is capable of operating in a high resolution mode? That is, with resolution better than 1 amu

resolution (peak width at 10% peak height) or better than an equivalent m/∆m of 300. To understand the capabilities and limitations of high

resolution, a consideration of the “mass” of the proton and neutron need

to be made, and the loss of mass as these fundamental particles combine to form nuclei where some of the mass is converted to binding energy.

While the manufactures of high resolution instrumentation would like buyers to believe that all possible mono- and poly-atomic ions can be

separated, that is simply not true. There are however some interesting applications of high resolution instrumentation.

The mass of each neutron and proton which makes up a nuclei each have

a mass of approximately 1 amu or 1 Dalton each. However, when neutrons and protons combine to form a nuclei, there is a loss of mass

(mass defect) as some of the mass is converted to binding energy of the nuclei. The loss of mass is in accordance with the Einsteinian law that “E”



equals “m c squared”. Thus the exact mass of an ion is not an integer

number which is equal to the total count of the number of neutrons and

protons. Further, there is not a constant binding energy due to each neutron and proton as shown in Fig. 45.

Fig. 45 Plot of the "exact mass" divided by the "integer mass" vs. the "integer

mass" [14]

The Figure is a plot of the “exact mass” divided by the “integer mass” vs.

the “integer mass”. Clearly seen is the extra stability at mass 4He. Not as clear in this figure, is the extra stability at mass 12C, 16O, and 20Ne. Note

the minimum which is found at mass 56Fe, the heaviest “major” element, which reflects that light elements “fuse” to form heavier nuclei while

heavy elements undergo “fission” forming lighter nuclei, with Fe the “goal”.

The “mass defect” varies with mass. In theory, it is possible, knowing the

exact mass of an ion, to determine the element. For example 14C (important for archaeological dating) has a mass of 13,996758 Dalton,

which is different form 14N (the major interference in the determination of 14C by mass spectrometric means) which has a larger mass of 13,996926

Dalton. The mass difference is 0,000168 Dalton, so a resolving power of

83 000 (14/0,000168) would be the minimum required to separate these two ions. Noting that standard double focussing sector instrument have a

high resolving power of around 10000 (which is a lot higher than quadrupole instruments), the determination of 14C is not practical on this

kind of mass spectrometer. An important conclusion is that all interferences cannot be resolved using high resolution standard double

focussing sector instruments, and this includes important elemental ion overlaps. The existence of the minimum at 56Fe has the unfortunate

consequence that two nuclei to the left of the minimum can have a combined mass very close to one nucleus to the right. For example 51V40Ar has an exact mass of 90,9063456 and could only be separated from 91Zr (90,9056442) with an unobtainable resolving power of 130 000.

In conclusion high resolution mass spectrometer have a powerful ability to resolve some overlaps of two ions having the same nominal mass, but not



all overlaps can be resolved. Note also that the calculation of the

minimum required resolving power (resolution) here assumes the unlikely

occurrence that the overlapping ions are of the same intensity. In the more likely situation in which one peak is much larger than the other,

even higher resolution would be required. [14]

3.7.2.2. Abundance sensitivity

We can see in Fig. 46 that the tails of the spectral peaks drop off more rapidly at the high mass end of the peak compared with the low mass

end. The overall peak shape, particularly its low mass and high mass tail, is determined by the abundance sensitivity of the quadrupole, which is

affected by a combination of factors including:

• Design of the rods • Frequency of the power supply

• Operating vacuum

Fig. 46 Spectral peak drop off [34]

Even though they are all important, probably the biggest impacts on

abundance sensitivity are the motion and kinetic energy of the ions as

they enter and exit the quadrupole. If one looks at the Mathieu stability plot in Fig. 42, it can be seen that the stability boundaries of each mass

are less defined (not so sharp) on the low mass side than they are on the high mass side. As a result, the characteristics of ion motion at the low

mass boundary are different from the high mass boundary and are therefore reflected in poorer abundance sensitivity at the low mass side

compared with the high mass side. In addition, the velocity (and therefore the kinetic energy) of the ions entering the quadrupole will

affect the ion motion and, as a result, will have a direct impact on the abundance sensitivity. For that reason, factors that affect the kinetic

energy of the ions, like high plasma potential and the use of lens components to accelerate the ion beam, will degrade the instrument’s

abundance sensitivity. These are the fundamental reasons why the peak shape is not symmetrical with a quadrupole and explains why there is

always a pronounced shoulder at the low mass side of the peak compared

to the high mass side — as represented in Fig. 46, which shows the theoretical peak shape of a nominal mass M. We can see that the shape



of the peak at one mass lower (M - 1) is slightly different from the other

side of the peak at one mass higher (M + 1) than the mass M. For this

reason, the abundance sensitivity specification for all quadrupoles is always worse on the low mass side than on the high mass side and is

typically 1 X 10-6 at M - 1 and 1 X 10-7 at M + 1. In other words, an interfering peak of 1 million counts per second (cps) at M - 1 would

produce a background of 1 cps at M, while it would take an interference of 107 cps at M + 1 to produce a background of 1 cps at M.

Fig. 47 shows an example of the importance of abundance sensitivity.

Fig. 47a is a spectral scan of 50 ppm of the doubly charged europium ion 151Eu++ at 75,5 amu (a doubly charged ion is one with two positive

charges, as opposed to a normal singly charged positive ion, and exhibits a m/z peak at half its mass). We can see that the intensity of the peak is

so great that its tail overlaps the adjacent mass at 75 amu, which is the only available mass for the determination of arsenic. This is highlighted in

Fig. 47b, which shows an expanded view of the tail of the 151Eu++,

together with a scan of 1 ppb of As at mass 75. We can see very clearly that the 75As signal lies on the sloping tail of the 151Eu++ peak.

Measurement on a sloping background like this would result in a significant degradation in the arsenic detection limit, particularly as the

element is mono-isotopic and no alternative mass is available. This example shows the importance of a low abundance sensitivity specification

in ICP-MS. [Fig. 47]

Fig. 47 Example of the importance of abundance sensitivity [34]



3.8. Detector The detector is the instrument, where the ion beam is converted to a

computer-usable number, either in pulse counting (PC) or digital mode in which ions are counted, or in an analogue mode in which the ion beam

current (amperage) is converted to a potential (voltage) which is in turn converted to a number using some kind of analogue-to-digital conversion

hardware. Ideally the user would like to have a detector which can operate in either mode because the two different modes are optimum in

different intensity ranges. For low intensity signals, the digital mode is optimum, from the point of view of the best relative standard deviation of

the signals and minimum detection limits. At high ion count rates (greater than approximately 1 000 000 counts per second or cps) pulse pile up or

dead time corrections degrade quality. Hardware or software corrections may be applied, but the correction theory has problems as intensities get

too high, and eventually fails entirely. As a rule of thumb, below about

100 000 cps dead time is not significant, and the user should be careful at higher count rates that a correction has been applied. In any case at

sufficiently high count rates (several million cps), the sequence of pulses turn into one giant pulse and the digital mode is no longer operative. For

most hardware, the analogue mode starts at an ion beam intensity equivalent to about 1 000 cps. However at this signal intensity, the digital

mode is superior, a superiority which the digital mode maintains until near the upper range of the digital mode. The analogue mode is operational up

to somewhere close to 1 000 000 000 equivalent cps, limited by the largest potential (voltage) input which the analogue digital converter can

allow. [14]

3.8.1. Instrumentation

The first detectors were channel electron multipliers (channeltron) for pulse counting and Faraday cups for analogue counting, however

nowadays they have been replaced almost everywhere by the discrete dynode electron multiplier.

3.8.1.1. Discrete dynode electron multiplier Discrete dynode electron multiplier, which are often called active film

multipliers, work in a similar way to the channeltron, but use discrete dynodes to carry out the electron multiplication. Fig. 48 illustrates the

principles of operation of this device. The detector is usually positioned off-axis to minimize the background from stray radiation and neutral

species coming from the ion source. When an ion emerges from the

quadrupole, it sweeps through a curved path before it strikes the first dynode. On striking the first dynode, it liberates secondary electrons.

The electron-optic design of the dynode produces acceleration of these secondary electrons to the next dynode, where they generate more



electrons. This process is repeated at each dynode, generating a pulse of

electrons that is finally captured by the multiplier collector or anode.

Because of the materials used in the discrete dynode detector and the difference in the way electrons are generated, it is typically more sensitive

than channeltron technology.

Fig. 48 Discrete dynode electron multiplier [35]

Although most discrete dynode detectors are very similar in the way they work, there are subtle differences in the way the measurement circuitry

handles low and high ion-count rates. When ICP-MS was first commercialized, it could only handle as many as five orders of dynamic

range; however, when attempts were made to extend the dynamic range, certain problems were encountered. [35]

3.8.2. Extending the dynamic range Traditionally, ICP-MS using the pulse counting measurement is capable of

about five orders of linear dynamic range. This means that ICP-MS calibration curves, generally speaking, are linear from ppt levels to as

much as a few hundred parts-per-billion. However, a number of ways exist to extend the dynamic range of ICP-MS another three to four orders

of magnitude to work from sub-part-per-trillion levels, to as much as 100 ppm. Following is a brief overview of some of the different approaches

that have been used. [35]

3.8.2.1. Filtering the ion beam.

One of the first approaches to extend the dynamic range in ICP-MS was to filter the ion beam by putting a non-optimum voltage on one of the ion

lens components or the quadrupole itself to limit the number of ions reaching the detector. This voltage offset, which was set on an individual

mass basis, acted as an energy filter to electronically screen the ion beam



and reduce the subsequent ion signal to within a range covered by pulse-

counting ion detection. The main disadvantage with this approach was

that the operator had to have prior knowledge of the sample to know what voltage to apply to the high concentration masses. [35]

3.8.2.2. Using two detectors.

Another technique that was implemented in some of the early quadrupole

ICP-MS instrumentation was to use two different detectors, such as a channel electron multiplier to measure low current signals, and a Faraday

cup to measure high ion currents. This process worked reasonably well, but struggled with some applications because it required rapid switching

between the two detectors. The problem was that the ion beam had to be physically deflected to select the optimum detector. Not only did this

degrade the measurement duty cycle, but detector switching and stabilization times of several seconds also precluded fast transient signal

detection. [35]

3.8.2.3. Using two scans with one detector.

The more modern approach is to use just one detector to extend the dynamic range. By using the detector in both the pulse-counting and

analogue modes, high and low concentrations can be determined in the same sample. Three approaches use this type of detection system; two of

them involve carrying out two scans of the sample, while the third uses only one scan.

The first approach uses an electron multiplier operated in both digital and

analogue modes. Digital counting provides the highest sensitivity, while

operation in the analogue mode (achieved by reducing the high voltage applied to the detector) is used to reduce the sensitivity of the detector,

thus extending the concentration range for which ion signals can be measured. The system is implemented by scanning the spectrometer

twice for each sample. A first scan, in which the detector is operated in the analogue mode, provides signals for elements present at high

concentrations. A second scan, in which the detector voltage is switched to digital pulse counting mode, provides high sensitivity detection for

elements present at low levels. A major advantage of this technology is that users do not need to know in advance whether to use analogue or

digital detection because the system automatically scans all elements in both modes. However, its disadvantage is that two independent mass

scans are required to gather data across an extended signal range. This not only results in degraded measurement efficiency and slower analyses,

but it also means that the system cannot be used for fast transient signal

analysis of unknown samples because mode switching is generally too slow.



The second way of extending the dynamic range is similar to the first

approach, except that the first scan is used as an investigative tool to

examine the sample spectrum before analysis. This first pre-scan establishes the mass positions at which the analogue and pulse modes will

be used for subsequently collecting the spectral signal. The second analytical scan is then used for data collection; the system switches the

detector back and forth rapidly between pulse and analogue mode depending on the concentration of each analytical mass. The main

disadvantage of these two approaches is that two separate scans are required to measure high and low levels. With conventional nebulisation,

this isn't such a major problem except that it can impact sample throughput. However, it does become a concern when it comes to

working with transient peaks found in electro-thermal vaporization, flow injection, or laser sampling ICP-MS. Because these transient peaks often

last only a few seconds, all the available time must be spent measuring the masses of interest to get the best detection limits. When two scans

have to be made, valuable time is wasted, which is not contributing to

quality of the analytical signal. [35]

3.8.2.4. Using one scan with one detector. These limitations of using two scans led to the development of a third

approach, which we also used in our experiments, using a dual-stage

discrete dynode detector. This technology uses measurement circuitry that allows both high and low concentrations to be determined in one

scan. This is achieved by measuring the ion signal as an analogue signal at the midpoint dynode. When more than a threshold number of ions are

detected, the signal is processed through the analogue circuitry

Fig. 49 Dual-stage discrete dynode detector [35]



When fewer than the threshold number of ions are detected, the signal

cascades through the rest of the dynodes and is measured as a pulse

signal in the conventional way. This process, which is shown in Fig. 49, is completely automatic and means that both the analogue and the pulse

signals are collected simultaneously in one scan. The pulse counting mode is typically linear from zero to about 106 cps, while the analogue

circuitry is suitable from 104 to 109 cps. To normalize both ranges, a cross calibration is performed to cover concentration levels, which could

generate a pulse and an analogue signal.

Fig. 50 Cross calibration [35]

Fig. 51 Normalized cross calibration [35]

This is possible because the analogue and pulse outputs can be defined in identical terms of incoming pulse counts per second, based on knowing

the voltage at the first analogue stage, the output current, and a



conversion factor defined by the detection circuitry electronics. By

performing a cross calibration across the mass range, a dual-mode

detector of this type is capable of achieving approximately eight to nine orders of dynamic range in one simultaneous scan. Fig. 50 shows the

pulse-counting calibration curve (yellow) is linear up to 106 cps, and the analogue calibration curve (blue) is linear from 104 to 109 cps. Fig. 51

shows that after cross calibration, the two curves are normalized, which means the detector is suitable for concentration levels between 0,1 ppt

and 100 ppm — typically eight to nine orders of magnitude for most elements. There are subtle variations of this type of detection system,

but its major benefit is that it requires only one scan to determine both high and low concentrations. Therefore, it not only offers the potential to

improve sample throughput, it also means that the maximum data can be collected on a transient signal that only lasts a few seconds.



Experimental part

1. Introduction

As said in the literature part, the LA-ICP-MS was bought to have an extra tool in the research about corrosion and agglomeration problems in

fluidized bed boilers. So the task was to learn how to work with the LA-

ICP-MS so that it could be used to investigate corrosion and agglomeration samples. The goal of this experimental part is to continue

the measurements with the LA-ICP-MS which were already started in the thesis work of Susanna Arvilommi at the University of Oulu. [1] Although

previous measurements were already made, the technique is still quite new for the university and lots of improvements need to be made in order

to get better results. To test the instrument’s capabilities we measured NIST-2691 and NIST-1633b standard reference material with known

certified analytical concentrations. (Fig. 52 shows the concentrations of the elements we tried to measure) In the following pages NIST-2691 and

NIST-1633b will be referred to as std A and std B.

NIST-2691 NIST-1633b

std A std B

wt % 2 x std dev wt % 2 x std dev

Mg 3,12 +/- 0,08 0,482 +/- 0,008

Al 9,81 +/- 0,39 15,05 +/- 0,27

Si 16,83 +/- 0,12 23,02 +/- 0,08

P 0,51 +/- 0,02 0,23 +/- non certified

S 0,83 +/- 0,05 0,2075 +/- 0,0011

Ca 18,45 +/- 0,32 1,51 +/- 0,06

Ti 0,90 +/- 0,02 0,791 +/- 0,014

Fe 4,42 +/- 0,03 7,78 +/- 0,23

Fig. 52 NIST-2691 and NIST-1633b

The aim of this thesis was to measure these certified samples and improve the measurement procedures so that we would get the same

concentrations as declared by the standard reference material. Once good procedures are found they can be used to measure unknown corrosion

and agglomeration samples, so that corrosion and agglomeration in fluidized bed boilers can be further investigated. Changes in ways of

measuring mainly focussed on better laser parameters, the use of the

collision and reaction cell, and the use of internal standards. In the end we will also compare the results of several methods of bulk analysis made

on these samples.



2. Sample preparation

As mentioned before, besides form measuring with LA-ICP-MS, which was off course the main task in this thesis, we also made fusions with Li2B4O7

and Na2CO3 fluxes to check the concentrations of the reference materials. These fusions were digested in acid and then sent to the laboratory for

analysis. In this chapter we will discuss the sample preparation of the LA-ICP-MS samples and the fusion samples.

2.1. LA-ICP-MS samples The reference powder was turned into a solid cylindrical sample, by using

the Epofix kit from Struers. [26] 15 volume parts of resin were mixed with 2 volume parts of hardener in a paper cup and stirred carefully for at least

2 minutes. After 2 minutes the mixture was poured carefully over the

specimen in the mould. The sample was then placed into a vacuum chamber in which air was evacuated for about 5-10 minutes. After that

the sample was dried for 8 hours.

The next day we polished the sample with 3 different sandpapers from Buehler with different roughness (gritt p240, gritt p600 and gritt p1200),

with the last paper being off course the less rough. To make the polishing process easier we used 99% glycerine as a lubricant. Between every

change of sandpaper we washed the sample with technical alcohol and put it for a few minutes in an ultrasonic device. After the polishing it is

important not to clean the surface of the sample anymore with plain paper because this may scratch the surface again. The surface can be cleaned

by pouring technical alcohol over it and just leave it to dry.

2.2. Fusion samples We made 3 fusion samples for each standard and blank and this for both

fluxes (Na2CO3 and Li2B4O7). After preparation these samples were sent to the laboratory for analysis.

2.2.1. Na2CO3 and Na202 as flux Samples were prepared in Ni crucibles. First we weighted 0,05 g of

sample and mixed it in the crucible with 0,5 g Na2CO3 and 0,5 g Na2O2. When everything was mixed we covered the mixture with another 0,5 g of

Na2C03 without mixing it again. Then we put the crucibles in a 700 °C preheated oven and left it there for 30 minutes.

After the samples had cooled down in an exsiccator they were put into a

inert plastic jar with a 25 ml HCl (1:20) solution. The plastic jars were



mixed and heated by putting them into boiling water. When the flux was

digested the Ni crucible was washed with some HCl solution and 7,5 ml of

concentrated HCl was added to the solution to make sure that everything was digested. The digested flux was then transferred to a 100 ml plastic

flask.

2.2.2. Li2B4O7 as flux

Samples were prepared in Pt crucibles. First we weighted 0,05 g of sample and mixed it in the crucible with 0,5 g Li2B4O7. When everything

was mixed we covered the mixture with another 0,5 g of Li2B4O7 without mixing it again. Then we put the crucibles in a 1000 °C preheated oven

and left it there for 20 minutes.

After the samples had cooled down in an exsiccator they were put into an inert plastic jar with a 25 ml HNO3 (5%) solution. The plastic jars were

mixed and heated by putting them into boiling water. When the flux was digested the Pt crucible was washed with some HNO3 (5%) solution and

the digested flux was transferred to a 100 ml plastic flask.



3. Configuration of LA-ICP-MS

Before we start configuring the LA-ICP-MS it is important to realize what we are measuring. We must know that we are testing the capabilities of

this mass spectrometer by trying to measure the same concentrations as were certified by the NIST standards. These certified concentrations are

bulk concentrations, so this means that we have to configure the LA-ICP-MS for optimal bulk measurements. Because the laser ablates with a spot

size between 5 µm and 160 µm it is important that the analysed sample is

as homogeneous as possible to be sure that the ablated material is

representative for the whole sample. Unfortunately the samples we analyzed were not very homogeneous so that configuration of data

acquisition parameters was essential. First we will discuss a few parameters which needed to be configured and then we will mention the

applied configuration.

3.1. Laser parameters As mentioned in the literature part we used a Nd:YAG laser with its 5th

harmonic in the UV (213 nm). Basically there are 4 major laser parameters which should be configured: scan speed (µm/s), output (%),

pulse repetition rate (Hz) and spot size (µm). For bulk measurements it is

recommended that spot size and scan speed is as high as possible

because then a bigger sample surface is analysed and the more representative the measurement should be for the entire sample.

However out of experience big values are not always applicable since a good peak/background ratio in the ICP-MS must be achieved and too big

count rates could damage the detector. Besides, as also mentioned before, ablation characteristics are still poorly understood and different

laser parameters for different elements are usually necessary. That’s why it’s worth optimizing the laser parameters before the analysis of the

sample starts. It’s also worth mentioning that too high energy output can

result in catastrophic ablation, which should off course be avoided.

For our measurements we used most of the time a line raster, which resulted in a relatively stable signal in the ICP-MS and made it possible to

cover ablation surface which is big enough so that the analyzed surface could have been considered representative for the whole sample.

Although the laser parameters differ from experiment to experiment, the

laser parameter for the blank measurement was always the same, as shown in Fig. 53.



Blank

scan speed 1 µm/sec

output 0 %

rep. rate 1 Hz

spot size 5 µm

Fig. 53 Laser parameters for blank measurements

3.2. ICP-MS parameters

Fig. 54 and Fig. 58 show the ICP-MS configuration we used in our experiments. As one can see in these figures, they are divided in two

groups; major and minor parameters. The major parameters are assumed to have the biggest influence on the background and peak levels.

Basically both major and minor parameters need to be configured so that the highest peak/background is obtained. For the measurements without

collision and reaction cell we used the same configuration as in the previous thesis of Susanna Arvilommi. [1] For the measurements with the

collision and reaction cell we had to change the parameters to get an optimal peak/background. It’s worth noticing that for the collision and

reaction cell experiments, the hexapole bias is set a little bit more positive. This slows down the ions in the CCT so that they stay longer in

the CCT and as a consequence collision and reaction is more complete. In

this way interference is further reduced. Notice that the pole bias (quadrupole potential) is negative in comparison with the hexapole bias

(CCT potential), this means that there is no kinetic discrimination applied. Further experiments need to be made to investigate the influence of

kinetic discrimination on the peak/background.

After that, we have to decide if we’re going to do a continuous or peak jumping scan. In continuous mode the entire spectrum is analyzed and in

peak jumping mode only certain elements are analyzed. Because continuous scans take a lot of time, we always measured in peak jumping

mode. The peak jumping approach is where the quadrupole power supply is driven to a discrete position on the peak (normally the peak point) and

allowed to settle; a measurement is then taken for a fixed amount of time. The amount of time that the quadrupole needs to settle is called the

settling time, and is normally fixed because it is a function of the

quadrupole and detector electronics. When the electronics are “settled” they start to “dwell” on the peak and take measurements for a fixed

period of time. This procedure is repeated for every selected peak (element) and is called a “sweep”. So if i.e. 300 sweeps are made, the

quadruple goes 300 times to every element to settle and to dwell. Usually the same dwell time is selected for every element so the time which is

needed for the entire analysis is given by:

( )settlingdwellelementssweepst +××=



4. Measurements with LA-ICP-MS

4.1. Detection of 34S without CCT or internal

standard The first experiments we carried out were the detection of 34S without CCT

or internal standard. Fig. 54 shows the ICP-MS configuration parameters we used for this experiment.

Major

Extraction Lens 1 Focus D 1 Pole bias Hexapole

bias Nebulizer

Sampling

depth

-712 2,0 16,0 -30,9 -2,4 -2,0 1,07 56

Minor

Lens 2 Lens 3 Forward

power Horizontal Vertical D 2 DA Cool Auxiliary

-45,0 -190,7 1100 80 614 -124 -43,4 13,0 0,79

Fig. 54 ICP-MS configuration for detection of 34S without CCT or internal

standard

We tried two different rasters for this experiment; we first tried to analyze

using a line raster, because that was the raster which was also used in

earlier experiments. After a few experiments we decided to try also a dot raster; the idea was that if we would measure more than 25 different dots

spread out over a bigger surface, the measurements should be more representative for the sample. The last two measurements (ajo42.tee and

ajo43.tee) have both more sweeps and longer dwell time, we thought that this would give better results because more sweeps and a higher dwell

time means that a larger surface is analyzed. Applied acquisition parameters are shown in Fig. 55.

Line raster ajo37.tee ajo38.tee ajo39.tee ajo40.tee

scan speed (µm/sec) 10 10 50 20

output (%) 55 55 55 55

rep. rate (Hz) 20 20 20 20

spot size (µm) 60 60 60 60

MS dwell time ms 30 30 30 30

MS sweeps 300 300 300 300

Dot raster (100x100µm) ajo41.tee ajo42.tee ajo43.tee

output (%) 55 55 60

rep. rate (Hz) 20 20 20

spot size (µm) 60 60 60

dwell time s 2 1 1

intersite pause s 2 1 1

MS dwell time ms 30 40 40

MS sweeps 300 400 400

Fig. 55 Acquisition parameters for 34S



Fig. 56 shows ajo37.tee till ajo43.tee, these are the measurements of 34S.

Every measurement consists of several runs (between 5 and 10 depending

on how much time we had to measure), every run on its own consisted of 10 sub-runs. A sub-run was defined as a number of sweeps (300 or 400)

and an amount of dwell time (30 or 40 ms for every element analyzed). (Which experiment used which amount of sweeps or dwell time can be

seen in Fig. 55)

std A std B Results

34S

ajo37.tee ajo41.tee ajo42.tee ajo43.tee ajo38.tee ajo39.tee ajo40.tee

run 1 0,768 0,823 0,881 0,680 0,385 0,178 0,219

run 2 0,973 0,692 0,833 0,776 0,272 0,235 0,217

run 3 0,749 0,976 0,776 1,034 0,143 0,210 0,250

run 4 0,713 1,411 0,684 0,781 0,119 0,249 0,236

run 5 0,670 1,002 0,709 0,822 0,086 0,138 0,231

run 6 0,694 1,063 0,681 0,792 0,140 0,227

run 7 0,695 0,931 0,752 0,776 0,161 0,202

run 8 0,837 0,973 0,697 0,767 0,140

run 9 0,580 0,449

run 10 0,163

average 0,742 0,984 0,752 0,804 0,201 0,206 0,226

std dev 0,112 0,208 0,074 0,102 0,125 0,094 0,015

Fig. 56 Results 34S in weight %

The NIST standards give us certified sulphur concentrations for std A (0,83 wt%) and std B (0,2075 wt%) as shown in Fig. 52. The results of

std A suggest that higher dwell time and more sweeps result in both better average and better standard deviation. (ajo42.tee and ajo43.tee)

There was no significant difference between the dot raster and the line raster. A careful reader will notice that the measured concentrations for

every run within one measurement can differ significantly form each other. One may think that these are wrong measurements, however this

is not necessarily true since our sample was rather heterogeneous. The fact that runs differ just describes the heterogeneousness of the sample.

However, although these measurements are in the right direction the

critical reader will agree that there is still some room for improvements, both average and standard deviation should be better. Probably two

changes could significantly improve the results.

First of all the use of the collision and reaction cell should be able to lower

the background level and lead to a better peak-background ratio. As seen in Fig. 57, the background level in the last measurements were quite high

with respect to the peak level; a peak/background of 10 should be much better than the now poorly attained peak/background of 2. Fig. 57 shows

the count rates of the blank measurement (=background) and the count rates of the first three run’s (=peak). It can be clearly seen that the

peak/background is about 1,5 which is too low.



ajo 43 std A 34

Sy = 7171,8x + 11699

R2 = 0,8852

0

5000

10000

15000

20000

25000

0 0,2 0,4 0,6 0,8 1

Weight %

Co

un

t ra

tes

pe

r s

ec

on

d

(cp

s)

Fig. 57 Calibration line ajo 43 std A 34S

Secondly, the use of internal standards should correct for matrix effects

and signal drift in the ICP-MS and should have a positive effect on our results. However if we wish to use an internal standard it is important

that we know at least one concentration exactly. For now this is not a problem since we use certified reference materials with off course known

bulk concentrations. But still we need to be careful when we’re using one element as an internal standard because our sample is not perfectly

homogeneous. So whenever we use an internal standard it becomes even more important that the analyzed surface is representative for the whole

sample, otherwise the internal standard deteriorates the results!

The use of the collision and reaction cell and the internal standard will be

discussed in the next chapter.



4.2. Measurements using CCT and internal

standard In the second part of our research we tried to use the collision and

reaction cell technology (CCT). As described in the literature part the CCT reduces interferences so that background levels are lowered and a better

peak/background ratio is obtained. As reaction gas we used a mixture of H2 and He with a flow rate of 7 – 8 ml/min. Because our LA-ICP-MS broke

down a few weeks ago and because it’s still in maintenance we were not able to measure std B. So the results presented in this chapter only apply

on std A. Fig. 58 Shows the ICP-MS configuration using CCT.

Major

Extraction Lens 1 Focus D 1 Pole bias Hexapole

bias Nebulizer

Sampling

depth

-512 0,0 8,0 -46,0 -1,2 3,3 1,07 56

Minor

Lens 2 Lens 3 Forward

power Horizontal Vertical D 2 DA Cool Auxiliary

-23,0 -16,0 1100 67 606 -125 -40,4 11,1 0,78

Fig. 58 ICP-MS configuration using CCT

Fig. 59 shows the acquisition parameters we used in our experiment using the collision and reaction cell technology. During CCTajo3.tee we decided

to change back to the line raster because we wanted to use the same raster as in the previous thesis [1] in which the same elements were

measured but without the CCT. This way we could compare these results in more equal conditions. However in the future it’s worth trying the dot

raster again, as it might give better results after all. It is also important

to know that if we use the dot raster we should do experiments in transient mode and not in continuous mode because the signal becomes

too transient especially when short dwell times are used.

Dot raster (100x100µm) CCTajo2.tee CCTajo3.tee

output (%) 75 95

rep. rate (Hz) 15 15

spot size (µm) 80 140

dwell time s 1 1

intersite pause s 1 1

MS dwell time ms 40 40

MS sweeps 400 400

Line raster CCTajo3.tee CCTajo4.tee CCTajo5.tee CCTajo6.tee CCTajo7.tee

scan speed (µm/sec) 20 20 20 40 40

output (%) 95 95 60 70 100

rep. rate (Hz) 15 15 15 20 20

spot size (µm) 140 140 80 120 160

MS dwell time ms 40 40 40 40 40

MS sweeps 400 400 400 400 400

Fig. 59 Acquisition parameters using CCT



Since we changed the raster in CCTajo3.tee during measurements we will

not discus this measurement further, because changing the laser

parameters during measuring could mess up our results. As in previous chapter every measurement consisted of 8 runs, every run on its own

consisted of 10 sub-runs. A sub-run was defined as a number of sweeps (400) and an amount of dwell time (40 ms for every element analyzed).

In every figure the measured concentrations are found in the column signal, i.e. column “40Ca signal” gives us the measured concentrations

from 40Ca. Then this “40Ca signal” is compared with the certified concentration of 40Ca, the certified concentration can be found in the

column “40Ca std”. The ratio of these two values will be used to correct the other concentrations. So 40Ca is used as internal standard. The

correction is carried out by multiplying the measured concentration (i.e. “56Fe signal”) with the correction factor (“40Ca std/40Ca signal”). The

corrected concentrations can be found in the column “56Fe int std”. Fig. 60 till Fig. 64 shows us the results of our experiments.

CCTajo2.tee

std A

40Ca

std

40Ca

signal

40Ca std /

40Ca signal

44Ca

signal

44Ca

int std

56Fe

signal

56Fe

int std

run 1 18,45 19,51 0,95 18,81 17,79 4,377 4,139

run 2 18,45 17,09 1,08 17,26 18,63 3,740 4,038

run 3 18,45 18,74 0,98 19,27 18,97 5,183 5,103

run 4 18,45 15,90 1,16 15,98 18,54 4,166 4,834

run 5 18,45 16,52 1,12 15,57 17,39 3,492 3,900

run 6 18,45 16,48 1,12 16,60 18,58 3,571 3,998

run 7 18,45 16,66 1,11 17,51 19,39 3,383 3,746

run 8 18,45 16,96 1,09 16,95 18,44 3,736 4,064

average 17,244 18,467 3,956 4,228

std dev 1,283 0,630 0,599 0,478

Fig. 60 Detection of 44Ca and 56Fe with 40Ca as internal standard

CCTajo4.tee

std A

40Ca

std

40Ca

signal

40Ca std /

40Ca signal

56Fe

signal

56Fe

int std

run 1 18,45 18,40 1,00 4,544 4,556

run 2 18,45 18,17 1,02 4,294 4,360

run 3 18,45 18,78 0,98 4,422 4,344

run 4 18,45 18,02 1,02 4,276 4,378

run 5 18,45 20,01 0,92 4,771 4,399

run 6 18,45 23,44 0,79 5,864 4,616

run 7 18,45 22,12 0,83 5,077 4,235

run 8 18,45 24,12 0,76 5,993 4,584

average 4,905 4,434

std dev 0,685 0,135

Fig. 61 Detection of 56Fe with 40Ca as internal standard



CCTajo5.tee

std A

40Ca

std

40Ca

signal

40Ca std /

40Ca signal

56Fe

signal

56Fe

int std

run 1 18,45 18,51 1,00 4,192 4,178

run 2 18,45 25,42 0,73 6,172 4,480

run 3 18,45 18,39 1,00 4,648 4,663

run 4 18,45 18,45 1,00 4,266 4,266

run 5 18,45 19,44 0,95 4,280 4,062

run 6 18,45 18,74 0,98 4,825 4,750

run 7 18,45 22,08 0,84 5,211 4,354

run 8 18,45 25,98 0,71 6,488 4,608

average 5,010 4,420

std dev 0,886 0,245

Fig. 62 Detection of 56Fe with 40Ca as internal standard

CCTajo6.tee

std A

40Ca

std

40Ca

signal

40Ca std /

40Ca signal

24Mg

signal

24Mg

int std

27Al

signal

27Al

int std

48Ti

signal

48Ti

int std

56Fe

signal

56Fe

int std

run 1 18,45 17,56 1,05 3,005 3,157 9,527 10,010 0,772 0,811 4,222 4,436

run 2 18,45 18,84 0,98 3,177 3,111 9,851 9,647 0,960 0,940 4,452 4,360

run 3 18,45 18,95 0,97 3,178 3,094 10,050 9,785 0,968 0,942 4,586 4,465

run 4 18,45 18,01 1,02 3,040 3,114 8,849 9,065 0,856 0,877 4,661 4,775

run 5 18,45 20,28 0,91 3,448 3,137 10,350 9,416 0,969 0,882 5,072 4,614

run 6 18,45 19,06 0,97 3,227 3,124 10,270 9,941 0,920 0,891 4,703 4,552

run 7 18,45 18,07 1,02 3,004 3,067 8,872 9,059 0,888 0,907 4,304 4,395

run 8 18,45 18,75 0,98 3,121 3,071 8,987 8,843 0,924 0,909 4,532 4,459

average 3,150 3,109 9,595 9,471 0,907 0,895 4,567 4,507

std dev 0,147 0,031 0,627 0,443 0,068 0,042 0,263 0,136

Fig. 63 Detection of 24Mg, 27Al, 48Ti and 56Fe with 40Ca as internal standard

CCTajo7.tee

std A

40Ca

std

40Ca

signal

40Ca std /

40Ca signal

28Si

signal

28Si

int std

31P

signal

31P

int std

34S

signal

34S

int std

44Ca

signal

44Ca

int std

run 1 18,45 17,44 1,06 15,670 16,577 0,490 0,518 0,827 0,875 17,500 18,513

run 2 18,45 19,12 0,96 16,680 16,096 0,524 0,506 0,845 0,815 19,210 18,537

run 3 18,45 18,79 0,98 18,140 17,812 0,516 0,507 0,818 0,803 18,640 18,303

run 4 18,45 19,60 0,94 17,070 16,068 0,516 0,486 0,821 0,773 19,740 18,582

run 5 18,45 20,30 0,91 17,720 16,105 0,518 0,471 0,852 0,774 20,200 18,359

run 6 18,45 20,24 0,91 17,590 16,034 0,539 0,491 0,763 0,696 19,730 17,985

run 7 18,45 22,21 0,83 18,730 15,559 0,551 0,458 0,859 0,714 22,640 18,807

run 8 18,45 20,32 0,91 17,970 16,316 0,519 0,471 0,776 0,705 20,740 18,831

average 17,446 16,321 0,522 0,488 0,820 0,769 19,800 18,490

std dev 0,955 0,667 0,018 0,021 0,035 0,062 1,517 0,277

Fig. 64 Detection of 28Si, 31P, 34S and 44Ca with 40Ca as internal standard

Before proceeding it’s important to know that the last two measurements (Fig. 63 and Fig. 64) were carried out with a bad sensitivity. The

sensitivity was two times lower than usual which probably had a negative



influence on our results. But because time was precious, we decided to

measure anyway, the results can be used but the reader should know that

results would probably be better when the measurements were carried out with normal sensitivity. Normally we would redo the measurements with

normal sensitivity but because the ICP-MS was in maintenance for quit some time this was not possible.

First of all it is clear that the CCT lowered the background tremendously,

this resulted in better peak/background and as a consequence in lower detection limits and better calibration lines. As a comparison we can see

the difference of a measurement carried out with CCT (Fig. 66) and without CCT (Fig. 65). Notice that the count rates in these figures were

not corrected with the internal standard.

ajo 19 std A 24

Mgy = 8326x + 7091

R2 = 0,9828

0

5000

10000

15000

20000

25000

30000

35000

40000

0 0,5 1 1,5 2 2,5 3 3,5

Weight %

Co

un

t ra

tes

pe

r s

ec

on

d

(cp

s)

Fig. 65 Calibration line ajo 19 std A 24Mg

CCTajo 6 std A 24

Mgy = 41287x + 225,34

R2 = 0,9973

0

20000

40000

60000

80000

100000

120000

140000

160000

0 0,5 1 1,5 2 2,5 3 3,5

Weight %

Co

un

t ra

tes

pe

r s

ec

on

d

(cp

s)

Fig. 66 Calibration line CCTajo 6 std A 24Mg

Secondly we noticed that the use of the internal standard has a significant positive influence on the quality of our results, as both the average and

standard deviation have improved significantly for al the measured elements (24Mg, 28Si, 27Al, 44Ca, 48Ti and 56Fe) except from 31P and 34S.

We can see that both the average and the standard deviation have



improved and that the certified concentrations are all within the standard

deviation. There are several reasons which could explain why an

improvement is noticed for 24Mg, 28Si, 27Al, 44Ca, 48Ti and 56Fe and not for 31P and 34S. When we look to the mean count-rates (cps) of all these

elements we can see that the elements which have better results all had a peak/background between 10 and 1000 while 31P and 34S only had a

peak/background of 5. (with peak count rates not corrected with the internal standard) (Fig. 67). Even though a peak/background of 5 is much

better than the peak/background of 1,5 which we attained in the experiments without CCT (Fig. 57) it is a advisable to measure with a

peak/background of at least 10. So maybe when measurements were carried out with normal sensitivity a better peak/background might be

attained.

CCTajo 7 std A 34

Sy = 256,64x + 43,91

R2 = 0,9992

0

50

100

150

200

250

300

350

0 0,2 0,4 0,6 0,8 1

Weight %

Co

un

t ra

tes

pe

r s

ec

on

d

(cp

s)

Fig. 67 Calibration line CCTajo 7 std A 34S

The peak/background might also be improved by measuring with kinetic discrimination, it is possible to further lower the background levels by

applying a kinetic discrimination which rejects new formed interferences which are formed in the CCT and have a lower kinetic energy then the

analyte. Kinetic discrimination is applied by making the pole bias (quadrupole potential) slightly more positive than the hexapole bias (CCT

potential), interferences might be rejected and background might be further lowered resulting in a better peak/background.

It is also possible that different elements need different internal

standards, because the internal standard and the analyzed element should have the same ablation behaviour. So it’s also worth trying measuring

with another internal standard.



4.3. Comparison different bulk analysis Fig. 68 and Fig. 69 compare the results of several bulk analyses on std A,

we used 4 different ways of sample preparation and analysis. As described before, we made fusion samples with Na2O2/Na2CO3 and Li2B4O7

fluxes. We also still had microwave digestion samples from a previous thesis [1], which we have sent to the analytical laboratory again. So both

the fusion samples and the microwave samples were analysed in the university’s analytical laboratory. These results were then compared with

our own measurements with the LA-ICP-MS and the certified concentrations. Unfortunately we didn’t have time to measure std B and

LLT 5 with the LA-ICP-MS since the mass spectrometer broke down and is for the moment in maintenance. However we do have analyzed std B and

LLT 5 with fusion samples and microwave samples, so in the future these results can be used to compare them with the LA-ICP-MS results. The

concentrations from the LA-ICP-MS were obtained from CCTajo6.tee and

CCTajo7.tee.

Std A

0,00

0,50

1,00

1,50

2,00

2,50

3,00

3,50

Mg S P Ti

We

igh

t %

Na2O2 flux Li2B4O7 flux Microwave LA-ICP-MS NIST-2691

Fig. 68 Comparison several bulk analysis on std A detecting Mg, S, P and Ti



Std A

0,00

5,00

10,00

15,00

20,00

25,00

30,00

35,00

Al Si Ca Fe

Na2O2 flux Li2B4O7 flux Microw ave LA-ICP-MS NIST-2691

Fig. 69 Comparison several bulk analysis on std A detecting Al, Si, Ca and Fe

The results from Fig. 68 and Fig. 69 show us that the LA-ICP-MS gives better results (besides from S analysis) than fusion of microwave samples

if we consider the certified concentrations as correct. Notice that the microwave samples had an unusual Si peak, which is probably explained

by the use of glassware instead of plastic during the sample preparation.



5. Conclusions

• Bigger spot size and higher scan speed resulted in more representative measurements and better results, however the highest scan speed and

the biggest spot size is not always applicable. • More sweeps and longer dwell time resulted in more representative

measurements and better results, however too many sweeps and too long dwell time makes the measurement time too long.

• The use of the CCT lowered the background tremendously, this resulted in better peak/background and as a consequence to lower detection

limits and better calibration lines. We were able to measure 24Mg, 28Si, 27Al, 44Ca, 48Ti, 56Fe, 31P and 34S. We didn’t succeed in configuring the

ICP-MS for 19K detection, we were unable to get a satisfying peak/background. However it’s worth trying to measure with kinetic

discrimination and if this doesn’t work, an other reaction gas (NH3)

might be a solution. • The use of the internal standard had a significant positive influence on

the quality of our results, as both the average and standard deviation have improved significantly for all the measured elements (24Mg, 28Si, 27Al, 44Ca, 48Ti and 56Fe) except from 31P and 34S. Kinetic discrimination or another internal standard might solve this problem.

• The LA-ICP-MS gives most of the time much better bulk concentrations than fusion of microwave samples, if we consider the certified

concentrations as correct.



References

1. Arvilommi Susanna, Polttoon liittyvien näytteiden karakterisointi LA-ICP-MS:lla, Oulu, University of Oulu, 2004, thesis

2. CVI laser LLC, High Energy Harmonic Separators, internet, (16 maart 2004), (http://www.cvilaser.com/static/tech_heharmonic.asp)

3. Engelen J., Hoofdstuk 2: fundamentele eigenschappen van de laser: werkingsprincipe, internet, (13 march 2004),

(http://canada.esat.kuleuven.ac.be/docarch/onderwijs/vakken/H245/HFDST_2_Werkingsprincipes.pdf)

4. Goldwasser S.M., Sam’s laser FAQ’s, internet, (15 maart 2004), (http://www.repairfaq.org/sam/laserssl.htm#ssleoq)

5. Gormally J., Laser tutorial, internet, (11 march 2004), (http://members.aol.com/WSRNet/tut/ut1.htm)

6. Gray A.L., ‘Solid sample introduction by laser ablation for inductively

coupled plasma source mass spectrometry’, Analyst, 110 (1985), p 551-556

7. Günther D., Heinrich C.A., Enhanced sensityvity in laser ablation-ICP mass spectrometry using helium-argon mixtures as aerosol carrier, J.

Anal. At. Spectrom, 14 (1999), p 1363-1368 8. Günther D., Jackson S. E., Longerich H.P, ‘Review: Laser ablation and

arc/spark solid sample introduction into inductively coupled plasma mass spectrometers’, Atomic Spectroscopy, 54 (1999), p 381-409

9. Günther D., Longerich H.P, Forsythe L., Jackson S.E., Laser ablation microprobe-inductively coupled plasma-mass spectrometry (LAM-ICP-

MS), Am. Laboratory, 27 (1995), p 24-29 10. Jackson S. E., Longerich H.P, Dunning G.R., Fryer B.J., The

application of laser-ablation microprobe-inductively coupled plasma-mass spectrometry (LAM-ICP-MS) to in situ trace-element

determinations in minerals, Can. Mineral., 30 (1992), p 1049-1064

11. Jackson S. E., The application of Nd:YAG lasers in LA-ICP-MS, Laser-Ablation-ICPMS in the Earth Sciences: principles and applications, 29

(2001), p 29-45 12. Jackson S., Pearson N.J., Griffin W.L., In situ isotope ratio

determination using LA-magnetic sector-ICP-MS, Laser-Ablation-ICPMS in the Earth Sciences: principles and applications, 29 (2001), p

105-119 13. Jarvis K.P., Dingwell C.B., Rocholl A., Stoll B., Hoffmann A.W.,

Handbook of inductively coupled plasma mass spectrometry, London, Blackie, 1992

14. Longerich H. P., Diegor W., Introduction to mass spectrometry, Laser-Ablation-ICPMS in the Earth Sciences: principles and applications, 29

(2001), p 1-19 15. Longerich H.P, Jackson S. E., Fryer B.J., Strong D.F., The laser

ablation microprobe-inductively coupled plasma-mass spectrometer,

Geosci. Can., 20 (1993), p 21-27 16. Lux J., Electro-optical measurements (Kerr, Pockels, and Faraday),

internet, (15 maart 2004) (http://home.earthlink.net/~jimlux/hv/eo.htm)



17. Lux J., Marx Generators, internet, (15 maart 2004),

(http://home.earthlink.net/~jimlux/hv/marx.htm)

18. Mason P.R.D., Expanding the capabilities of laser-ablation ICP-MS with collision and reaction cells, Laser-Ablation-ICPMS in the Earth

Sciences: principles and applications, 29 (2001), p 63-81 19. Moenke L.-Blankenburg, Laser Microanalysis, Toronto, John Wiley,

1989 20. Montaser A., Inductively coupled plasma mass spectrometry, New

York, Wiley-VCH, 1998 21. Nave R., Parallel light from a laser, internet, (13 march 2004),

(http://hyperphysics.phy-astr.gsu.edu/hbase/optmod/qualig.html#c5) 22. New Wave Research Inc., UP Series Nd:YAG Laser Ablation Systems:

Operator’s Manual (UP-266, UP-266 IV, UP -213, UP – 213 IV), Fremont, 2001 (Operator’s Manual)

23. Newport Corporation, wave plates, internet, (27 maart 2004), (http://www.newport.com/Support/Tutorials/Optics/o8.asp)

24. Pearce N.G.J., Perkins W.T., Abell I., Duller G.A., Fuge R., Mineral

microanalysis by laser ablation inductively coupled plasma mass spectrometry, J. Anal. At. Spectrom., 7 (1992), p 53-57

25. Perry H.R., Green D.W., Perry’s Chemical Engineers’ handbook, U.S.A., The McGraw-Hill Companies, 1999

26. Struers, Epofix intstruction, July 2000 27. Thermo Elemental, X series ICP-MS training course lectures, Winsford

Cheshire, June 2002 28. Thomas R., Beginner’s guide to ICP-MS part I, Spectroscopy, 16, April

2001, p 38-42 29. Thomas R., Beginner’s guide to ICP-MS part II: The sample-

introduction system, Spectroscopy, 16, Mai 2001, p 56-60 30. Thomas R., Beginner’s guide to ICP-MS part III: The plasma source,

Spectroscopy, 16, June 2001, p 26-30 31. Thomas R., Beginner’s guide to ICP-MS part IV: The interface region,

Spectroscopy, 16, July 2001, p 26-34

32. Thomas R., Beginner’s guide to ICP-MS part IX: Mass analyzers: collision/reaction cell technology, 17, February 2002, p 42-48

33. Thomas R., Beginner’s guide to ICP-MS part V: The ion focusing system, 16, September 2001, p 38-44

34. Thomas R., Beginner’s guide to ICP-MS part VI: The mass analyzer, 16, October 2001, p 44-48

35. Thomas R., Beginner’s guide to ICP-MS part X: Detectors, 17, April 2002, p 34-39

36. Van Geenhoven G., Sevenhans W., Kuppens J., Vaste stof fysica, Hoboken, KIHA, 2001 (course KIHA industrieel ingenieur 2de

kandidatuur) 37. Watson J., Lasers, internet, (15 maart 2004),

(http://vcs.abdn.ac.uk/ENGINEERING/lasers/lasers.html) 38. Wijnants M., Organische chemie: Deel 1 structuur opheldering van

organische verbindingen, Hoboken, KIHA, 2002 (course KIHA

industrieel ingenieur 3de jaar)

Date post:	18-Jun-2020
Category:	Documents
Upload:	others
View:	7 times
Download:	0 times

Laser Ablation - Inductively Coupled Plasma - Mass...

Documents